diff options
| author | Nicholas Car <nicholas.car@surroundaustralia.com> | 2020-05-16 21:00:24 +1000 |
|---|---|---|
| committer | Nicholas Car <nicholas.car@surroundaustralia.com> | 2020-05-16 21:00:24 +1000 |
| commit | 0be6f6039479ce29cf71b11e76be54e186130036 (patch) | |
| tree | 897d208525a7e8952fb496151db074d49dcdeb3a /rdflib/plugins | |
| parent | 2a8d70824e1b4caf0c606074a44ac3a15fa72718 (diff) | |
| download | rdflib-0be6f6039479ce29cf71b11e76be54e186130036.tar.gz | |
blacked all python files
Diffstat (limited to 'rdflib/plugins')
41 files changed, 2638 insertions, 1879 deletions
diff --git a/rdflib/plugins/memory.py b/rdflib/plugins/memory.py index 1c1161e9..c102799f 100644 --- a/rdflib/plugins/memory.py +++ b/rdflib/plugins/memory.py @@ -6,7 +6,7 @@ import random from rdflib.store import Store, NO_STORE, VALID_STORE -__all__ = ['Memory', 'IOMemory'] +__all__ = ["Memory", "IOMemory"] ANY = Any = None @@ -96,14 +96,12 @@ class Memory(Store): if predicate in subjectDictionary: if object != ANY: # subject+predicate+object is given if object in subjectDictionary[predicate]: - yield (subject, predicate, object), \ - self.__contexts() + yield (subject, predicate, object), self.__contexts() else: # given object not found pass else: # subject+predicate is given, object unbound for o in subjectDictionary[predicate].keys(): - yield (subject, predicate, o), \ - self.__contexts() + yield (subject, predicate, o), self.__contexts() else: # given predicate not found pass else: # subject given, predicate unbound @@ -196,6 +194,7 @@ class IOMemory(Store): slow. """ + context_aware = True formula_aware = True graph_aware = True @@ -222,11 +221,12 @@ class IOMemory(Store): self.__obj2int = {None: None} # maps objects to integer keys # Indexes for each triple part, and a list of contexts for each triple - self.__subjectIndex = {} # key: sid val: set(enctriples) + self.__subjectIndex = {} # key: sid val: set(enctriples) self.__predicateIndex = {} # key: pid val: set(enctriples) - self.__objectIndex = {} # key: oid val: set(enctriples) - self.__tripleContexts = { - } # key: enctriple val: {cid1: quoted, cid2: quoted ...} + self.__objectIndex = {} # key: oid val: set(enctriples) + self.__tripleContexts = ( + {} + ) # key: enctriple val: {cid1: quoted, cid2: quoted ...} self.__contextTriples = {None: set()} # key: cid val: set(enctriples) # all contexts used in store (unencoded) @@ -294,16 +294,20 @@ class IOMemory(Store): del self.__tripleContexts[enctriple] - if not req_cid is None and \ - req_cid in self.__contextTriples and \ - len(self.__contextTriples[req_cid]) == 0: + if ( + not req_cid is None + and req_cid in self.__contextTriples + and len(self.__contextTriples[req_cid]) == 0 + ): # all triples are removed out of this context # and it's not the default context so delete it del self.__contextTriples[req_cid] - if triplepat == (None, None, None) and \ - context in self.__all_contexts and \ - not self.graph_aware: + if ( + triplepat == (None, None, None) + and context in self.__all_contexts + and not self.graph_aware + ): # remove the whole context self.__all_contexts.remove(context) @@ -322,9 +326,11 @@ class IOMemory(Store): # optimize "triple in graph" case (all parts given) if sid is not None and pid is not None and oid is not None: - if sid in self.__subjectIndex and \ - enctriple in self.__subjectIndex[sid] and \ - self.__tripleHasContext(enctriple, cid): + if ( + sid in self.__subjectIndex + and enctriple in self.__subjectIndex[sid] + and self.__tripleHasContext(enctriple, cid) + ): return ((triplein, self.__contexts(enctriple)) for i in [0]) else: return self.__emptygen() @@ -353,9 +359,11 @@ class IOMemory(Store): else: enctriples = sets[0].copy() - return ((self.__decodeTriple(enctriple), self.__contexts(enctriple)) - for enctriple in enctriples - if self.__tripleHasContext(enctriple, cid)) + return ( + (self.__decodeTriple(enctriple), self.__contexts(enctriple)) + for enctriple in enctriples + if self.__tripleHasContext(enctriple, cid) + ) def contexts(self, triple=None): if triple is None or triple == (None, None, None): @@ -402,8 +410,7 @@ class IOMemory(Store): if enctriple not in self.__tripleContexts: # triple exists with default ctx info # start with a copy of the default ctx info - self.__tripleContexts[ - enctriple] = self.__defaultContexts.copy() + self.__tripleContexts[enctriple] = self.__defaultContexts.copy() self.__tripleContexts[enctriple][cid] = quoted if not quoted: @@ -446,12 +453,11 @@ class IOMemory(Store): def __tripleHasContext(self, enctriple, cid): """return True iff the triple exists in the given context""" ctxs = self.__tripleContexts.get(enctriple, self.__defaultContexts) - return (cid in ctxs) + return cid in ctxs def __removeTripleContext(self, enctriple, cid): """remove the context from the triple""" - ctxs = self.__tripleContexts.get( - enctriple, self.__defaultContexts).copy() + ctxs = self.__tripleContexts.get(enctriple, self.__defaultContexts).copy() del ctxs[cid] if ctxs == self.__defaultContexts: del self.__tripleContexts[enctriple] @@ -491,7 +497,11 @@ class IOMemory(Store): def __contexts(self, enctriple): """return a generator for all the non-quoted contexts (unencoded) the encoded triple appears in""" - return (self.__int2obj.get(cid) for cid in self.__getTripleContexts(enctriple, skipQuoted=True) if cid is not None) + return ( + self.__int2obj.get(cid) + for cid in self.__getTripleContexts(enctriple, skipQuoted=True) + if cid is not None + ) def __emptygen(self): """return an empty generator""" diff --git a/rdflib/plugins/parsers/notation3.py b/rdflib/plugins/parsers/notation3.py index 44a25adc..3bc2169f 100755 --- a/rdflib/plugins/parsers/notation3.py +++ b/rdflib/plugins/parsers/notation3.py @@ -48,9 +48,17 @@ from rdflib.graph import QuotedGraph, ConjunctiveGraph, Graph from rdflib.compat import long_type from rdflib.compat import narrow_build -__all__ = ['BadSyntax', 'N3Parser', 'TurtleParser', - "splitFragP", "join", "base", - "runNamespace", "uniqueURI", "hexify"] +__all__ = [ + "BadSyntax", + "N3Parser", + "TurtleParser", + "splitFragP", + "join", + "base", + "runNamespace", + "uniqueURI", + "hexify", +] from rdflib.parser import Parser @@ -74,7 +82,7 @@ def splitFragP(uriref, punct=0): if i >= 0: return uriref[:i], uriref[i:] else: - return uriref, '' + return uriref, "" def join(here, there): @@ -112,65 +120,67 @@ def join(here, there): u'http://example.org/#Andr\\xe9' """ -# assert(here.find("#") < 0), \ -# "Base may not contain hash: '%s'" % here # why must caller splitFrag? + # assert(here.find("#") < 0), \ + # "Base may not contain hash: '%s'" % here # why must caller splitFrag? - slashl = there.find('/') - colonl = there.find(':') + slashl = there.find("/") + colonl = there.find(":") # join(base, 'foo:/') -- absolute if colonl >= 0 and (slashl < 0 or colonl < slashl): return there - bcolonl = here.find(':') - assert(bcolonl >= 0), \ - "Base uri '%s' is not absolute" % here # else it's not absolute + bcolonl = here.find(":") + assert bcolonl >= 0, ( + "Base uri '%s' is not absolute" % here + ) # else it's not absolute path, frag = splitFragP(there) if not path: return here + frag # join('mid:foo@example', '../foo') bzzt - if here[bcolonl + 1:bcolonl + 2] != '/': + if here[bcolonl + 1 : bcolonl + 2] != "/": raise ValueError( - ("Base <%s> has no slash after " - "colon - with relative '%s'.") % (here, there)) + ("Base <%s> has no slash after " "colon - with relative '%s'.") + % (here, there) + ) - if here[bcolonl + 1:bcolonl + 3] == '//': - bpath = here.find('/', bcolonl + 3) + if here[bcolonl + 1 : bcolonl + 3] == "//": + bpath = here.find("/", bcolonl + 3) else: bpath = bcolonl + 1 # join('http://xyz', 'foo') if bpath < 0: bpath = len(here) - here = here + '/' + here = here + "/" # join('http://xyz/', '//abc') => 'http://abc' - if there[:2] == '//': - return here[:bcolonl + 1] + there + if there[:2] == "//": + return here[: bcolonl + 1] + there # join('http://xyz/', '/abc') => 'http://xyz/abc' - if there[:1] == '/': + if there[:1] == "/": return here[:bpath] + there - slashr = here.rfind('/') + slashr = here.rfind("/") while 1: - if path[:2] == './': + if path[:2] == "./": path = path[2:] - if path == '.': - path = '' - elif path[:3] == '../' or path == '..': + if path == ".": + path = "" + elif path[:3] == "../" or path == "..": path = path[3:] - i = here.rfind('/', bpath, slashr) + i = here.rfind("/", bpath, slashr) if i >= 0: - here = here[:i + 1] + here = here[: i + 1] slashr = i else: break - return here[:slashr + 1] + path + frag + return here[: slashr + 1] + path + frag def base(): @@ -190,7 +200,7 @@ def _fixslash(s): """ Fix windowslike filename to unixlike - (#ifdef WINDOWS)""" s = s.replace("\\", "/") if s[0] != "/" and s[1] == ":": - s = s[2:] # @@@ Hack when drive letter present + s = s[2:] # @@@ Hack when drive letter present return s @@ -211,7 +221,7 @@ ANONYMOUS = 3 XMLLITERAL = 25 Logic_NS = "http://www.w3.org/2000/10/swap/log#" -NODE_MERGE_URI = Logic_NS + "is" # Pseudo-property indicating node merging +NODE_MERGE_URI = Logic_NS + "is" # Pseudo-property indicating node merging forSomeSym = Logic_NS + "forSome" forAllSym = Logic_NS + "forAll" @@ -222,7 +232,7 @@ DAML_sameAs_URI = OWL_NS + "sameAs" parsesTo_URI = Logic_NS + "parsesTo" RDF_spec = "http://www.w3.org/TR/REC-rdf-syntax/" -List_NS = RDF_NS_URI # From 20030808 +List_NS = RDF_NS_URI # From 20030808 _Old_Logic_NS = "http://www.w3.org/2000/10/swap/log.n3#" N3_first = (SYMBOL, List_NS + "first") @@ -238,12 +248,13 @@ runNamespaceValue = None def runNamespace(): "Return a URI suitable as a namespace for run-local objects" - # @@@ include hostname (privacy?) (hash it?) + # @@@ include hostname (privacy?) (hash it?) global runNamespaceValue if runNamespaceValue is None: - runNamespaceValue = join(base(), _unique_id()) + '#' + runNamespaceValue = join(base(), _unique_id()) + "#" return runNamespaceValue + nextu = 0 @@ -262,20 +273,21 @@ chatty_flag = 50 def BecauseOfData(*args, **kargs): - # print args, kargs + # print args, kargs pass def becauseSubexpression(*args, **kargs): - # print args, kargs + # print args, kargs pass + N3_forSome_URI = forSomeSym N3_forAll_URI = forAllSym # Magic resources we know about -ADDED_HASH = "#" # Stop where we use this in case we want to remove it! +ADDED_HASH = "#" # Stop where we use this in case we want to remove it! # This is the hash on namespace URIs RDF_type = (SYMBOL, RDF_type_URI) @@ -289,20 +301,20 @@ DOUBLE_DATATYPE = _XSD_PFX + "double" FLOAT_DATATYPE = _XSD_PFX + "float" INTEGER_DATATYPE = _XSD_PFX + "integer" -option_noregen = 0 # If set, do not regenerate genids on output +option_noregen = 0 # If set, do not regenerate genids on output # @@ I18n - the notname chars need extending for well known unicode non-text # characters. The XML spec switched to assuming unknown things were name # characaters. # _namechars = string.lowercase + string.uppercase + string.digits + '_-' -_notQNameChars = \ - "\t\r\n !\"#$&'()*,+/;<=>?@[\\]^`{|}~" # else valid qname :-/ +_notQNameChars = "\t\r\n !\"#$&'()*,+/;<=>?@[\\]^`{|}~" # else valid qname :-/ _notKeywordsChars = _notQNameChars + "." -_notNameChars = _notQNameChars + ":" # Assume anything else valid name :-/ -_rdfns = 'http://www.w3.org/1999/02/22-rdf-syntax-ns#' +_notNameChars = _notQNameChars + ":" # Assume anything else valid name :-/ +_rdfns = "http://www.w3.org/1999/02/22-rdf-syntax-ns#" + +hexChars = "ABCDEFabcdef0123456789" +escapeChars = "(_~.-!$&'()*+,;=/?#@%)" # valid for \ escapes in localnames -hexChars = 'ABCDEFabcdef0123456789' -escapeChars = "(_~.-!$&'()*+,;=/?#@%)" # valid for \ escapes in localnames def unicodeExpand(m): try: @@ -310,81 +322,90 @@ def unicodeExpand(m): except: raise Exception("Invalid unicode code point: " + m.group(1)) + if narrow_build: + def unicodeExpand(m): try: return chr(int(m.group(1), 16)) except ValueError: warnings.warn( - 'Encountered a unicode char > 0xFFFF in a narrow python build. ' - 'Trying to degrade gracefully, but this can cause problems ' - 'later when working with the string:\n%s' % m.group(0)) - return codecs.decode(m.group(0), 'unicode_escape') + "Encountered a unicode char > 0xFFFF in a narrow python build. " + "Trying to degrade gracefully, but this can cause problems " + "later when working with the string:\n%s" % m.group(0) + ) + return codecs.decode(m.group(0), "unicode_escape") -unicodeEscape4 = re.compile( - r'\\u([0-9a-fA-F]{4})') -unicodeEscape8 = re.compile( - r'\\U([0-9a-fA-F]{8})') +unicodeEscape4 = re.compile(r"\\u([0-9a-fA-F]{4})") +unicodeEscape8 = re.compile(r"\\U([0-9a-fA-F]{8})") -N3CommentCharacter = "#" # For unix script # ! compatabilty +N3CommentCharacter = "#" # For unix script # ! compatabilty ########################################## Parse string to sink # # Regular expressions: -eol = re.compile( - r'[ \t]*(#[^\n]*)?\r?\n') # end of line, poss. w/comment -eof = re.compile( - r'[ \t]*(#[^\n]*)?$') # end of file, poss. w/comment -ws = re.compile(r'[ \t]*') # Whitespace not including NL -signed_integer = re.compile(r'[-+]?[0-9]+') # integer -integer_syntax = re.compile(r'[-+]?[0-9]+') -decimal_syntax = re.compile(r'[-+]?[0-9]*\.[0-9]+') -exponent_syntax = re.compile(r'[-+]?(?:[0-9]+\.[0-9]*|\.[0-9]+|[0-9]+)(?:e|E)[-+]?[0-9]+') -digitstring = re.compile(r'[0-9]+') # Unsigned integer +eol = re.compile(r"[ \t]*(#[^\n]*)?\r?\n") # end of line, poss. w/comment +eof = re.compile(r"[ \t]*(#[^\n]*)?$") # end of file, poss. w/comment +ws = re.compile(r"[ \t]*") # Whitespace not including NL +signed_integer = re.compile(r"[-+]?[0-9]+") # integer +integer_syntax = re.compile(r"[-+]?[0-9]+") +decimal_syntax = re.compile(r"[-+]?[0-9]*\.[0-9]+") +exponent_syntax = re.compile( + r"[-+]?(?:[0-9]+\.[0-9]*|\.[0-9]+|[0-9]+)(?:e|E)[-+]?[0-9]+" +) +digitstring = re.compile(r"[0-9]+") # Unsigned integer interesting = re.compile(r"""[\\\r\n\"\']""") -langcode = re.compile(r'[a-zA-Z0-9]+(-[a-zA-Z0-9]+)*') +langcode = re.compile(r"[a-zA-Z0-9]+(-[a-zA-Z0-9]+)*") class SinkParser: - def __init__(self, store, openFormula=None, thisDoc="", baseURI=None, - genPrefix="", why=None, turtle=False): + def __init__( + self, + store, + openFormula=None, + thisDoc="", + baseURI=None, + genPrefix="", + why=None, + turtle=False, + ): """ note: namespace names should *not* end in # ; the # will get added during qname processing """ self._bindings = {} if thisDoc != "": - assert ':' in thisDoc, "Document URI not absolute: <%s>" % thisDoc - self._bindings[""] = thisDoc + "#" # default + assert ":" in thisDoc, "Document URI not absolute: <%s>" % thisDoc + self._bindings[""] = thisDoc + "#" # default self._store = store if genPrefix: store.setGenPrefix(genPrefix) # pass it on self._thisDoc = thisDoc - self.lines = 0 # for error handling - self.startOfLine = 0 # For calculating character number + self.lines = 0 # for error handling + self.startOfLine = 0 # For calculating character number self._genPrefix = genPrefix - self.keywords = ['a', 'this', 'bind', 'has', 'is', 'of', - 'true', 'false'] - self.keywordsSet = 0 # Then only can others be considerd qnames + self.keywords = ["a", "this", "bind", "has", "is", "of", "true", "false"] + self.keywordsSet = 0 # Then only can others be considerd qnames self._anonymousNodes = {} - # Dict of anon nodes already declared ln: Term + # Dict of anon nodes already declared ln: Term self._variables = {} self._parentVariables = {} - self._reason = why # Why the parser was asked to parse this + self._reason = why # Why the parser was asked to parse this - self.turtle = turtle # raise exception when encountering N3 extensions + self.turtle = turtle # raise exception when encountering N3 extensions # Turtle allows single or double quotes around strings, whereas N3 # only allows double quotes. self.string_delimiters = ('"', "'") if turtle else ('"',) - self._reason2 = None # Why these triples - # was: diag.tracking + self._reason2 = None # Why these triples + # was: diag.tracking if tracking: self._reason2 = BecauseOfData( - store.newSymbol(thisDoc), because=self._reason) + store.newSymbol(thisDoc), because=self._reason + ) if baseURI: self._baseURI = baseURI @@ -394,7 +415,7 @@ class SinkParser: else: self._baseURI = None - assert not self._baseURI or ':' in self._baseURI + assert not self._baseURI or ":" in self._baseURI if not self._genPrefix: if self._thisDoc: @@ -424,21 +445,20 @@ class SinkParser: _L1C1. It used to be used only for tracking, but for tests in general it makes the canonical ordering of bnodes repeatable.""" - return "%s_L%iC%i" % (self._genPrefix, self.lines, - i - self.startOfLine + 1) + return "%s_L%iC%i" % (self._genPrefix, self.lines, i - self.startOfLine + 1) def formula(self): return self._formula def loadStream(self, stream): - return self.loadBuf(stream.read()) # Not ideal + return self.loadBuf(stream.read()) # Not ideal def loadBuf(self, buf): """Parses a buffer and returns its top level formula""" self.startDoc() self.feed(buf) - return self.endDoc() # self._formula + return self.endDoc() # self._formula def feed(self, octets): """Feed an octet stream tothe parser @@ -450,9 +470,9 @@ class SinkParser: parser, it should be straightforward to recover.""" if not isinstance(octets, str): - s = octets.decode('utf-8') - # NB already decoded, so \ufeff - if len(s) > 0 and s[0] == codecs.BOM_UTF8.decode('utf-8'): + s = octets.decode("utf-8") + # NB already decoded, so \ufeff + if len(s) > 0 and s[0] == codecs.BOM_UTF8.decode("utf-8"): s = s[1:] else: s = octets @@ -465,15 +485,14 @@ class SinkParser: i = self.directiveOrStatement(s, j) if i < 0: - #print("# next char: %s" % s[j]) - self.BadSyntax(s, j, - "expected directive or statement") + # print("# next char: %s" % s[j]) + self.BadSyntax(s, j, "expected directive or statement") def directiveOrStatement(self, argstr, h): i = self.skipSpace(argstr, h) if i < 0: - return i # EOF + return i # EOF if self.turtle: j = self.sparqlDirective(argstr, i) @@ -490,8 +509,8 @@ class SinkParser: return j - # @@I18N - # _namechars = string.lowercase + string.uppercase + string.digits + '_-' + # @@I18N + # _namechars = string.lowercase + string.uppercase + string.digits + '_-' def tok(self, tok, argstr, i, colon=False): """Check for keyword. Space must have been stripped on entry and @@ -502,15 +521,17 @@ class SinkParser: """ assert tok[0] not in _notNameChars # not for punctuation - if argstr[i:i + 1] == "@": + if argstr[i : i + 1] == "@": i = i + 1 else: if tok not in self.keywords: return -1 # No, this has neither keywords declaration nor "@" - if (argstr[i:i + len(tok)] == tok - and ( argstr[i + len(tok)] in _notKeywordsChars) - or (colon and argstr[i+len(tok)] == ':')): + if ( + argstr[i : i + len(tok)] == tok + and (argstr[i + len(tok)] in _notKeywordsChars) + or (colon and argstr[i + len(tok)] == ":") + ): i = i + len(tok) return i else: @@ -524,109 +545,114 @@ class SinkParser: assert tok[0] not in _notNameChars # not for punctuation - if (argstr[i:i + len(tok)].lower() == tok.lower() - and (argstr[i + len(tok)] in _notQNameChars)): + if argstr[i : i + len(tok)].lower() == tok.lower() and ( + argstr[i + len(tok)] in _notQNameChars + ): i = i + len(tok) return i else: return -1 - def directive(self, argstr, i): j = self.skipSpace(argstr, i) if j < 0: return j # eof res = [] - j = self.tok('bind', argstr, i) # implied "#". Obsolete. + j = self.tok("bind", argstr, i) # implied "#". Obsolete. if j > 0: - self.BadSyntax(argstr, i, - "keyword bind is obsolete: use @prefix") + self.BadSyntax(argstr, i, "keyword bind is obsolete: use @prefix") - j = self.tok('keywords', argstr, i) + j = self.tok("keywords", argstr, i) if j > 0: if self.turtle: self.BadSyntax(argstr, i, "Found 'keywords' when in Turtle mode.") i = self.commaSeparatedList(argstr, j, res, self.bareWord) if i < 0: - self.BadSyntax(argstr, i, - "'@keywords' needs comma separated list of words") + self.BadSyntax( + argstr, i, "'@keywords' needs comma separated list of words" + ) self.setKeywords(res[:]) return i - j = self.tok('forAll', argstr, i) + j = self.tok("forAll", argstr, i) if j > 0: if self.turtle: self.BadSyntax(argstr, i, "Found 'forAll' when in Turtle mode.") i = self.commaSeparatedList(argstr, j, res, self.uri_ref2) if i < 0: - self.BadSyntax(argstr, i, - "Bad variable list after @forAll") + self.BadSyntax(argstr, i, "Bad variable list after @forAll") for x in res: - # self._context.declareUniversal(x) + # self._context.declareUniversal(x) if x not in self._variables or x in self._parentVariables: self._variables[x] = self._context.newUniversal(x) return i - j = self.tok('forSome', argstr, i) + j = self.tok("forSome", argstr, i) if j > 0: if self.turtle: self.BadSyntax(argstr, i, "Found 'forSome' when in Turtle mode.") - i = self. commaSeparatedList(argstr, j, res, self.uri_ref2) + i = self.commaSeparatedList(argstr, j, res, self.uri_ref2) if i < 0: - self.BadSyntax(argstr, i, - "Bad variable list after @forSome") + self.BadSyntax(argstr, i, "Bad variable list after @forSome") for x in res: self._context.declareExistential(x) return i - j = self.tok('prefix', argstr, i, colon=True) # no implied "#" + j = self.tok("prefix", argstr, i, colon=True) # no implied "#" if j >= 0: t = [] i = self.qname(argstr, j, t) if i < 0: - self.BadSyntax(argstr, j, - "expected qname after @prefix") + self.BadSyntax(argstr, j, "expected qname after @prefix") j = self.uri_ref2(argstr, i, t) if j < 0: - self.BadSyntax(argstr, i, - "expected <uriref> after @prefix _qname_") + self.BadSyntax(argstr, i, "expected <uriref> after @prefix _qname_") ns = self.uriOf(t[1]) if self._baseURI: ns = join(self._baseURI, ns) elif ":" not in ns: - self.BadSyntax(argstr, j, - "With no base URI, cannot use " + - "relative URI in @prefix <" + ns + ">") - assert ':' in ns # must be absolute + self.BadSyntax( + argstr, + j, + "With no base URI, cannot use " + + "relative URI in @prefix <" + + ns + + ">", + ) + assert ":" in ns # must be absolute self._bindings[t[0][0]] = ns self.bind(t[0][0], hexify(ns)) return j - j = self.tok('base', argstr, i) # Added 2007/7/7 + j = self.tok("base", argstr, i) # Added 2007/7/7 if j >= 0: t = [] i = self.uri_ref2(argstr, j, t) if i < 0: - self.BadSyntax(argstr, j, - "expected <uri> after @base ") + self.BadSyntax(argstr, j, "expected <uri> after @base ") ns = self.uriOf(t[0]) if self._baseURI: ns = join(self._baseURI, ns) else: - self.BadSyntax(argstr, j, - "With no previous base URI, cannot use " + - "relative URI in @base <" + ns + ">") - assert ':' in ns # must be absolute + self.BadSyntax( + argstr, + j, + "With no previous base URI, cannot use " + + "relative URI in @base <" + + ns + + ">", + ) + assert ":" in ns # must be absolute self._baseURI = ns return i - return -1 # Not a directive, could be something else. + return -1 # Not a directive, could be something else. def sparqlDirective(self, argstr, i): @@ -639,55 +665,60 @@ class SinkParser: if j < 0: return j # eof - j = self.sparqlTok('PREFIX', argstr, i) + j = self.sparqlTok("PREFIX", argstr, i) if j >= 0: t = [] i = self.qname(argstr, j, t) if i < 0: - self.BadSyntax(argstr, j, - "expected qname after @prefix") + self.BadSyntax(argstr, j, "expected qname after @prefix") j = self.uri_ref2(argstr, i, t) if j < 0: - self.BadSyntax(argstr, i, - "expected <uriref> after @prefix _qname_") + self.BadSyntax(argstr, i, "expected <uriref> after @prefix _qname_") ns = self.uriOf(t[1]) if self._baseURI: ns = join(self._baseURI, ns) elif ":" not in ns: - self.BadSyntax(argstr, j, - "With no base URI, cannot use " + - "relative URI in @prefix <" + ns + ">") - assert ':' in ns # must be absolute + self.BadSyntax( + argstr, + j, + "With no base URI, cannot use " + + "relative URI in @prefix <" + + ns + + ">", + ) + assert ":" in ns # must be absolute self._bindings[t[0][0]] = ns self.bind(t[0][0], hexify(ns)) return j - j = self.sparqlTok('BASE', argstr, i) + j = self.sparqlTok("BASE", argstr, i) if j >= 0: t = [] i = self.uri_ref2(argstr, j, t) if i < 0: - self.BadSyntax(argstr, j, - "expected <uri> after @base ") + self.BadSyntax(argstr, j, "expected <uri> after @base ") ns = self.uriOf(t[0]) if self._baseURI: ns = join(self._baseURI, ns) else: - self.BadSyntax(argstr, j, - "With no previous base URI, cannot use " + - "relative URI in @base <" + ns + ">") - assert ':' in ns # must be absolute + self.BadSyntax( + argstr, + j, + "With no previous base URI, cannot use " + + "relative URI in @base <" + + ns + + ">", + ) + assert ":" in ns # must be absolute self._baseURI = ns return i - return -1 # Not a directive, could be something else. - + return -1 # Not a directive, could be something else. def bind(self, qn, uri): - assert isinstance( - uri, bytes), "Any unicode must be %x-encoded already" + assert isinstance(uri, bytes), "Any unicode must be %x-encoded already" if qn == "": self._store.setDefaultNamespace(uri) else: @@ -702,31 +733,29 @@ class SinkParser: self.keywordsSet = 1 def startDoc(self): - # was: self._store.startDoc() + # was: self._store.startDoc() self._store.startDoc(self._formula) def endDoc(self): """Signal end of document and stop parsing. returns formula""" - self._store.endDoc(self._formula) # don't canonicalize yet + self._store.endDoc(self._formula) # don't canonicalize yet return self._formula def makeStatement(self, quadruple): - # $$$$$$$$$$$$$$$$$$$$$ - # print "# Parser output: ", `quadruple` + # $$$$$$$$$$$$$$$$$$$$$ + # print "# Parser output: ", `quadruple` self._store.makeStatement(quadruple, why=self._reason2) def statement(self, argstr, i): r = [] - i = self.object( - argstr, i, r) # Allow literal for subject - extends RDF + i = self.object(argstr, i, r) # Allow literal for subject - extends RDF if i < 0: return i j = self.property_list(argstr, i, r[0]) if j < 0: - self.BadSyntax( - argstr, i, "expected propertylist") + self.BadSyntax(argstr, i, "expected propertylist") return j def subject(self, argstr, i, res): @@ -748,77 +777,73 @@ class SinkParser: r = [] - j = self.tok('has', argstr, i) + j = self.tok("has", argstr, i) if j >= 0: if self.turtle: self.BadSyntax(argstr, i, "Found 'has' keyword in Turtle mode") i = self.prop(argstr, j, r) if i < 0: - self.BadSyntax(argstr, j, - "expected property after 'has'") - res.append(('->', r[0])) + self.BadSyntax(argstr, j, "expected property after 'has'") + res.append(("->", r[0])) return i - j = self.tok('is', argstr, i) + j = self.tok("is", argstr, i) if j >= 0: if self.turtle: self.BadSyntax(argstr, i, "Found 'is' keyword in Turtle mode") i = self.prop(argstr, j, r) if i < 0: - self.BadSyntax(argstr, j, - "expected <property> after 'is'") + self.BadSyntax(argstr, j, "expected <property> after 'is'") j = self.skipSpace(argstr, i) if j < 0: - self.BadSyntax(argstr, i, - "End of file found, expected property after 'is'") + self.BadSyntax( + argstr, i, "End of file found, expected property after 'is'" + ) i = j - j = self.tok('of', argstr, i) + j = self.tok("of", argstr, i) if j < 0: - self.BadSyntax(argstr, i, - "expected 'of' after 'is' <prop>") - res.append(('<-', r[0])) + self.BadSyntax(argstr, i, "expected 'of' after 'is' <prop>") + res.append(("<-", r[0])) return j - j = self.tok('a', argstr, i) + j = self.tok("a", argstr, i) if j >= 0: - res.append(('->', RDF_type)) + res.append(("->", RDF_type)) return j - if argstr[i:i + 2] == "<=": + if argstr[i : i + 2] == "<=": if self.turtle: - self.BadSyntax(argstr, i, - "Found '<=' in Turtle mode. ") + self.BadSyntax(argstr, i, "Found '<=' in Turtle mode. ") - res.append(('<-', self._store.newSymbol(Logic_NS + "implies"))) + res.append(("<-", self._store.newSymbol(Logic_NS + "implies"))) return i + 2 - if argstr[i:i + 1] == "=": + if argstr[i : i + 1] == "=": if self.turtle: self.BadSyntax(argstr, i, "Found '=' in Turtle mode") - if argstr[i + 1:i + 2] == ">": - res.append(('->', self._store.newSymbol(Logic_NS + "implies"))) + if argstr[i + 1 : i + 2] == ">": + res.append(("->", self._store.newSymbol(Logic_NS + "implies"))) return i + 2 - res.append(('->', DAML_sameAs)) + res.append(("->", DAML_sameAs)) return i + 1 - if argstr[i:i + 2] == ":=": + if argstr[i : i + 2] == ":=": if self.turtle: self.BadSyntax(argstr, i, "Found ':=' in Turtle mode") - # patch file relates two formulae, uses this @@ really? - res.append(('->', Logic_NS + "becomes")) + # patch file relates two formulae, uses this @@ really? + res.append(("->", Logic_NS + "becomes")) return i + 2 j = self.prop(argstr, i, r) if j >= 0: - res.append(('->', r[0])) + res.append(("->", r[0])) return j - if argstr[i:i + 2] == ">-" or argstr[i:i + 2] == "<-": - self.BadSyntax(argstr, j, - ">- ... -> syntax is obsolete.") + if argstr[i : i + 2] == ">-" or argstr[i : i + 2] == "<-": + self.BadSyntax(argstr, j, ">- ... -> syntax is obsolete.") return -1 @@ -836,16 +861,15 @@ class SinkParser: """ j = self.nodeOrLiteral(argstr, i, res) if j < 0: - return j # nope + return j # nope - while argstr[j:j + 1] in "!^": # no spaces, must follow exactly (?) - ch = argstr[j:j + 1] + while argstr[j : j + 1] in "!^": # no spaces, must follow exactly (?) + ch = argstr[j : j + 1] subj = res.pop() obj = self.blankNode(uri=self.here(j)) j = self.node(argstr, j + 1, res) if j < 0: - self.BadSyntax(argstr, j, - "EOF found in middle of path syntax") + self.BadSyntax(argstr, j, "EOF found in middle of path syntax") pred = res.pop() if ch == "^": # Reverse traverse self.makeStatement((self._context, pred, obj, subj)) @@ -874,18 +898,19 @@ class SinkParser: if j < 0: return j # eof i = j - ch = argstr[i:i + 1] # Quick 1-character checks first: + ch = argstr[i : i + 1] # Quick 1-character checks first: if ch == "[": bnodeID = self.here(i) j = self.skipSpace(argstr, i + 1) if j < 0: - self.BadSyntax(argstr, i, - "EOF after '['") + self.BadSyntax(argstr, i, "EOF after '['") # Hack for "is" binding name to anon node - if argstr[j:j + 1] == "=": + if argstr[j : j + 1] == "=": if self.turtle: - self.BadSyntax(argstr, j, "Found '[=' or '[ =' when in turtle mode.") + self.BadSyntax( + argstr, j, "Found '[=' or '[ =' when in turtle mode." + ) i = j + 1 objs = [] j = self.objectList(argstr, i, objs) @@ -893,33 +918,31 @@ class SinkParser: subj = objs[0] if len(objs) > 1: for obj in objs: - self.makeStatement((self._context, - DAML_sameAs, subj, obj)) + self.makeStatement((self._context, DAML_sameAs, subj, obj)) j = self.skipSpace(argstr, j) if j < 0: - self.BadSyntax(argstr, i, - "EOF when objectList expected after [ = ") - if argstr[j:j + 1] == ";": + self.BadSyntax( + argstr, i, "EOF when objectList expected after [ = " + ) + if argstr[j : j + 1] == ";": j = j + 1 else: - self.BadSyntax(argstr, i, - "objectList expected after [= ") + self.BadSyntax(argstr, i, "objectList expected after [= ") if subj is None: subj = self.blankNode(uri=bnodeID) i = self.property_list(argstr, j, subj) if i < 0: - self.BadSyntax(argstr, j, - "property_list expected") + self.BadSyntax(argstr, j, "property_list expected") j = self.skipSpace(argstr, i) if j < 0: - self.BadSyntax(argstr, i, - "EOF when ']' expected after [ <propertyList>") - if argstr[j:j + 1] != "]": - self.BadSyntax(argstr, j, - "']' expected") + self.BadSyntax( + argstr, i, "EOF when ']' expected after [ <propertyList>" + ) + if argstr[j : j + 1] != "]": + self.BadSyntax(argstr, j, "']' expected") res.append(subj) return j + 1 @@ -927,8 +950,8 @@ class SinkParser: # if self.turtle: # self.BadSyntax(argstr, i, # "found '{' while in Turtle mode, Formulas not supported!") - ch2 = argstr[i + 1:i + 2] - if ch2 == '$': + ch2 = argstr[i + 1 : i + 2] + if ch2 == "$": # a set i += 1 j = i + 1 @@ -937,27 +960,23 @@ class SinkParser: while 1: i = self.skipSpace(argstr, j) if i < 0: - self.BadSyntax(argstr, i, - "needed '$}', found end.") - if argstr[i:i + 2] == '$}': + self.BadSyntax(argstr, i, "needed '$}', found end.") + if argstr[i : i + 2] == "$}": j = i + 2 break if not first_run: - if argstr[i:i + 1] == ',': + if argstr[i : i + 1] == ",": i += 1 else: - self.BadSyntax( - argstr, i, "expected: ','") + self.BadSyntax(argstr, i, "expected: ','") else: first_run = False item = [] - j = self.item( - argstr, i, item) # @@@@@ should be path, was object + j = self.item(argstr, i, item) # @@@@@ should be path, was object if j < 0: - self.BadSyntax(argstr, i, - "expected item in set or '$}'") + self.BadSyntax(argstr, i, "expected item in set or '$}'") List.append(self._store.intern(item[0])) res.append(self._store.newSet(List, self._context)) return j @@ -980,17 +999,15 @@ class SinkParser: while 1: i = self.skipSpace(argstr, j) if i < 0: - self.BadSyntax( - argstr, i, "needed '}', found end.") + self.BadSyntax(argstr, i, "needed '}', found end.") - if argstr[i:i + 1] == "}": + if argstr[i : i + 1] == "}": j = i + 1 break j = self.directiveOrStatement(argstr, i) if j < 0: - self.BadSyntax( - argstr, i, "expected statement or '}'") + self.BadSyntax(argstr, i, "expected statement or '}'") self._anonymousNodes = parentAnonymousNodes self._variables = self._parentVariables @@ -998,13 +1015,13 @@ class SinkParser: self._context = self._parentContext self._reason2 = reason2 self._parentContext = oldParentContext - res.append(subj.close()) # No use until closed + res.append(subj.close()) # No use until closed return j if ch == "(": thing_type = self._store.newList - ch2 = argstr[i + 1:i + 2] - if ch2 == '$': + ch2 = argstr[i + 1 : i + 2] + if ch2 == "$": thing_type = self._store.newSet i += 1 j = i + 1 @@ -1013,34 +1030,34 @@ class SinkParser: while 1: i = self.skipSpace(argstr, j) if i < 0: - self.BadSyntax( - argstr, i, "needed ')', found end.") - if argstr[i:i + 1] == ')': + self.BadSyntax(argstr, i, "needed ')', found end.") + if argstr[i : i + 1] == ")": j = i + 1 break item = [] - j = self.item( - argstr, i, item) # @@@@@ should be path, was object + j = self.item(argstr, i, item) # @@@@@ should be path, was object if j < 0: - self.BadSyntax(argstr, i, - "expected item in list or ')'") + self.BadSyntax(argstr, i, "expected item in list or ')'") List.append(self._store.intern(item[0])) res.append(thing_type(List, self._context)) return j - j = self.tok('this', argstr, i) # This context + j = self.tok("this", argstr, i) # This context if j >= 0: - self.BadSyntax(argstr, i, - "Keyword 'this' was ancient N3. Now use " + - "@forSome and @forAll keywords.") - - # booleans - j = self.tok('true', argstr, i) + self.BadSyntax( + argstr, + i, + "Keyword 'this' was ancient N3. Now use " + + "@forSome and @forAll keywords.", + ) + + # booleans + j = self.tok("true", argstr, i) if j >= 0: res.append(True) return j - j = self.tok('false', argstr, i) + j = self.tok("false", argstr, i) if j >= 0: res.append(False) return j @@ -1057,23 +1074,24 @@ class SinkParser: Leaves the terminating punctuation in the buffer """ while 1: - while 1: # skip repeat ; + while 1: # skip repeat ; j = self.skipSpace(argstr, i) if j < 0: - self.BadSyntax(argstr, i, - "EOF found when expected verb in property list") - if argstr[j]!=';': break - i = j+1 + self.BadSyntax( + argstr, i, "EOF found when expected verb in property list" + ) + if argstr[j] != ";": + break + i = j + 1 - if argstr[j:j + 2] == ":-": + if argstr[j : j + 2] == ":-": if self.turtle: self.BadSyntax(argstr, j, "Found in ':-' in Turtle mode") i = j + 2 res = [] j = self.node(argstr, i, res, subj) if j < 0: - self.BadSyntax(argstr, i, - "bad {} or () or [] node after :- ") + self.BadSyntax(argstr, i, "bad {} or () or [] node after :- ") i = j continue i = j @@ -1085,20 +1103,18 @@ class SinkParser: objs = [] i = self.objectList(argstr, j, objs) if i < 0: - self.BadSyntax(argstr, j, - "objectList expected") + self.BadSyntax(argstr, j, "objectList expected") for obj in objs: dira, sym = v[0] - if dira == '->': + if dira == "->": self.makeStatement((self._context, sym, subj, obj)) else: self.makeStatement((self._context, sym, obj, subj)) j = self.skipSpace(argstr, i) if j < 0: - self.BadSyntax(argstr, j, - "EOF found in list of objects") - if argstr[i:i + 1] != ";": + self.BadSyntax(argstr, j, "EOF found in list of objects") + if argstr[i : i + 1] != ";": return i i = i + 1 # skip semicolon and continue @@ -1108,10 +1124,9 @@ class SinkParser: """ i = self.skipSpace(argstr, j) if i < 0: - self.BadSyntax(argstr, i, - "EOF found expecting comma sep list") + self.BadSyntax(argstr, i, "EOF found expecting comma sep list") if argstr[i] == ".": - return j # empty list is OK + return j # empty list is OK i = what(argstr, i, res) if i < 0: return -1 @@ -1120,15 +1135,14 @@ class SinkParser: j = self.skipSpace(argstr, i) if j < 0: return j # eof - ch = argstr[j:j + 1] + ch = argstr[j : j + 1] if ch != ",": if ch != ".": return -1 - return j # Found but not swallowed "." + return j # Found but not swallowed "." i = what(argstr, j + 1, res) if i < 0: - self.BadSyntax(argstr, i, - "bad list content") + self.BadSyntax(argstr, i, "bad list content") def objectList(self, argstr, i, res): i = self.object(argstr, i, res) @@ -1137,10 +1151,9 @@ class SinkParser: while 1: j = self.skipSpace(argstr, i) if j < 0: - self.BadSyntax(argstr, j, - "EOF found after object") - if argstr[j:j + 1] != ",": - return j # Found something else! + self.BadSyntax(argstr, j, "EOF found after object") + if argstr[j : j + 1] != ",": + return j # Found something else! i = self.object(argstr, j + 1, res) if i < 0: return i @@ -1149,14 +1162,13 @@ class SinkParser: j = self.skipSpace(argstr, i) if j < 0: return j # eof - if argstr[j:j + 1] == ".": - return j + 1 # skip - if argstr[j:j + 1] == "}": - return j # don't skip it - if argstr[j:j + 1] == "]": + if argstr[j : j + 1] == ".": + return j + 1 # skip + if argstr[j : j + 1] == "}": + return j # don't skip it + if argstr[j : j + 1] == "]": return j - self.BadSyntax(argstr, j, - "expected '.' or '}' or ']' at end of statement") + self.BadSyntax(argstr, j, "expected '.' or '}' or ']' at end of statement") def uri_ref2(self, argstr, i, res): """Generate uri from n3 representation. @@ -1182,8 +1194,7 @@ class SinkParser: if not self.turtle and pfx == "": ns = join(self._baseURI or "", "#") else: - self.BadSyntax(argstr, i, - "Prefix \"%s:\" not bound" % (pfx)) + self.BadSyntax(argstr, i, 'Prefix "%s:" not bound' % (pfx)) symb = self._store.newSymbol(ns + ln) if symb in self._variables: res.append(self._variables[symb]) @@ -1217,11 +1228,11 @@ class SinkParser: if self._baseURI: uref = join(self._baseURI, uref) # was: uripath.join else: - assert ":" in uref, \ - "With no base URI, cannot deal with relative URIs" - if argstr[i - 1:i] == "#" and not uref[-1:] == "#": - uref = uref + \ - "#" # She meant it! Weirdness in urlparse? + assert ( + ":" in uref + ), "With no base URI, cannot deal with relative URIs" + if argstr[i - 1 : i] == "#" and not uref[-1:] == "#": + uref = uref + "#" # She meant it! Weirdness in urlparse? symb = self._store.newSymbol(uref) if symb in self._variables: res.append(self._variables[symb]) @@ -1229,17 +1240,15 @@ class SinkParser: res.append(symb) return i + 1 i = i + 1 - self.BadSyntax(argstr, j, - "unterminated URI reference") + self.BadSyntax(argstr, j, "unterminated URI reference") elif self.keywordsSet: v = [] j = self.bareWord(argstr, i, v) if j < 0: - return -1 # Forget varibles as a class, only in context. + return -1 # Forget varibles as a class, only in context. if v[0] in self.keywords: - self.BadSyntax(argstr, i, - 'Keyword "%s" not allowed here.' % v[0]) + self.BadSyntax(argstr, i, 'Keyword "%s" not allowed here.' % v[0]) res.append(self._store.newSymbol(self._bindings[""] + v[0])) return j else: @@ -1253,7 +1262,7 @@ class SinkParser: if m is None: break self.lines = self.lines + 1 - i = m.end() # Point to first character unmatched + i = m.end() # Point to first character unmatched self.startOfLine = i m = ws.match(argstr, i) if m is not None: @@ -1271,30 +1280,31 @@ class SinkParser: if j < 0: return -1 - if argstr[j:j + 1] != "?": + if argstr[j : j + 1] != "?": return -1 j = j + 1 i = j if argstr[j] in "0123456789-": - self.BadSyntax(argstr, j, - "Varible name can't start with '%s'" % argstr[j]) + self.BadSyntax(argstr, j, "Varible name can't start with '%s'" % argstr[j]) while i < len(argstr) and argstr[i] not in _notKeywordsChars: i = i + 1 if self._parentContext is None: varURI = self._store.newSymbol(self._baseURI + "#" + argstr[j:i]) if varURI not in self._variables: self._variables[varURI] = self._context.newUniversal( - varURI, why=self._reason2) + varURI, why=self._reason2 + ) res.append(self._variables[varURI]) return i - # @@ was: - # self.BadSyntax(argstr, j, - # "Can't use ?xxx syntax for variable in outermost level: %s" - # % argstr[j-1:i]) + # @@ was: + # self.BadSyntax(argstr, j, + # "Can't use ?xxx syntax for variable in outermost level: %s" + # % argstr[j-1:i]) varURI = self._store.newSymbol(self._baseURI + "#" + argstr[j:i]) if varURI not in self._parentVariables: self._parentVariables[varURI] = self._parentContext.newUniversal( - varURI, why=self._reason2) + varURI, why=self._reason2 + ) res.append(self._parentVariables[varURI]) return i @@ -1340,16 +1350,17 @@ class SinkParser: if argstr[i - 1] == ".": # qname cannot end with "." ln = ln[:-1] - if not ln: return -1 + if not ln: + return -1 i -= 1 else: # First character is non-alpha - ln = '' # Was: None - TBL (why? useful?) + ln = "" # Was: None - TBL (why? useful?) - if i < len(argstr) and argstr[i] == ':': + if i < len(argstr) and argstr[i] == ":": pfx = ln # bnodes names have different rules - if pfx == '_': + if pfx == "_": allowedChars = _notNameChars else: allowedChars = _notQNameChars @@ -1357,10 +1368,10 @@ class SinkParser: i = i + 1 lastslash = False # start = i # TODO first char . - ln = '' + ln = "" while i < len(argstr): c = argstr[i] - if not lastslash and c == '\\': + if not lastslash and c == "\\": lastslash = True i += 1 @@ -1368,12 +1379,25 @@ class SinkParser: if lastslash: if c not in escapeChars: - raise BadSyntax(self._thisDoc, self.line, argstr, i, - "illegal escape "+c) - elif c=='%': - if argstr[i+1] not in hexChars or argstr[i+2] not in hexChars: - raise BadSyntax(self._thisDoc, self.line, argstr, i, - "illegal hex escape "+c) + raise BadSyntax( + self._thisDoc, + self.line, + argstr, + i, + "illegal escape " + c, + ) + elif c == "%": + if ( + argstr[i + 1] not in hexChars + or argstr[i + 2] not in hexChars + ): + raise BadSyntax( + self._thisDoc, + self.line, + argstr, + i, + "illegal hex escape " + c, + ) ln = ln + c i = i + 1 @@ -1383,22 +1407,22 @@ class SinkParser: if lastslash: raise BadSyntax( - self._thisDoc, self.line, argstr, i, - "qname cannot end with \\") - + self._thisDoc, self.line, argstr, i, "qname cannot end with \\" + ) - if argstr[i-1]=='.': + if argstr[i - 1] == ".": # localname cannot end in . ln = ln[:-1] - if not ln: return -1 + if not ln: + return -1 i -= 1 res.append((pfx, ln)) return i - else: # delimiter was not ":" + else: # delimiter was not ":" if ln and self.keywordsSet and ln not in self.keywords: - res.append(('', ln)) + res.append(("", ln)) return i return -1 @@ -1414,7 +1438,7 @@ class SinkParser: i = j if argstr[i] in self.string_delimiters: - if argstr[i:i + 3] == argstr[i] * 3: + if argstr[i : i + 3] == argstr[i] * 3: delim = argstr[i] * 3 else: delim = argstr[i] @@ -1462,7 +1486,7 @@ class SinkParser: # return -1 ## or fall through? if argstr[i] in self.string_delimiters: - if argstr[i:i + 3] == argstr[i] * 3: + if argstr[i : i + 3] == argstr[i] * 3: delim = argstr[i] * 3 else: delim = argstr[i] @@ -1471,17 +1495,20 @@ class SinkParser: dt = None j, s = self.strconst(argstr, i, delim) lang = None - if argstr[j:j + 1] == "@": # Language? + if argstr[j : j + 1] == "@": # Language? m = langcode.match(argstr, j + 1) if m is None: raise BadSyntax( - self._thisDoc, startline, argstr, i, - "Bad language code syntax on string " + - "literal, after @") + self._thisDoc, + startline, + argstr, + i, + "Bad language code syntax on string " + "literal, after @", + ) i = m.end() - lang = argstr[j + 1:i] + lang = argstr[j + 1 : i] j = i - if argstr[j:j + 2] == "^^": + if argstr[j : j + 2] == "^^": res2 = [] j = self.uri_ref2(argstr, j + 2, res2) # Read datatype URI dt = res2[0] @@ -1493,7 +1520,7 @@ class SinkParser: def uriOf(self, sym): if isinstance(sym, tuple): return sym[1] # old system for --pipe - # return sym.uriref() # cwm api + # return sym.uriref() # cwm api return sym def strconst(self, argstr, i, delim): @@ -1504,35 +1531,39 @@ class SinkParser: delim2, delim3, delim4, delim5 = delim1 * 2, delim1 * 3, delim1 * 4, delim1 * 5 j = i - ustr = u"" # Empty unicode string + ustr = u"" # Empty unicode string startline = self.lines # Remember where for error messages while j < len(argstr): if argstr[j] == delim1: if delim == delim1: # done when delim is " or ' i = j + 1 return i, ustr - if delim == delim3: # done when delim is """ or ''' and, respectively ... - if argstr[j:j + 5] == delim5: # ... we have "" or '' before + if ( + delim == delim3 + ): # done when delim is """ or ''' and, respectively ... + if argstr[j : j + 5] == delim5: # ... we have "" or '' before i = j + 5 ustr = ustr + delim2 return i, ustr - if argstr[j:j + 4] == delim4: # ... we have " or ' before + if argstr[j : j + 4] == delim4: # ... we have " or ' before i = j + 4 ustr = ustr + delim1 return i, ustr - if argstr[j:j + 3] == delim3: # current " or ' is part of delim + if argstr[j : j + 3] == delim3: # current " or ' is part of delim i = j + 3 return i, ustr - # we are inside of the string and current char is " or ' + # we are inside of the string and current char is " or ' j = j + 1 ustr = ustr + delim1 continue - m = interesting.search(argstr, j) # was argstr[j:]. - # Note for pos param to work, MUST be compiled ... re bug? + m = interesting.search(argstr, j) # was argstr[j:]. + # Note for pos param to work, MUST be compiled ... re bug? assert m, "Quote expected in string at ^ in %s^%s" % ( - argstr[j - 20:j], argstr[j:j + 20]) # at least need a quote + argstr[j - 20 : j], + argstr[j : j + 20], + ) # at least need a quote i = m.start() try: @@ -1543,12 +1574,15 @@ class SinkParser: err = err + (" %02x" % ord(c)) streason = sys.exc_info()[1].__str__() raise BadSyntax( - self._thisDoc, startline, argstr, j, - "Unicode error appending characters" + - " %s to string, because\n\t%s" - % (err, streason)) + self._thisDoc, + startline, + argstr, + j, + "Unicode error appending characters" + + " %s to string, because\n\t%s" % (err, streason), + ) - # print "@@@ i = ",i, " j=",j, "m.end=", m.end() + # print "@@@ i = ",i, " j=",j, "m.end=", m.end() ch = argstr[i] if ch == delim1: @@ -1561,8 +1595,12 @@ class SinkParser: elif ch in "\r\n": if delim == delim1: raise BadSyntax( - self._thisDoc, startline, argstr, i, - "newline found in string literal") + self._thisDoc, + startline, + argstr, + i, + "newline found in string literal", + ) self.lines = self.lines + 1 ustr = ustr + ch j = i + 1 @@ -1570,14 +1608,18 @@ class SinkParser: elif ch == "\\": j = i + 1 - ch = argstr[j:j + 1] # Will be empty if string ends + ch = argstr[j : j + 1] # Will be empty if string ends if not ch: raise BadSyntax( - self._thisDoc, startline, argstr, i, - "unterminated string literal (2)") - k = 'abfrtvn\\"\''.find(ch) + self._thisDoc, + startline, + argstr, + i, + "unterminated string literal (2)", + ) + k = "abfrtvn\\\"'".find(ch) if k >= 0: - uch = '\a\b\f\r\t\v\n\\"\''[k] + uch = "\a\b\f\r\t\v\n\\\"'"[k] ustr = ustr + uch j = j + 1 elif ch == "u": @@ -1587,41 +1629,43 @@ class SinkParser: j, ch = self.UEscape(argstr, j + 1, startline) ustr = ustr + ch else: - self.BadSyntax(argstr, i, - "bad escape") + self.BadSyntax(argstr, i, "bad escape") - self.BadSyntax(argstr, i, - "unterminated string literal") + self.BadSyntax(argstr, i, "unterminated string literal") def _unicodeEscape(self, argstr, i, startline, reg, n, prefix): - if len(argstr)<i+n: + if len(argstr) < i + n: raise BadSyntax( - self._thisDoc, startline, argstr, i, - "unterminated string literal(3)") + self._thisDoc, startline, argstr, i, "unterminated string literal(3)" + ) try: - return i+n, reg.sub(unicodeExpand, '\\'+prefix+argstr[i:i+n]) + return i + n, reg.sub(unicodeExpand, "\\" + prefix + argstr[i : i + n]) except: raise BadSyntax( - self._thisDoc, startline, argstr, i, - "bad string literal hex escape: "+argstr[i:i+n]) + self._thisDoc, + startline, + argstr, + i, + "bad string literal hex escape: " + argstr[i : i + n], + ) def uEscape(self, argstr, i, startline): - return self._unicodeEscape(argstr, i, startline, unicodeEscape4, 4, 'u') + return self._unicodeEscape(argstr, i, startline, unicodeEscape4, 4, "u") def UEscape(self, argstr, i, startline): - return self._unicodeEscape(argstr, i, startline, unicodeEscape8, 8, 'U') + return self._unicodeEscape(argstr, i, startline, unicodeEscape8, 8, "U") def BadSyntax(self, argstr, i, msg): raise BadSyntax(self._thisDoc, self.lines, argstr, i, msg) + # If we are going to do operators then they should generate # [ is operator:plus of ( \1 \2 ) ] class BadSyntax(SyntaxError): def __init__(self, uri, lines, argstr, i, why): - self._str = argstr.encode( - 'utf-8') # Better go back to strings for errors + self._str = argstr.encode("utf-8") # Better go back to strings for errors self._i = i self._why = why self.lines = lines @@ -1641,16 +1685,21 @@ class BadSyntax(SyntaxError): else: post = "" - return 'at line %i of <%s>:\nBad syntax (%s) at ^ in:\n"%s%s^%s%s"' \ - % (self.lines + 1, self._uri, self._why, pre, - argstr[st:i], argstr[i:i + 60], post) + return 'at line %i of <%s>:\nBad syntax (%s) at ^ in:\n"%s%s^%s%s"' % ( + self.lines + 1, + self._uri, + self._why, + pre, + argstr[st:i], + argstr[i : i + 60], + post, + ) @property def message(self): return str(self) - ############################################################################### class Formula(object): number = 0 @@ -1663,25 +1712,24 @@ class Formula(object): self.existentials = {} self.universals = {} - self.quotedgraph = QuotedGraph( - store=parent.store, identifier=self.id()) + self.quotedgraph = QuotedGraph(store=parent.store, identifier=self.id()) def __str__(self): - return '_:Formula%s' % self.number + return "_:Formula%s" % self.number def id(self): - return BNode('_:Formula%s' % self.number) + return BNode("_:Formula%s" % self.number) def newBlankNode(self, uri=None, why=None): if uri is None: self.counter += 1 - bn = BNode('f%sb%s' % (self.uuid, self.counter)) + bn = BNode("f%sb%s" % (self.uuid, self.counter)) else: - bn = BNode(uri.split('#').pop().replace('_', 'b')) + bn = BNode(uri.split("#").pop().replace("_", "b")) return bn def newUniversal(self, uri, why=None): - return Variable(uri.split('#').pop()) + return Variable(uri.split("#").pop()) def declareExistential(self, x): self.existentials[x] = self.newBlankNode() @@ -1691,7 +1739,7 @@ class Formula(object): return self.quotedgraph -r_hibyte = re.compile(r'([\x80-\xff])') +r_hibyte = re.compile(r"([\x80-\xff])") class RDFSink(object): @@ -1716,9 +1764,9 @@ class RDFSink(object): return arg.newBlankNode(uri) elif isinstance(arg, Graph) or arg is None: self.counter += 1 - bn = BNode('n' + str(self.counter)) + bn = BNode("n" + str(self.counter)) else: - bn = BNode(str(arg[0]).split('#').pop().replace('_', 'b')) + bn = BNode(str(arg[0]).split("#").pop().replace("_", "b")) return bn def newLiteral(self, s, dt, lang): @@ -1728,18 +1776,12 @@ class RDFSink(object): return Literal(s, lang=lang) def newList(self, n, f): - nil = self.newSymbol( - 'http://www.w3.org/1999/02/22-rdf-syntax-ns#nil' - ) + nil = self.newSymbol("http://www.w3.org/1999/02/22-rdf-syntax-ns#nil") if not n: return nil - first = self.newSymbol( - 'http://www.w3.org/1999/02/22-rdf-syntax-ns#first' - ) - rest = self.newSymbol( - 'http://www.w3.org/1999/02/22-rdf-syntax-ns#rest' - ) + first = self.newSymbol("http://www.w3.org/1999/02/22-rdf-syntax-ns#first") + rest = self.newSymbol("http://www.w3.org/1999/02/22-rdf-syntax-ns#rest") af = a = self.newBlankNode(f) for ne in n[:-1]: @@ -1755,12 +1797,12 @@ class RDFSink(object): return set(args) def setDefaultNamespace(self, *args): - return ':'.join(repr(n) for n in args) + return ":".join(repr(n) for n in args) def makeStatement(self, quadruple, why=None): f, p, s, o = quadruple - if hasattr(p, 'formula'): + if hasattr(p, "formula"): raise Exception("Formula used as predicate") s = self.normalise(f, s) @@ -1768,14 +1810,14 @@ class RDFSink(object): o = self.normalise(f, o) if f == self.rootFormula: - # print s, p, o, '.' + # print s, p, o, '.' self.graph.add((s, p, o)) elif isinstance(f, Formula): f.quotedgraph.add((s, p, o)) else: - f.add((s,p,o)) + f.add((s, p, o)) - # return str(quadruple) + # return str(quadruple) def normalise(self, f, n): if isinstance(n, tuple): @@ -1791,8 +1833,8 @@ class RDFSink(object): if isinstance(n, Decimal): value = str(n) - if value == '-0': - value = '0' + if value == "-0": + value = "0" s = Literal(value, datatype=DECIMAL_DATATYPE) return s @@ -1804,11 +1846,11 @@ class RDFSink(object): if n in f.existentials: return f.existentials[n] - # if isinstance(n, Var): - # if f.universals.has_key(n): - # return f.universals[n] - # f.universals[n] = f.newBlankNode() - # return f.universals[n] + # if isinstance(n, Var): + # if f.universals.has_key(n): + # return f.universals[n] + # f.universals[n] = f.newBlankNode() + # return f.universals[n] return n @@ -1841,7 +1883,7 @@ def hexify(ustr): """ # s1=ustr.encode('utf-8') s = "" - for ch in ustr: # .encode('utf-8'): + for ch in ustr: # .encode('utf-8'): if ord(ch) > 126 or ord(ch) < 33: ch = "%%%02X" % ord(ch) else: @@ -1865,13 +1907,13 @@ class TurtleParser(Parser): if encoding not in [None, "utf-8"]: raise Exception( - ("N3/Turtle files are always utf-8 encoded, ", - "I was passed: %s") % encoding) + ("N3/Turtle files are always utf-8 encoded, ", "I was passed: %s") + % encoding + ) sink = RDFSink(graph) - baseURI = graph.absolutize( - source.getPublicId() or source.getSystemId() or "") + baseURI = graph.absolutize(source.getPublicId() or source.getSystemId() or "") p = SinkParser(sink, baseURI=baseURI, turtle=turtle) p.loadStream(source.getByteStream()) @@ -1893,38 +1935,40 @@ class N3Parser(TurtleParser): pass def parse(self, source, graph, encoding="utf-8"): - # we're currently being handed a Graph, not a ConjunctiveGraph + # we're currently being handed a Graph, not a ConjunctiveGraph assert graph.store.context_aware # is this implied by formula_aware assert graph.store.formula_aware conj_graph = ConjunctiveGraph(store=graph.store) conj_graph.default_context = graph # TODO: CG __init__ should have a - # default_context arg - # TODO: update N3Processor so that it can use conj_graph as the sink + # default_context arg + # TODO: update N3Processor so that it can use conj_graph as the sink conj_graph.namespace_manager = graph.namespace_manager TurtleParser.parse(self, source, conj_graph, encoding, turtle=False) -def _test(): # pragma: no cover +def _test(): # pragma: no cover import doctest + doctest.testmod() # if __name__ == '__main__': # _test() -def main(): # pragma: no cover + +def main(): # pragma: no cover g = ConjunctiveGraph() sink = RDFSink(g) - base_uri = 'file://' + os.path.join(os.getcwd(), sys.argv[1]) + base_uri = "file://" + os.path.join(os.getcwd(), sys.argv[1]) p = SinkParser(sink, baseURI=base_uri) - p._bindings[''] = p._baseURI + '#' + p._bindings[""] = p._baseURI + "#" p.startDoc() - f = open(sys.argv[1], 'rb') + f = open(sys.argv[1], "rb") rdbytes = f.read() f.close() @@ -1934,7 +1978,8 @@ def main(): # pragma: no cover print(t) -if __name__ == '__main__': + +if __name__ == "__main__": main() # ends diff --git a/rdflib/plugins/parsers/nquads.py b/rdflib/plugins/parsers/nquads.py index 8bc415bd..0c29fc4c 100644 --- a/rdflib/plugins/parsers/nquads.py +++ b/rdflib/plugins/parsers/nquads.py @@ -36,26 +36,26 @@ from rdflib.plugins.parsers.ntriples import ParseError from rdflib.plugins.parsers.ntriples import r_tail from rdflib.plugins.parsers.ntriples import r_wspace -__all__ = ['NQuadsParser'] +__all__ = ["NQuadsParser"] class NQuadsParser(NTriplesParser): - def parse(self, inputsource, sink, **kwargs): """Parse f as an N-Triples file.""" - assert sink.store.context_aware, ("NQuadsParser must be given" - " a context aware store.") + assert sink.store.context_aware, ( + "NQuadsParser must be given" " a context aware store." + ) self.sink = ConjunctiveGraph(store=sink.store, identifier=sink.identifier) source = inputsource.getByteStream() - if not hasattr(source, 'read'): + if not hasattr(source, "read"): raise ParseError("Item to parse must be a file-like object.") - source = getreader('utf-8')(source) + source = getreader("utf-8")(source) self.file = source - self.buffer = '' + self.buffer = "" while True: self.line = __line = self.readline() if self.line is None: @@ -69,7 +69,7 @@ class NQuadsParser(NTriplesParser): def parseline(self): self.eat(r_wspace) - if (not self.line) or self.line.startswith(('#')): + if (not self.line) or self.line.startswith(("#")): return # The line is empty or a comment subject = self.subject() diff --git a/rdflib/plugins/parsers/nt.py b/rdflib/plugins/parsers/nt.py index 783488af..d7d3b336 100644 --- a/rdflib/plugins/parsers/nt.py +++ b/rdflib/plugins/parsers/nt.py @@ -1,7 +1,7 @@ from rdflib.parser import Parser from rdflib.plugins.parsers.ntriples import NTriplesParser -__all__ = ['NTSink', 'NTParser'] +__all__ = ["NTSink", "NTParser"] class NTSink(object): diff --git a/rdflib/plugins/parsers/ntriples.py b/rdflib/plugins/parsers/ntriples.py index 0724a86e..9398c8de 100644 --- a/rdflib/plugins/parsers/ntriples.py +++ b/rdflib/plugins/parsers/ntriples.py @@ -22,18 +22,18 @@ from rdflib.compat import decodeUnicodeEscape from io import BytesIO -__all__ = ['unquote', 'uriquote', 'Sink', 'NTriplesParser'] +__all__ = ["unquote", "uriquote", "Sink", "NTriplesParser"] uriref = r'<([^:]+:[^\s"<>]*)>' literal = r'"([^"\\]*(?:\\.[^"\\]*)*)"' -litinfo = r'(?:@([a-zA-Z]+(?:-[a-zA-Z0-9]+)*)|\^\^' + uriref + r')?' +litinfo = r"(?:@([a-zA-Z]+(?:-[a-zA-Z0-9]+)*)|\^\^" + uriref + r")?" -r_line = re.compile(r'([^\r\n]*)(?:\r\n|\r|\n)') -r_wspace = re.compile(r'[ \t]*') -r_wspaces = re.compile(r'[ \t]+') -r_tail = re.compile(r'[ \t]*\.[ \t]*(#.*)?') +r_line = re.compile(r"([^\r\n]*)(?:\r\n|\r|\n)") +r_wspace = re.compile(r"[ \t]*") +r_wspaces = re.compile(r"[ \t]+") +r_tail = re.compile(r"[ \t]*\.[ \t]*(#.*)?") r_uriref = re.compile(uriref) -r_nodeid = re.compile(r'_:([A-Za-z0-9_:]([-A-Za-z0-9_:\.]*[-A-Za-z0-9_:])?)') +r_nodeid = re.compile(r"_:([A-Za-z0-9_:]([-A-Za-z0-9_:\.]*[-A-Za-z0-9_:])?)") r_literal = re.compile(literal + litinfo) bufsiz = 2048 @@ -57,11 +57,10 @@ class Sink(object): print(s, p, o) -quot = {'t': u'\t', 'n': u'\n', 'r': u'\r', '"': u'"', '\\': - u'\\'} -r_safe = re.compile(r'([\x20\x21\x23-\x5B\x5D-\x7E]+)') +quot = {"t": u"\t", "n": u"\n", "r": u"\r", '"': u'"', "\\": u"\\"} +r_safe = re.compile(r"([\x20\x21\x23-\x5B\x5D-\x7E]+)") r_quot = re.compile(r'\\(t|n|r|"|\\)') -r_uniquot = re.compile(r'\\u([0-9A-F]{4})|\\U([0-9A-F]{8})') +r_uniquot = re.compile(r"\\u([0-9A-F]{4})|\\U([0-9A-F]{8})") def unquote(s): @@ -71,7 +70,7 @@ def unquote(s): if isinstance(s, str): # nquads s = decodeUnicodeEscape(s) else: - s = s.decode('unicode-escape') + s = s.decode("unicode-escape") return s else: @@ -79,7 +78,7 @@ def unquote(s): while s: m = r_safe.match(s) if m: - s = s[m.end():] + s = s[m.end() :] result.append(m.group(1)) continue @@ -91,28 +90,27 @@ def unquote(s): m = r_uniquot.match(s) if m: - s = s[m.end():] + s = s[m.end() :] u, U = m.groups() codepoint = int(u or U, 16) if codepoint > 0x10FFFF: raise ParseError("Disallowed codepoint: %08X" % codepoint) result.append(chr(codepoint)) - elif s.startswith('\\'): + elif s.startswith("\\"): raise ParseError("Illegal escape at: %s..." % s[:10]) else: raise ParseError("Illegal literal character: %r" % s[0]) - return u''.join(result) + return u"".join(result) -r_hibyte = re.compile(r'([\x80-\xFF])') +r_hibyte = re.compile(r"([\x80-\xFF])") def uriquote(uri): if not validate: return uri else: - return r_hibyte.sub( - lambda m: '%%%02X' % ord(m.group(1)), uri) + return r_hibyte.sub(lambda m: "%%%02X" % ord(m.group(1)), uri) class NTriplesParser(object): @@ -134,14 +132,14 @@ class NTriplesParser(object): def parse(self, f): """Parse f as an N-Triples file.""" - if not hasattr(f, 'read'): + if not hasattr(f, "read"): raise ParseError("Item to parse must be a file-like object.") # since N-Triples 1.1 files can and should be utf-8 encoded - f = codecs.getreader('utf-8')(f) + f = codecs.getreader("utf-8")(f) self.file = f - self.buffer = '' + self.buffer = "" while True: self.line = self.readline() if self.line is None: @@ -174,7 +172,7 @@ class NTriplesParser(object): while True: m = r_line.match(self.buffer) if m: # the more likely prospect - self.buffer = self.buffer[m.end():] + self.buffer = self.buffer[m.end() :] return m.group(1) else: buffer = self.file.read(bufsiz) @@ -187,7 +185,7 @@ class NTriplesParser(object): def parseline(self): self.eat(r_wspace) - if (not self.line) or self.line.startswith('#'): + if (not self.line) or self.line.startswith("#"): return # The line is empty or a comment subject = self.subject() @@ -212,7 +210,7 @@ class NTriplesParser(object): # print(dir(pattern)) # print repr(self.line), type(self.line) raise ParseError("Failed to eat %s at %s" % (pattern.pattern, self.line)) - self.line = self.line[m.end():] + self.line = self.line[m.end() :] return m def subject(self): @@ -235,7 +233,7 @@ class NTriplesParser(object): return objt def uriref(self): - if self.peek('<'): + if self.peek("<"): uri = self.eat(r_uriref).group(1) uri = unquote(uri) uri = uriquote(uri) @@ -243,7 +241,7 @@ class NTriplesParser(object): return False def nodeid(self): - if self.peek('_'): + if self.peek("_"): # Fix for https://github.com/RDFLib/rdflib/issues/204 bnode_id = self.eat(r_nodeid).group(1) new_id = self._bnode_ids.get(bnode_id, None) @@ -277,6 +275,7 @@ class NTriplesParser(object): return Literal(lit, lang, dtype) return False + # # Obsolete, unused # def parseURI(uri): # import urllib diff --git a/rdflib/plugins/parsers/rdfxml.py b/rdflib/plugins/parsers/rdfxml.py index 17554ba1..976edf2c 100644 --- a/rdflib/plugins/parsers/rdfxml.py +++ b/rdflib/plugins/parsers/rdfxml.py @@ -15,22 +15,30 @@ from rdflib.term import Literal from rdflib.exceptions import ParserError, Error from rdflib.parser import Parser -__all__ = ['create_parser', 'BagID', 'ElementHandler', - 'RDFXMLHandler', 'RDFXMLParser'] +__all__ = ["create_parser", "BagID", "ElementHandler", "RDFXMLHandler", "RDFXMLParser"] RDFNS = RDF # http://www.w3.org/TR/rdf-syntax-grammar/#eventterm-attribute-URI # A mapping from unqualified terms to their qualified version. -UNQUALIFIED = {"about": RDF.about, - "ID": RDF.ID, - "type": RDF.type, - "resource": RDF.resource, - "parseType": RDF.parseType} +UNQUALIFIED = { + "about": RDF.about, + "ID": RDF.ID, + "type": RDF.type, + "resource": RDF.resource, + "parseType": RDF.parseType, +} # http://www.w3.org/TR/rdf-syntax-grammar/#coreSyntaxTerms -CORE_SYNTAX_TERMS = [RDF.RDF, RDF.ID, RDF.about, RDF.parseType, - RDF.resource, RDF.nodeID, RDF.datatype] +CORE_SYNTAX_TERMS = [ + RDF.RDF, + RDF.ID, + RDF.about, + RDF.parseType, + RDF.resource, + RDF.nodeID, + RDF.datatype, +] # http://www.w3.org/TR/rdf-syntax-grammar/#syntaxTerms SYNTAX_TERMS = CORE_SYNTAX_TERMS + [RDF.Description, RDF.li] @@ -39,15 +47,16 @@ SYNTAX_TERMS = CORE_SYNTAX_TERMS + [RDF.Description, RDF.li] OLD_TERMS = [ URIRef("http://www.w3.org/1999/02/22-rdf-syntax-ns#aboutEach"), URIRef("http://www.w3.org/1999/02/22-rdf-syntax-ns#aboutEachPrefix"), - URIRef("http://www.w3.org/1999/02/22-rdf-syntax-ns#bagID")] + URIRef("http://www.w3.org/1999/02/22-rdf-syntax-ns#bagID"), +] -NODE_ELEMENT_EXCEPTIONS = CORE_SYNTAX_TERMS + [RDF.li, ] + OLD_TERMS +NODE_ELEMENT_EXCEPTIONS = CORE_SYNTAX_TERMS + [RDF.li,] + OLD_TERMS NODE_ELEMENT_ATTRIBUTES = [RDF.ID, RDF.nodeID, RDF.about] -PROPERTY_ELEMENT_EXCEPTIONS = \ - CORE_SYNTAX_TERMS + [RDF.Description, ] + OLD_TERMS -PROPERTY_ATTRIBUTE_EXCEPTIONS = \ +PROPERTY_ELEMENT_EXCEPTIONS = CORE_SYNTAX_TERMS + [RDF.Description,] + OLD_TERMS +PROPERTY_ATTRIBUTE_EXCEPTIONS = ( CORE_SYNTAX_TERMS + [RDF.Description, RDF.li] + OLD_TERMS +) PROPERTY_ELEMENT_ATTRIBUTES = [RDF.ID, RDF.resource, RDF.nodeID] XMLNS = "http://www.w3.org/XML/1998/namespace" @@ -56,7 +65,7 @@ LANG = (XMLNS, "lang") class BagID(URIRef): - __slots__ = ['li'] + __slots__ = ["li"] def __init__(self, val): super(URIRef, self).__init__(val) @@ -64,13 +73,26 @@ class BagID(URIRef): def next_li(self): self.li += 1 - return RDFNS['_%s' % self.li] + return RDFNS["_%s" % self.li] class ElementHandler(object): - __slots__ = ['start', 'char', 'end', 'li', 'id', - 'base', 'subject', 'predicate', 'object', - 'list', 'language', 'datatype', 'declared', 'data'] + __slots__ = [ + "start", + "char", + "end", + "li", + "id", + "base", + "subject", + "predicate", + "object", + "list", + "language", + "datatype", + "declared", + "data", + ] def __init__(self): self.start = None @@ -89,11 +111,10 @@ class ElementHandler(object): def next_li(self): self.li += 1 - return RDFNS['_%s' % self.li] + return RDFNS["_%s" % self.li] class RDFXMLHandler(handler.ContentHandler): - def __init__(self, store): self.store = store self.preserve_bnode_ids = False @@ -103,7 +124,10 @@ class RDFXMLHandler(handler.ContentHandler): document_element = ElementHandler() document_element.start = self.document_element_start document_element.end = lambda name, qname: None - self.stack = [None, document_element, ] + self.stack = [ + None, + document_element, + ] self.ids = {} # remember IDs we have already seen self.bnode = {} self._ns_contexts = [{}] # contains uri -> prefix dicts @@ -137,16 +161,14 @@ class RDFXMLHandler(handler.ContentHandler): if parent and parent.base: base = urljoin(parent.base, base) else: - systemId = self.locator.getPublicId() \ - or self.locator.getSystemId() + systemId = self.locator.getPublicId() or self.locator.getSystemId() if systemId: base = urljoin(systemId, base) else: if parent: base = parent.base if base is None: - systemId = self.locator.getPublicId() \ - or self.locator.getSystemId() + systemId = self.locator.getPublicId() or self.locator.getSystemId() if systemId: base, frag = urldefrag(systemId) current.base = base @@ -181,25 +203,30 @@ class RDFXMLHandler(handler.ContentHandler): def error(self, message): locator = self.locator - info = "%s:%s:%s: " % (locator.getSystemId(), - locator.getLineNumber(), - locator.getColumnNumber()) + info = "%s:%s:%s: " % ( + locator.getSystemId(), + locator.getLineNumber(), + locator.getColumnNumber(), + ) raise ParserError(info + message) def get_current(self): return self.stack[-2] + # Create a read only property called current so that self.current # give the current element handler. current = property(get_current) def get_next(self): return self.stack[-1] + # Create a read only property that gives the element handler to be # used for the next element. next = property(get_next) def get_parent(self): return self.stack[-3] + # Create a read only property that gives the current parent # element handler parent = property(get_parent) @@ -233,7 +260,7 @@ class RDFXMLHandler(handler.ContentHandler): def document_element_start(self, name, qname, attrs): if name[0] and URIRef("".join(name)) == RDF.RDF: # Cheap hack so 2to3 doesn't turn it into __next__ - next = getattr(self, 'next') + next = getattr(self, "next") next.start = self.node_element_start next.end = self.node_element_end else: @@ -248,7 +275,7 @@ class RDFXMLHandler(handler.ContentHandler): absolutize = self.absolutize # Cheap hack so 2to3 doesn't turn it into __next__ - next = getattr(self, 'next') + next = getattr(self, "next") next.start = self.property_element_start next.end = self.property_element_end @@ -257,27 +284,21 @@ class RDFXMLHandler(handler.ContentHandler): if RDF.ID in atts: if RDF.about in atts or RDF.nodeID in atts: - self.error( - "Can have at most one of rdf:ID, rdf:about, and rdf:nodeID" - ) + self.error("Can have at most one of rdf:ID, rdf:about, and rdf:nodeID") id = atts[RDF.ID] if not is_ncname(id): self.error("rdf:ID value is not a valid NCName: %s" % id) subject = absolutize("#%s" % id) if subject in self.ids: - self.error( - "two elements cannot use the same ID: '%s'" % subject) + self.error("two elements cannot use the same ID: '%s'" % subject) self.ids[subject] = 1 # IDs can only appear once within a document elif RDF.nodeID in atts: if RDF.ID in atts or RDF.about in atts: - self.error( - "Can have at most one of rdf:ID, rdf:about, and rdf:nodeID" - ) + self.error("Can have at most one of rdf:ID, rdf:about, and rdf:nodeID") nodeID = atts[RDF.nodeID] if not is_ncname(nodeID): - self.error( - "rdf:nodeID value is not a valid NCName: %s" % nodeID) + self.error("rdf:nodeID value is not a valid NCName: %s" % nodeID) if self.preserve_bnode_ids is False: if nodeID in self.bnode: subject = self.bnode[nodeID] @@ -288,9 +309,7 @@ class RDFXMLHandler(handler.ContentHandler): subject = BNode(nodeID) elif RDF.about in atts: if RDF.ID in atts or RDF.nodeID in atts: - self.error( - "Can have at most one of rdf:ID, rdf:about, and rdf:nodeID" - ) + self.error("Can have at most one of rdf:ID, rdf:about, and rdf:nodeID") subject = absolutize(atts[RDF.about]) else: subject = BNode() @@ -330,7 +349,9 @@ class RDFXMLHandler(handler.ContentHandler): if self.parent.object and self.current != self.stack[2]: - self.error("Repeat node-elements inside property elements: %s"%"".join(name)) + self.error( + "Repeat node-elements inside property elements: %s" % "".join(name) + ) self.parent.object = self.current.subject @@ -340,7 +361,7 @@ class RDFXMLHandler(handler.ContentHandler): absolutize = self.absolutize # Cheap hack so 2to3 doesn't turn it into __next__ - next = getattr(self, 'next') + next = getattr(self, "next") object = None current.data = None current.list = None @@ -366,17 +387,14 @@ class RDFXMLHandler(handler.ContentHandler): nodeID = atts.get(RDF.nodeID, None) parse_type = atts.get(RDF.parseType, None) if resource is not None and nodeID is not None: - self.error( - "Property element cannot have both rdf:nodeID and rdf:resource" - ) + self.error("Property element cannot have both rdf:nodeID and rdf:resource") if resource is not None: object = absolutize(resource) next.start = self.node_element_start next.end = self.node_element_end elif nodeID is not None: if not is_ncname(nodeID): - self.error( - "rdf:nodeID value is not a valid NCName: %s" % nodeID) + self.error("rdf:nodeID value is not a valid NCName: %s" % nodeID) if self.preserve_bnode_ids is False: if nodeID in self.bnode: object = self.bnode[nodeID] @@ -401,16 +419,16 @@ class RDFXMLHandler(handler.ContentHandler): elif parse_type == "Collection": current.char = None object = current.list = RDF.nil # BNode() - # self.parent.subject + # self.parent.subject next.start = self.node_element_start next.end = self.list_node_element_end else: # if parse_type=="Literal": - # All other values are treated as Literal - # See: http://www.w3.org/TR/rdf-syntax-grammar/ - # parseTypeOtherPropertyElt + # All other values are treated as Literal + # See: http://www.w3.org/TR/rdf-syntax-grammar/ + # parseTypeOtherPropertyElt object = Literal("", datatype=RDF.XMLLiteral) current.char = self.literal_element_char - current.declared = {XMLNS: 'xml'} + current.declared = {XMLNS: "xml"} next.start = self.literal_element_start next.char = self.literal_element_char next.end = self.literal_element_end @@ -466,18 +484,17 @@ class RDFXMLHandler(handler.ContentHandler): literalLang = current.language if current.datatype is not None: literalLang = None - current.object = Literal( - current.data, literalLang, current.datatype) + current.object = Literal(current.data, literalLang, current.datatype) current.data = None if self.next.end == self.list_node_element_end: if current.object != RDF.nil: self.store.add((current.list, RDF.rest, RDF.nil)) if current.object is not None: - self.store.add( - (self.parent.subject, current.predicate, current.object)) + self.store.add((self.parent.subject, current.predicate, current.object)) if current.id is not None: - self.add_reified(current.id, (self.parent.subject, - current.predicate, current.object)) + self.add_reified( + current.id, (self.parent.subject, current.predicate, current.object) + ) current.subject = None def list_node_element_end(self, name, qname): @@ -513,9 +530,9 @@ class RDFXMLHandler(handler.ContentHandler): if not name[0] in current.declared: current.declared[name[0]] = prefix if prefix: - current.object += (' xmlns:%s="%s"' % (prefix, name[0])) + current.object += ' xmlns:%s="%s"' % (prefix, name[0]) else: - current.object += (' xmlns="%s"' % name[0]) + current.object += ' xmlns="%s"' % name[0] else: current.object = "<%s" % name[1] @@ -526,7 +543,7 @@ class RDFXMLHandler(handler.ContentHandler): name = current.declared[name[0]] + ":" + name[1] else: name = name[1] - current.object += (' %s=%s' % (name, quoteattr(value))) + current.object += " %s=%s" % (name, quoteattr(value)) current.object += ">" def literal_element_char(self, data): @@ -549,8 +566,7 @@ def create_parser(target, store): try: # Workaround for bug in expatreader.py. Needed when # expatreader is trying to guess a prefix. - parser.start_namespace_decl( - "xml", "http://www.w3.org/XML/1998/namespace") + parser.start_namespace_decl("xml", "http://www.w3.org/XML/1998/namespace") except AttributeError: pass # Not present in Jython (at least) parser.setFeature(handler.feature_namespaces, 1) @@ -563,7 +579,6 @@ def create_parser(target, store): class RDFXMLParser(Parser): - def __init__(self): pass diff --git a/rdflib/plugins/parsers/trig.py b/rdflib/plugins/parsers/trig.py index f4c3ff1b..96c94503 100644 --- a/rdflib/plugins/parsers/trig.py +++ b/rdflib/plugins/parsers/trig.py @@ -5,18 +5,18 @@ from rdflib.parser import Parser from .notation3 import SinkParser, RDFSink -def becauseSubGraph(*args, **kwargs): pass +def becauseSubGraph(*args, **kwargs): + pass class TrigSinkParser(SinkParser): - def directiveOrStatement(self, argstr, h): - #import pdb; pdb.set_trace() + # import pdb; pdb.set_trace() i = self.skipSpace(argstr, h) if i < 0: - return i # EOF + return i # EOF j = self.graph(argstr, i) if j >= 0: @@ -46,12 +46,11 @@ class TrigSinkParser(SinkParser): if j >= 0: return j - if argstr[i] == '[': + if argstr[i] == "[": j = self.skipSpace(argstr, i + 1) if j < 0: - self.BadSyntax(argstr, i, - "Expected ] got EOF") - if argstr[j] == ']': + self.BadSyntax(argstr, i, "Expected ] got EOF") + if argstr[j] == "]": res.append(self.blankNode()) return j + 1 return -1 @@ -66,8 +65,8 @@ class TrigSinkParser(SinkParser): raise Exception if it looks like a graph, but isn't. """ - #import pdb; pdb.set_trace() - j = self.sparqlTok('GRAPH', argstr, i) # optional GRAPH keyword + # import pdb; pdb.set_trace() + j = self.sparqlTok("GRAPH", argstr, i) # optional GRAPH keyword if j >= 0: i = j @@ -81,10 +80,9 @@ class TrigSinkParser(SinkParser): j = self.skipSpace(argstr, i) if j < 0: - self.BadSyntax(argstr, i, - "EOF found when expected graph") + self.BadSyntax(argstr, i, "EOF found when expected graph") - if argstr[j:j + 1] == "=": # optional = for legacy support + if argstr[j : j + 1] == "=": # optional = for legacy support i = self.skipSpace(argstr, j + 1) if i < 0: @@ -92,7 +90,7 @@ class TrigSinkParser(SinkParser): else: i = j - if argstr[i:i + 1] != "{": + if argstr[i : i + 1] != "{": return -1 # the node wasn't part of a graph j = i + 1 @@ -106,17 +104,15 @@ class TrigSinkParser(SinkParser): while 1: i = self.skipSpace(argstr, j) if i < 0: - self.BadSyntax( - argstr, i, "needed '}', found end.") + self.BadSyntax(argstr, i, "needed '}', found end.") - if argstr[i:i + 1] == "}": + if argstr[i : i + 1] == "}": j = i + 1 break j = self.directiveOrStatement(argstr, i) if j < 0: - self.BadSyntax( - argstr, i, "expected statement or '}'") + self.BadSyntax(argstr, i, "expected statement or '}'") self._context = self._parentContext self._reason2 = reason2 @@ -138,22 +134,23 @@ class TrigParser(Parser): if encoding not in [None, "utf-8"]: raise Exception( - ("TriG files are always utf-8 encoded, ", - "I was passed: %s") % encoding) + ("TriG files are always utf-8 encoded, ", "I was passed: %s") % encoding + ) # we're currently being handed a Graph, not a ConjunctiveGraph assert graph.store.context_aware, "TriG Parser needs a context-aware store!" conj_graph = ConjunctiveGraph(store=graph.store, identifier=graph.identifier) conj_graph.default_context = graph # TODO: CG __init__ should have a - # default_context arg + # default_context arg # TODO: update N3Processor so that it can use conj_graph as the sink conj_graph.namespace_manager = graph.namespace_manager sink = RDFSink(conj_graph) baseURI = conj_graph.absolutize( - source.getPublicId() or source.getSystemId() or "") + source.getPublicId() or source.getSystemId() or "" + ) p = TrigSinkParser(sink, baseURI=baseURI, turtle=True) p.loadStream(source.getByteStream()) diff --git a/rdflib/plugins/parsers/trix.py b/rdflib/plugins/parsers/trix.py index ffd883fe..56819514 100644 --- a/rdflib/plugins/parsers/trix.py +++ b/rdflib/plugins/parsers/trix.py @@ -14,7 +14,7 @@ from xml.sax.saxutils import handler from xml.sax import make_parser from xml.sax.handler import ErrorHandler -__all__ = ['create_parser', 'TriXHandler', 'TriXParser'] +__all__ = ["create_parser", "TriXHandler", "TriXParser"] TRIXNS = Namespace("http://www.w3.org/2004/03/trix/trix-1/") @@ -56,7 +56,8 @@ class TriXHandler(handler.ContentHandler): if name[0] != str(TRIXNS): self.error( "Only elements in the TriX namespace are allowed. %s!=%s" - % (name[0], TRIXNS)) + % (name[0], TRIXNS) + ) if name[1] == "TriX": if self.state == 0: @@ -143,46 +144,55 @@ class TriXHandler(handler.ContentHandler): if name[0] != str(TRIXNS): self.error( "Only elements in the TriX namespace are allowed. %s!=%s" - % (name[0], TRIXNS)) + % (name[0], TRIXNS) + ) if name[1] == "uri": if self.state == 3: - self.graph = Graph(store=self.store, - identifier=URIRef(self.chars.strip())) + self.graph = Graph( + store=self.store, identifier=URIRef(self.chars.strip()) + ) self.state = 2 elif self.state == 4: self.triple += [URIRef(self.chars.strip())] else: self.error( - "Illegal internal self.state - This should never " + - "happen if the SAX parser ensures XML syntax correctness") + "Illegal internal self.state - This should never " + + "happen if the SAX parser ensures XML syntax correctness" + ) elif name[1] == "id": if self.state == 3: - self.graph = Graph(self.store, identifier=self.get_bnode( - self.chars.strip())) + self.graph = Graph( + self.store, identifier=self.get_bnode(self.chars.strip()) + ) self.state = 2 elif self.state == 4: self.triple += [self.get_bnode(self.chars.strip())] else: self.error( - "Illegal internal self.state - This should never " + - "happen if the SAX parser ensures XML syntax correctness") + "Illegal internal self.state - This should never " + + "happen if the SAX parser ensures XML syntax correctness" + ) elif name[1] == "plainLiteral" or name[1] == "typedLiteral": if self.state == 4: - self.triple += [Literal( - self.chars, lang=self.lang, datatype=self.datatype)] + self.triple += [ + Literal(self.chars, lang=self.lang, datatype=self.datatype) + ] else: self.error( - "This should never happen if the SAX parser " + - "ensures XML syntax correctness") + "This should never happen if the SAX parser " + + "ensures XML syntax correctness" + ) elif name[1] == "triple": if self.state == 4: if len(self.triple) != 3: - self.error("Triple has wrong length, got %d elements: %s" % - (len(self.triple), self.triple)) + self.error( + "Triple has wrong length, got %d elements: %s" + % (len(self.triple), self.triple) + ) self.graph.add(self.triple) # self.store.store.add(self.triple,context=self.graph) @@ -190,8 +200,9 @@ class TriXHandler(handler.ContentHandler): self.state = 2 else: self.error( - "This should never happen if the SAX parser " + - "ensures XML syntax correctness") + "This should never happen if the SAX parser " + + "ensures XML syntax correctness" + ) elif name[1] == "graph": self.graph = None @@ -228,7 +239,8 @@ class TriXHandler(handler.ContentHandler): info = "%s:%s:%s: " % ( locator.getSystemId(), locator.getLineNumber(), - locator.getColumnNumber()) + locator.getColumnNumber(), + ) raise ParserError(info + message) @@ -237,8 +249,7 @@ def create_parser(store): try: # Workaround for bug in expatreader.py. Needed when # expatreader is trying to guess a prefix. - parser.start_namespace_decl( - "xml", "http://www.w3.org/XML/1998/namespace") + parser.start_namespace_decl("xml", "http://www.w3.org/XML/1998/namespace") except AttributeError: pass # Not present in Jython (at least) parser.setFeature(handler.feature_namespaces, 1) @@ -255,8 +266,9 @@ class TriXParser(Parser): pass def parse(self, source, sink, **args): - assert sink.store.context_aware, ( - "TriXParser must be given a context aware store.") + assert ( + sink.store.context_aware + ), "TriXParser must be given a context aware store." self._parser = create_parser(sink.store) content_handler = self._parser.getContentHandler() diff --git a/rdflib/plugins/serializers/n3.py b/rdflib/plugins/serializers/n3.py index c5efc735..6c4e2ec4 100644 --- a/rdflib/plugins/serializers/n3.py +++ b/rdflib/plugins/serializers/n3.py @@ -3,10 +3,9 @@ Notation 3 (N3) RDF graph serializer for RDFLib. """ from rdflib.graph import Graph from rdflib.namespace import Namespace, OWL -from rdflib.plugins.serializers.turtle import ( - TurtleSerializer, SUBJECT, OBJECT) +from rdflib.plugins.serializers.turtle import TurtleSerializer, SUBJECT, OBJECT -__all__ = ['N3Serializer'] +__all__ = ["N3Serializer"] SWAP_LOG = Namespace("http://www.w3.org/2000/10/swap/log#") @@ -17,10 +16,7 @@ class N3Serializer(TurtleSerializer): def __init__(self, store, parent=None): super(N3Serializer, self).__init__(store) - self.keywords.update({ - OWL.sameAs: '=', - SWAP_LOG.implies: '=>' - }) + self.keywords.update({OWL.sameAs: "=", SWAP_LOG.implies: "=>"}) self.parent = parent def reset(self): @@ -33,8 +29,9 @@ class N3Serializer(TurtleSerializer): self.parent.subjectDone(subject) def isDone(self, subject): - return (super(N3Serializer, self).isDone(subject) - and (not self.parent or self.parent.isDone(subject))) + return super(N3Serializer, self).isDone(subject) and ( + not self.parent or self.parent.isDone(subject) + ) def startDocument(self): super(N3Serializer, self).startDocument() @@ -88,8 +85,7 @@ class N3Serializer(TurtleSerializer): properties = self.buildPredicateHash(subject) if len(properties) == 0: return False - return (self.s_clause(subject) - or super(N3Serializer, self).statement(subject)) + return self.s_clause(subject) or super(N3Serializer, self).statement(subject) def path(self, node, position, newline=False): if not self.p_clause(node, position): @@ -97,10 +93,10 @@ class N3Serializer(TurtleSerializer): def s_clause(self, subject): if isinstance(subject, Graph): - self.write('\n' + self.indent()) + self.write("\n" + self.indent()) self.p_clause(subject, SUBJECT) self.predicateList(subject) - self.write(' .') + self.write(" .") return True else: return False @@ -109,13 +105,13 @@ class N3Serializer(TurtleSerializer): if isinstance(node, Graph): self.subjectDone(node) if position is OBJECT: - self.write(' ') - self.write('{') + self.write(" ") + self.write("{") self.depth += 1 serializer = N3Serializer(node, parent=self) serializer.serialize(self.stream) self.depth -= 1 - self.write(self.indent() + '}') + self.write(self.indent() + "}") return True else: return False diff --git a/rdflib/plugins/serializers/nquads.py b/rdflib/plugins/serializers/nquads.py index a193e125..70c414cd 100644 --- a/rdflib/plugins/serializers/nquads.py +++ b/rdflib/plugins/serializers/nquads.py @@ -5,16 +5,15 @@ from rdflib.serializer import Serializer from rdflib.plugins.serializers.nt import _quoteLiteral -__all__ = ['NQuadsSerializer'] +__all__ = ["NQuadsSerializer"] class NQuadsSerializer(Serializer): - def __init__(self, store): if not store.context_aware: raise Exception( - "NQuads serialization only makes " - "sense for context-aware stores!") + "NQuads serialization only makes " "sense for context-aware stores!" + ) super(NQuadsSerializer, self).__init__(store) @@ -26,19 +25,24 @@ class NQuadsSerializer(Serializer): encoding = self.encoding for context in self.store.contexts(): for triple in context: - stream.write(_nq_row( - triple, context.identifier).encode(encoding, "replace")) + stream.write( + _nq_row(triple, context.identifier).encode(encoding, "replace") + ) stream.write("\n".encode("latin-1")) def _nq_row(triple, context): if isinstance(triple[2], Literal): - return u"%s %s %s %s .\n" % (triple[0].n3(), - triple[1].n3(), - _quoteLiteral(triple[2]), - context.n3()) + return u"%s %s %s %s .\n" % ( + triple[0].n3(), + triple[1].n3(), + _quoteLiteral(triple[2]), + context.n3(), + ) else: - return u"%s %s %s %s .\n" % (triple[0].n3(), - triple[1].n3(), - triple[2].n3(), - context.n3()) + return u"%s %s %s %s .\n" % ( + triple[0].n3(), + triple[1].n3(), + triple[2].n3(), + context.n3(), + ) diff --git a/rdflib/plugins/serializers/nt.py b/rdflib/plugins/serializers/nt.py index 95a88ae3..94632155 100644 --- a/rdflib/plugins/serializers/nt.py +++ b/rdflib/plugins/serializers/nt.py @@ -9,7 +9,7 @@ from rdflib.serializer import Serializer import warnings import codecs -__all__ = ['NTSerializer'] +__all__ = ["NTSerializer"] class NTSerializer(Serializer): @@ -19,7 +19,7 @@ class NTSerializer(Serializer): def __init__(self, store): Serializer.__init__(self, store) - self.encoding = 'ascii' # n-triples are ascii encoded + self.encoding = "ascii" # n-triples are ascii encoded def serialize(self, stream, base=None, encoding=None, **args): if base is not None: @@ -48,35 +48,33 @@ def _nt_row(triple): return u"%s %s %s .\n" % ( triple[0].n3(), triple[1].n3(), - _quoteLiteral(triple[2])) + _quoteLiteral(triple[2]), + ) else: - return u"%s %s %s .\n" % (triple[0].n3(), - triple[1].n3(), - triple[2].n3()) + return u"%s %s %s .\n" % (triple[0].n3(), triple[1].n3(), triple[2].n3()) def _quoteLiteral(l): - ''' + """ a simpler version of term.Literal.n3() - ''' + """ encoded = _quote_encode(l) if l.language: if l.datatype: raise Exception("Literal has datatype AND language!") - return '%s@%s' % (encoded, l.language) + return "%s@%s" % (encoded, l.language) elif l.datatype: - return '%s^^<%s>' % (encoded, l.datatype) + return "%s^^<%s>" % (encoded, l.datatype) else: - return '%s' % encoded + return "%s" % encoded def _quote_encode(l): - return '"%s"' % l.replace('\\', '\\\\')\ - .replace('\n', '\\n')\ - .replace('"', '\\"')\ - .replace('\r', '\\r') + return '"%s"' % l.replace("\\", "\\\\").replace("\n", "\\n").replace( + '"', '\\"' + ).replace("\r", "\\r") def _nt_unicode_error_resolver(err): @@ -86,11 +84,11 @@ def _nt_unicode_error_resolver(err): def _replace_single(c): c = ord(c) - fmt = u'\\u%04X' if c <= 0xFFFF else u'\\U%08X' + fmt = u"\\u%04X" if c <= 0xFFFF else u"\\U%08X" return fmt % c - string = err.object[err.start:err.end] + string = err.object[err.start : err.end] return ("".join(_replace_single(c) for c in string), err.end) -codecs.register_error('_rdflib_nt_escape', _nt_unicode_error_resolver) +codecs.register_error("_rdflib_nt_escape", _nt_unicode_error_resolver) diff --git a/rdflib/plugins/serializers/rdfxml.py b/rdflib/plugins/serializers/rdfxml.py index d3a705d2..425c0a7e 100644 --- a/rdflib/plugins/serializers/rdfxml.py +++ b/rdflib/plugins/serializers/rdfxml.py @@ -14,11 +14,10 @@ import xml.dom.minidom from .xmlwriter import ESCAPE_ENTITIES -__all__ = ['fix', 'XMLSerializer', 'PrettyXMLSerializer'] +__all__ = ["fix", "XMLSerializer", "PrettyXMLSerializer"] class XMLSerializer(Serializer): - def __init__(self, store): super(XMLSerializer, self).__init__(store) @@ -50,18 +49,17 @@ class XMLSerializer(Serializer): self.__stream = stream self.__serialized = {} encoding = self.encoding - self.write = write = lambda uni: stream.write( - uni.encode(encoding, 'replace')) + self.write = write = lambda uni: stream.write(uni.encode(encoding, "replace")) # startDocument write('<?xml version="1.0" encoding="%s"?>\n' % self.encoding) # startRDF - write('<rdf:RDF\n') + write("<rdf:RDF\n") # If provided, write xml:base attribute for the RDF if "xml_base" in args: - write(' xml:base="%s"\n' % args['xml_base']) + write(' xml:base="%s"\n' % args["xml_base"]) elif self.base: write(' xml:base="%s"\n' % self.base) # TODO: @@ -75,7 +73,7 @@ class XMLSerializer(Serializer): write(' xmlns:%s="%s"\n' % (prefix, namespace)) else: write(' xmlns="%s"\n' % namespace) - write('>\n') + write(">\n") # write out triples by subject for subject in self.store.subjects(): @@ -98,8 +96,7 @@ class XMLSerializer(Serializer): element_name = "rdf:Description" if isinstance(subject, BNode): - write('%s<%s rdf:nodeID="%s"' % ( - indent, element_name, subject)) + write('%s<%s rdf:nodeID="%s"' % (indent, element_name, subject)) else: uri = quoteattr(self.relativize(subject)) write("%s<%s rdf:about=%s" % (indent, element_name, uri)) @@ -107,8 +104,7 @@ class XMLSerializer(Serializer): if (subject, None, None) in self.store: write(">\n") - for predicate, object in self.store.predicate_objects( - subject): + for predicate, object in self.store.predicate_objects(subject): self.predicate(predicate, object, depth + 1) write("%s</%s>\n" % (indent, element_name)) @@ -129,22 +125,24 @@ class XMLSerializer(Serializer): if object.datatype: attributes += ' rdf:datatype="%s"' % object.datatype - write("%s<%s%s>%s</%s>\n" % - (indent, qname, attributes, - escape(object, ESCAPE_ENTITIES), qname)) + write( + "%s<%s%s>%s</%s>\n" + % (indent, qname, attributes, escape(object, ESCAPE_ENTITIES), qname) + ) else: if isinstance(object, BNode): - write('%s<%s rdf:nodeID="%s"/>\n' % - (indent, qname, object)) + write('%s<%s rdf:nodeID="%s"/>\n' % (indent, qname, object)) else: - write("%s<%s rdf:resource=%s/>\n" % - (indent, qname, quoteattr(self.relativize(object)))) + write( + "%s<%s rdf:resource=%s/>\n" + % (indent, qname, quoteattr(self.relativize(object))) + ) XMLLANG = "http://www.w3.org/XML/1998/namespacelang" XMLBASE = "http://www.w3.org/XML/1998/namespacebase" -OWL_NS = Namespace('http://www.w3.org/2002/07/owl#') +OWL_NS = Namespace("http://www.w3.org/2002/07/owl#") # TODO: @@ -157,7 +155,6 @@ def fix(val): class PrettyXMLSerializer(Serializer): - def __init__(self, store, max_depth=3): super(PrettyXMLSerializer, self).__init__(store) self.forceRDFAbout = set() @@ -177,8 +174,7 @@ class PrettyXMLSerializer(Serializer): self.writer = writer = XMLWriter(stream, nm, encoding) namespaces = {} - possible = set(store.predicates()).union( - store.objects(None, RDF.type)) + possible = set(store.predicates()).union(store.objects(None, RDF.type)) for predicate in possible: prefix, namespace, local = nm.compute_qname_strict(predicate) @@ -247,6 +243,7 @@ class PrettyXMLSerializer(Serializer): writer.push(element) if isinstance(subject, BNode): + def subj_as_obj_more_than(ceil): return True # more_than(store.triples((None, None, subject)), ceil) @@ -282,8 +279,9 @@ class PrettyXMLSerializer(Serializer): if object.language: writer.attribute(XMLLANG, object.language) - if (object.datatype == RDF.XMLLiteral - and isinstance(object.value, xml.dom.minidom.Document)): + if object.datatype == RDF.XMLLiteral and isinstance( + object.value, xml.dom.minidom.Document + ): writer.attribute(RDF.parseType, "Literal") writer.text(u"") writer.stream.write(object) @@ -302,17 +300,20 @@ class PrettyXMLSerializer(Serializer): else: if first(store.objects(object, RDF.first)): # may not have type - # RDF.List + # RDF.List self.__serialized[object] = 1 # Warn that any assertions on object other than # RDF.first and RDF.rest are ignored... including RDF.List import warnings + warnings.warn( - "Assertions on %s other than RDF.first " % repr(object) + - "and RDF.rest are ignored ... including RDF.List", - UserWarning, stacklevel=2) + "Assertions on %s other than RDF.first " % repr(object) + + "and RDF.rest are ignored ... including RDF.List", + UserWarning, + stacklevel=2, + ) writer.attribute(RDF.parseType, "Collection") col = Collection(store, object) @@ -326,9 +327,11 @@ class PrettyXMLSerializer(Serializer): if not isinstance(item, URIRef): self.__serialized[item] = 1 else: - if first(store.triples_choices( - (object, RDF.type, [OWL_NS.Class, RDFS.Class]))) \ - and isinstance(object, URIRef): + if first( + store.triples_choices( + (object, RDF.type, [OWL_NS.Class, RDFS.Class]) + ) + ) and isinstance(object, URIRef): writer.attribute(RDF.resource, self.relativize(object)) elif depth <= self.max_depth: @@ -336,9 +339,11 @@ class PrettyXMLSerializer(Serializer): elif isinstance(object, BNode): - if not object in self.__serialized \ - and (object, None, None) in store \ - and len(list(store.subjects(object=object))) == 1: + if ( + not object in self.__serialized + and (object, None, None) in store + and len(list(store.subjects(object=object))) == 1 + ): # inline blank nodes if they haven't been serialized yet # and are only referenced once (regardless of depth) self.subject(object, depth + 1) diff --git a/rdflib/plugins/serializers/trig.py b/rdflib/plugins/serializers/trig.py index 755587dc..432224e0 100644 --- a/rdflib/plugins/serializers/trig.py +++ b/rdflib/plugins/serializers/trig.py @@ -8,13 +8,13 @@ from collections import defaultdict from rdflib.plugins.serializers.turtle import TurtleSerializer, _GEN_QNAME_FOR_DT, VERB from rdflib.term import BNode, Literal -__all__ = ['TrigSerializer'] +__all__ = ["TrigSerializer"] class TrigSerializer(TurtleSerializer): short_name = "trig" - indentString = 4 * u' ' + indentString = 4 * u" " def __init__(self, store): if store.context_aware: @@ -38,14 +38,17 @@ class TrigSerializer(TurtleSerializer): for triple in context: self.preprocessTriple(triple) - self._contexts[context] = (self.orderSubjects(), self._subjects, self._references) + self._contexts[context] = ( + self.orderSubjects(), + self._subjects, + self._references, + ) def reset(self): super(TrigSerializer, self).reset() self._contexts = {} - def serialize(self, stream, base=None, encoding=None, - spacious=None, **args): + def serialize(self, stream, base=None, encoding=None, spacious=None, **args): self.reset() self.stream = stream # if base is given here, use that, if not and a base is set for the graph use that @@ -72,7 +75,7 @@ class TrigSerializer(TurtleSerializer): self._subjects = subjects if self.default_context and store.identifier == self.default_context: - self.write(self.indent() + '\n{') + self.write(self.indent() + "\n{") else: if isinstance(store.identifier, BNode): iri = store.identifier.n3() @@ -80,7 +83,7 @@ class TrigSerializer(TurtleSerializer): iri = self.getQName(store.identifier) if iri is None: iri = store.identifier.n3() - self.write(self.indent() + '\n%s {' % iri) + self.write(self.indent() + "\n%s {" % iri) self.depth += 1 for subject in ordered_subjects: @@ -89,9 +92,9 @@ class TrigSerializer(TurtleSerializer): if firstTime: firstTime = False if self.statement(subject) and not firstTime: - self.write('\n') + self.write("\n") self.depth -= 1 - self.write('}\n') + self.write("}\n") self.endDocument() stream.write("\n".encode("latin-1")) diff --git a/rdflib/plugins/serializers/trix.py b/rdflib/plugins/serializers/trix.py index e6651c70..f6115bf8 100644 --- a/rdflib/plugins/serializers/trix.py +++ b/rdflib/plugins/serializers/trix.py @@ -7,7 +7,7 @@ from rdflib.namespace import Namespace from rdflib.graph import Graph, ConjunctiveGraph -__all__ = ['TriXSerializer'] +__all__ = ["TriXSerializer"] # TODO: Move this somewhere central TRIXNS = Namespace("http://www.w3.org/2004/03/trix/trix-1/") @@ -19,7 +19,8 @@ class TriXSerializer(Serializer): super(TriXSerializer, self).__init__(store) if not store.context_aware: raise Exception( - "TriX serialization only makes sense for context-aware stores") + "TriX serialization only makes sense for context-aware stores" + ) def serialize(self, stream, base=None, encoding=None, **args): @@ -49,10 +50,11 @@ class TriXSerializer(Serializer): def _writeGraph(self, graph): self.writer.push(TRIXNS[u"graph"]) if graph.base: - self.writer.attribute("http://www.w3.org/XML/1998/namespacebase", graph.base) + self.writer.attribute( + "http://www.w3.org/XML/1998/namespacebase", graph.base + ) if isinstance(graph.identifier, URIRef): - self.writer.element( - TRIXNS[u"uri"], content=str(graph.identifier)) + self.writer.element(TRIXNS[u"uri"], content=str(graph.identifier)) for triple in graph.triples((None, None, None)): self._writeTriple(triple) @@ -62,23 +64,22 @@ class TriXSerializer(Serializer): self.writer.push(TRIXNS[u"triple"]) for component in triple: if isinstance(component, URIRef): - self.writer.element(TRIXNS[u"uri"], - content=str(component)) + self.writer.element(TRIXNS[u"uri"], content=str(component)) elif isinstance(component, BNode): - self.writer.element(TRIXNS[u"id"], - content=str(component)) + self.writer.element(TRIXNS[u"id"], content=str(component)) elif isinstance(component, Literal): if component.datatype: - self.writer.element(TRIXNS[u"typedLiteral"], - content=str(component), - attributes={TRIXNS[u"datatype"]: - str(component.datatype)}) + self.writer.element( + TRIXNS[u"typedLiteral"], + content=str(component), + attributes={TRIXNS[u"datatype"]: str(component.datatype)}, + ) elif component.language: - self.writer.element(TRIXNS[u"plainLiteral"], - content=str(component), - attributes={XMLNS[u"lang"]: - str(component.language)}) + self.writer.element( + TRIXNS[u"plainLiteral"], + content=str(component), + attributes={XMLNS[u"lang"]: str(component.language)}, + ) else: - self.writer.element(TRIXNS[u"plainLiteral"], - content=str(component)) + self.writer.element(TRIXNS[u"plainLiteral"], content=str(component)) self.writer.pop() diff --git a/rdflib/plugins/serializers/turtle.py b/rdflib/plugins/serializers/turtle.py index b89ff2d8..52693a8c 100644 --- a/rdflib/plugins/serializers/turtle.py +++ b/rdflib/plugins/serializers/turtle.py @@ -11,7 +11,7 @@ from rdflib.exceptions import Error from rdflib.serializer import Serializer from rdflib.namespace import RDF, RDFS -__all__ = ['RecursiveSerializer', 'TurtleSerializer'] +__all__ = ["RecursiveSerializer", "TurtleSerializer"] def _object_comparator(a, b): @@ -52,16 +52,20 @@ class RecursiveSerializer(Serializer): def addNamespace(self, prefix, uri): if prefix in self.namespaces and self.namespaces[prefix] != uri: - raise Exception("Trying to override namespace prefix %s => %s, but it's already bound to %s" % (prefix, uri, self.namespaces[prefix])) + raise Exception( + "Trying to override namespace prefix %s => %s, but it's already bound to %s" + % (prefix, uri, self.namespaces[prefix]) + ) self.namespaces[prefix] = uri def checkSubject(self, subject): """Check to see if the subject should be serialized yet""" - if ((self.isDone(subject)) + if ( + (self.isDone(subject)) or (subject not in self._subjects) or ((subject in self._topLevels) and (self.depth > 1)) - or (isinstance(subject, URIRef) and - (self.depth >= self.maxDepth))): + or (isinstance(subject, URIRef) and (self.depth >= self.maxDepth)) + ): return False return True @@ -83,9 +87,10 @@ class RecursiveSerializer(Serializer): seen[member] = True recursable = [ - (isinstance(subject, BNode), - self._references[subject], subject) - for subject in self._subjects if subject not in seen] + (isinstance(subject, BNode), self._references[subject], subject) + for subject in self._subjects + if subject not in seen + ] recursable.sort() subjects.extend([subject for (isbnode, refs, subject) in recursable]) @@ -111,7 +116,7 @@ class RecursiveSerializer(Serializer): self._topLevels = {} if self.roundtrip_prefixes: - if hasattr(self.roundtrip_prefixes, '__iter__'): + if hasattr(self.roundtrip_prefixes, "__iter__"): for prefix, ns in self.store.namespaces(): if prefix in self.roundtrip_prefixes: self.addNamespace(prefix, ns) @@ -163,7 +168,7 @@ class RecursiveSerializer(Serializer): def write(self, text): """Write text in given encoding.""" - self.stream.write(text.encode(self.encoding, 'replace')) + self.stream.write(text.encode(self.encoding, "replace")) SUBJECT = 0 @@ -177,14 +182,12 @@ _SPACIOUS_OUTPUT = False class TurtleSerializer(RecursiveSerializer): short_name = "turtle" - indentString = ' ' + indentString = " " def __init__(self, store): self._ns_rewrite = {} super(TurtleSerializer, self).__init__(store) - self.keywords = { - RDF.type: 'a' - } + self.keywords = {RDF.type: "a"} self.reset() self.stream = None self._spacious = _SPACIOUS_OUTPUT @@ -199,8 +202,9 @@ class TurtleSerializer(RecursiveSerializer): # so we need to keep track of ns rewrites we made so far. - if (prefix > '' and prefix[0] == '_') \ - or self.namespaces.get(prefix, namespace) != namespace: + if (prefix > "" and prefix[0] == "_") or self.namespaces.get( + prefix, namespace + ) != namespace: if prefix not in self._ns_rewrite: p = "p" + prefix @@ -219,8 +223,7 @@ class TurtleSerializer(RecursiveSerializer): self._started = False self._ns_rewrite = {} - def serialize(self, stream, base=None, encoding=None, - spacious=None, **args): + def serialize(self, stream, base=None, encoding=None, spacious=None, **args): self.reset() self.stream = stream # if base is given here, use that, if not and a base is set for the graph use that @@ -244,7 +247,7 @@ class TurtleSerializer(RecursiveSerializer): if firstTime: firstTime = False if self.statement(subject) and not firstTime: - self.write('\n') + self.write("\n") self.endDocument() stream.write("\n".encode("latin-1")) @@ -278,7 +281,7 @@ class TurtleSerializer(RecursiveSerializer): pfx = self.store.store.prefix(uri) if pfx is not None: - parts = (pfx, uri, '') + parts = (pfx, uri, "") else: # nothing worked return None @@ -291,95 +294,99 @@ class TurtleSerializer(RecursiveSerializer): prefix = self.addNamespace(prefix, namespace) - return u'%s:%s' % (prefix, local) + return u"%s:%s" % (prefix, local) def startDocument(self): self._started = True ns_list = sorted(self.namespaces.items()) if self.base: - self.write(self.indent() + '@base <%s> .\n' % self.base) + self.write(self.indent() + "@base <%s> .\n" % self.base) for prefix, uri in ns_list: - self.write(self.indent() + '@prefix %s: <%s> .\n' % (prefix, uri)) + self.write(self.indent() + "@prefix %s: <%s> .\n" % (prefix, uri)) if ns_list and self._spacious: - self.write('\n') + self.write("\n") def endDocument(self): if self._spacious: - self.write('\n') + self.write("\n") def statement(self, subject): self.subjectDone(subject) return self.s_squared(subject) or self.s_default(subject) def s_default(self, subject): - self.write('\n' + self.indent()) + self.write("\n" + self.indent()) self.path(subject, SUBJECT) self.predicateList(subject) - self.write(' .') + self.write(" .") return True def s_squared(self, subject): if (self._references[subject] > 0) or not isinstance(subject, BNode): return False - self.write('\n' + self.indent() + '[]') + self.write("\n" + self.indent() + "[]") self.predicateList(subject) - self.write(' .') + self.write(" .") return True def path(self, node, position, newline=False): - if not (self.p_squared(node, position, newline) - or self.p_default(node, position, newline)): - raise Error("Cannot serialize node '%s'" % (node, )) + if not ( + self.p_squared(node, position, newline) + or self.p_default(node, position, newline) + ): + raise Error("Cannot serialize node '%s'" % (node,)) def p_default(self, node, position, newline=False): if position != SUBJECT and not newline: - self.write(' ') + self.write(" ") self.write(self.label(node, position)) return True def label(self, node, position): if node == RDF.nil: - return '()' + return "()" if position is VERB and node in self.keywords: return self.keywords[node] if isinstance(node, Literal): return node._literal_n3( use_plain=True, - qname_callback=lambda dt: self.getQName( - dt, _GEN_QNAME_FOR_DT)) + qname_callback=lambda dt: self.getQName(dt, _GEN_QNAME_FOR_DT), + ) else: node = self.relativize(node) return self.getQName(node, position == VERB) or node.n3() def p_squared(self, node, position, newline=False): - if (not isinstance(node, BNode) - or node in self._serialized - or self._references[node] > 1 - or position == SUBJECT): + if ( + not isinstance(node, BNode) + or node in self._serialized + or self._references[node] > 1 + or position == SUBJECT + ): return False if not newline: - self.write(' ') + self.write(" ") if self.isValidList(node): # this is a list - self.write('(') + self.write("(") self.depth += 1 # 2 self.doList(node) self.depth -= 1 # 2 - self.write(' )') + self.write(" )") else: self.subjectDone(node) self.depth += 2 # self.write('[\n' + self.indent()) - self.write('[') + self.write("[") self.depth -= 1 # self.predicateList(node, newline=True) self.predicateList(node, newline=False) # self.write('\n' + self.indent() + ']') - self.write(' ]') + self.write(" ]") self.depth -= 1 return True @@ -394,8 +401,7 @@ class TurtleSerializer(RecursiveSerializer): except: return False while l: - if l != RDF.nil and len( - list(self.store.predicate_objects(l))) != 2: + if l != RDF.nil and len(list(self.store.predicate_objects(l))) != 2: return False l = self.store.value(l, RDF.rest) return True @@ -416,7 +422,7 @@ class TurtleSerializer(RecursiveSerializer): self.verb(propList[0], newline=newline) self.objectList(properties[propList[0]]) for predicate in propList[1:]: - self.write(' ;\n' + self.indent(1)) + self.write(" ;\n" + self.indent(1)) self.verb(predicate, newline=True) self.objectList(properties[predicate]) @@ -431,6 +437,6 @@ class TurtleSerializer(RecursiveSerializer): self.depth += depthmod self.path(objects[0], OBJECT) for obj in objects[1:]: - self.write(',\n' + self.indent(1)) + self.write(",\n" + self.indent(1)) self.path(obj, OBJECT, newline=True) self.depth -= depthmod diff --git a/rdflib/plugins/serializers/xmlwriter.py b/rdflib/plugins/serializers/xmlwriter.py index de720e8c..b6f0acb5 100644 --- a/rdflib/plugins/serializers/xmlwriter.py +++ b/rdflib/plugins/serializers/xmlwriter.py @@ -1,19 +1,15 @@ import codecs from xml.sax.saxutils import quoteattr, escape -__all__ = ['XMLWriter'] +__all__ = ["XMLWriter"] -ESCAPE_ENTITIES = { - '\r': ' ' -} +ESCAPE_ENTITIES = {"\r": " "} class XMLWriter(object): - def __init__(self, stream, namespace_manager, encoding=None, - decl=1, extra_ns=None): - encoding = encoding or 'utf-8' - encoder, decoder, stream_reader, stream_writer = \ - codecs.lookup(encoding) + def __init__(self, stream, namespace_manager, encoding=None, decl=1, extra_ns=None): + encoding = encoding or "utf-8" + encoder, decoder, stream_reader, stream_writer = codecs.lookup(encoding) self.stream = stream = stream_writer(stream) if decl: stream.write('<?xml version="1.0" encoding="%s"?>' % encoding) @@ -24,6 +20,7 @@ class XMLWriter(object): def __get_indent(self): return " " * len(self.element_stack) + indent = property(__get_indent) def __close_start_tag(self): @@ -103,8 +100,8 @@ class XMLWriter(object): for pre, ns in self.extra_ns.items(): if uri.startswith(ns): if pre != "": - return ":".join(pre, uri[len(ns):]) + return ":".join(pre, uri[len(ns) :]) else: - return uri[len(ns):] + return uri[len(ns) :] return self.nm.qname_strict(uri) diff --git a/rdflib/plugins/sleepycat.py b/rdflib/plugins/sleepycat.py index 2d08b45b..7729969e 100644 --- a/rdflib/plugins/sleepycat.py +++ b/rdflib/plugins/sleepycat.py @@ -8,15 +8,17 @@ from urllib.request import pathname2url def bb(u): - return u.encode('utf-8') + return u.encode("utf-8") try: from bsddb import db + has_bsddb = True except ImportError: try: from bsddb3 import db + has_bsddb = True except ImportError: has_bsddb = False @@ -36,7 +38,7 @@ if has_bsddb: logger = logging.getLogger(__name__) -__all__ = ['Sleepycat'] +__all__ = ["Sleepycat"] class Sleepycat(Store): @@ -48,8 +50,7 @@ class Sleepycat(Store): def __init__(self, configuration=None, identifier=None): if not has_bsddb: - raise ImportError( - "Unable to import bsddb/bsddb3, store is unusable.") + raise ImportError("Unable to import bsddb/bsddb3, store is unusable.") self.__open = False self.__identifier = identifier super(Sleepycat, self).__init__(configuration) @@ -58,6 +59,7 @@ class Sleepycat(Store): def __get_identifier(self): return self.__identifier + identifier = property(__get_identifier) def _init_db_environment(self, homeDir, create=True): @@ -108,11 +110,13 @@ class Sleepycat(Store): dbsetflags = 0 # create and open the DBs - self.__indicies = [None, ] * 3 - self.__indicies_info = [None, ] * 3 + self.__indicies = [None,] * 3 + self.__indicies_info = [None,] * 3 for i in range(0, 3): - index_name = to_key_func( - i)(("s".encode("latin-1"), "p".encode("latin-1"), "o".encode("latin-1")), "c".encode("latin-1")).decode() + index_name = to_key_func(i)( + ("s".encode("latin-1"), "p".encode("latin-1"), "o".encode("latin-1")), + "c".encode("latin-1"), + ).decode() index = db.DB(db_env) index.set_flags(dbsetflags) index.open(index_name, dbname, dbtype, dbopenflags, dbmode) @@ -148,13 +152,15 @@ class Sleepycat(Store): yield triple[i % 3] i += 1 yield "" + return get_prefix lookup[i] = ( self.__indicies[start], get_prefix_func(start, start + len), from_key_func(start), - results_from_key_func(start, self._from_string)) + results_from_key_func(start, self._from_string), + ) self.__lookup_dict = lookup @@ -187,6 +193,7 @@ class Sleepycat(Store): def __sync_run(self): from time import sleep, time + try: min_seconds, max_seconds = 10, 300 while self.__open: @@ -194,12 +201,11 @@ class Sleepycat(Store): t0 = t1 = time() self.__needs_sync = False while self.__open: - sleep(.1) + sleep(0.1) if self.__needs_sync: t1 = time() self.__needs_sync = False - if time() - t1 > min_seconds \ - or time() - t0 > max_seconds: + if time() - t1 > min_seconds or time() - t0 > max_seconds: self.__needs_sync = False logger.debug("sync") self.sync() @@ -254,7 +260,8 @@ class Sleepycat(Store): self.__contexts.put(bb(c), "", txn=txn) contexts_value = cspo.get( - bb("%s^%s^%s^%s^" % ("", s, p, o)), txn=txn) or "".encode("latin-1") + bb("%s^%s^%s^%s^" % ("", s, p, o)), txn=txn + ) or "".encode("latin-1") contexts = set(contexts_value.split("^".encode("latin-1"))) contexts.add(bb(c)) contexts_value = "^".encode("latin-1").join(contexts) @@ -264,12 +271,9 @@ class Sleepycat(Store): cpos.put(bb("%s^%s^%s^%s^" % (c, p, o, s)), "", txn=txn) cosp.put(bb("%s^%s^%s^%s^" % (c, o, s, p)), "", txn=txn) if not quoted: - cspo.put(bb( - "%s^%s^%s^%s^" % ("", s, p, o)), contexts_value, txn=txn) - cpos.put(bb( - "%s^%s^%s^%s^" % ("", p, o, s)), contexts_value, txn=txn) - cosp.put(bb( - "%s^%s^%s^%s^" % ("", o, s, p)), contexts_value, txn=txn) + cspo.put(bb("%s^%s^%s^%s^" % ("", s, p, o)), contexts_value, txn=txn) + cpos.put(bb("%s^%s^%s^%s^" % ("", p, o, s)), contexts_value, txn=txn) + cosp.put(bb("%s^%s^%s^%s^" % ("", o, s, p)), contexts_value, txn=txn) self.__needs_sync = True @@ -277,7 +281,11 @@ class Sleepycat(Store): s, p, o = spo cspo, cpos, cosp = self.__indicies contexts_value = cspo.get( - "^".encode("latin-1").join(["".encode("latin-1"), s, p, o, "".encode("latin-1")]), txn=txn) or "".encode("latin-1") + "^".encode("latin-1").join( + ["".encode("latin-1"), s, p, o, "".encode("latin-1")] + ), + txn=txn, + ) or "".encode("latin-1") contexts = set(contexts_value.split("^".encode("latin-1"))) contexts.discard(c) contexts_value = "^".encode("latin-1").join(contexts) @@ -286,7 +294,11 @@ class Sleepycat(Store): if not quoted: if contexts_value: for i, _to_key, _from_key in self.__indicies_info: - i.put(_to_key((s, p, o), "".encode("latin-1")), contexts_value, txn=txn) + i.put( + _to_key((s, p, o), "".encode("latin-1")), + contexts_value, + txn=txn, + ) else: for i, _to_key, _from_key in self.__indicies_info: try: @@ -304,23 +316,25 @@ class Sleepycat(Store): if context == self: context = None - if subject is not None \ - and predicate is not None \ - and object is not None \ - and context is not None: + if ( + subject is not None + and predicate is not None + and object is not None + and context is not None + ): s = _to_string(subject, txn=txn) p = _to_string(predicate, txn=txn) o = _to_string(object, txn=txn) c = _to_string(context, txn=txn) - value = self.__indicies[0].get(bb("%s^%s^%s^%s^" % - (c, s, p, o)), txn=txn) + value = self.__indicies[0].get(bb("%s^%s^%s^%s^" % (c, s, p, o)), txn=txn) if value is not None: self.__remove((bb(s), bb(p), bb(o)), bb(c), txn=txn) self.__needs_sync = True else: cspo, cpos, cosp = self.__indicies index, prefix, from_key, results_from_key = self.__lookup( - (subject, predicate, object), context, txn=txn) + (subject, predicate, object), context, txn=txn + ) cursor = index.cursor(txn=txn) try: @@ -336,7 +350,7 @@ class Sleepycat(Store): try: cursor.set_range(key) # Hack to stop 2to3 converting this to next(cursor) - current = getattr(cursor, 'next')() + current = getattr(cursor, "next")() except db.DBNotFoundError: current = None cursor.close() @@ -362,7 +376,8 @@ class Sleepycat(Store): # remove((None, None, None), c) try: self.__contexts.delete( - bb(_to_string(context, txn=txn)), txn=txn) + bb(_to_string(context, txn=txn)), txn=txn + ) except db.DBNotFoundError: pass @@ -380,7 +395,8 @@ class Sleepycat(Store): # _from_string = self._from_string ## UNUSED index, prefix, from_key, results_from_key = self.__lookup( - (subject, predicate, object), context, txn=txn) + (subject, predicate, object), context, txn=txn + ) cursor = index.cursor(txn=txn) try: @@ -394,14 +410,13 @@ class Sleepycat(Store): try: cursor.set_range(key) # Cheap hack so 2to3 doesn't convert to next(cursor) - current = getattr(cursor, 'next')() + current = getattr(cursor, "next")() except db.DBNotFoundError: current = None cursor.close() if key and key.startswith(prefix): contexts_value = index.get(key, txn=txn) - yield results_from_key( - key, subject, predicate, object, contexts_value) + yield results_from_key(key, subject, predicate, object, contexts_value) else: break @@ -425,7 +440,7 @@ class Sleepycat(Store): if key.startswith(prefix): count += 1 # Hack to stop 2to3 converting this to next(cursor) - current = getattr(cursor, 'next')() + current = getattr(cursor, "next")() else: break cursor.close() @@ -444,14 +459,14 @@ class Sleepycat(Store): prefix = prefix.encode("utf-8") ns = self.__namespace.get(prefix, None) if ns is not None: - return URIRef(ns.decode('utf-8')) + return URIRef(ns.decode("utf-8")) return None def prefix(self, namespace): namespace = namespace.encode("utf-8") prefix = self.__prefix.get(namespace, None) if prefix is not None: - return prefix.decode('utf-8') + return prefix.decode("utf-8") return None def namespaces(self): @@ -460,9 +475,9 @@ class Sleepycat(Store): current = cursor.first() while current: prefix, namespace = current - results.append((prefix.decode('utf-8'), namespace.decode('utf-8'))) + results.append((prefix.decode("utf-8"), namespace.decode("utf-8"))) # Hack to stop 2to3 converting this to next(cursor) - current = getattr(cursor, 'next')() + current = getattr(cursor, "next")() cursor.close() for prefix, namespace in results: yield prefix, URIRef(namespace) @@ -476,8 +491,7 @@ class Sleepycat(Store): s = _to_string(s) p = _to_string(p) o = _to_string(o) - contexts = self.__indicies[0].get(bb( - "%s^%s^%s^%s^" % ("", s, p, o))) + contexts = self.__indicies[0].get(bb("%s^%s^%s^%s^" % ("", s, p, o))) if contexts: for c in contexts.split("^".encode("latin-1")): if c: @@ -495,7 +509,7 @@ class Sleepycat(Store): try: cursor.set_range(key) # Hack to stop 2to3 converting this to next(cursor) - current = getattr(cursor, 'next')() + current = getattr(cursor, "next")() except db.DBNotFoundError: current = None cursor.close() @@ -544,8 +558,7 @@ class Sleepycat(Store): index, prefix_func, from_key, results_from_key = self.__lookup_dict[i] # print (subject, predicate, object), context, prefix_func, index # #DEBUG - prefix = bb( - "^".join(prefix_func((subject, predicate, object), context))) + prefix = bb("^".join(prefix_func((subject, predicate, object), context))) return index, prefix, from_key, results_from_key @@ -553,10 +566,15 @@ def to_key_func(i): def to_key(triple, context): "Takes a string; returns key" return "^".encode("latin-1").join( - (context, - triple[i % 3], - triple[(i + 1) % 3], - triple[(i + 2) % 3], "".encode("latin-1"))) # "" to tac on the trailing ^ + ( + context, + triple[i % 3], + triple[(i + 1) % 3], + triple[(i + 2) % 3], + "".encode("latin-1"), + ) + ) # "" to tac on the trailing ^ + return to_key @@ -564,11 +582,13 @@ def from_key_func(i): def from_key(key): "Takes a key; returns string" parts = key.split("^".encode("latin-1")) - return \ - parts[0], \ - parts[(3 - i + 0) % 3 + 1], \ - parts[(3 - i + 1) % 3 + 1], \ - parts[(3 - i + 2) % 3 + 1] + return ( + parts[0], + parts[(3 - i + 0) % 3 + 1], + parts[(3 - i + 1) % 3 + 1], + parts[(3 - i + 2) % 3 + 1], + ) + return from_key @@ -590,8 +610,11 @@ def results_from_key_func(i, from_string): o = from_string(parts[(3 - i + 2) % 3 + 1]) else: o = object - return (s, p, o), ( - from_string(c) for c in contexts_value.split("^".encode("latin-1")) if c) + return ( + (s, p, o), + (from_string(c) for c in contexts_value.split("^".encode("latin-1")) if c), + ) + return from_key diff --git a/rdflib/plugins/sparql/__init__.py b/rdflib/plugins/sparql/__init__.py index bc1227f2..9efbd87f 100644 --- a/rdflib/plugins/sparql/__init__.py +++ b/rdflib/plugins/sparql/__init__.py @@ -28,7 +28,7 @@ NotImplementedError if they cannot handle a certain part """ -PLUGIN_ENTRY_POINT = 'rdf.plugins.sparqleval' +PLUGIN_ENTRY_POINT = "rdf.plugins.sparqleval" from . import parser from . import operators diff --git a/rdflib/plugins/sparql/aggregates.py b/rdflib/plugins/sparql/aggregates.py index 11144778..8c70aeb1 100644 --- a/rdflib/plugins/sparql/aggregates.py +++ b/rdflib/plugins/sparql/aggregates.py @@ -39,7 +39,6 @@ class Accumulator(object): class Counter(Accumulator): - def __init__(self, aggregation): super(Counter, self).__init__(aggregation) self.value = 0 @@ -71,16 +70,14 @@ class Counter(Accumulator): def type_safe_numbers(*args): - if ( - any(isinstance(arg, float) for arg in args) and - any(isinstance(arg, Decimal) for arg in args) + if any(isinstance(arg, float) for arg in args) and any( + isinstance(arg, Decimal) for arg in args ): return map(float, args) return args class Sum(Accumulator): - def __init__(self, aggregation): super(Sum, self).__init__(aggregation) self.value = 0 @@ -107,7 +104,6 @@ class Sum(Accumulator): class Average(Accumulator): - def __init__(self, aggregation): super(Average, self).__init__(aggregation) self.counter = 0 @@ -171,13 +167,11 @@ class Extremum(Accumulator): class Minimum(Extremum): - def compare(self, val1, val2): return min(val1, val2, key=_val) class Maximum(Extremum): - def compare(self, val1, val2): return max(val1, val2, key=_val) @@ -205,7 +199,6 @@ class Sample(Accumulator): class GroupConcat(Accumulator): - def __init__(self, aggregation): super(GroupConcat, self).__init__(aggregation) # only GROUPCONCAT needs to have a list as accumlator diff --git a/rdflib/plugins/sparql/algebra.py b/rdflib/plugins/sparql/algebra.py index 00a6d0b2..f84e51c9 100644 --- a/rdflib/plugins/sparql/algebra.py +++ b/rdflib/plugins/sparql/algebra.py @@ -1,4 +1,3 @@ - """ Converting the 'parse-tree' output of pyparsing to a SPARQL Algebra expression @@ -19,9 +18,11 @@ from rdflib import Literal, Variable, URIRef, BNode from rdflib.plugins.sparql.sparql import Prologue, Query from rdflib.plugins.sparql.parserutils import CompValue, Expr from rdflib.plugins.sparql.operators import ( - and_, TrueFilter, simplify as simplifyFilters) -from rdflib.paths import ( - InvPath, AlternativePath, SequencePath, MulPath, NegatedPath) + and_, + TrueFilter, + simplify as simplifyFilters, +) +from rdflib.paths import InvPath, AlternativePath, SequencePath, MulPath, NegatedPath from pyparsing import ParseResults @@ -29,63 +30,69 @@ from pyparsing import ParseResults # --------------------------- # Some convenience methods def OrderBy(p, expr): - return CompValue('OrderBy', p=p, expr=expr) + return CompValue("OrderBy", p=p, expr=expr) def ToMultiSet(p): - return CompValue('ToMultiSet', p=p) + return CompValue("ToMultiSet", p=p) def Union(p1, p2): - return CompValue('Union', p1=p1, p2=p2) + return CompValue("Union", p1=p1, p2=p2) def Join(p1, p2): - return CompValue('Join', p1=p1, p2=p2) + return CompValue("Join", p1=p1, p2=p2) def Minus(p1, p2): - return CompValue('Minus', p1=p1, p2=p2) + return CompValue("Minus", p1=p1, p2=p2) def Graph(term, graph): - return CompValue('Graph', term=term, p=graph) + return CompValue("Graph", term=term, p=graph) def BGP(triples=None): - return CompValue('BGP', triples=triples or []) + return CompValue("BGP", triples=triples or []) def LeftJoin(p1, p2, expr): - return CompValue('LeftJoin', p1=p1, p2=p2, expr=expr) + return CompValue("LeftJoin", p1=p1, p2=p2, expr=expr) def Filter(expr, p): - return CompValue('Filter', expr=expr, p=p) + return CompValue("Filter", expr=expr, p=p) def Extend(p, expr, var): - return CompValue('Extend', p=p, expr=expr, var=var) + return CompValue("Extend", p=p, expr=expr, var=var) def Values(res): - return CompValue('values', res=res) + return CompValue("values", res=res) def Project(p, PV): - return CompValue('Project', p=p, PV=PV) + return CompValue("Project", p=p, PV=PV) def Group(p, expr=None): - return CompValue('Group', p=p, expr=expr) + return CompValue("Group", p=p, expr=expr) def _knownTerms(triple, varsknown, varscount): - return (len([x for x in triple if x not in varsknown and - isinstance(x, (Variable, BNode))]), - -sum(varscount.get(x, 0) for x in triple), - not isinstance(triple[2], Literal), - ) + return ( + len( + [ + x + for x in triple + if x not in varsknown and isinstance(x, (Variable, BNode)) + ] + ), + -sum(varscount.get(x, 0) for x in triple), + not isinstance(triple[2], Literal), + ) def reorderTriples(l): @@ -115,8 +122,7 @@ def reorderTriples(l): # we sort by decorate/undecorate, since we need the value of the sort keys while i < len(l): - l[i:] = sorted((_knownTerms(x[ - 1], varsknown, varscount), x[1]) for x in l[i:]) + l[i:] = sorted((_knownTerms(x[1], varsknown, varscount), x[1]) for x in l[i:]) t = l[i][0][0] # top block has this many terms bound j = 0 while i + j < len(l) and l[i + j][0][0] == t: @@ -132,9 +138,8 @@ def triples(l): l = reduce(lambda x, y: x + y, l) if (len(l) % 3) != 0: - raise Exception('these aint triples') - return reorderTriples((l[x], l[x + 1], l[x + 2]) - for x in range(0, len(l), 3)) + raise Exception("these aint triples") + return reorderTriples((l[x], l[x + 1], l[x + 2]) for x in range(0, len(l), 3)) def translatePName(p, prologue): @@ -142,11 +147,12 @@ def translatePName(p, prologue): Expand prefixed/relative URIs """ if isinstance(p, CompValue): - if p.name == 'pname': + if p.name == "pname": return prologue.absolutize(p) - if p.name == 'literal': - return Literal(p.string, lang=p.lang, - datatype=prologue.absolutize(p.datatype)) + if p.name == "literal": + return Literal( + p.string, lang=p.lang, datatype=prologue.absolutize(p.datatype) + ) elif isinstance(p, URIRef): return prologue.absolutize(p) @@ -157,39 +163,39 @@ def translatePath(p): """ if isinstance(p, CompValue): - if p.name == 'PathAlternative': + if p.name == "PathAlternative": if len(p.part) == 1: return p.part[0] else: return AlternativePath(*p.part) - elif p.name == 'PathSequence': + elif p.name == "PathSequence": if len(p.part) == 1: return p.part[0] else: return SequencePath(*p.part) - elif p.name == 'PathElt': + elif p.name == "PathElt": if not p.mod: return p.part else: if isinstance(p.part, list): if len(p.part) != 1: - raise Exception('Denkfehler!') + raise Exception("Denkfehler!") return MulPath(p.part[0], p.mod) else: return MulPath(p.part, p.mod) - elif p.name == 'PathEltOrInverse': + elif p.name == "PathEltOrInverse": if isinstance(p.part, list): if len(p.part) != 1: - raise Exception('Denkfehler!') + raise Exception("Denkfehler!") return InvPath(p.part[0]) else: return InvPath(p.part) - elif p.name == 'PathNegatedPropertySet': + elif p.name == "PathNegatedPropertySet": if isinstance(p.part, list): return NegatedPath(AlternativePath(*p.part)) else: @@ -204,9 +210,9 @@ def translateExists(e): def _c(n): if isinstance(n, CompValue): - if n.name in ('Builtin_EXISTS', 'Builtin_NOTEXISTS'): + if n.name in ("Builtin_EXISTS", "Builtin_NOTEXISTS"): n.graph = translateGroupGraphPattern(n.graph) - if n.graph.name == 'Filter': + if n.graph.name == "Filter": # filters inside (NOT) EXISTS can see vars bound outside n.graph.no_isolated_scope = True @@ -229,7 +235,7 @@ def collectAndRemoveFilters(parts): i = 0 while i < len(parts): p = parts[i] - if p.name == 'Filter': + if p.name == "Filter": filters.append(translateExists(p.expr)) parts.pop(i) else: @@ -254,8 +260,7 @@ def translateGroupOrUnionGraphPattern(graphPattern): def translateGraphGraphPattern(graphPattern): - return Graph(graphPattern.term, - translateGroupGraphPattern(graphPattern.graph)) + return Graph(graphPattern.term, translateGroupGraphPattern(graphPattern.graph)) def translateInlineData(graphPattern): @@ -267,7 +272,7 @@ def translateGroupGraphPattern(graphPattern): http://www.w3.org/TR/sparql11-query/#convertGraphPattern """ - if graphPattern.name == 'SubSelect': + if graphPattern.name == "SubSelect": return ToMultiSet(translate(graphPattern)[0]) if not graphPattern.part: @@ -277,9 +282,9 @@ def translateGroupGraphPattern(graphPattern): g = [] for p in graphPattern.part: - if p.name == 'TriplesBlock': + if p.name == "TriplesBlock": # merge adjacent TripleBlocks - if not (g and g[-1].name == 'BGP'): + if not (g and g[-1].name == "BGP"): g.append(BGP()) g[-1]["triples"] += triples(p.triples) else: @@ -287,30 +292,31 @@ def translateGroupGraphPattern(graphPattern): G = BGP() for p in g: - if p.name == 'OptionalGraphPattern': + if p.name == "OptionalGraphPattern": A = translateGroupGraphPattern(p.graph) - if A.name == 'Filter': + if A.name == "Filter": G = LeftJoin(G, A.p, A.expr) else: G = LeftJoin(G, A, TrueFilter) - elif p.name == 'MinusGraphPattern': + elif p.name == "MinusGraphPattern": G = Minus(p1=G, p2=translateGroupGraphPattern(p.graph)) - elif p.name == 'GroupOrUnionGraphPattern': + elif p.name == "GroupOrUnionGraphPattern": G = Join(p1=G, p2=translateGroupOrUnionGraphPattern(p)) - elif p.name == 'GraphGraphPattern': + elif p.name == "GraphGraphPattern": G = Join(p1=G, p2=translateGraphGraphPattern(p)) - elif p.name == 'InlineData': + elif p.name == "InlineData": G = Join(p1=G, p2=translateInlineData(p)) - elif p.name == 'ServiceGraphPattern': + elif p.name == "ServiceGraphPattern": G = Join(p1=G, p2=p) - elif p.name in ('BGP', 'Extend'): + elif p.name in ("BGP", "Extend"): G = Join(p1=G, p2=p) - elif p.name == 'Bind': + elif p.name == "Bind": G = Extend(G, p.expr, p.var) else: - raise Exception('Unknown part in GroupGraphPattern: %s - %s' % - (type(p), p.name)) + raise Exception( + "Unknown part in GroupGraphPattern: %s - %s" % (type(p), p.name) + ) if filters: G = Filter(expr=filters, p=G) @@ -372,9 +378,7 @@ def _traverseAgg(e, visitor=lambda n, v: None): return visitor(e, res) -def traverse( - tree, visitPre=lambda n: None, - visitPost=lambda n: None, complete=None): +def traverse(tree, visitPre=lambda n: None, visitPost=lambda n: None, complete=None): """ Traverse tree, visit each node with visit function visit function may raise StopTraversal to stop traversal @@ -397,7 +401,7 @@ def _hasAggregate(x): """ if isinstance(x, CompValue): - if x.name.startswith('Aggregate_'): + if x.name.startswith("Aggregate_"): raise StopTraversal(True) @@ -409,9 +413,9 @@ def _aggs(e, A): # TODO: nested Aggregates? - if isinstance(e, CompValue) and e.name.startswith('Aggregate_'): + if isinstance(e, CompValue) and e.name.startswith("Aggregate_"): A.append(e) - aggvar = Variable('__agg_%d__' % len(A)) + aggvar = Variable("__agg_%d__" % len(A)) e["res"] = aggvar return aggvar @@ -426,7 +430,7 @@ def _findVars(x, res): if x.name == "Bind": res.add(x.var) return x # stop recursion and finding vars in the expr - elif x.name == 'SubSelect': + elif x.name == "SubSelect": if x.projection: res.update(v.var or v.evar for v in x.projection) return x @@ -443,13 +447,16 @@ def _addVars(x, children): x["_vars"] = set() elif x.name == "Extend": # vars only used in the expr for a bind should not be included - x["_vars"] = reduce(operator.or_, [child for child, - part in zip(children, x) if part != 'expr'], set()) + x["_vars"] = reduce( + operator.or_, + [child for child, part in zip(children, x) if part != "expr"], + set(), + ) else: x["_vars"] = set(reduce(operator.or_, children, set())) - if x.name == 'SubSelect': + if x.name == "SubSelect": if x.projection: s = set(v.var or v.evar for v in x.projection) else: @@ -470,7 +477,7 @@ def _sample(e, v=None): if isinstance(e, CompValue) and e.name.startswith("Aggregate_"): return e # do not replace vars in aggregates if isinstance(e, Variable) and v != e: - return CompValue('Aggregate_Sample', vars=e) + return CompValue("Aggregate_Sample", vars=e) def _simplifyFilters(e): @@ -505,11 +512,11 @@ def translateAggregates(q, M): if q.projection: for v in q.projection: if v.var: - rv = Variable('__agg_%d__' % (len(A) + 1)) - A.append(CompValue('Aggregate_Sample', vars=v.var, res=rv)) + rv = Variable("__agg_%d__" % (len(A) + 1)) + A.append(CompValue("Aggregate_Sample", vars=v.var, res=rv)) E.append((rv, v.var)) - return CompValue('AggregateJoin', A=A, p=M), E + return CompValue("AggregateJoin", A=A, p=M), E def translateValues(v): @@ -554,17 +561,22 @@ def translate(q): conditions = [] # convert "GROUP BY (?expr as ?var)" to an Extend for c in q.groupby.condition: - if isinstance(c, CompValue) and c.name == 'GroupAs': + if isinstance(c, CompValue) and c.name == "GroupAs": M = Extend(M, c.expr, c.var) c = c.var conditions.append(c) M = Group(p=M, expr=conditions) aggregate = True - elif traverse(q.having, _hasAggregate, complete=False) or \ - traverse(q.orderby, _hasAggregate, complete=False) or \ - any(traverse(x.expr, _hasAggregate, complete=False) - for x in q.projection or [] if x.evar): + elif ( + traverse(q.having, _hasAggregate, complete=False) + or traverse(q.orderby, _hasAggregate, complete=False) + or any( + traverse(x.expr, _hasAggregate, complete=False) + for x in q.projection or [] + if x.evar + ) + ): # if any aggregate is used, implicit group by M = Group(p=M) aggregate = True @@ -604,17 +616,22 @@ def translate(q): # ORDER BY if q.orderby: - M = OrderBy(M, [CompValue('OrderCondition', expr=c.expr, - order=c.order) for c in q.orderby.condition]) + M = OrderBy( + M, + [ + CompValue("OrderCondition", expr=c.expr, order=c.order) + for c in q.orderby.condition + ], + ) # PROJECT M = Project(M, PV) if q.modifier: - if q.modifier == 'DISTINCT': - M = CompValue('Distinct', p=M) - elif q.modifier == 'REDUCED': - M = CompValue('Reduced', p=M) + if q.modifier == "DISTINCT": + M = CompValue("Distinct", p=M) + elif q.modifier == "REDUCED": + M = CompValue("Reduced", p=M) if q.limitoffset: offset = 0 @@ -622,10 +639,11 @@ def translate(q): offset = q.limitoffset.offset.toPython() if q.limitoffset.limit is not None: - M = CompValue('Slice', p=M, start=offset, - length=q.limitoffset.limit.toPython()) + M = CompValue( + "Slice", p=M, start=offset, length=q.limitoffset.limit.toPython() + ) else: - M = CompValue('Slice', p=M, start=offset) + M = CompValue("Slice", p=M, start=offset) return M, PV @@ -633,12 +651,12 @@ def translate(q): def simplify(n): """Remove joins to empty BGPs""" if isinstance(n, CompValue): - if n.name == 'Join': - if n.p1.name == 'BGP' and len(n.p1.triples) == 0: + if n.name == "Join": + if n.p1.name == "BGP" and len(n.p1.triples) == 0: return n.p2 - if n.p2.name == 'BGP' and len(n.p2.triples) == 0: + if n.p2.name == "BGP" and len(n.p2.triples) == 0: return n.p1 - elif n.name == 'BGP': + elif n.name == "BGP": n["triples"] = reorderTriples(n.triples) return n @@ -651,10 +669,10 @@ def analyse(n, children): """ if isinstance(n, CompValue): - if n.name == 'Join': + if n.name == "Join": n["lazy"] = all(children) return False - elif n.name in ('Slice', 'Distinct'): + elif n.name in ("Slice", "Distinct"): return False else: return all(children) @@ -674,9 +692,9 @@ def translatePrologue(p, base, initNs=None, prologue=None): prologue.bind(k, v) for x in p: - if x.name == 'Base': + if x.name == "Base": prologue.base = x.iri - elif x.name == 'PrefixDecl': + elif x.name == "PrefixDecl": prologue.bind(x.prefix, prologue.absolutize(x.iri)) return prologue @@ -699,26 +717,24 @@ def translateQuads(quads): def translateUpdate1(u, prologue): - if u.name in ('Load', 'Clear', 'Drop', 'Create'): + if u.name in ("Load", "Clear", "Drop", "Create"): pass # no translation needed - elif u.name in ('Add', 'Move', 'Copy'): + elif u.name in ("Add", "Move", "Copy"): pass - elif u.name in ('InsertData', 'DeleteData', 'DeleteWhere'): + elif u.name in ("InsertData", "DeleteData", "DeleteWhere"): t, q = translateQuads(u.quads) u["quads"] = q u["triples"] = t - if u.name in ('DeleteWhere', 'DeleteData'): + if u.name in ("DeleteWhere", "DeleteData"): pass # TODO: check for bnodes in triples - elif u.name == 'Modify': + elif u.name == "Modify": if u.delete: - u.delete["triples"], u.delete[ - "quads"] = translateQuads(u.delete.quads) + u.delete["triples"], u.delete["quads"] = translateQuads(u.delete.quads) if u.insert: - u.insert["triples"], u.insert[ - "quads"] = translateQuads(u.insert.quads) + u.insert["triples"], u.insert["quads"] = translateQuads(u.insert.quads) u["where"] = translateGroupGraphPattern(u.where) else: - raise Exception('Unknown type of update operation: %s' % u) + raise Exception("Unknown type of update operation: %s" % u) u.prologue = prologue return u @@ -737,8 +753,7 @@ def translateUpdate(q, base=None, initNs=None): prologue = translatePrologue(p, base, initNs, prologue) # absolutize/resolve prefixes - u = traverse( - u, visitPost=functools.partial(translatePName, prologue=prologue)) + u = traverse(u, visitPost=functools.partial(translatePName, prologue=prologue)) u = _traverse(u, _simplifyFilters) u = traverse(u, visitPost=translatePath) @@ -761,17 +776,16 @@ def translateQuery(q, base=None, initNs=None): # absolutize/resolve prefixes q[1] = traverse( - q[1], visitPost=functools.partial(translatePName, prologue=prologue)) + q[1], visitPost=functools.partial(translatePName, prologue=prologue) + ) P, PV = translate(q[1]) datasetClause = q[1].datasetClause - if q[1].name == 'ConstructQuery': + if q[1].name == "ConstructQuery": template = triples(q[1].template) if q[1].template else None - res = CompValue(q[1].name, p=P, - template=template, - datasetClause=datasetClause) + res = CompValue(q[1].name, p=P, template=template, datasetClause=datasetClause) else: res = CompValue(q[1].name, p=P, datasetClause=datasetClause, PV=PV) @@ -792,9 +806,9 @@ def pprintAlgebra(q): if not isinstance(p, CompValue): print(p) return - print("%s(" % (p.name, )) + print("%s(" % (p.name,)) for k in p: - print("%s%s =" % (ind, k,), end=' ') + print("%s%s =" % (ind, k,), end=" ") pp(p[k], ind + " ") print("%s)" % ind) @@ -806,7 +820,7 @@ def pprintAlgebra(q): pp(x) -if __name__ == '__main__': +if __name__ == "__main__": import sys from rdflib.plugins.sparql import parser import os.path diff --git a/rdflib/plugins/sparql/datatypes.py b/rdflib/plugins/sparql/datatypes.py index 1e8475e0..5ab8c92f 100644 --- a/rdflib/plugins/sparql/datatypes.py +++ b/rdflib/plugins/sparql/datatypes.py @@ -5,17 +5,45 @@ Utility functions for supporting the XML Schema Datatypes hierarchy from rdflib import XSD XSD_DTs = set( - (XSD.integer, XSD.decimal, XSD.float, XSD.double, XSD.string, - XSD.boolean, XSD.dateTime, XSD.nonPositiveInteger, XSD.negativeInteger, - XSD.long, XSD.int, XSD.short, XSD.byte, XSD.nonNegativeInteger, - XSD.unsignedLong, XSD.unsignedInt, XSD.unsignedShort, XSD.unsignedByte, - XSD.positiveInteger, XSD.date)) + ( + XSD.integer, + XSD.decimal, + XSD.float, + XSD.double, + XSD.string, + XSD.boolean, + XSD.dateTime, + XSD.nonPositiveInteger, + XSD.negativeInteger, + XSD.long, + XSD.int, + XSD.short, + XSD.byte, + XSD.nonNegativeInteger, + XSD.unsignedLong, + XSD.unsignedInt, + XSD.unsignedShort, + XSD.unsignedByte, + XSD.positiveInteger, + XSD.date, + ) +) _sub_types = { XSD.integer: [ - XSD.nonPositiveInteger, XSD.negativeInteger, XSD.long, XSD.int, - XSD.short, XSD.byte, XSD.nonNegativeInteger, XSD.positiveInteger, - XSD.unsignedLong, XSD.unsignedInt, XSD.unsignedShort, XSD.unsignedByte], + XSD.nonPositiveInteger, + XSD.negativeInteger, + XSD.long, + XSD.int, + XSD.short, + XSD.byte, + XSD.nonNegativeInteger, + XSD.positiveInteger, + XSD.unsignedLong, + XSD.unsignedInt, + XSD.unsignedShort, + XSD.unsignedByte, + ], } _super_types = {} @@ -25,21 +53,22 @@ for superdt in XSD_DTs: # we only care about float, double, integer, decimal _typePromotionMap = { - XSD.float: {XSD.integer: XSD.float, - XSD.decimal: XSD.float, - XSD.double: XSD.double}, - - XSD.double: {XSD.integer: XSD.double, - XSD.float: XSD.double, - XSD.decimal: XSD.double}, - - XSD.decimal: {XSD.integer: XSD.decimal, - XSD.float: XSD.float, - XSD.double: XSD.double}, - - XSD.integer: {XSD.decimal: XSD.decimal, - XSD.float: XSD.float, - XSD.double: XSD.double} + XSD.float: {XSD.integer: XSD.float, XSD.decimal: XSD.float, XSD.double: XSD.double}, + XSD.double: { + XSD.integer: XSD.double, + XSD.float: XSD.double, + XSD.decimal: XSD.double, + }, + XSD.decimal: { + XSD.integer: XSD.decimal, + XSD.float: XSD.float, + XSD.double: XSD.double, + }, + XSD.integer: { + XSD.decimal: XSD.decimal, + XSD.float: XSD.float, + XSD.double: XSD.double, + }, } @@ -53,5 +82,4 @@ def type_promotion(t1, t2): try: return _typePromotionMap[t1][t2] except KeyError: - raise TypeError( - 'Operators cannot combine datatypes %s and %s' % (t1, t2)) + raise TypeError("Operators cannot combine datatypes %s and %s" % (t1, t2)) diff --git a/rdflib/plugins/sparql/evaluate.py b/rdflib/plugins/sparql/evaluate.py index fc9b0c30..43b3d0b0 100644 --- a/rdflib/plugins/sparql/evaluate.py +++ b/rdflib/plugins/sparql/evaluate.py @@ -24,9 +24,22 @@ from rdflib import Variable, Graph, BNode, URIRef, Literal from rdflib.plugins.sparql import CUSTOM_EVALS from rdflib.plugins.sparql.parserutils import value from rdflib.plugins.sparql.sparql import ( - QueryContext, AlreadyBound, FrozenBindings, Bindings, SPARQLError) + QueryContext, + AlreadyBound, + FrozenBindings, + Bindings, + SPARQLError, +) from rdflib.plugins.sparql.evalutils import ( - _filter, _eval, _join, _diff, _minus, _fillTemplate, _ebv, _val) + _filter, + _eval, + _join, + _diff, + _minus, + _fillTemplate, + _ebv, + _val, +) from rdflib.plugins.sparql.aggregates import Aggregator from rdflib.plugins.sparql.algebra import Join, ToMultiSet, Values @@ -143,9 +156,10 @@ def evalLeftJoin(ctx, join): # check that we would have had no OPTIONAL matches # even without prior bindings... p1_vars = join.p1._vars - if p1_vars is None \ - or not any(_ebv(join.expr, b) for b in - evalPart(ctx.thaw(a.remember(p1_vars)), join.p2)): + if p1_vars is None or not any( + _ebv(join.expr, b) + for b in evalPart(ctx.thaw(a.remember(p1_vars)), join.p2) + ): yield a @@ -153,7 +167,10 @@ def evalLeftJoin(ctx, join): def evalFilter(ctx, part): # TODO: Deal with dict returned from evalPart! for c in evalPart(ctx, part.p): - if _ebv(part.expr, c.forget(ctx, _except=part._vars) if not part.no_isolated_scope else c): + if _ebv( + part.expr, + c.forget(ctx, _except=part._vars) if not part.no_isolated_scope else c, + ): yield c @@ -161,8 +178,9 @@ def evalGraph(ctx, part): if ctx.dataset is None: raise Exception( - "Non-conjunctive-graph doesn't know about " + - "graphs. Try a query without GRAPH.") + "Non-conjunctive-graph doesn't know about " + + "graphs. Try a query without GRAPH." + ) ctx = ctx.clone() graph = ctx[part.term] @@ -191,7 +209,7 @@ def evalValues(ctx, part): c = ctx.push() try: for k, v in r.items(): - if v != 'UNDEF': + if v != "UNDEF": c[k] = v except AlreadyBound: continue @@ -201,7 +219,7 @@ def evalValues(ctx, part): def evalMultiset(ctx, part): - if part.p.name == 'values': + if part.p.name == "values": return evalValues(ctx, part) return evalPart(ctx, part.p) @@ -216,91 +234,102 @@ def evalPart(ctx, part): except NotImplementedError: pass # the given custome-function did not handle this part - if part.name == 'BGP': + if part.name == "BGP": # Reorder triples patterns by number of bound nodes in the current ctx # Do patterns with more bound nodes first - triples = sorted(part.triples, key=lambda t: len([n for n in t if ctx[n] is None])) + triples = sorted( + part.triples, key=lambda t: len([n for n in t if ctx[n] is None]) + ) return evalBGP(ctx, triples) - elif part.name == 'Filter': + elif part.name == "Filter": return evalFilter(ctx, part) - elif part.name == 'Join': + elif part.name == "Join": return evalJoin(ctx, part) - elif part.name == 'LeftJoin': + elif part.name == "LeftJoin": return evalLeftJoin(ctx, part) - elif part.name == 'Graph': + elif part.name == "Graph": return evalGraph(ctx, part) - elif part.name == 'Union': + elif part.name == "Union": return evalUnion(ctx, part) - elif part.name == 'ToMultiSet': + elif part.name == "ToMultiSet": return evalMultiset(ctx, part) - elif part.name == 'Extend': + elif part.name == "Extend": return evalExtend(ctx, part) - elif part.name == 'Minus': + elif part.name == "Minus": return evalMinus(ctx, part) - elif part.name == 'Project': + elif part.name == "Project": return evalProject(ctx, part) - elif part.name == 'Slice': + elif part.name == "Slice": return evalSlice(ctx, part) - elif part.name == 'Distinct': + elif part.name == "Distinct": return evalDistinct(ctx, part) - elif part.name == 'Reduced': + elif part.name == "Reduced": return evalReduced(ctx, part) - elif part.name == 'OrderBy': + elif part.name == "OrderBy": return evalOrderBy(ctx, part) - elif part.name == 'Group': + elif part.name == "Group": return evalGroup(ctx, part) - elif part.name == 'AggregateJoin': + elif part.name == "AggregateJoin": return evalAggregateJoin(ctx, part) - elif part.name == 'SelectQuery': + elif part.name == "SelectQuery": return evalSelectQuery(ctx, part) - elif part.name == 'AskQuery': + elif part.name == "AskQuery": return evalAskQuery(ctx, part) - elif part.name == 'ConstructQuery': + elif part.name == "ConstructQuery": return evalConstructQuery(ctx, part) - elif part.name == 'ServiceGraphPattern': + elif part.name == "ServiceGraphPattern": return evalServiceQuery(ctx, part) - #raise Exception('ServiceGraphPattern not implemented') + # raise Exception('ServiceGraphPattern not implemented') - elif part.name == 'DescribeQuery': - raise Exception('DESCRIBE not implemented') + elif part.name == "DescribeQuery": + raise Exception("DESCRIBE not implemented") else: - raise Exception('I dont know: %s' % part.name) + raise Exception("I dont know: %s" % part.name) + def evalServiceQuery(ctx, part): res = {} - match = re.match('^service <(.*)>[ \n]*{(.*)}[ \n]*$', - part.get('service_string', ''), re.DOTALL | re.I) + match = re.match( + "^service <(.*)>[ \n]*{(.*)}[ \n]*$", + part.get("service_string", ""), + re.DOTALL | re.I, + ) if match: service_url = match.group(1) service_query = _buildQueryStringForServiceCall(ctx, match) - query_settings = {'query': service_query, - 'output': 'json'} - headers = {'accept' : 'application/sparql-results+json', - 'user-agent': 'rdflibForAnUser'} + query_settings = {"query": service_query, "output": "json"} + headers = { + "accept": "application/sparql-results+json", + "user-agent": "rdflibForAnUser", + } # GET is easier to cache so prefer that if the query is not to long if len(service_query) < 600: response = requests.get(service_url, params=query_settings, headers=headers) else: - response = requests.post(service_url, params=query_settings, headers=headers) + response = requests.post( + service_url, params=query_settings, headers=headers + ) if response.status_code == 200: - json = response.json(); - variables = res["vars_"] = json['head']['vars'] + json = response.json() + variables = res["vars_"] = json["head"]["vars"] # or just return the bindings? - res = json['results']['bindings'] + res = json["results"]["bindings"] if len(res) > 0: for r in res: for bound in _yieldBindingsFromServiceCallResult(ctx, r, variables): yield bound else: - raise Exception("Service: %s responded with code: %s", service_url, response.status_code); + raise Exception( + "Service: %s responded with code: %s", service_url, response.status_code + ) """ @@ -309,6 +338,8 @@ def evalServiceQuery(ctx, part): Re-adds prefixes if added and sets the base. Wraps it in select if needed. """ + + def _buildQueryStringForServiceCall(ctx, match): service_query = match.group(2) @@ -316,18 +347,20 @@ def _buildQueryStringForServiceCall(ctx, match): parser.parseQuery(service_query) except ParseException: # This could be because we don't have a select around the service call. - service_query = 'SELECT REDUCED * WHERE {' + service_query + '}' + service_query = "SELECT REDUCED * WHERE {" + service_query + "}" for p in ctx.prologue.namespace_manager.store.namespaces(): - service_query = 'PREFIX ' + p[0] + ':' + p[1].n3() + ' ' + service_query + service_query = "PREFIX " + p[0] + ":" + p[1].n3() + " " + service_query # re add the base if one was defined base = ctx.prologue.base if base is not None and len(base) > 0: - service_query = 'BASE <' + base + '> ' + service_query - sol = ctx.solution(); + service_query = "BASE <" + base + "> " + service_query + sol = ctx.solution() if len(sol) > 0: - variables = ' '.join(map(lambda v:v.n3(), sol)) - variables_bound = ' '.join(map(lambda v: ctx.get(v).n3(), sol)) - service_query = service_query + 'VALUES (' + variables + ') {(' + variables_bound + ')}' + variables = " ".join(map(lambda v: v.n3(), sol)) + variables_bound = " ".join(map(lambda v: ctx.get(v).n3(), sol)) + service_query = ( + service_query + "VALUES (" + variables + ") {(" + variables_bound + ")}" + ) return service_query @@ -335,14 +368,18 @@ def _yieldBindingsFromServiceCallResult(ctx, r, variables): res_dict = {} for var in variables: if var in r and r[var]: - if r[var]['type'] == "uri": + if r[var]["type"] == "uri": res_dict[Variable(var)] = URIRef(r[var]["value"]) - elif r[var]['type'] == "bnode": + elif r[var]["type"] == "bnode": res_dict[Variable(var)] = BNode(r[var]["value"]) - elif r[var]['type'] == "literal" and 'datatype' in r[var]: - res_dict[Variable(var)] = Literal(r[var]["value"], datatype=r[var]['datatype']) - elif r[var]['type'] == "literal" and 'xml:lang' in r[var]: - res_dict[Variable(var)] = Literal(r[var]["value"], lang=r[var]['xml:lang']) + elif r[var]["type"] == "literal" and "datatype" in r[var]: + res_dict[Variable(var)] = Literal( + r[var]["value"], datatype=r[var]["datatype"] + ) + elif r[var]["type"] == "literal" and "xml:lang" in r[var]: + res_dict[Variable(var)] = Literal( + r[var]["value"], lang=r[var]["xml:lang"] + ) yield FrozenBindings(ctx, res_dict) @@ -389,8 +426,10 @@ def evalOrderBy(ctx, part): for e in reversed(part.expr): - reverse = bool(e.order and e.order == 'DESC') - res = sorted(res, key=lambda x: _val(value(x, e.expr, variables=True)), reverse=reverse) + reverse = bool(e.order and e.order == "DESC") + res = sorted( + res, key=lambda x: _val(value(x, e.expr, variables=True)), reverse=reverse + ) return res @@ -398,7 +437,11 @@ def evalOrderBy(ctx, part): def evalSlice(ctx, slice): res = evalPart(ctx, slice.p) - return itertools.islice(res, slice.start, slice.start + slice.length if slice.length is not None else None) + return itertools.islice( + res, + slice.start, + slice.start + slice.length if slice.length is not None else None, + ) def evalReduced(ctx, part): @@ -506,8 +549,9 @@ def evalQuery(graph, query, initBindings, base=None): if main.datasetClause: if ctx.dataset is None: raise Exception( - "Non-conjunctive-graph doesn't know about " + - "graphs! Try a query without FROM (NAMED).") + "Non-conjunctive-graph doesn't know about " + + "graphs! Try a query without FROM (NAMED)." + ) ctx = ctx.clone() # or push/pop? diff --git a/rdflib/plugins/sparql/evalutils.py b/rdflib/plugins/sparql/evalutils.py index 25353fe0..8bf1981d 100644 --- a/rdflib/plugins/sparql/evalutils.py +++ b/rdflib/plugins/sparql/evalutils.py @@ -49,8 +49,7 @@ def _ebv(expr, ctx): except SPARQLError: return False # filter error == False elif isinstance(expr, CompValue): - raise Exception( - "Weird - filter got a CompValue without evalfn! %r" % expr) + raise Exception("Weird - filter got a CompValue without evalfn! %r" % expr) elif isinstance(expr, Variable): try: return EBV(ctx[expr]) @@ -73,8 +72,7 @@ def _eval(expr, ctx, raise_not_bound_error=True): else: return None elif isinstance(expr, CompValue): - raise Exception( - "Weird - _eval got a CompValue without evalfn! %r" % expr) + raise Exception("Weird - _eval got a CompValue without evalfn! %r" % expr) else: raise Exception("Cannot eval thing: %s (%s)" % (expr, type(expr))) @@ -101,12 +99,11 @@ def _fillTemplate(template, solution): _o = solution.get(o) # instantiate new bnodes for each solution - _s, _p, _o = [bnodeMap[x] if isinstance( - x, BNode) else y for x, y in zip(t, (_s, _p, _o))] + _s, _p, _o = [ + bnodeMap[x] if isinstance(x, BNode) else y for x, y in zip(t, (_s, _p, _o)) + ] - if _s is not None and \ - _p is not None and \ - _o is not None: + if _s is not None and _p is not None and _o is not None: yield (_s, _p, _o) diff --git a/rdflib/plugins/sparql/operators.py b/rdflib/plugins/sparql/operators.py index 52558aa9..8f644a1f 100644 --- a/rdflib/plugins/sparql/operators.py +++ b/rdflib/plugins/sparql/operators.py @@ -49,7 +49,7 @@ def Builtin_IRI(expr, ctx): if isinstance(a, Literal): return ctx.prologue.absolutize(URIRef(a)) - raise SPARQLError('IRI function only accepts URIRefs or Literals/Strings!') + raise SPARQLError("IRI function only accepts URIRefs or Literals/Strings!") def Builtin_isBLANK(expr, ctx): @@ -85,8 +85,7 @@ def Builtin_BNODE(expr, ctx): if isinstance(a, Literal): return ctx.bnodes[a] # defaultdict does the right thing - raise SPARQLError( - 'BNode function only accepts no argument or literal/string') + raise SPARQLError("BNode function only accepts no argument or literal/string") def Builtin_ABS(expr, ctx): @@ -158,11 +157,10 @@ def Builtin_COALESCE(expr, ctx): """ http://www.w3.org/TR/sparql11-query/#func-coalesce """ - for x in expr.get('arg', variables=True): + for x in expr.get("arg", variables=True): if x is not None and not isinstance(x, (SPARQLError, Variable)): return x - raise SPARQLError( - "COALESCE got no arguments that did not evaluate to an error") + raise SPARQLError("COALESCE got no arguments that did not evaluate to an error") def Builtin_CEIL(expr, ctx): @@ -214,8 +212,7 @@ def Builtin_REGEX(expr, ctx): if flags: # Maps XPath REGEX flags (http://www.w3.org/TR/xpath-functions/#flags) # to Python's re flags - flagMap = dict( - [('i', re.IGNORECASE), ('s', re.DOTALL), ('m', re.MULTILINE)]) + flagMap = dict([("i", re.IGNORECASE), ("s", re.DOTALL), ("m", re.MULTILINE)]) cFlag = reduce(pyop.or_, [flagMap.get(f, 0) for f in flags]) return Literal(bool(re.search(str(pattern), text, cFlag))) @@ -231,7 +228,7 @@ def Builtin_REPLACE(expr, ctx): flags = expr.flags # python uses \1, xpath/sparql uses $1 - replacement = re.sub('\\$([0-9]*)', r'\\\1', replacement) + replacement = re.sub("\\$([0-9]*)", r"\\\1", replacement) def _r(m): @@ -245,7 +242,7 @@ def Builtin_REPLACE(expr, ctx): # the match object is replaced with a wrapper that # returns "" instead of None for unmatched groups - class _m(): + class _m: def __init__(self, m): self.m = m self.string = m.string @@ -259,8 +256,7 @@ def Builtin_REPLACE(expr, ctx): if flags: # Maps XPath REGEX flags (http://www.w3.org/TR/xpath-functions/#flags) # to Python's re flags - flagMap = dict( - [('i', re.IGNORECASE), ('s', re.DOTALL), ('m', re.MULTILINE)]) + flagMap = dict([("i", re.IGNORECASE), ("s", re.DOTALL), ("m", re.MULTILINE)]) cFlag = reduce(pyop.or_, [flagMap.get(f, 0) for f in flags]) # @@FIXME@@ either datatype OR lang, NOT both @@ -269,8 +265,11 @@ def Builtin_REPLACE(expr, ctx): # python versions. see comments above in _r(m). compat_r = str(replacement) if sys.version_info[:2] >= (3, 5) else _r - return Literal(re.sub(str(pattern), compat_r, text, cFlag), - datatype=text.datatype, lang=text.language) + return Literal( + re.sub(str(pattern), compat_r, text, cFlag), + datatype=text.datatype, + lang=text.language, + ) def Builtin_STRDT(expr, ctx): @@ -288,7 +287,7 @@ def Builtin_STRLANG(expr, ctx): s = string(expr.arg1) if s.language or s.datatype: - raise SPARQLError('STRLANG expects a simple literal') + raise SPARQLError("STRLANG expects a simple literal") # TODO: normalisation of lang tag to lower-case # should probably happen in literal __init__ @@ -308,8 +307,7 @@ def Builtin_CONCAT(expr, ctx): lang = set(x.language for x in expr.arg) lang = lang.pop() if len(lang) == 1 else None - return Literal("".join(string(x) - for x in expr.arg), datatype=dt, lang=lang) + return Literal("".join(string(x) for x in expr.arg), datatype=dt, lang=lang) def _compatibleStrings(a, b): @@ -317,7 +315,7 @@ def _compatibleStrings(a, b): string(b) if b.language and a.language != b.language: - raise SPARQLError('incompatible arguments to str functions') + raise SPARQLError("incompatible arguments to str functions") def Builtin_STRSTARTS(expr, ctx): @@ -373,7 +371,7 @@ def Builtin_STRAFTER(expr, ctx): if i == -1: return Literal("") else: - return Literal(a[i + len(b):], lang=a.language, datatype=a.datatype) + return Literal(a[i + len(b) :], lang=a.language, datatype=a.datatype) def Builtin_CONTAINS(expr, ctx): @@ -491,7 +489,7 @@ def Builtin_TIMEZONE(e, ctx): """ dt = datetime(e.arg) if not dt.tzinfo: - raise SPARQLError('datatime has no timezone: %r' % dt) + raise SPARQLError("datatime has no timezone: %r" % dt) delta = dt.tzinfo.utcoffset(ctx.now) @@ -508,11 +506,13 @@ def Builtin_TIMEZONE(e, ctx): m = (s - h * 60 * 60) / 60 s = s - h * 60 * 60 - m * 60 - tzdelta = "%sP%sT%s%s%s" % (neg, - "%dD" % d if d else "", - "%dH" % h if h else "", - "%dM" % m if m else "", - "%dS" % s if not d and not h and not m else "") + tzdelta = "%sP%sT%s%s%s" % ( + neg, + "%dD" % d if d else "", + "%dH" % h if h else "", + "%dM" % m if m else "", + "%dS" % s if not d and not h and not m else "", + ) return Literal(tzdelta, datatype=XSD.dayTimeDuration) @@ -549,7 +549,7 @@ def Builtin_LANG(e, ctx): def Builtin_DATATYPE(e, ctx): l = e.arg if not isinstance(l, Literal): - raise SPARQLError('Can only get datatype of literal: %r' % l) + raise SPARQLError("Can only get datatype of literal: %r" % l) if l.language: return RDF_langString if not l.datatype and not l.language: @@ -567,7 +567,7 @@ def Builtin_BOUND(e, ctx): """ http://www.w3.org/TR/sparql11-query/#func-bound """ - n = e.get('arg', variables=True) + n = e.get("arg", variables=True) return Literal(not isinstance(n, Variable)) @@ -576,7 +576,7 @@ def Builtin_EXISTS(e, ctx): # damn... from rdflib.plugins.sparql.evaluate import evalPart - exists = e.name == 'Builtin_EXISTS' + exists = e.name == "Builtin_EXISTS" ctx = ctx.ctx.thaw(ctx) # hmm for x in evalPart(ctx, e.graph): @@ -605,9 +605,11 @@ def custom_function(uri, override=False, raw=False): """ Decorator version of :func:`register_custom_function`. """ + def decorator(func): register_custom_function(uri, func, override=override, raw=raw) return func + return decorator @@ -624,7 +626,7 @@ def Function(e, ctx): pair = _CUSTOM_FUNCTIONS.get(e.iri) if pair is None: # no such function is registered - raise SPARQLError('Unknown function %r' % e.iri) + raise SPARQLError("Unknown function %r" % e.iri) func, raw = pair if raw: # function expects expression and context @@ -658,21 +660,17 @@ def default_cast(e, ctx): if isinstance(x, (URIRef, Literal)): return Literal(x, datatype=XSD.string) else: - raise SPARQLError( - "Cannot cast term %r of type %r" % (x, type(x))) + raise SPARQLError("Cannot cast term %r of type %r" % (x, type(x))) if not isinstance(x, Literal): - raise SPARQLError( - "Can only cast Literals to non-string data-types") + raise SPARQLError("Can only cast Literals to non-string data-types") if x.datatype and not x.datatype in XSD_DTs: - raise SPARQLError( - "Cannot cast literal with unknown datatype: %r" % x.datatype) + raise SPARQLError("Cannot cast literal with unknown datatype: %r" % x.datatype) if e.iri == XSD.dateTime: if x.datatype and x.datatype not in (XSD.dateTime, XSD.string): - raise SPARQLError( - "Cannot cast %r to XSD:dateTime" % x.datatype) + raise SPARQLError("Cannot cast %r to XSD:dateTime" % x.datatype) try: return Literal(isodate.parse_datetime(x), datatype=e.iri) except: @@ -742,12 +740,12 @@ def MultiplicativeExpression(e, ctx): if type(f) == float: res = float(res) - if op == '*': + if op == "*": res *= f else: res /= f except (InvalidOperation, ZeroDivisionError): - raise SPARQLError('divide by 0') + raise SPARQLError("divide by 0") return Literal(res) @@ -775,7 +773,7 @@ def AdditiveExpression(e, ctx): dt = type_promotion(dt, term.datatype) - if op == '+': + if op == "+": res += n else: res -= n @@ -794,18 +792,22 @@ def RelationalExpression(e, ctx): if other is None: return expr - ops = dict([('>', lambda x, y: x.__gt__(y)), - ('<', lambda x, y: x.__lt__(y)), - ('=', lambda x, y: x.eq(y)), - ('!=', lambda x, y: x.neq(y)), - ('>=', lambda x, y: x.__ge__(y)), - ('<=', lambda x, y: x.__le__(y)), - ('IN', pyop.contains), - ('NOT IN', lambda x, y: not pyop.contains(x, y))]) + ops = dict( + [ + (">", lambda x, y: x.__gt__(y)), + ("<", lambda x, y: x.__lt__(y)), + ("=", lambda x, y: x.eq(y)), + ("!=", lambda x, y: x.neq(y)), + (">=", lambda x, y: x.__ge__(y)), + ("<=", lambda x, y: x.__le__(y)), + ("IN", pyop.contains), + ("NOT IN", lambda x, y: not pyop.contains(x, y)), + ] + ) - if op in ('IN', 'NOT IN'): + if op in ("IN", "NOT IN"): - res = (op == 'NOT IN') + res = op == "NOT IN" error = False @@ -823,33 +825,37 @@ def RelationalExpression(e, ctx): else: raise error - if not op in ('=', '!=', 'IN', 'NOT IN'): + if not op in ("=", "!=", "IN", "NOT IN"): if not isinstance(expr, Literal): raise SPARQLError( - "Compare other than =, != of non-literals is an error: %r" % - expr) + "Compare other than =, != of non-literals is an error: %r" % expr + ) if not isinstance(other, Literal): raise SPARQLError( - "Compare other than =, != of non-literals is an error: %r" % - other) + "Compare other than =, != of non-literals is an error: %r" % other + ) else: if not isinstance(expr, Node): - raise SPARQLError('I cannot compare this non-node: %r' % expr) + raise SPARQLError("I cannot compare this non-node: %r" % expr) if not isinstance(other, Node): - raise SPARQLError('I cannot compare this non-node: %r' % other) + raise SPARQLError("I cannot compare this non-node: %r" % other) if isinstance(expr, Literal) and isinstance(other, Literal): - if expr.datatype is not None and expr.datatype not in XSD_DTs and other.datatype is not None and other.datatype not in XSD_DTs: + if ( + expr.datatype is not None + and expr.datatype not in XSD_DTs + and other.datatype is not None + and other.datatype not in XSD_DTs + ): # in SPARQL for non-XSD DT Literals we can only do =,!= - if op not in ('=', '!='): - raise SPARQLError( - 'Can only do =,!= comparisons of non-XSD Literals') + if op not in ("=", "!="): + raise SPARQLError("Can only do =,!= comparisons of non-XSD Literals") try: r = ops[op](expr, other) if r == NotImplemented: - raise SPARQLError('Error when comparing') + raise SPARQLError("Error when comparing") except TypeError as te: raise SPARQLError(*te.args) return Literal(r) @@ -897,18 +903,22 @@ def ConditionalOrExpression(e, ctx): def not_(arg): - return Expr('UnaryNot', UnaryNot, expr=arg) + return Expr("UnaryNot", UnaryNot, expr=arg) def and_(*args): if len(args) == 1: return args[0] - return Expr('ConditionalAndExpression', ConditionalAndExpression, - expr=args[0], other=list(args[1:])) + return Expr( + "ConditionalAndExpression", + ConditionalAndExpression, + expr=args[0], + other=list(args[1:]), + ) -TrueFilter = Expr('TrueFilter', lambda _1, _2: Literal(True)) +TrueFilter = Expr("TrueFilter", lambda _1, _2: Literal(True)) def simplify(expr): @@ -919,7 +929,7 @@ def simplify(expr): return list(map(simplify, expr)) if not isinstance(expr, CompValue): return expr - if expr.name.endswith('Expression'): + if expr.name.endswith("Expression"): if expr.other is None: return simplify(expr.expr) @@ -941,8 +951,7 @@ def datetime(e): if not isinstance(e, Literal): raise SPARQLError("Non-literal passed as datetime: %r" % e) if not e.datatype == XSD.dateTime: - raise SPARQLError( - "Literal with wrong datatype passed as datetime: %r" % e) + raise SPARQLError("Literal with wrong datatype passed as datetime: %r" % e) return e.toPython() @@ -954,8 +963,7 @@ def string(s): if not isinstance(s, Literal): raise SPARQLError("Non-literal passes as string: %r" % s) if s.datatype and s.datatype != XSD.string: - raise SPARQLError( - "Non-string datatype-literal passes as string: %r" % s) + raise SPARQLError("Non-string datatype-literal passes as string: %r" % s) return s @@ -970,13 +978,24 @@ def numeric(expr): if not isinstance(expr, Literal): raise SPARQLTypeError("%r is not a literal!" % expr) - if expr.datatype not in (XSD.float, XSD.double, - XSD.decimal, XSD.integer, - XSD.nonPositiveInteger, XSD.negativeInteger, - XSD.nonNegativeInteger, XSD.positiveInteger, - XSD.unsignedLong, XSD.unsignedInt, - XSD.unsignedShort, XSD.unsignedByte, - XSD.long, XSD.int, XSD.short, XSD.byte): + if expr.datatype not in ( + XSD.float, + XSD.double, + XSD.decimal, + XSD.integer, + XSD.nonPositiveInteger, + XSD.negativeInteger, + XSD.nonNegativeInteger, + XSD.positiveInteger, + XSD.unsignedLong, + XSD.unsignedInt, + XSD.unsignedShort, + XSD.unsignedByte, + XSD.long, + XSD.int, + XSD.short, + XSD.byte, + ): raise SPARQLTypeError("%r does not have a numeric datatype!" % expr) return expr.toPython() @@ -1011,14 +1030,18 @@ def EBV(rt): # Type error, see: http://www.w3.org/TR/rdf-sparql-query/#ebv raise SPARQLTypeError( "http://www.w3.org/TR/rdf-sparql-query/#ebv - ' + \ - 'Could not determine the EBV for : %r" % rt) + 'Could not determine the EBV for : %r" + % rt + ) else: return bool(pyRT) else: raise SPARQLTypeError( "http://www.w3.org/TR/rdf-sparql-query/#ebv - ' + \ - 'Only literals have Boolean values! %r" % rt) + 'Only literals have Boolean values! %r" + % rt + ) def _lang_range_check(range, lang): @@ -1038,6 +1061,7 @@ def _lang_range_check(range, lang): .. __:http://dev.w3.org/2004/PythonLib-IH/RDFClosure/RestrictedDatatype.py """ + def _match(r, l): """ Matching of a range and language item: either range is a wildcard @@ -1046,10 +1070,10 @@ def _lang_range_check(range, lang): @param l: language tag item @rtype: boolean """ - return r == '*' or r == l + return r == "*" or r == l - rangeList = range.strip().lower().split('-') - langList = lang.strip().lower().split('-') + rangeList = range.strip().lower().split("-") + langList = lang.strip().lower().split("-") if not _match(rangeList[0], langList[0]): return False if len(rangeList) > len(langList): diff --git a/rdflib/plugins/sparql/parser.py b/rdflib/plugins/sparql/parser.py index cc785f4c..e8c37a2e 100644 --- a/rdflib/plugins/sparql/parser.py +++ b/rdflib/plugins/sparql/parser.py @@ -10,10 +10,22 @@ import sys import re from pyparsing import ( - Literal, Regex, Optional, OneOrMore, ZeroOrMore, Forward, - ParseException, Suppress, Combine, restOfLine, Group, - ParseResults, delimitedList) + Literal, + Regex, + Optional, + OneOrMore, + ZeroOrMore, + Forward, + ParseException, + Suppress, + Combine, + restOfLine, + Group, + ParseResults, + delimitedList, +) from pyparsing import CaselessKeyword as Keyword # watch out :) + # from pyparsing import Keyword as CaseSensitiveKeyword from .parserutils import Comp, Param, ParamList @@ -51,9 +63,9 @@ def expandTriples(terms): print("Terms", terms) l = len(terms) for i, t in enumerate(terms): - if t == ',': + if t == ",": res.extend([res[-3], res[-2]]) - elif t == ';': + elif t == ";": if i + 1 == len(terms) or terms[i + 1] == ";" or terms[i + 1] == ".": continue # this semicolon is spurious res.append(res[0]) @@ -70,12 +82,13 @@ def expandTriples(terms): res.append(t[0]) elif isinstance(t, ParseResults): res += t.asList() - elif t != '.': + elif t != ".": res.append(t) if DEBUG: print(len(res), t) if DEBUG: import json + print(json.dumps(res, indent=2)) return res @@ -87,6 +100,7 @@ def expandTriples(terms): except: if DEBUG: import traceback + traceback.print_exc() raise @@ -139,13 +153,16 @@ def expandCollection(terms): # SPARQL Grammar from http://www.w3.org/TR/sparql11-query/#grammar # ------ TERMINALS -------------- # [139] IRIREF ::= '<' ([^<>"{}|^`\]-[#x00-#x20])* '>' -IRIREF = Combine(Suppress('<') + Regex(r'[^<>"{}|^`\\%s]*' % ''.join( - '\\x%02X' % i for i in range(33))) + Suppress('>')) +IRIREF = Combine( + Suppress("<") + + Regex(r'[^<>"{}|^`\\%s]*' % "".join("\\x%02X" % i for i in range(33))) + + Suppress(">") +) IRIREF.setParseAction(lambda x: rdflib.URIRef(x[0])) # [164] P_CHARS_BASE ::= [A-Z] | [a-z] | [#x00C0-#x00D6] | [#x00D8-#x00F6] | [#x00F8-#x02FF] | [#x0370-#x037D] | [#x037F-#x1FFF] | [#x200C-#x200D] | [#x2070-#x218F] | [#x2C00-#x2FEF] | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD] | [#x10000-#xEFFFF] -if sys.maxunicode == 0xffff: +if sys.maxunicode == 0xFFFF: # this is narrow python build (default on windows/osx) # this means that unicode code points over 0xffff are stored # as several characters, which in turn means that regex character @@ -162,52 +179,54 @@ if sys.maxunicode == 0xffff: # # in py3.3 this is fixed - PN_CHARS_BASE_re = u'A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD' + PN_CHARS_BASE_re = u"A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD" else: # wide python build - PN_CHARS_BASE_re = u'A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD\U00010000-\U000EFFFF' + PN_CHARS_BASE_re = u"A-Za-z\u00C0-\u00D6\u00D8-\u00F6\u00F8-\u02FF\u0370-\u037D\u037F-\u1FFF\u200C-\u200D\u2070-\u218F\u2C00-\u2FEF\u3001-\uD7FF\uF900-\uFDCF\uFDF0-\uFFFD\U00010000-\U000EFFFF" # [165] PN_CHARS_U ::= PN_CHARS_BASE | '_' -PN_CHARS_U_re = '_' + PN_CHARS_BASE_re +PN_CHARS_U_re = "_" + PN_CHARS_BASE_re # [167] PN_CHARS ::= PN_CHARS_U | '-' | [0-9] | #x00B7 | [#x0300-#x036F] | [#x203F-#x2040] -PN_CHARS_re = u'\\-0-9\u00B7\u0300-\u036F\u203F-\u2040' + PN_CHARS_U_re +PN_CHARS_re = u"\\-0-9\u00B7\u0300-\u036F\u203F-\u2040" + PN_CHARS_U_re # PN_CHARS = Regex(u'[%s]'%PN_CHARS_re, flags=re.U) # [168] PN_PREFIX ::= PN_CHARS_BASE ((PN_CHARS|'.')* PN_CHARS)? -PN_PREFIX = Regex(u'[%s](?:[%s\\.]*[%s])?' % (PN_CHARS_BASE_re, - PN_CHARS_re, PN_CHARS_re), flags=re.U) +PN_PREFIX = Regex( + u"[%s](?:[%s\\.]*[%s])?" % (PN_CHARS_BASE_re, PN_CHARS_re, PN_CHARS_re), flags=re.U +) # [140] PNAME_NS ::= PN_PREFIX? ':' -PNAME_NS = Optional( - Param('prefix', PN_PREFIX)) + Suppress(':').leaveWhitespace() +PNAME_NS = Optional(Param("prefix", PN_PREFIX)) + Suppress(":").leaveWhitespace() # [173] PN_LOCAL_ESC ::= '\' ( '_' | '~' | '.' | '-' | '!' | '$' | '&' | "'" | '(' | ')' | '*' | '+' | ',' | ';' | '=' | '/' | '?' | '#' | '@' | '%' ) -PN_LOCAL_ESC_re = '\\\\[_~\\.\\-!$&"\'()*+,;=/?#@%]' +PN_LOCAL_ESC_re = "\\\\[_~\\.\\-!$&\"'()*+,;=/?#@%]" # PN_LOCAL_ESC = Regex(PN_LOCAL_ESC_re) # regex'd -#PN_LOCAL_ESC.setParseAction(lambda x: x[0][1:]) +# PN_LOCAL_ESC.setParseAction(lambda x: x[0][1:]) # [172] HEX ::= [0-9] | [A-F] | [a-f] # HEX = Regex('[0-9A-Fa-f]') # not needed # [171] PERCENT ::= '%' HEX HEX -PERCENT_re = '%[0-9a-fA-F]{2}' +PERCENT_re = "%[0-9a-fA-F]{2}" # PERCENT = Regex(PERCENT_re) # regex'd -#PERCENT.setParseAction(lambda x: chr(int(x[0][1:], 16))) +# PERCENT.setParseAction(lambda x: chr(int(x[0][1:], 16))) # [170] PLX ::= PERCENT | PN_LOCAL_ESC -PLX_re = '(%s|%s)' % (PN_LOCAL_ESC_re, PERCENT_re) +PLX_re = "(%s|%s)" % (PN_LOCAL_ESC_re, PERCENT_re) # PLX = PERCENT | PN_LOCAL_ESC # regex'd # [169] PN_LOCAL ::= (PN_CHARS_U | ':' | [0-9] | PLX ) ((PN_CHARS | '.' | ':' | PLX)* (PN_CHARS | ':' | PLX) )? -PN_LOCAL = Regex(u"""([%(PN_CHARS_U)s:0-9]|%(PLX)s) +PN_LOCAL = Regex( + u"""([%(PN_CHARS_U)s:0-9]|%(PLX)s) (([%(PN_CHARS)s\\.:]|%(PLX)s)* - ([%(PN_CHARS)s:]|%(PLX)s) )?""" % dict(PN_CHARS_U=PN_CHARS_U_re, - PN_CHARS=PN_CHARS_re, - PLX=PLX_re), flags=re.X | re.UNICODE) + ([%(PN_CHARS)s:]|%(PLX)s) )?""" + % dict(PN_CHARS_U=PN_CHARS_U_re, PN_CHARS=PN_CHARS_re, PLX=PLX_re), + flags=re.X | re.UNICODE, +) def _hexExpand(match): @@ -218,71 +237,72 @@ PN_LOCAL.setParseAction(lambda x: re.sub("(%s)" % PERCENT_re, _hexExpand, x[0])) # [141] PNAME_LN ::= PNAME_NS PN_LOCAL -PNAME_LN = PNAME_NS + Param('localname', PN_LOCAL.leaveWhitespace()) +PNAME_LN = PNAME_NS + Param("localname", PN_LOCAL.leaveWhitespace()) # [142] BLANK_NODE_LABEL ::= '_:' ( PN_CHARS_U | [0-9] ) ((PN_CHARS|'.')* PN_CHARS)? -BLANK_NODE_LABEL = Regex(u'_:[0-9%s](?:[\\.%s]*[%s])?' % ( - PN_CHARS_U_re, PN_CHARS_re, PN_CHARS_re), flags=re.U) +BLANK_NODE_LABEL = Regex( + u"_:[0-9%s](?:[\\.%s]*[%s])?" % (PN_CHARS_U_re, PN_CHARS_re, PN_CHARS_re), + flags=re.U, +) BLANK_NODE_LABEL.setParseAction(lambda x: rdflib.BNode(x[0][2:])) # [166] VARNAME ::= ( PN_CHARS_U | [0-9] ) ( PN_CHARS_U | [0-9] | #x00B7 | [#x0300-#x036F] | [#x203F-#x2040] )* -VARNAME = Regex(u'[%s0-9][%s0-9\u00B7\u0300-\u036F\u203F-\u2040]*' % ( - PN_CHARS_U_re, PN_CHARS_U_re), flags=re.U) +VARNAME = Regex( + u"[%s0-9][%s0-9\u00B7\u0300-\u036F\u203F-\u2040]*" % (PN_CHARS_U_re, PN_CHARS_U_re), + flags=re.U, +) # [143] VAR1 ::= '?' VARNAME -VAR1 = Combine(Suppress('?') + VARNAME) +VAR1 = Combine(Suppress("?") + VARNAME) # [144] VAR2 ::= '$' VARNAME -VAR2 = Combine(Suppress('$') + VARNAME) +VAR2 = Combine(Suppress("$") + VARNAME) # [145] LANGTAG ::= '@' [a-zA-Z]+ ('-' [a-zA-Z0-9]+)* -LANGTAG = Combine(Suppress('@') + Regex('[a-zA-Z]+(?:-[a-zA-Z0-9]+)*')) +LANGTAG = Combine(Suppress("@") + Regex("[a-zA-Z]+(?:-[a-zA-Z0-9]+)*")) # [146] INTEGER ::= [0-9]+ INTEGER = Regex(r"[0-9]+") # INTEGER.setResultsName('integer') -INTEGER.setParseAction( - lambda x: rdflib.Literal(x[0], datatype=rdflib.XSD.integer)) +INTEGER.setParseAction(lambda x: rdflib.Literal(x[0], datatype=rdflib.XSD.integer)) # [155] EXPONENT ::= [eE] [+-]? [0-9]+ -EXPONENT_re = '[eE][+-]?[0-9]+' +EXPONENT_re = "[eE][+-]?[0-9]+" # [147] DECIMAL ::= [0-9]* '.' [0-9]+ -DECIMAL = Regex(r'[0-9]*\.[0-9]+') # (?![eE]) +DECIMAL = Regex(r"[0-9]*\.[0-9]+") # (?![eE]) # DECIMAL.setResultsName('decimal') -DECIMAL.setParseAction( - lambda x: rdflib.Literal(x[0], datatype=rdflib.XSD.decimal)) +DECIMAL.setParseAction(lambda x: rdflib.Literal(x[0], datatype=rdflib.XSD.decimal)) # [148] DOUBLE ::= [0-9]+ '.' [0-9]* EXPONENT | '.' ([0-9])+ EXPONENT | ([0-9])+ EXPONENT -DOUBLE = Regex( - r'[0-9]+\.[0-9]*%(e)s|\.([0-9])+%(e)s|[0-9]+%(e)s' % {'e': EXPONENT_re}) +DOUBLE = Regex(r"[0-9]+\.[0-9]*%(e)s|\.([0-9])+%(e)s|[0-9]+%(e)s" % {"e": EXPONENT_re}) # DOUBLE.setResultsName('double') -DOUBLE.setParseAction( - lambda x: rdflib.Literal(x[0], datatype=rdflib.XSD.double)) +DOUBLE.setParseAction(lambda x: rdflib.Literal(x[0], datatype=rdflib.XSD.double)) # [149] INTEGER_POSITIVE ::= '+' INTEGER -INTEGER_POSITIVE = Suppress('+') + INTEGER.copy().leaveWhitespace() -INTEGER_POSITIVE.setParseAction(lambda x: rdflib.Literal( - "+" + x[0], datatype=rdflib.XSD.integer)) +INTEGER_POSITIVE = Suppress("+") + INTEGER.copy().leaveWhitespace() +INTEGER_POSITIVE.setParseAction( + lambda x: rdflib.Literal("+" + x[0], datatype=rdflib.XSD.integer) +) # [150] DECIMAL_POSITIVE ::= '+' DECIMAL -DECIMAL_POSITIVE = Suppress('+') + DECIMAL.copy().leaveWhitespace() +DECIMAL_POSITIVE = Suppress("+") + DECIMAL.copy().leaveWhitespace() # [151] DOUBLE_POSITIVE ::= '+' DOUBLE -DOUBLE_POSITIVE = Suppress('+') + DOUBLE.copy().leaveWhitespace() +DOUBLE_POSITIVE = Suppress("+") + DOUBLE.copy().leaveWhitespace() # [152] INTEGER_NEGATIVE ::= '-' INTEGER -INTEGER_NEGATIVE = Suppress('-') + INTEGER.copy().leaveWhitespace() +INTEGER_NEGATIVE = Suppress("-") + INTEGER.copy().leaveWhitespace() INTEGER_NEGATIVE.setParseAction(lambda x: neg(x[0])) # [153] DECIMAL_NEGATIVE ::= '-' DECIMAL -DECIMAL_NEGATIVE = Suppress('-') + DECIMAL.copy().leaveWhitespace() +DECIMAL_NEGATIVE = Suppress("-") + DECIMAL.copy().leaveWhitespace() DECIMAL_NEGATIVE.setParseAction(lambda x: neg(x[0])) # [154] DOUBLE_NEGATIVE ::= '-' DOUBLE -DOUBLE_NEGATIVE = Suppress('-') + DOUBLE.copy().leaveWhitespace() +DOUBLE_NEGATIVE = Suppress("-") + DOUBLE.copy().leaveWhitespace() DOUBLE_NEGATIVE.setParseAction(lambda x: neg(x[0])) # [160] ECHAR ::= '\' [tbnrf\"'] @@ -294,57 +314,58 @@ DOUBLE_NEGATIVE.setParseAction(lambda x: neg(x[0])) # ) + ZeroOrMore( ~ Literal("'\\") | ECHAR ) ) + "'''" STRING_LITERAL_LONG1 = Regex(u"'''((?:'|'')?(?:[^'\\\\]|\\\\['ntbrf\\\\]))*'''") STRING_LITERAL_LONG1.setParseAction( - lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][3:-3]))) + lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][3:-3])) +) # [159] STRING_LITERAL_LONG2 ::= '"""' ( ( '"' | '""' )? ( [^"\] | ECHAR ) )* '"""' # STRING_LITERAL_LONG2 = Literal('"""') + ( Optional( Literal('"') | '""' # ) + ZeroOrMore( ~ Literal('"\\') | ECHAR ) ) + '"""' STRING_LITERAL_LONG2 = Regex(u'"""(?:(?:"|"")?(?:[^"\\\\]|\\\\["ntbrf\\\\]))*"""') STRING_LITERAL_LONG2.setParseAction( - lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][3:-3]))) + lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][3:-3])) +) # [156] STRING_LITERAL1 ::= "'" ( ([^#x27#x5C#xA#xD]) | ECHAR )* "'" # STRING_LITERAL1 = Literal("'") + ZeroOrMore( # Regex(u'[^\u0027\u005C\u000A\u000D]',flags=re.U) | ECHAR ) + "'" -STRING_LITERAL1 = Regex( - u"'(?:[^'\\n\\r\\\\]|\\\\['ntbrf\\\\])*'(?!')", flags=re.U) +STRING_LITERAL1 = Regex(u"'(?:[^'\\n\\r\\\\]|\\\\['ntbrf\\\\])*'(?!')", flags=re.U) STRING_LITERAL1.setParseAction( - lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][1:-1]))) + lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][1:-1])) +) # [157] STRING_LITERAL2 ::= '"' ( ([^#x22#x5C#xA#xD]) | ECHAR )* '"' # STRING_LITERAL2 = Literal('"') + ZeroOrMore ( # Regex(u'[^\u0022\u005C\u000A\u000D]',flags=re.U) | ECHAR ) + '"' -STRING_LITERAL2 = Regex( - u'"(?:[^"\\n\\r\\\\]|\\\\["ntbrf\\\\])*"(?!")', flags=re.U) +STRING_LITERAL2 = Regex(u'"(?:[^"\\n\\r\\\\]|\\\\["ntbrf\\\\])*"(?!")', flags=re.U) STRING_LITERAL2.setParseAction( - lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][1:-1]))) + lambda x: rdflib.Literal(decodeUnicodeEscape(x[0][1:-1])) +) # [161] NIL ::= '(' WS* ')' -NIL = Literal('(') + ')' +NIL = Literal("(") + ")" NIL.setParseAction(lambda x: rdflib.RDF.nil) # [162] WS ::= #x20 | #x9 | #xD | #xA # Not needed? # WS = #x20 | #x9 | #xD | #xA # [163] ANON ::= '[' WS* ']' -ANON = Literal('[') + ']' +ANON = Literal("[") + "]" ANON.setParseAction(lambda x: rdflib.BNode()) # A = CaseSensitiveKeyword('a') -A = Literal('a') +A = Literal("a") A.setParseAction(lambda x: rdflib.RDF.type) # ------ NON-TERMINALS -------------- # [5] BaseDecl ::= 'BASE' IRIREF -BaseDecl = Comp('Base', Keyword('BASE') + Param('iri', IRIREF)) +BaseDecl = Comp("Base", Keyword("BASE") + Param("iri", IRIREF)) # [6] PrefixDecl ::= 'PREFIX' PNAME_NS IRIREF -PrefixDecl = Comp( - 'PrefixDecl', Keyword('PREFIX') + PNAME_NS + Param('iri', IRIREF)) +PrefixDecl = Comp("PrefixDecl", Keyword("PREFIX") + PNAME_NS + Param("iri", IRIREF)) # [4] Prologue ::= ( BaseDecl | PrefixDecl )* Prologue = Group(ZeroOrMore(BaseDecl | PrefixDecl)) @@ -354,7 +375,7 @@ Var = VAR1 | VAR2 Var.setParseAction(lambda x: rdflib.term.Variable(x[0])) # [137] PrefixedName ::= PNAME_LN | PNAME_NS -PrefixedName = Comp('pname', PNAME_LN | PNAME_NS) +PrefixedName = Comp("pname", PNAME_LN | PNAME_NS) # [136] iri ::= IRIREF | PrefixedName iri = IRIREF | PrefixedName @@ -364,8 +385,14 @@ String = STRING_LITERAL_LONG1 | STRING_LITERAL_LONG2 | STRING_LITERAL1 | STRING_ # [129] RDFLiteral ::= String ( LANGTAG | ( '^^' iri ) )? -RDFLiteral = Comp('literal', Param('string', String) + Optional(Param( - 'lang', LANGTAG.leaveWhitespace()) | Literal('^^').leaveWhitespace() + Param('datatype', iri).leaveWhitespace())) +RDFLiteral = Comp( + "literal", + Param("string", String) + + Optional( + Param("lang", LANGTAG.leaveWhitespace()) + | Literal("^^").leaveWhitespace() + Param("datatype", iri).leaveWhitespace() + ), +) # [132] NumericLiteralPositive ::= INTEGER_POSITIVE | DECIMAL_POSITIVE | DOUBLE_POSITIVE NumericLiteralPositive = DOUBLE_POSITIVE | DECIMAL_POSITIVE | INTEGER_POSITIVE @@ -377,11 +404,14 @@ NumericLiteralNegative = DOUBLE_NEGATIVE | DECIMAL_NEGATIVE | INTEGER_NEGATIVE NumericLiteralUnsigned = DOUBLE | DECIMAL | INTEGER # [130] NumericLiteral ::= NumericLiteralUnsigned | NumericLiteralPositive | NumericLiteralNegative -NumericLiteral = NumericLiteralUnsigned | NumericLiteralPositive | NumericLiteralNegative +NumericLiteral = ( + NumericLiteralUnsigned | NumericLiteralPositive | NumericLiteralNegative +) # [134] BooleanLiteral ::= 'true' | 'false' -BooleanLiteral = Keyword('true').setParseAction(lambda: rdflib.Literal(True)) |\ - Keyword('false').setParseAction(lambda: rdflib.Literal(False)) +BooleanLiteral = Keyword("true").setParseAction(lambda: rdflib.Literal(True)) | Keyword( + "false" +).setParseAction(lambda: rdflib.Literal(False)) # [138] BlankNode ::= BLANK_NODE_LABEL | ANON BlankNode = BLANK_NODE_LABEL | ANON @@ -396,19 +426,23 @@ VarOrTerm = Var | GraphTerm VarOrIri = Var | iri # [46] GraphRef ::= 'GRAPH' iri -GraphRef = Keyword('GRAPH') + Param('graphiri', iri) +GraphRef = Keyword("GRAPH") + Param("graphiri", iri) # [47] GraphRefAll ::= GraphRef | 'DEFAULT' | 'NAMED' | 'ALL' -GraphRefAll = GraphRef | Param('graphiri', Keyword('DEFAULT')) | Param( - 'graphiri', Keyword('NAMED')) | Param('graphiri', Keyword('ALL')) +GraphRefAll = ( + GraphRef + | Param("graphiri", Keyword("DEFAULT")) + | Param("graphiri", Keyword("NAMED")) + | Param("graphiri", Keyword("ALL")) +) # [45] GraphOrDefault ::= 'DEFAULT' | 'GRAPH'? iri -GraphOrDefault = ParamList('graph', Keyword( - 'DEFAULT')) | Optional(Keyword('GRAPH')) + ParamList('graph', iri) +GraphOrDefault = ParamList("graph", Keyword("DEFAULT")) | Optional( + Keyword("GRAPH") +) + ParamList("graph", iri) # [65] DataBlockValue ::= iri | RDFLiteral | NumericLiteral | BooleanLiteral | 'UNDEF' -DataBlockValue = iri | RDFLiteral | NumericLiteral | BooleanLiteral | Keyword( - 'UNDEF') +DataBlockValue = iri | RDFLiteral | NumericLiteral | BooleanLiteral | Keyword("UNDEF") # [78] Verb ::= VarOrIri | A Verb = VarOrIri | A @@ -432,37 +466,58 @@ GraphNodePath = VarOrTerm | TriplesNodePath # [93] PathMod ::= '?' | '*' | '+' -PathMod = Literal('?') | '*' | '+' +PathMod = Literal("?") | "*" | "+" # [96] PathOneInPropertySet ::= iri | A | '^' ( iri | A ) -PathOneInPropertySet = iri | A | Comp('InversePath', '^' + (iri | A)) +PathOneInPropertySet = iri | A | Comp("InversePath", "^" + (iri | A)) Path = Forward() # [95] PathNegatedPropertySet ::= PathOneInPropertySet | '(' ( PathOneInPropertySet ( '|' PathOneInPropertySet )* )? ')' -PathNegatedPropertySet = Comp('PathNegatedPropertySet', ParamList('part', PathOneInPropertySet) | '(' + Optional( - ParamList('part', PathOneInPropertySet) + ZeroOrMore('|' + ParamList('part', PathOneInPropertySet))) + ')') +PathNegatedPropertySet = Comp( + "PathNegatedPropertySet", + ParamList("part", PathOneInPropertySet) + | "(" + + Optional( + ParamList("part", PathOneInPropertySet) + + ZeroOrMore("|" + ParamList("part", PathOneInPropertySet)) + ) + + ")", +) # [94] PathPrimary ::= iri | A | '!' PathNegatedPropertySet | '(' Path ')' | 'DISTINCT' '(' Path ')' -PathPrimary = iri | A | Suppress('!') + PathNegatedPropertySet | Suppress('(') + Path + Suppress( - ')') | Comp('DistinctPath', Keyword('DISTINCT') + '(' + Param('part', Path) + ')') +PathPrimary = ( + iri + | A + | Suppress("!") + PathNegatedPropertySet + | Suppress("(") + Path + Suppress(")") + | Comp("DistinctPath", Keyword("DISTINCT") + "(" + Param("part", Path) + ")") +) # [91] PathElt ::= PathPrimary Optional(PathMod) -PathElt = Comp('PathElt', Param( - 'part', PathPrimary) + Optional(Param('mod', PathMod.leaveWhitespace()))) +PathElt = Comp( + "PathElt", + Param("part", PathPrimary) + Optional(Param("mod", PathMod.leaveWhitespace())), +) # [92] PathEltOrInverse ::= PathElt | '^' PathElt -PathEltOrInverse = PathElt | Suppress( - '^') + Comp('PathEltOrInverse', Param('part', PathElt)) +PathEltOrInverse = PathElt | Suppress("^") + Comp( + "PathEltOrInverse", Param("part", PathElt) +) # [90] PathSequence ::= PathEltOrInverse ( '/' PathEltOrInverse )* -PathSequence = Comp('PathSequence', ParamList('part', PathEltOrInverse) + - ZeroOrMore('/' + ParamList('part', PathEltOrInverse))) +PathSequence = Comp( + "PathSequence", + ParamList("part", PathEltOrInverse) + + ZeroOrMore("/" + ParamList("part", PathEltOrInverse)), +) # [89] PathAlternative ::= PathSequence ( '|' PathSequence )* -PathAlternative = Comp('PathAlternative', ParamList('part', PathSequence) + - ZeroOrMore('|' + ParamList('part', PathSequence))) +PathAlternative = Comp( + "PathAlternative", + ParamList("part", PathSequence) + ZeroOrMore("|" + ParamList("part", PathSequence)), +) # [88] Path ::= PathAlternative Path <<= PathAlternative @@ -474,127 +529,172 @@ VerbPath = Path ObjectPath = GraphNodePath # [86] ObjectListPath ::= ObjectPath ( ',' ObjectPath )* -ObjectListPath = ObjectPath + ZeroOrMore(',' + ObjectPath) +ObjectListPath = ObjectPath + ZeroOrMore("," + ObjectPath) GroupGraphPattern = Forward() # [102] Collection ::= '(' OneOrMore(GraphNode) ')' -Collection = Suppress('(') + OneOrMore(GraphNode) + Suppress(')') +Collection = Suppress("(") + OneOrMore(GraphNode) + Suppress(")") Collection.setParseAction(expandCollection) # [103] CollectionPath ::= '(' OneOrMore(GraphNodePath) ')' -CollectionPath = Suppress('(') + OneOrMore(GraphNodePath) + Suppress(')') +CollectionPath = Suppress("(") + OneOrMore(GraphNodePath) + Suppress(")") CollectionPath.setParseAction(expandCollection) # [80] Object ::= GraphNode Object = GraphNode # [79] ObjectList ::= Object ( ',' Object )* -ObjectList = Object + ZeroOrMore(',' + Object) +ObjectList = Object + ZeroOrMore("," + Object) # [83] PropertyListPathNotEmpty ::= ( VerbPath | VerbSimple ) ObjectListPath ( ';' ( ( VerbPath | VerbSimple ) ObjectList )? )* -PropertyListPathNotEmpty = (VerbPath | VerbSimple) + ObjectListPath + ZeroOrMore( - ';' + Optional((VerbPath | VerbSimple) + ObjectListPath)) +PropertyListPathNotEmpty = ( + (VerbPath | VerbSimple) + + ObjectListPath + + ZeroOrMore(";" + Optional((VerbPath | VerbSimple) + ObjectListPath)) +) # [82] PropertyListPath ::= Optional(PropertyListPathNotEmpty) PropertyListPath = Optional(PropertyListPathNotEmpty) # [77] PropertyListNotEmpty ::= Verb ObjectList ( ';' ( Verb ObjectList )? )* -PropertyListNotEmpty = Verb + ObjectList + ZeroOrMore(';' + Optional(Verb + - ObjectList)) +PropertyListNotEmpty = Verb + ObjectList + ZeroOrMore(";" + Optional(Verb + ObjectList)) # [76] PropertyList ::= Optional(PropertyListNotEmpty) PropertyList = Optional(PropertyListNotEmpty) # [99] BlankNodePropertyList ::= '[' PropertyListNotEmpty ']' -BlankNodePropertyList = Group( - Suppress('[') + PropertyListNotEmpty + Suppress(']')) +BlankNodePropertyList = Group(Suppress("[") + PropertyListNotEmpty + Suppress("]")) BlankNodePropertyList.setParseAction(expandBNodeTriples) # [101] BlankNodePropertyListPath ::= '[' PropertyListPathNotEmpty ']' BlankNodePropertyListPath = Group( - Suppress('[') + PropertyListPathNotEmpty + Suppress(']')) + Suppress("[") + PropertyListPathNotEmpty + Suppress("]") +) BlankNodePropertyListPath.setParseAction(expandBNodeTriples) # [98] TriplesNode ::= Collection | BlankNodePropertyList -TriplesNode <<= (Collection | BlankNodePropertyList) +TriplesNode <<= Collection | BlankNodePropertyList # [100] TriplesNodePath ::= CollectionPath | BlankNodePropertyListPath -TriplesNodePath <<= (CollectionPath | BlankNodePropertyListPath) +TriplesNodePath <<= CollectionPath | BlankNodePropertyListPath # [75] TriplesSameSubject ::= VarOrTerm PropertyListNotEmpty | TriplesNode PropertyList -TriplesSameSubject = VarOrTerm + PropertyListNotEmpty | TriplesNode + \ - PropertyList +TriplesSameSubject = VarOrTerm + PropertyListNotEmpty | TriplesNode + PropertyList TriplesSameSubject.setParseAction(expandTriples) # [52] TriplesTemplate ::= TriplesSameSubject ( '.' Optional(TriplesTemplate) )? TriplesTemplate = Forward() -TriplesTemplate <<= (ParamList('triples', TriplesSameSubject) + Optional( - Suppress('.') + Optional(TriplesTemplate))) +TriplesTemplate <<= ParamList("triples", TriplesSameSubject) + Optional( + Suppress(".") + Optional(TriplesTemplate) +) # [51] QuadsNotTriples ::= 'GRAPH' VarOrIri '{' Optional(TriplesTemplate) '}' -QuadsNotTriples = Comp('QuadsNotTriples', Keyword('GRAPH') + Param( - 'term', VarOrIri) + '{' + Optional(TriplesTemplate) + '}') +QuadsNotTriples = Comp( + "QuadsNotTriples", + Keyword("GRAPH") + Param("term", VarOrIri) + "{" + Optional(TriplesTemplate) + "}", +) # [50] Quads ::= Optional(TriplesTemplate) ( QuadsNotTriples '.'? Optional(TriplesTemplate) )* -Quads = Comp('Quads', Optional(TriplesTemplate) + ZeroOrMore(ParamList( - 'quadsNotTriples', QuadsNotTriples) + Optional(Suppress('.')) + Optional(TriplesTemplate))) +Quads = Comp( + "Quads", + Optional(TriplesTemplate) + + ZeroOrMore( + ParamList("quadsNotTriples", QuadsNotTriples) + + Optional(Suppress(".")) + + Optional(TriplesTemplate) + ), +) # [48] QuadPattern ::= '{' Quads '}' -QuadPattern = '{' + Param('quads', Quads) + '}' +QuadPattern = "{" + Param("quads", Quads) + "}" # [49] QuadData ::= '{' Quads '}' -QuadData = '{' + Param('quads', Quads) + '}' +QuadData = "{" + Param("quads", Quads) + "}" # [81] TriplesSameSubjectPath ::= VarOrTerm PropertyListPathNotEmpty | TriplesNodePath PropertyListPath -TriplesSameSubjectPath = VarOrTerm + \ - PropertyListPathNotEmpty | TriplesNodePath + PropertyListPath +TriplesSameSubjectPath = ( + VarOrTerm + PropertyListPathNotEmpty | TriplesNodePath + PropertyListPath +) TriplesSameSubjectPath.setParseAction(expandTriples) # [55] TriplesBlock ::= TriplesSameSubjectPath ( '.' Optional(TriplesBlock) )? TriplesBlock = Forward() -TriplesBlock <<= (ParamList('triples', TriplesSameSubjectPath) + Optional( - Suppress('.') + Optional(TriplesBlock))) +TriplesBlock <<= ParamList("triples", TriplesSameSubjectPath) + Optional( + Suppress(".") + Optional(TriplesBlock) +) # [66] MinusGraphPattern ::= 'MINUS' GroupGraphPattern MinusGraphPattern = Comp( - 'MinusGraphPattern', Keyword('MINUS') + Param('graph', GroupGraphPattern)) + "MinusGraphPattern", Keyword("MINUS") + Param("graph", GroupGraphPattern) +) # [67] GroupOrUnionGraphPattern ::= GroupGraphPattern ( 'UNION' GroupGraphPattern )* -GroupOrUnionGraphPattern = Comp('GroupOrUnionGraphPattern', ParamList( - 'graph', GroupGraphPattern) + ZeroOrMore(Keyword('UNION') + ParamList('graph', GroupGraphPattern))) +GroupOrUnionGraphPattern = Comp( + "GroupOrUnionGraphPattern", + ParamList("graph", GroupGraphPattern) + + ZeroOrMore(Keyword("UNION") + ParamList("graph", GroupGraphPattern)), +) Expression = Forward() # [72] ExpressionList ::= NIL | '(' Expression ( ',' Expression )* ')' -ExpressionList = NIL | Group( - Suppress('(') + delimitedList(Expression) + Suppress(')')) +ExpressionList = NIL | Group(Suppress("(") + delimitedList(Expression) + Suppress(")")) # [122] RegexExpression ::= 'REGEX' '(' Expression ',' Expression ( ',' Expression )? ')' -RegexExpression = Comp('Builtin_REGEX', Keyword('REGEX') + '(' + Param('text', Expression) + ',' + Param( - 'pattern', Expression) + Optional(',' + Param('flags', Expression)) + ')') +RegexExpression = Comp( + "Builtin_REGEX", + Keyword("REGEX") + + "(" + + Param("text", Expression) + + "," + + Param("pattern", Expression) + + Optional("," + Param("flags", Expression)) + + ")", +) RegexExpression.setEvalFn(op.Builtin_REGEX) # [123] SubstringExpression ::= 'SUBSTR' '(' Expression ',' Expression ( ',' Expression )? ')' -SubstringExpression = Comp('Builtin_SUBSTR', Keyword('SUBSTR') + '(' + Param('arg', Expression) + ',' + Param( - 'start', Expression) + Optional(',' + Param('length', Expression)) + ')').setEvalFn(op.Builtin_SUBSTR) +SubstringExpression = Comp( + "Builtin_SUBSTR", + Keyword("SUBSTR") + + "(" + + Param("arg", Expression) + + "," + + Param("start", Expression) + + Optional("," + Param("length", Expression)) + + ")", +).setEvalFn(op.Builtin_SUBSTR) # [124] StrReplaceExpression ::= 'REPLACE' '(' Expression ',' Expression ',' Expression ( ',' Expression )? ')' -StrReplaceExpression = Comp('Builtin_REPLACE', Keyword('REPLACE') + '(' + Param('arg', Expression) + ',' + Param( - 'pattern', Expression) + ',' + Param('replacement', Expression) + Optional(',' + Param('flags', Expression)) + ')').setEvalFn(op.Builtin_REPLACE) +StrReplaceExpression = Comp( + "Builtin_REPLACE", + Keyword("REPLACE") + + "(" + + Param("arg", Expression) + + "," + + Param("pattern", Expression) + + "," + + Param("replacement", Expression) + + Optional("," + Param("flags", Expression)) + + ")", +).setEvalFn(op.Builtin_REPLACE) # [125] ExistsFunc ::= 'EXISTS' GroupGraphPattern -ExistsFunc = Comp('Builtin_EXISTS', Keyword('EXISTS') + Param( - 'graph', GroupGraphPattern)).setEvalFn(op.Builtin_EXISTS) +ExistsFunc = Comp( + "Builtin_EXISTS", Keyword("EXISTS") + Param("graph", GroupGraphPattern) +).setEvalFn(op.Builtin_EXISTS) # [126] NotExistsFunc ::= 'NOT' 'EXISTS' GroupGraphPattern -NotExistsFunc = Comp('Builtin_NOTEXISTS', Keyword('NOT') + Keyword( - 'EXISTS') + Param('graph', GroupGraphPattern)).setEvalFn(op.Builtin_EXISTS) +NotExistsFunc = Comp( + "Builtin_NOTEXISTS", + Keyword("NOT") + Keyword("EXISTS") + Param("graph", GroupGraphPattern), +).setEvalFn(op.Builtin_EXISTS) # [127] Aggregate ::= 'COUNT' '(' 'DISTINCT'? ( '*' | Expression ) ')' @@ -605,17 +705,33 @@ NotExistsFunc = Comp('Builtin_NOTEXISTS', Keyword('NOT') + Keyword( # | 'SAMPLE' '(' Optional('DISTINCT') Expression ')' # | 'GROUP_CONCAT' '(' Optional('DISTINCT') Expression ( ';' 'SEPARATOR' '=' String )? ')' -_Distinct = Optional(Keyword('DISTINCT')) -_AggregateParams = '(' + Param( - 'distinct', _Distinct) + Param('vars', Expression) + ')' - -Aggregate = Comp('Aggregate_Count', Keyword('COUNT') + '(' + Param('distinct', _Distinct) + Param('vars', '*' | Expression) + ')')\ - | Comp('Aggregate_Sum', Keyword('SUM') + _AggregateParams)\ - | Comp('Aggregate_Min', Keyword('MIN') + _AggregateParams)\ - | Comp('Aggregate_Max', Keyword('MAX') + _AggregateParams)\ - | Comp('Aggregate_Avg', Keyword('AVG') + _AggregateParams)\ - | Comp('Aggregate_Sample', Keyword('SAMPLE') + _AggregateParams)\ - | Comp('Aggregate_GroupConcat', Keyword('GROUP_CONCAT') + '(' + Param('distinct', _Distinct) + Param('vars', Expression) + Optional(';' + Keyword('SEPARATOR') + '=' + Param('separator', String)) + ')') +_Distinct = Optional(Keyword("DISTINCT")) +_AggregateParams = "(" + Param("distinct", _Distinct) + Param("vars", Expression) + ")" + +Aggregate = ( + Comp( + "Aggregate_Count", + Keyword("COUNT") + + "(" + + Param("distinct", _Distinct) + + Param("vars", "*" | Expression) + + ")", + ) + | Comp("Aggregate_Sum", Keyword("SUM") + _AggregateParams) + | Comp("Aggregate_Min", Keyword("MIN") + _AggregateParams) + | Comp("Aggregate_Max", Keyword("MAX") + _AggregateParams) + | Comp("Aggregate_Avg", Keyword("AVG") + _AggregateParams) + | Comp("Aggregate_Sample", Keyword("SAMPLE") + _AggregateParams) + | Comp( + "Aggregate_GroupConcat", + Keyword("GROUP_CONCAT") + + "(" + + Param("distinct", _Distinct) + + Param("vars", Expression) + + Optional(";" + Keyword("SEPARATOR") + "=" + Param("separator", String)) + + ")", + ) +) # [121] BuiltInCall ::= Aggregate # | 'STR' '(' + Expression + ')' @@ -673,93 +789,271 @@ Aggregate = Comp('Aggregate_Count', Keyword('COUNT') + '(' + Param('distinct', _ # | ExistsFunc # | NotExistsFunc -BuiltInCall = Aggregate \ - | Comp('Builtin_STR', Keyword('STR') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_STR) \ - | Comp('Builtin_LANG', Keyword('LANG') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_LANG) \ - | Comp('Builtin_LANGMATCHES', Keyword('LANGMATCHES') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_LANGMATCHES) \ - | Comp('Builtin_DATATYPE', Keyword('DATATYPE') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_DATATYPE) \ - | Comp('Builtin_BOUND', Keyword('BOUND') + '(' + Param('arg', Var) + ')').setEvalFn(op.Builtin_BOUND) \ - | Comp('Builtin_IRI', Keyword('IRI') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_IRI) \ - | Comp('Builtin_URI', Keyword('URI') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_IRI) \ - | Comp('Builtin_BNODE', Keyword('BNODE') + ('(' + Param('arg', Expression) + ')' | NIL)).setEvalFn(op.Builtin_BNODE) \ - | Comp('Builtin_RAND', Keyword('RAND') + NIL).setEvalFn(op.Builtin_RAND) \ - | Comp('Builtin_ABS', Keyword('ABS') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_ABS) \ - | Comp('Builtin_CEIL', Keyword('CEIL') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_CEIL) \ - | Comp('Builtin_FLOOR', Keyword('FLOOR') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_FLOOR) \ - | Comp('Builtin_ROUND', Keyword('ROUND') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_ROUND) \ - | Comp('Builtin_CONCAT', Keyword('CONCAT') + Param('arg', ExpressionList)).setEvalFn(op.Builtin_CONCAT) \ - | SubstringExpression \ - | Comp('Builtin_STRLEN', Keyword('STRLEN') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_STRLEN) \ - | StrReplaceExpression \ - | Comp('Builtin_UCASE', Keyword('UCASE') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_UCASE) \ - | Comp('Builtin_LCASE', Keyword('LCASE') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_LCASE) \ - | Comp('Builtin_ENCODE_FOR_URI', Keyword('ENCODE_FOR_URI') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_ENCODE_FOR_URI) \ - | Comp('Builtin_CONTAINS', Keyword('CONTAINS') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_CONTAINS) \ - | Comp('Builtin_STRSTARTS', Keyword('STRSTARTS') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_STRSTARTS) \ - | Comp('Builtin_STRENDS', Keyword('STRENDS') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_STRENDS) \ - | Comp('Builtin_STRBEFORE', Keyword('STRBEFORE') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_STRBEFORE) \ - | Comp('Builtin_STRAFTER', Keyword('STRAFTER') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_STRAFTER) \ - | Comp('Builtin_YEAR', Keyword('YEAR') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_YEAR) \ - | Comp('Builtin_MONTH', Keyword('MONTH') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_MONTH) \ - | Comp('Builtin_DAY', Keyword('DAY') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_DAY) \ - | Comp('Builtin_HOURS', Keyword('HOURS') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_HOURS) \ - | Comp('Builtin_MINUTES', Keyword('MINUTES') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_MINUTES) \ - | Comp('Builtin_SECONDS', Keyword('SECONDS') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_SECONDS) \ - | Comp('Builtin_TIMEZONE', Keyword('TIMEZONE') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_TIMEZONE) \ - | Comp('Builtin_TZ', Keyword('TZ') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_TZ) \ - | Comp('Builtin_NOW', Keyword('NOW') + NIL).setEvalFn(op.Builtin_NOW) \ - | Comp('Builtin_UUID', Keyword('UUID') + NIL).setEvalFn(op.Builtin_UUID) \ - | Comp('Builtin_STRUUID', Keyword('STRUUID') + NIL).setEvalFn(op.Builtin_STRUUID) \ - | Comp('Builtin_MD5', Keyword('MD5') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_MD5) \ - | Comp('Builtin_SHA1', Keyword('SHA1') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_SHA1) \ - | Comp('Builtin_SHA256', Keyword('SHA256') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_SHA256) \ - | Comp('Builtin_SHA384', Keyword('SHA384') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_SHA384) \ - | Comp('Builtin_SHA512', Keyword('SHA512') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_SHA512) \ - | Comp('Builtin_COALESCE', Keyword('COALESCE') + Param('arg', ExpressionList)).setEvalFn(op.Builtin_COALESCE) \ - | Comp('Builtin_IF', Keyword('IF') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ',' + Param('arg3', Expression) + ')').setEvalFn(op.Builtin_IF) \ - | Comp('Builtin_STRLANG', Keyword('STRLANG') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_STRLANG) \ - | Comp('Builtin_STRDT', Keyword('STRDT') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_STRDT) \ - | Comp('Builtin_sameTerm', Keyword('sameTerm') + '(' + Param('arg1', Expression) + ',' + Param('arg2', Expression) + ')').setEvalFn(op.Builtin_sameTerm) \ - | Comp('Builtin_isIRI', Keyword('isIRI') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_isIRI) \ - | Comp('Builtin_isURI', Keyword('isURI') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_isIRI) \ - | Comp('Builtin_isBLANK', Keyword('isBLANK') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_isBLANK) \ - | Comp('Builtin_isLITERAL', Keyword('isLITERAL') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_isLITERAL) \ - | Comp('Builtin_isNUMERIC', Keyword('isNUMERIC') + '(' + Param('arg', Expression) + ')').setEvalFn(op.Builtin_isNUMERIC) \ - | RegexExpression \ - | ExistsFunc \ +BuiltInCall = ( + Aggregate + | Comp( + "Builtin_STR", Keyword("STR") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_STR) + | Comp( + "Builtin_LANG", Keyword("LANG") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_LANG) + | Comp( + "Builtin_LANGMATCHES", + Keyword("LANGMATCHES") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_LANGMATCHES) + | Comp( + "Builtin_DATATYPE", Keyword("DATATYPE") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_DATATYPE) + | Comp("Builtin_BOUND", Keyword("BOUND") + "(" + Param("arg", Var) + ")").setEvalFn( + op.Builtin_BOUND + ) + | Comp( + "Builtin_IRI", Keyword("IRI") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_IRI) + | Comp( + "Builtin_URI", Keyword("URI") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_IRI) + | Comp( + "Builtin_BNODE", Keyword("BNODE") + ("(" + Param("arg", Expression) + ")" | NIL) + ).setEvalFn(op.Builtin_BNODE) + | Comp("Builtin_RAND", Keyword("RAND") + NIL).setEvalFn(op.Builtin_RAND) + | Comp( + "Builtin_ABS", Keyword("ABS") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_ABS) + | Comp( + "Builtin_CEIL", Keyword("CEIL") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_CEIL) + | Comp( + "Builtin_FLOOR", Keyword("FLOOR") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_FLOOR) + | Comp( + "Builtin_ROUND", Keyword("ROUND") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_ROUND) + | Comp( + "Builtin_CONCAT", Keyword("CONCAT") + Param("arg", ExpressionList) + ).setEvalFn(op.Builtin_CONCAT) + | SubstringExpression + | Comp( + "Builtin_STRLEN", Keyword("STRLEN") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_STRLEN) + | StrReplaceExpression + | Comp( + "Builtin_UCASE", Keyword("UCASE") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_UCASE) + | Comp( + "Builtin_LCASE", Keyword("LCASE") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_LCASE) + | Comp( + "Builtin_ENCODE_FOR_URI", + Keyword("ENCODE_FOR_URI") + "(" + Param("arg", Expression) + ")", + ).setEvalFn(op.Builtin_ENCODE_FOR_URI) + | Comp( + "Builtin_CONTAINS", + Keyword("CONTAINS") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_CONTAINS) + | Comp( + "Builtin_STRSTARTS", + Keyword("STRSTARTS") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_STRSTARTS) + | Comp( + "Builtin_STRENDS", + Keyword("STRENDS") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_STRENDS) + | Comp( + "Builtin_STRBEFORE", + Keyword("STRBEFORE") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_STRBEFORE) + | Comp( + "Builtin_STRAFTER", + Keyword("STRAFTER") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_STRAFTER) + | Comp( + "Builtin_YEAR", Keyword("YEAR") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_YEAR) + | Comp( + "Builtin_MONTH", Keyword("MONTH") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_MONTH) + | Comp( + "Builtin_DAY", Keyword("DAY") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_DAY) + | Comp( + "Builtin_HOURS", Keyword("HOURS") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_HOURS) + | Comp( + "Builtin_MINUTES", Keyword("MINUTES") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_MINUTES) + | Comp( + "Builtin_SECONDS", Keyword("SECONDS") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_SECONDS) + | Comp( + "Builtin_TIMEZONE", Keyword("TIMEZONE") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_TIMEZONE) + | Comp( + "Builtin_TZ", Keyword("TZ") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_TZ) + | Comp("Builtin_NOW", Keyword("NOW") + NIL).setEvalFn(op.Builtin_NOW) + | Comp("Builtin_UUID", Keyword("UUID") + NIL).setEvalFn(op.Builtin_UUID) + | Comp("Builtin_STRUUID", Keyword("STRUUID") + NIL).setEvalFn(op.Builtin_STRUUID) + | Comp( + "Builtin_MD5", Keyword("MD5") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_MD5) + | Comp( + "Builtin_SHA1", Keyword("SHA1") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_SHA1) + | Comp( + "Builtin_SHA256", Keyword("SHA256") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_SHA256) + | Comp( + "Builtin_SHA384", Keyword("SHA384") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_SHA384) + | Comp( + "Builtin_SHA512", Keyword("SHA512") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_SHA512) + | Comp( + "Builtin_COALESCE", Keyword("COALESCE") + Param("arg", ExpressionList) + ).setEvalFn(op.Builtin_COALESCE) + | Comp( + "Builtin_IF", + Keyword("IF") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + "," + + Param("arg3", Expression) + + ")", + ).setEvalFn(op.Builtin_IF) + | Comp( + "Builtin_STRLANG", + Keyword("STRLANG") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_STRLANG) + | Comp( + "Builtin_STRDT", + Keyword("STRDT") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_STRDT) + | Comp( + "Builtin_sameTerm", + Keyword("sameTerm") + + "(" + + Param("arg1", Expression) + + "," + + Param("arg2", Expression) + + ")", + ).setEvalFn(op.Builtin_sameTerm) + | Comp( + "Builtin_isIRI", Keyword("isIRI") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_isIRI) + | Comp( + "Builtin_isURI", Keyword("isURI") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_isIRI) + | Comp( + "Builtin_isBLANK", Keyword("isBLANK") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_isBLANK) + | Comp( + "Builtin_isLITERAL", Keyword("isLITERAL") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_isLITERAL) + | Comp( + "Builtin_isNUMERIC", Keyword("isNUMERIC") + "(" + Param("arg", Expression) + ")" + ).setEvalFn(op.Builtin_isNUMERIC) + | RegexExpression + | ExistsFunc | NotExistsFunc +) # [71] ArgList ::= NIL | '(' 'DISTINCT'? Expression ( ',' Expression )* ')' -ArgList = NIL | '(' + Param('distinct', _Distinct) + delimitedList( - ParamList('expr', Expression)) + ')' +ArgList = ( + NIL + | "(" + + Param("distinct", _Distinct) + + delimitedList(ParamList("expr", Expression)) + + ")" +) # [128] iriOrFunction ::= iri Optional(ArgList) -iriOrFunction = (Comp( - 'Function', Param('iri', iri) + ArgList).setEvalFn(op.Function)) | iri +iriOrFunction = ( + Comp("Function", Param("iri", iri) + ArgList).setEvalFn(op.Function) +) | iri # [70] FunctionCall ::= iri ArgList -FunctionCall = Comp( - 'Function', Param('iri', iri) + ArgList).setEvalFn(op.Function) +FunctionCall = Comp("Function", Param("iri", iri) + ArgList).setEvalFn(op.Function) # [120] BrackettedExpression ::= '(' Expression ')' -BrackettedExpression = Suppress('(') + Expression + Suppress(')') +BrackettedExpression = Suppress("(") + Expression + Suppress(")") # [119] PrimaryExpression ::= BrackettedExpression | BuiltInCall | iriOrFunction | RDFLiteral | NumericLiteral | BooleanLiteral | Var -PrimaryExpression = BrackettedExpression | BuiltInCall | iriOrFunction | RDFLiteral | NumericLiteral | BooleanLiteral | Var +PrimaryExpression = ( + BrackettedExpression + | BuiltInCall + | iriOrFunction + | RDFLiteral + | NumericLiteral + | BooleanLiteral + | Var +) # [118] UnaryExpression ::= '!' PrimaryExpression # | '+' PrimaryExpression # | '-' PrimaryExpression # | PrimaryExpression -UnaryExpression = Comp('UnaryNot', '!' + Param('expr', PrimaryExpression)).setEvalFn(op.UnaryNot) \ - | Comp('UnaryPlus', '+' + Param('expr', PrimaryExpression)).setEvalFn(op.UnaryPlus) \ - | Comp('UnaryMinus', '-' + Param('expr', PrimaryExpression)).setEvalFn(op.UnaryMinus) \ +UnaryExpression = ( + Comp("UnaryNot", "!" + Param("expr", PrimaryExpression)).setEvalFn(op.UnaryNot) + | Comp("UnaryPlus", "+" + Param("expr", PrimaryExpression)).setEvalFn(op.UnaryPlus) + | Comp("UnaryMinus", "-" + Param("expr", PrimaryExpression)).setEvalFn( + op.UnaryMinus + ) | PrimaryExpression +) # [117] MultiplicativeExpression ::= UnaryExpression ( '*' UnaryExpression | '/' UnaryExpression )* -MultiplicativeExpression = Comp('MultiplicativeExpression', Param('expr', UnaryExpression) + ZeroOrMore(ParamList('op', '*') + ParamList( - 'other', UnaryExpression) | ParamList('op', '/') + ParamList('other', UnaryExpression))).setEvalFn(op.MultiplicativeExpression) +MultiplicativeExpression = Comp( + "MultiplicativeExpression", + Param("expr", UnaryExpression) + + ZeroOrMore( + ParamList("op", "*") + ParamList("other", UnaryExpression) + | ParamList("op", "/") + ParamList("other", UnaryExpression) + ), +).setEvalFn(op.MultiplicativeExpression) # [116] AdditiveExpression ::= MultiplicativeExpression ( '+' MultiplicativeExpression | '-' MultiplicativeExpression | ( NumericLiteralPositive | NumericLiteralNegative ) ( ( '*' UnaryExpression ) | ( '/' UnaryExpression ) )* )* @@ -770,36 +1064,55 @@ MultiplicativeExpression = Comp('MultiplicativeExpression', Param('expr', UnaryE # tokenizing and parsing -AdditiveExpression = Comp('AdditiveExpression', Param('expr', MultiplicativeExpression) + - ZeroOrMore(ParamList('op', '+') + ParamList('other', MultiplicativeExpression) | - ParamList('op', '-') + ParamList('other', MultiplicativeExpression))).setEvalFn(op.AdditiveExpression) +AdditiveExpression = Comp( + "AdditiveExpression", + Param("expr", MultiplicativeExpression) + + ZeroOrMore( + ParamList("op", "+") + ParamList("other", MultiplicativeExpression) + | ParamList("op", "-") + ParamList("other", MultiplicativeExpression) + ), +).setEvalFn(op.AdditiveExpression) # [115] NumericExpression ::= AdditiveExpression NumericExpression = AdditiveExpression # [114] RelationalExpression ::= NumericExpression ( '=' NumericExpression | '!=' NumericExpression | '<' NumericExpression | '>' NumericExpression | '<=' NumericExpression | '>=' NumericExpression | 'IN' ExpressionList | 'NOT' 'IN' ExpressionList )? -RelationalExpression = Comp('RelationalExpression', Param('expr', NumericExpression) + Optional( - Param('op', '=') + Param('other', NumericExpression) | - Param('op', '!=') + Param('other', NumericExpression) | - Param('op', '<') + Param('other', NumericExpression) | - Param('op', '>') + Param('other', NumericExpression) | - Param('op', '<=') + Param('other', NumericExpression) | - Param('op', '>=') + Param('other', NumericExpression) | - Param('op', Keyword('IN')) + Param('other', ExpressionList) | - Param('op', Combine(Keyword('NOT') + Keyword('IN'), adjacent=False, joinString=" ")) + Param('other', ExpressionList))).setEvalFn(op.RelationalExpression) +RelationalExpression = Comp( + "RelationalExpression", + Param("expr", NumericExpression) + + Optional( + Param("op", "=") + Param("other", NumericExpression) + | Param("op", "!=") + Param("other", NumericExpression) + | Param("op", "<") + Param("other", NumericExpression) + | Param("op", ">") + Param("other", NumericExpression) + | Param("op", "<=") + Param("other", NumericExpression) + | Param("op", ">=") + Param("other", NumericExpression) + | Param("op", Keyword("IN")) + Param("other", ExpressionList) + | Param( + "op", + Combine(Keyword("NOT") + Keyword("IN"), adjacent=False, joinString=" "), + ) + + Param("other", ExpressionList) + ), +).setEvalFn(op.RelationalExpression) # [113] ValueLogical ::= RelationalExpression ValueLogical = RelationalExpression # [112] ConditionalAndExpression ::= ValueLogical ( '&&' ValueLogical )* -ConditionalAndExpression = Comp('ConditionalAndExpression', Param('expr', ValueLogical) + ZeroOrMore( - '&&' + ParamList('other', ValueLogical))).setEvalFn(op.ConditionalAndExpression) +ConditionalAndExpression = Comp( + "ConditionalAndExpression", + Param("expr", ValueLogical) + ZeroOrMore("&&" + ParamList("other", ValueLogical)), +).setEvalFn(op.ConditionalAndExpression) # [111] ConditionalOrExpression ::= ConditionalAndExpression ( '||' ConditionalAndExpression )* -ConditionalOrExpression = Comp('ConditionalOrExpression', Param('expr', ConditionalAndExpression) + ZeroOrMore( - '||' + ParamList('other', ConditionalAndExpression))).setEvalFn(op.ConditionalOrExpression) +ConditionalOrExpression = Comp( + "ConditionalOrExpression", + Param("expr", ConditionalAndExpression) + + ZeroOrMore("||" + ParamList("other", ConditionalAndExpression)), +).setEvalFn(op.ConditionalOrExpression) # [110] Expression ::= ConditionalOrExpression Expression <<= ConditionalOrExpression @@ -809,7 +1122,7 @@ Expression <<= ConditionalOrExpression Constraint = BrackettedExpression | BuiltInCall | FunctionCall # [68] Filter ::= 'FILTER' Constraint -Filter = Comp('Filter', Keyword('FILTER') + Param('expr', Constraint)) +Filter = Comp("Filter", Keyword("FILTER") + Param("expr", Constraint)) # [16] SourceSelector ::= iri @@ -819,128 +1132,217 @@ SourceSelector = iri DefaultGraphClause = SourceSelector # [15] NamedGraphClause ::= 'NAMED' SourceSelector -NamedGraphClause = Keyword('NAMED') + Param('named', SourceSelector) +NamedGraphClause = Keyword("NAMED") + Param("named", SourceSelector) # [13] DatasetClause ::= 'FROM' ( DefaultGraphClause | NamedGraphClause ) -DatasetClause = Comp('DatasetClause', Keyword( - 'FROM') + (Param('default', DefaultGraphClause) | NamedGraphClause)) +DatasetClause = Comp( + "DatasetClause", + Keyword("FROM") + (Param("default", DefaultGraphClause) | NamedGraphClause), +) # [20] GroupCondition ::= BuiltInCall | FunctionCall | '(' Expression ( 'AS' Var )? ')' | Var -GroupCondition = BuiltInCall | FunctionCall | Comp('GroupAs', '(' + Param( - 'expr', Expression) + Optional(Keyword('AS') + Param('var', Var)) + ')') | Var +GroupCondition = ( + BuiltInCall + | FunctionCall + | Comp( + "GroupAs", + "(" + + Param("expr", Expression) + + Optional(Keyword("AS") + Param("var", Var)) + + ")", + ) + | Var +) # [19] GroupClause ::= 'GROUP' 'BY' GroupCondition+ -GroupClause = Comp('GroupClause', Keyword('GROUP') + Keyword( - 'BY') + OneOrMore(ParamList('condition', GroupCondition))) +GroupClause = Comp( + "GroupClause", + Keyword("GROUP") + + Keyword("BY") + + OneOrMore(ParamList("condition", GroupCondition)), +) -_Silent = Optional(Param('silent', Keyword('SILENT'))) +_Silent = Optional(Param("silent", Keyword("SILENT"))) # [31] Load ::= 'LOAD' 'SILENT'? iri ( 'INTO' GraphRef )? -Load = Comp('Load', Keyword('LOAD') + _Silent + Param('iri', iri) + - Optional(Keyword('INTO') + GraphRef)) +Load = Comp( + "Load", + Keyword("LOAD") + + _Silent + + Param("iri", iri) + + Optional(Keyword("INTO") + GraphRef), +) # [32] Clear ::= 'CLEAR' 'SILENT'? GraphRefAll -Clear = Comp('Clear', Keyword('CLEAR') + _Silent + GraphRefAll) +Clear = Comp("Clear", Keyword("CLEAR") + _Silent + GraphRefAll) # [33] Drop ::= 'DROP' _Silent GraphRefAll -Drop = Comp('Drop', Keyword('DROP') + _Silent + GraphRefAll) +Drop = Comp("Drop", Keyword("DROP") + _Silent + GraphRefAll) # [34] Create ::= 'CREATE' _Silent GraphRef -Create = Comp('Create', Keyword('CREATE') + _Silent + GraphRef) +Create = Comp("Create", Keyword("CREATE") + _Silent + GraphRef) # [35] Add ::= 'ADD' _Silent GraphOrDefault 'TO' GraphOrDefault -Add = Comp('Add', Keyword( - 'ADD') + _Silent + GraphOrDefault + Keyword('TO') + GraphOrDefault) +Add = Comp( + "Add", Keyword("ADD") + _Silent + GraphOrDefault + Keyword("TO") + GraphOrDefault +) # [36] Move ::= 'MOVE' _Silent GraphOrDefault 'TO' GraphOrDefault -Move = Comp('Move', Keyword( - 'MOVE') + _Silent + GraphOrDefault + Keyword('TO') + GraphOrDefault) +Move = Comp( + "Move", Keyword("MOVE") + _Silent + GraphOrDefault + Keyword("TO") + GraphOrDefault +) # [37] Copy ::= 'COPY' _Silent GraphOrDefault 'TO' GraphOrDefault -Copy = Comp('Copy', Keyword( - 'COPY') + _Silent + GraphOrDefault + Keyword('TO') + GraphOrDefault) +Copy = Comp( + "Copy", Keyword("COPY") + _Silent + GraphOrDefault + Keyword("TO") + GraphOrDefault +) # [38] InsertData ::= 'INSERT DATA' QuadData -InsertData = Comp('InsertData', Keyword('INSERT') + Keyword('DATA') + QuadData) +InsertData = Comp("InsertData", Keyword("INSERT") + Keyword("DATA") + QuadData) # [39] DeleteData ::= 'DELETE DATA' QuadData -DeleteData = Comp('DeleteData', Keyword('DELETE') + Keyword('DATA') + QuadData) +DeleteData = Comp("DeleteData", Keyword("DELETE") + Keyword("DATA") + QuadData) # [40] DeleteWhere ::= 'DELETE WHERE' QuadPattern -DeleteWhere = Comp( - 'DeleteWhere', Keyword('DELETE') + Keyword('WHERE') + QuadPattern) +DeleteWhere = Comp("DeleteWhere", Keyword("DELETE") + Keyword("WHERE") + QuadPattern) # [42] DeleteClause ::= 'DELETE' QuadPattern -DeleteClause = Comp('DeleteClause', Keyword('DELETE') + QuadPattern) +DeleteClause = Comp("DeleteClause", Keyword("DELETE") + QuadPattern) # [43] InsertClause ::= 'INSERT' QuadPattern -InsertClause = Comp('InsertClause', Keyword('INSERT') + QuadPattern) +InsertClause = Comp("InsertClause", Keyword("INSERT") + QuadPattern) # [44] UsingClause ::= 'USING' ( iri | 'NAMED' iri ) -UsingClause = Comp('UsingClause', Keyword('USING') + ( - Param('default', iri) | Keyword('NAMED') + Param('named', iri))) +UsingClause = Comp( + "UsingClause", + Keyword("USING") + (Param("default", iri) | Keyword("NAMED") + Param("named", iri)), +) # [41] Modify ::= ( 'WITH' iri )? ( DeleteClause Optional(InsertClause) | InsertClause ) ZeroOrMore(UsingClause) 'WHERE' GroupGraphPattern -Modify = Comp('Modify', Optional(Keyword('WITH') + Param('withClause', iri)) + (Param('delete', DeleteClause) + Optional(Param( - 'insert', InsertClause)) | Param('insert', InsertClause)) + ZeroOrMore(ParamList('using', UsingClause)) + Keyword('WHERE') + Param('where', GroupGraphPattern)) +Modify = Comp( + "Modify", + Optional(Keyword("WITH") + Param("withClause", iri)) + + ( + Param("delete", DeleteClause) + Optional(Param("insert", InsertClause)) + | Param("insert", InsertClause) + ) + + ZeroOrMore(ParamList("using", UsingClause)) + + Keyword("WHERE") + + Param("where", GroupGraphPattern), +) # [30] Update1 ::= Load | Clear | Drop | Add | Move | Copy | Create | InsertData | DeleteData | DeleteWhere | Modify -Update1 = Load | Clear | Drop | Add | Move | Copy | Create | InsertData | DeleteData | DeleteWhere | Modify +Update1 = ( + Load + | Clear + | Drop + | Add + | Move + | Copy + | Create + | InsertData + | DeleteData + | DeleteWhere + | Modify +) # [63] InlineDataOneVar ::= Var '{' ZeroOrMore(DataBlockValue) '}' -InlineDataOneVar = ParamList( - 'var', Var) + '{' + ZeroOrMore(ParamList('value', DataBlockValue)) + '}' +InlineDataOneVar = ( + ParamList("var", Var) + "{" + ZeroOrMore(ParamList("value", DataBlockValue)) + "}" +) # [64] InlineDataFull ::= ( NIL | '(' ZeroOrMore(Var) ')' ) '{' ( '(' ZeroOrMore(DataBlockValue) ')' | NIL )* '}' -InlineDataFull = (NIL | '(' + ZeroOrMore(ParamList('var', Var)) + ')') + '{' + ZeroOrMore( - ParamList('value', Group(Suppress('(') + ZeroOrMore(DataBlockValue) + Suppress(')') | NIL))) + '}' +InlineDataFull = ( + (NIL | "(" + ZeroOrMore(ParamList("var", Var)) + ")") + + "{" + + ZeroOrMore( + ParamList( + "value", + Group(Suppress("(") + ZeroOrMore(DataBlockValue) + Suppress(")") | NIL), + ) + ) + + "}" +) # [62] DataBlock ::= InlineDataOneVar | InlineDataFull DataBlock = InlineDataOneVar | InlineDataFull # [28] ValuesClause ::= ( 'VALUES' DataBlock )? -ValuesClause = Optional(Param( - 'valuesClause', Comp('ValuesClause', Keyword('VALUES') + DataBlock))) +ValuesClause = Optional( + Param("valuesClause", Comp("ValuesClause", Keyword("VALUES") + DataBlock)) +) # [74] ConstructTriples ::= TriplesSameSubject ( '.' Optional(ConstructTriples) )? ConstructTriples = Forward() -ConstructTriples <<= (ParamList('template', TriplesSameSubject) + Optional( - Suppress('.') + Optional(ConstructTriples))) +ConstructTriples <<= ParamList("template", TriplesSameSubject) + Optional( + Suppress(".") + Optional(ConstructTriples) +) # [73] ConstructTemplate ::= '{' Optional(ConstructTriples) '}' -ConstructTemplate = Suppress('{') + Optional(ConstructTriples) + Suppress('}') +ConstructTemplate = Suppress("{") + Optional(ConstructTriples) + Suppress("}") # [57] OptionalGraphPattern ::= 'OPTIONAL' GroupGraphPattern -OptionalGraphPattern = Comp('OptionalGraphPattern', Keyword( - 'OPTIONAL') + Param('graph', GroupGraphPattern)) +OptionalGraphPattern = Comp( + "OptionalGraphPattern", Keyword("OPTIONAL") + Param("graph", GroupGraphPattern) +) # [58] GraphGraphPattern ::= 'GRAPH' VarOrIri GroupGraphPattern -GraphGraphPattern = Comp('GraphGraphPattern', Keyword( - 'GRAPH') + Param('term', VarOrIri) + Param('graph', GroupGraphPattern)) +GraphGraphPattern = Comp( + "GraphGraphPattern", + Keyword("GRAPH") + Param("term", VarOrIri) + Param("graph", GroupGraphPattern), +) # [59] ServiceGraphPattern ::= 'SERVICE' _Silent VarOrIri GroupGraphPattern -ServiceGraphPattern = Comp('ServiceGraphPattern', Keyword( - 'SERVICE') + _Silent + Param('term', VarOrIri) + Param('graph', GroupGraphPattern)) +ServiceGraphPattern = Comp( + "ServiceGraphPattern", + Keyword("SERVICE") + + _Silent + + Param("term", VarOrIri) + + Param("graph", GroupGraphPattern), +) # [60] Bind ::= 'BIND' '(' Expression 'AS' Var ')' -Bind = Comp('Bind', Keyword('BIND') + '(' + Param( - 'expr', Expression) + Keyword('AS') + Param('var', Var) + ')') +Bind = Comp( + "Bind", + Keyword("BIND") + + "(" + + Param("expr", Expression) + + Keyword("AS") + + Param("var", Var) + + ")", +) # [61] InlineData ::= 'VALUES' DataBlock -InlineData = Comp('InlineData', Keyword('VALUES') + DataBlock) +InlineData = Comp("InlineData", Keyword("VALUES") + DataBlock) # [56] GraphPatternNotTriples ::= GroupOrUnionGraphPattern | OptionalGraphPattern | MinusGraphPattern | GraphGraphPattern | ServiceGraphPattern | Filter | Bind | InlineData -GraphPatternNotTriples = GroupOrUnionGraphPattern | OptionalGraphPattern | MinusGraphPattern | GraphGraphPattern | ServiceGraphPattern | Filter | Bind | InlineData +GraphPatternNotTriples = ( + GroupOrUnionGraphPattern + | OptionalGraphPattern + | MinusGraphPattern + | GraphGraphPattern + | ServiceGraphPattern + | Filter + | Bind + | InlineData +) # [54] GroupGraphPatternSub ::= Optional(TriplesBlock) ( GraphPatternNotTriples '.'? Optional(TriplesBlock) )* -GroupGraphPatternSub = Comp('GroupGraphPatternSub', Optional(ParamList('part', Comp('TriplesBlock', TriplesBlock))) + ZeroOrMore( - ParamList('part', GraphPatternNotTriples) + Optional('.') + Optional(ParamList('part', Comp('TriplesBlock', TriplesBlock))))) +GroupGraphPatternSub = Comp( + "GroupGraphPatternSub", + Optional(ParamList("part", Comp("TriplesBlock", TriplesBlock))) + + ZeroOrMore( + ParamList("part", GraphPatternNotTriples) + + Optional(".") + + Optional(ParamList("part", Comp("TriplesBlock", TriplesBlock))) + ), +) # ---------------- @@ -948,70 +1350,151 @@ GroupGraphPatternSub = Comp('GroupGraphPatternSub', Optional(ParamList('part', C HavingCondition = Constraint # [21] HavingClause ::= 'HAVING' HavingCondition+ -HavingClause = Comp('HavingClause', Keyword( - 'HAVING') + OneOrMore(ParamList('condition', HavingCondition))) +HavingClause = Comp( + "HavingClause", + Keyword("HAVING") + OneOrMore(ParamList("condition", HavingCondition)), +) # [24] OrderCondition ::= ( ( 'ASC' | 'DESC' ) BrackettedExpression ) # | ( Constraint | Var ) -OrderCondition = Comp('OrderCondition', Param('order', Keyword('ASC') | Keyword( - 'DESC')) + Param('expr', BrackettedExpression) | Param('expr', Constraint | Var)) +OrderCondition = Comp( + "OrderCondition", + Param("order", Keyword("ASC") | Keyword("DESC")) + + Param("expr", BrackettedExpression) + | Param("expr", Constraint | Var), +) # [23] OrderClause ::= 'ORDER' 'BY' OneOrMore(OrderCondition) -OrderClause = Comp('OrderClause', Keyword('ORDER') + Keyword( - 'BY') + OneOrMore(ParamList('condition', OrderCondition))) +OrderClause = Comp( + "OrderClause", + Keyword("ORDER") + + Keyword("BY") + + OneOrMore(ParamList("condition", OrderCondition)), +) # [26] LimitClause ::= 'LIMIT' INTEGER -LimitClause = Keyword('LIMIT') + Param('limit', INTEGER) +LimitClause = Keyword("LIMIT") + Param("limit", INTEGER) # [27] OffsetClause ::= 'OFFSET' INTEGER -OffsetClause = Keyword('OFFSET') + Param('offset', INTEGER) +OffsetClause = Keyword("OFFSET") + Param("offset", INTEGER) # [25] LimitOffsetClauses ::= LimitClause Optional(OffsetClause) | OffsetClause Optional(LimitClause) -LimitOffsetClauses = Comp('LimitOffsetClauses', LimitClause + Optional( - OffsetClause) | OffsetClause + Optional(LimitClause)) +LimitOffsetClauses = Comp( + "LimitOffsetClauses", + LimitClause + Optional(OffsetClause) | OffsetClause + Optional(LimitClause), +) # [18] SolutionModifier ::= GroupClause? HavingClause? OrderClause? LimitOffsetClauses? -SolutionModifier = Optional(Param('groupby', GroupClause)) + Optional(Param('having', HavingClause)) + Optional( - Param('orderby', OrderClause)) + Optional(Param('limitoffset', LimitOffsetClauses)) +SolutionModifier = ( + Optional(Param("groupby", GroupClause)) + + Optional(Param("having", HavingClause)) + + Optional(Param("orderby", OrderClause)) + + Optional(Param("limitoffset", LimitOffsetClauses)) +) # [9] SelectClause ::= 'SELECT' ( 'DISTINCT' | 'REDUCED' )? ( ( Var | ( '(' Expression 'AS' Var ')' ) )+ | '*' ) -SelectClause = Keyword('SELECT') + Optional(Param('modifier', Keyword('DISTINCT') | Keyword('REDUCED'))) + (OneOrMore(ParamList('projection', Comp('vars', - Param('var', Var) | (Literal('(') + Param('expr', Expression) + Keyword('AS') + Param('evar', Var) + ')')))) | '*') +SelectClause = ( + Keyword("SELECT") + + Optional(Param("modifier", Keyword("DISTINCT") | Keyword("REDUCED"))) + + ( + OneOrMore( + ParamList( + "projection", + Comp( + "vars", + Param("var", Var) + | ( + Literal("(") + + Param("expr", Expression) + + Keyword("AS") + + Param("evar", Var) + + ")" + ), + ), + ) + ) + | "*" + ) +) # [17] WhereClause ::= 'WHERE'? GroupGraphPattern -WhereClause = Optional(Keyword('WHERE')) + Param('where', GroupGraphPattern) +WhereClause = Optional(Keyword("WHERE")) + Param("where", GroupGraphPattern) # [8] SubSelect ::= SelectClause WhereClause SolutionModifier ValuesClause -SubSelect = Comp('SubSelect', SelectClause + WhereClause + - SolutionModifier + ValuesClause) +SubSelect = Comp( + "SubSelect", SelectClause + WhereClause + SolutionModifier + ValuesClause +) # [53] GroupGraphPattern ::= '{' ( SubSelect | GroupGraphPatternSub ) '}' -GroupGraphPattern <<= ( - Suppress('{') + (SubSelect | GroupGraphPatternSub) + Suppress('}')) +GroupGraphPattern <<= Suppress("{") + (SubSelect | GroupGraphPatternSub) + Suppress("}") # [7] SelectQuery ::= SelectClause DatasetClause* WhereClause SolutionModifier -SelectQuery = Comp('SelectQuery', SelectClause + ZeroOrMore(ParamList( - 'datasetClause', DatasetClause)) + WhereClause + SolutionModifier + ValuesClause) +SelectQuery = Comp( + "SelectQuery", + SelectClause + + ZeroOrMore(ParamList("datasetClause", DatasetClause)) + + WhereClause + + SolutionModifier + + ValuesClause, +) # [10] ConstructQuery ::= 'CONSTRUCT' ( ConstructTemplate DatasetClause* WhereClause SolutionModifier | DatasetClause* 'WHERE' '{' TriplesTemplate? '}' SolutionModifier ) # NOTE: The CONSTRUCT WHERE alternative has unnescessarily many Comp/Param pairs # to allow it to through the same algebra translation process -ConstructQuery = Comp('ConstructQuery', Keyword('CONSTRUCT') + (ConstructTemplate + ZeroOrMore(ParamList('datasetClause', DatasetClause)) + WhereClause + SolutionModifier + ValuesClause | ZeroOrMore(ParamList( - 'datasetClause', DatasetClause)) + Keyword('WHERE') + '{' + Optional(Param('where', Comp('FakeGroupGraphPatten', ParamList('part', Comp('TriplesBlock', TriplesTemplate))))) + '}' + SolutionModifier + ValuesClause)) +ConstructQuery = Comp( + "ConstructQuery", + Keyword("CONSTRUCT") + + ( + ConstructTemplate + + ZeroOrMore(ParamList("datasetClause", DatasetClause)) + + WhereClause + + SolutionModifier + + ValuesClause + | ZeroOrMore(ParamList("datasetClause", DatasetClause)) + + Keyword("WHERE") + + "{" + + Optional( + Param( + "where", + Comp( + "FakeGroupGraphPatten", + ParamList("part", Comp("TriplesBlock", TriplesTemplate)), + ), + ) + ) + + "}" + + SolutionModifier + + ValuesClause + ), +) # [12] AskQuery ::= 'ASK' DatasetClause* WhereClause SolutionModifier -AskQuery = Comp('AskQuery', Keyword('ASK') + Param('datasetClause', ZeroOrMore( - DatasetClause)) + WhereClause + SolutionModifier + ValuesClause) +AskQuery = Comp( + "AskQuery", + Keyword("ASK") + + Param("datasetClause", ZeroOrMore(DatasetClause)) + + WhereClause + + SolutionModifier + + ValuesClause, +) # [11] DescribeQuery ::= 'DESCRIBE' ( VarOrIri+ | '*' ) DatasetClause* WhereClause? SolutionModifier -DescribeQuery = Comp('DescribeQuery', Keyword('DESCRIBE') + (OneOrMore(ParamList('var', VarOrIri)) | '*') + Param( - 'datasetClause', ZeroOrMore(DatasetClause)) + Optional(WhereClause) + SolutionModifier + ValuesClause) +DescribeQuery = Comp( + "DescribeQuery", + Keyword("DESCRIBE") + + (OneOrMore(ParamList("var", VarOrIri)) | "*") + + Param("datasetClause", ZeroOrMore(DatasetClause)) + + Optional(WhereClause) + + SolutionModifier + + ValuesClause, +) # [29] Update ::= Prologue ( Update1 ( ';' Update )? )? Update = Forward() -Update <<= (ParamList('prologue', Prologue) + Optional(ParamList('request', - Update1) + Optional(';' + Update))) +Update <<= ParamList("prologue", Prologue) + Optional( + ParamList("request", Update1) + Optional(";" + Update) +) # [2] Query ::= Prologue @@ -1021,17 +1504,16 @@ Update <<= (ParamList('prologue', Prologue) + Optional(ParamList('request', Query = Prologue + (SelectQuery | ConstructQuery | DescribeQuery | AskQuery) # [3] UpdateUnit ::= Update -UpdateUnit = Comp('Update', Update) +UpdateUnit = Comp("Update", Update) # [1] QueryUnit ::= Query QueryUnit = Query -QueryUnit.ignore('#' + restOfLine) -UpdateUnit.ignore('#' + restOfLine) +QueryUnit.ignore("#" + restOfLine) +UpdateUnit.ignore("#" + restOfLine) -expandUnicodeEscapes_re = re.compile( - r'\\u([0-9a-f]{4}(?:[0-9a-f]{4})?)', flags=re.I) +expandUnicodeEscapes_re = re.compile(r"\\u([0-9a-f]{4}(?:[0-9a-f]{4})?)", flags=re.I) def expandUnicodeEscapes(q): @@ -1050,28 +1532,29 @@ def expandUnicodeEscapes(q): def parseQuery(q): - if hasattr(q, 'read'): + if hasattr(q, "read"): q = q.read() if isinstance(q, bytes): - q = q.decode('utf-8') + q = q.decode("utf-8") q = expandUnicodeEscapes(q) return Query.parseString(q, parseAll=True) def parseUpdate(q): - if hasattr(q, 'read'): + if hasattr(q, "read"): q = q.read() if isinstance(q, bytes): - q = q.decode('utf-8') + q = q.decode("utf-8") q = expandUnicodeEscapes(q) return UpdateUnit.parseString(q, parseAll=True)[0] -if __name__ == '__main__': +if __name__ == "__main__": import sys + DEBUG = True try: q = Query.parseString(sys.argv[1]) diff --git a/rdflib/plugins/sparql/parserutils.py b/rdflib/plugins/sparql/parserutils.py index 29804eea..e67b754b 100644 --- a/rdflib/plugins/sparql/parserutils.py +++ b/rdflib/plugins/sparql/parserutils.py @@ -1,4 +1,3 @@ - from types import MethodType from collections import OrderedDict @@ -44,6 +43,7 @@ the resulting CompValue # Comp('Sum')( Param('x')(Number) + '+' + Param('y')(Number) ) + def value(ctx, val, variables=False, errors=False): """ utility function for evaluating something... @@ -172,7 +172,7 @@ class CompValue(OrderedDict): def __getattr__(self, a): # Hack hack: OrderedDict relies on this - if a in ('_OrderedDict__root', '_OrderedDict__end'): + if a in ("_OrderedDict__root", "_OrderedDict__end"): raise AttributeError try: return self[a] @@ -224,13 +224,13 @@ class Comp(TokenConverter): res._evalfn = MethodType(self.evalfn, res) else: res = CompValue(self.name) - if self.name == 'ServiceGraphPattern': + if self.name == "ServiceGraphPattern": # Then this must be a service graph pattern and have # already matched. # lets assume there is one, for now, then test for two later. sgp = originalTextFor(self.expr) service_string = sgp.searchString(instring)[0][0] - res['service_string'] = service_string + res["service_string"] = service_string for t in tokenList: if isinstance(t, ParamValue): @@ -250,38 +250,38 @@ class Comp(TokenConverter): return self -def prettify_parsetree(t, indent='', depth=0): +def prettify_parsetree(t, indent="", depth=0): out = [] if isinstance(t, ParseResults): for e in t.asList(): out.append(prettify_parsetree(e, indent, depth + 1)) for k, v in sorted(t.items()): - out.append("%s%s- %s:\n" % (indent, ' ' * depth, k)) + out.append("%s%s- %s:\n" % (indent, " " * depth, k)) out.append(prettify_parsetree(v, indent, depth + 1)) elif isinstance(t, CompValue): - out.append("%s%s> %s:\n" % (indent, ' ' * depth, t.name)) + out.append("%s%s> %s:\n" % (indent, " " * depth, t.name)) for k, v in t.items(): - out.append("%s%s- %s:\n" % (indent, ' ' * (depth + 1), k)) + out.append("%s%s- %s:\n" % (indent, " " * (depth + 1), k)) out.append(prettify_parsetree(v, indent, depth + 2)) elif isinstance(t, dict): for k, v in t.items(): - out.append("%s%s- %s:\n" % (indent, ' ' * (depth + 1), k)) + out.append("%s%s- %s:\n" % (indent, " " * (depth + 1), k)) out.append(prettify_parsetree(v, indent, depth + 2)) elif isinstance(t, list): for e in t: out.append(prettify_parsetree(e, indent, depth + 1)) else: - out.append("%s%s- %r\n" % (indent, ' ' * depth, t)) + out.append("%s%s- %r\n" % (indent, " " * depth, t)) return "".join(out) -if __name__ == '__main__': +if __name__ == "__main__": from pyparsing import Word, nums import sys Number = Word(nums) Number.setParseAction(lambda x: int(x[0])) - Plus = Comp('plus', Param('a', Number) + '+' + Param('b', Number)) + Plus = Comp("plus", Param("a", Number) + "+" + Param("b", Number)) Plus.setEvalFn(lambda self, ctx: self.a + self.b) r = Plus.parseString(sys.argv[1]) diff --git a/rdflib/plugins/sparql/processor.py b/rdflib/plugins/sparql/processor.py index 073a387e..84e8c823 100644 --- a/rdflib/plugins/sparql/processor.py +++ b/rdflib/plugins/sparql/processor.py @@ -1,4 +1,3 @@ - """ Code for tying SPARQL Engine into RDFLib @@ -7,7 +6,6 @@ These should be automatically registered with RDFLib """ - from rdflib.query import Processor, Result, UpdateProcessor from rdflib.plugins.sparql.sparql import Query @@ -33,12 +31,12 @@ def processUpdate(graph, updateString, initBindings={}, initNs={}, base=None): Process a SPARQL Update Request returns Nothing on success or raises Exceptions on error """ - evalUpdate(graph, translateUpdate( - parseUpdate(updateString), base, initNs), initBindings) + evalUpdate( + graph, translateUpdate(parseUpdate(updateString), base, initNs), initBindings + ) class SPARQLResult(Result): - def __init__(self, res): Result.__init__(self, res["type_"]) self.vars = res.get("vars_") @@ -59,13 +57,10 @@ class SPARQLUpdateProcessor(UpdateProcessor): class SPARQLProcessor(Processor): - def __init__(self, graph): self.graph = graph - def query( - self, strOrQuery, initBindings={}, - initNs={}, base=None, DEBUG=False): + def query(self, strOrQuery, initBindings={}, initNs={}, base=None, DEBUG=False): """ Evaluate a query with the given initial bindings, and initial namespaces. The given base is used to resolve relative URIs in diff --git a/rdflib/plugins/sparql/results/csvresults.py b/rdflib/plugins/sparql/results/csvresults.py index d354ccf5..c87b6ea7 100644 --- a/rdflib/plugins/sparql/results/csvresults.py +++ b/rdflib/plugins/sparql/results/csvresults.py @@ -21,11 +21,11 @@ class CSVResultParser(ResultParser): def parse(self, source, content_type=None): - r = Result('SELECT') + r = Result("SELECT") if isinstance(source.read(0), bytes): # if reading from source returns bytes do utf-8 decoding - source = codecs.getreader('utf-8')(source) + source = codecs.getreader("utf-8")(source) reader = csv.reader(source, delimiter=self.delim) r.vars = [Variable(x) for x in next(reader)] @@ -37,9 +37,11 @@ class CSVResultParser(ResultParser): return r def parseRow(self, row, v): - return dict((var, val) - for var, val in zip(v, [self.convertTerm(t) - for t in row]) if val is not None) + return dict( + (var, val) + for var, val in zip(v, [self.convertTerm(t) for t in row]) + if val is not None + ) def convertTerm(self, t): if t == "": @@ -52,22 +54,21 @@ class CSVResultParser(ResultParser): class CSVResultSerializer(ResultSerializer): - def __init__(self, result): ResultSerializer.__init__(self, result) self.delim = "," if result.type != "SELECT": - raise Exception( - "CSVSerializer can only serialize select query results") + raise Exception("CSVSerializer can only serialize select query results") - def serialize(self, stream, encoding='utf-8', **kwargs): + def serialize(self, stream, encoding="utf-8", **kwargs): # the serialiser writes bytes in the given encoding # in py3 csv.writer is unicode aware and writes STRINGS, # so we encode afterwards import codecs + stream = codecs.getwriter(encoding)(stream) out = csv.writer(stream, delimiter=self.delim) @@ -75,8 +76,9 @@ class CSVResultSerializer(ResultSerializer): vs = [self.serializeTerm(v, encoding) for v in self.result.vars] out.writerow(vs) for row in self.result.bindings: - out.writerow([self.serializeTerm( - row.get(v), encoding) for v in self.result.vars]) + out.writerow( + [self.serializeTerm(row.get(v), encoding) for v in self.result.vars] + ) def serializeTerm(self, term, encoding): if term is None: diff --git a/rdflib/plugins/sparql/results/graph.py b/rdflib/plugins/sparql/results/graph.py index c47daa72..13e256bb 100644 --- a/rdflib/plugins/sparql/results/graph.py +++ b/rdflib/plugins/sparql/results/graph.py @@ -1,18 +1,12 @@ from rdflib import Graph -from rdflib.query import ( - Result, - ResultParser, - ResultSerializer, - ResultException -) +from rdflib.query import Result, ResultParser, ResultSerializer, ResultException class GraphResultParser(ResultParser): - def parse(self, source, content_type): - res = Result('CONSTRUCT') # hmm - or describe?type_) + res = Result("CONSTRUCT") # hmm - or describe?type_) res.graph = Graph() res.graph.parse(source, format=content_type) diff --git a/rdflib/plugins/sparql/results/jsonresults.py b/rdflib/plugins/sparql/results/jsonresults.py index 6110c324..13a8da5e 100644 --- a/rdflib/plugins/sparql/results/jsonresults.py +++ b/rdflib/plugins/sparql/results/jsonresults.py @@ -1,7 +1,6 @@ import json -from rdflib.query import ( - Result, ResultException, ResultSerializer, ResultParser) +from rdflib.query import Result, ResultException, ResultSerializer, ResultParser from rdflib import Literal, URIRef, BNode, Variable @@ -18,23 +17,21 @@ Authors: Drew Perttula, Gunnar Aastrand Grimnes class JSONResultParser(ResultParser): - def parse(self, source, content_type=None): inp = source.read() if isinstance(inp, bytes): - inp = inp.decode('utf-8') + inp = inp.decode("utf-8") return JSONResult(json.loads(inp)) class JSONResultSerializer(ResultSerializer): - def __init__(self, result): ResultSerializer.__init__(self, result) def serialize(self, stream, encoding=None): res = {} - if self.result.type == 'ASK': + if self.result.type == "ASK": res["head"] = {} res["boolean"] = self.result.askAnswer else: @@ -42,8 +39,9 @@ class JSONResultSerializer(ResultSerializer): res["results"] = {} res["head"] = {} res["head"]["vars"] = self.result.vars - res["results"]["bindings"] = [self._bindingToJSON( - x) for x in self.result.bindings] + res["results"]["bindings"] = [ + self._bindingToJSON(x) for x in self.result.bindings + ] r = json.dumps(res, allow_nan=False, ensure_ascii=False) if encoding is not None: @@ -61,27 +59,26 @@ class JSONResultSerializer(ResultSerializer): class JSONResult(Result): - def __init__(self, json): self.json = json if "boolean" in json: - type_ = 'ASK' + type_ = "ASK" elif "results" in json: - type_ = 'SELECT' + type_ = "SELECT" else: - raise ResultException('No boolean or results in json!') + raise ResultException("No boolean or results in json!") Result.__init__(self, type_) - if type_ == 'ASK': - self.askAnswer = bool(json['boolean']) + if type_ == "ASK": + self.askAnswer = bool(json["boolean"]) else: self.bindings = self._get_bindings() self.vars = [Variable(x) for x in json["head"]["vars"]] def _get_bindings(self): ret = [] - for row in self.json['results']['bindings']: + for row in self.json["results"]["bindings"]: outRow = {} for k, v in row.items(): outRow[Variable(k)] = parseJsonTerm(v) @@ -97,36 +94,34 @@ def parseJsonTerm(d): { 'type': 'literal', 'value': 'drewp' } """ - t = d['type'] - if t == 'uri': - return URIRef(d['value']) - elif t == 'literal': - return Literal(d['value'], datatype=d.get('datatype'), lang=d.get('xml:lang')) - elif t == 'typed-literal': - return Literal(d['value'], datatype=URIRef(d['datatype'])) - elif t == 'bnode': - return BNode(d['value']) + t = d["type"] + if t == "uri": + return URIRef(d["value"]) + elif t == "literal": + return Literal(d["value"], datatype=d.get("datatype"), lang=d.get("xml:lang")) + elif t == "typed-literal": + return Literal(d["value"], datatype=URIRef(d["datatype"])) + elif t == "bnode": + return BNode(d["value"]) else: raise NotImplementedError("json term type %r" % t) def termToJSON(self, term): if isinstance(term, URIRef): - return {'type': 'uri', 'value': str(term)} + return {"type": "uri", "value": str(term)} elif isinstance(term, Literal): - r = {'type': 'literal', - 'value': str(term)} + r = {"type": "literal", "value": str(term)} if term.datatype is not None: - r['datatype'] = str(term.datatype) + r["datatype"] = str(term.datatype) if term.language is not None: - r['xml:lang'] = term.language + r["xml:lang"] = term.language return r elif isinstance(term, BNode): - return {'type': 'bnode', 'value': str(term)} + return {"type": "bnode", "value": str(term)} elif term is None: return None else: - raise ResultException( - 'Unknown term type: %s (%s)' % (term, type(term))) + raise ResultException("Unknown term type: %s (%s)" % (term, type(term))) diff --git a/rdflib/plugins/sparql/results/rdfresults.py b/rdflib/plugins/sparql/results/rdfresults.py index ac71ff1d..7f64bbf4 100644 --- a/rdflib/plugins/sparql/results/rdfresults.py +++ b/rdflib/plugins/sparql/results/rdfresults.py @@ -2,7 +2,7 @@ from rdflib import Graph, Namespace, RDF, Variable from rdflib.query import Result, ResultParser -RS = Namespace('http://www.w3.org/2001/sw/DataAccess/tests/result-set#') +RS = Namespace("http://www.w3.org/2001/sw/DataAccess/tests/result-set#") class RDFResultParser(ResultParser): @@ -11,7 +11,6 @@ class RDFResultParser(ResultParser): class RDFResult(Result): - def __init__(self, source, **kwargs): if not isinstance(source, Graph): @@ -24,7 +23,7 @@ class RDFResult(Result): # there better be only one :) if rs is None: - type_ = 'CONSTRUCT' + type_ = "CONSTRUCT" # use a new graph g = Graph() @@ -35,27 +34,27 @@ class RDFResult(Result): askAnswer = graph.value(rs, RS.boolean) if askAnswer is not None: - type_ = 'ASK' + type_ = "ASK" else: - type_ = 'SELECT' + type_ = "SELECT" Result.__init__(self, type_) - if type_ == 'SELECT': - self.vars = [Variable(v) for v in graph.objects(rs, - RS.resultVariable)] + if type_ == "SELECT": + self.vars = [Variable(v) for v in graph.objects(rs, RS.resultVariable)] self.bindings = [] for s in graph.objects(rs, RS.solution): sol = {} for b in graph.objects(s, RS.binding): - sol[Variable(graph.value( - b, RS.variable))] = graph.value(b, RS.value) + sol[Variable(graph.value(b, RS.variable))] = graph.value( + b, RS.value + ) self.bindings.append(sol) - elif type_ == 'ASK': + elif type_ == "ASK": self.askAnswer = askAnswer.value if askAnswer.value is None: - raise Exception('Malformed boolean in ask answer!') - elif type_ == 'CONSTRUCT': + raise Exception("Malformed boolean in ask answer!") + elif type_ == "CONSTRUCT": self.graph = g diff --git a/rdflib/plugins/sparql/results/tsvresults.py b/rdflib/plugins/sparql/results/tsvresults.py index 1395eaff..bdfa2d4a 100644 --- a/rdflib/plugins/sparql/results/tsvresults.py +++ b/rdflib/plugins/sparql/results/tsvresults.py @@ -1,4 +1,3 @@ - """ This implements the Tab Separated SPARQL Result Format @@ -8,14 +7,28 @@ It is implemented with pyparsing, reusing the elements from the SPARQL Parser import codecs from pyparsing import ( - Optional, ZeroOrMore, Literal, ParserElement, ParseException, Suppress, - FollowedBy, LineEnd) + Optional, + ZeroOrMore, + Literal, + ParserElement, + ParseException, + Suppress, + FollowedBy, + LineEnd, +) from rdflib.query import Result, ResultParser from rdflib.plugins.sparql.parser import ( - Var, STRING_LITERAL1, STRING_LITERAL2, IRIREF, BLANK_NODE_LABEL, - NumericLiteral, BooleanLiteral, LANGTAG) + Var, + STRING_LITERAL1, + STRING_LITERAL2, + IRIREF, + BLANK_NODE_LABEL, + NumericLiteral, + BooleanLiteral, + LANGTAG, +) from rdflib.plugins.sparql.parserutils import Comp, Param, CompValue from rdflib import Literal as RDFLiteral @@ -25,10 +38,14 @@ ParserElement.setDefaultWhitespaceChars(" \n") String = STRING_LITERAL1 | STRING_LITERAL2 -RDFLITERAL = Comp('literal', Param('string', String) + Optional( - Param('lang', LANGTAG.leaveWhitespace() - ) | Literal('^^').leaveWhitespace( - ) + Param('datatype', IRIREF).leaveWhitespace())) +RDFLITERAL = Comp( + "literal", + Param("string", String) + + Optional( + Param("lang", LANGTAG.leaveWhitespace()) + | Literal("^^").leaveWhitespace() + Param("datatype", IRIREF).leaveWhitespace() + ), +) NONE_VALUE = object() @@ -49,10 +66,10 @@ class TSVResultParser(ResultParser): if isinstance(source.read(0), bytes): # if reading from source returns bytes do utf-8 decoding - source = codecs.getreader('utf-8')(source) + source = codecs.getreader("utf-8")(source) try: - r = Result('SELECT') + r = Result("SELECT") header = source.readline() @@ -62,13 +79,12 @@ class TSVResultParser(ResultParser): line = source.readline() if not line: break - line = line.strip('\n') + line = line.strip("\n") if line == "": continue row = ROW.parseString(line, parseAll=True) - r.bindings.append( - dict(zip(r.vars, (self.convertTerm(x) for x in row)))) + r.bindings.append(dict(zip(r.vars, (self.convertTerm(x) for x in row)))) return r @@ -81,7 +97,7 @@ class TSVResultParser(ResultParser): if t is NONE_VALUE: return None if isinstance(t, CompValue): - if t.name == 'literal': + if t.name == "literal": return RDFLiteral(t.string, lang=t.lang, datatype=t.datatype) else: raise Exception("I dont know how to handle this: %s" % (t,)) @@ -89,9 +105,10 @@ class TSVResultParser(ResultParser): return t -if __name__ == '__main__': +if __name__ == "__main__": import sys - r = Result.parse(file(sys.argv[1]), format='tsv') + + r = Result.parse(file(sys.argv[1]), format="tsv") print(r.vars) print(r.bindings) # print r.serialize(format='json') diff --git a/rdflib/plugins/sparql/results/txtresults.py b/rdflib/plugins/sparql/results/txtresults.py index c42f24c4..426dd9a1 100644 --- a/rdflib/plugins/sparql/results/txtresults.py +++ b/rdflib/plugins/sparql/results/txtresults.py @@ -1,4 +1,3 @@ - from rdflib import URIRef, BNode, Literal from rdflib.query import ResultSerializer @@ -37,7 +36,7 @@ class TXTResultSerializer(ResultSerializer): h2 += 1 return " " * h1 + s + " " * h2 - if self.result.type != 'SELECT': + if self.result.type != "SELECT": raise Exception("Can only pretty print SELECT results!") if not self.result: @@ -46,14 +45,17 @@ class TXTResultSerializer(ResultSerializer): keys = sorted(self.result.vars) maxlen = [0] * len(keys) - b = [[_termString(r[k], namespace_manager) for k in keys] for r in self.result] + b = [ + [_termString(r[k], namespace_manager) for k in keys] + for r in self.result + ] for r in b: for i in range(len(keys)): maxlen[i] = max(maxlen[i], len(r[i])) - stream.write( - "|".join([c(k, maxlen[i]) for i, k in enumerate(keys)]) + "\n") + stream.write("|".join([c(k, maxlen[i]) for i, k in enumerate(keys)]) + "\n") stream.write("-" * (len(maxlen) + sum(maxlen)) + "\n") for r in sorted(b): - stream.write("|".join( - [t + " " * (i - len(t)) for i, t in zip(maxlen, r)]) + "\n") + stream.write( + "|".join([t + " " * (i - len(t)) for i, t in zip(maxlen, r)]) + "\n" + ) diff --git a/rdflib/plugins/sparql/results/xmlresults.py b/rdflib/plugins/sparql/results/xmlresults.py index cdb81f76..aa4f796f 100644 --- a/rdflib/plugins/sparql/results/xmlresults.py +++ b/rdflib/plugins/sparql/results/xmlresults.py @@ -8,17 +8,11 @@ from xml.sax.xmlreader import AttributesNSImpl from rdflib.compat import etree from rdflib import Literal, URIRef, BNode, Graph, Variable -from rdflib.query import ( - Result, - ResultParser, - ResultSerializer, - ResultException -) +from rdflib.query import Result, ResultParser, ResultSerializer, ResultException - -SPARQL_XML_NAMESPACE = u'http://www.w3.org/2005/sparql-results#' -RESULTS_NS_ET = '{%s}' % SPARQL_XML_NAMESPACE +SPARQL_XML_NAMESPACE = u"http://www.w3.org/2005/sparql-results#" +RESULTS_NS_ET = "{%s}" % SPARQL_XML_NAMESPACE log = logging.getLogger(__name__) @@ -35,7 +29,6 @@ Authors: Drew Perttula, Gunnar Aastrand Grimnes class XMLResultParser(ResultParser): - def parse(self, source, content_type=None): return XMLResult(source) @@ -49,31 +42,32 @@ class XMLResult(Result): except TypeError: tree = etree.parse(source) - boolean = tree.find(RESULTS_NS_ET + 'boolean') - results = tree.find(RESULTS_NS_ET + 'results') + boolean = tree.find(RESULTS_NS_ET + "boolean") + results = tree.find(RESULTS_NS_ET + "results") if boolean is not None: - type_ = 'ASK' + type_ = "ASK" elif results is not None: - type_ = 'SELECT' + type_ = "SELECT" else: - raise ResultException( - "No RDF result-bindings or boolean answer found!") + raise ResultException("No RDF result-bindings or boolean answer found!") Result.__init__(self, type_) - if type_ == 'SELECT': + if type_ == "SELECT": self.bindings = [] for result in results: r = {} for binding in result: - r[Variable(binding.get('name'))] = parseTerm(binding[0]) + r[Variable(binding.get("name"))] = parseTerm(binding[0]) self.bindings.append(r) - self.vars = [Variable(x.get("name")) - for x in tree.findall( - './%shead/%svariable' % ( - RESULTS_NS_ET, RESULTS_NS_ET))] + self.vars = [ + Variable(x.get("name")) + for x in tree.findall( + "./%shead/%svariable" % (RESULTS_NS_ET, RESULTS_NS_ET) + ) + ] else: self.askAnswer = boolean.text.lower().strip() == "true" @@ -83,36 +77,35 @@ def parseTerm(element): """rdflib object (Literal, URIRef, BNode) for the given elementtree element""" tag, text = element.tag, element.text - if tag == RESULTS_NS_ET + 'literal': + if tag == RESULTS_NS_ET + "literal": if text is None: - text = '' + text = "" datatype = None lang = None - if element.get('datatype', None): - datatype = URIRef(element.get('datatype')) + if element.get("datatype", None): + datatype = URIRef(element.get("datatype")) elif element.get("{%s}lang" % XML_NAMESPACE, None): lang = element.get("{%s}lang" % XML_NAMESPACE) ret = Literal(text, datatype=datatype, lang=lang) return ret - elif tag == RESULTS_NS_ET + 'uri': + elif tag == RESULTS_NS_ET + "uri": return URIRef(text) - elif tag == RESULTS_NS_ET + 'bnode': + elif tag == RESULTS_NS_ET + "bnode": return BNode(text) else: raise TypeError("unknown binding type %r" % element) class XMLResultSerializer(ResultSerializer): - def __init__(self, result): ResultSerializer.__init__(self, result) def serialize(self, stream, encoding="utf-8"): writer = SPARQLXMLWriter(stream, encoding) - if self.result.type == 'ASK': + if self.result.type == "ASK": writer.write_header([]) writer.write_ask(self.result.askAnswer) else: @@ -134,14 +127,14 @@ class SPARQLXMLWriter: Python saxutils-based SPARQL XML Writer """ - def __init__(self, output, encoding='utf-8'): + def __init__(self, output, encoding="utf-8"): writer = XMLGenerator(output, encoding) writer.startDocument() - writer.startPrefixMapping(u'', SPARQL_XML_NAMESPACE) - writer.startPrefixMapping(u'xml', XML_NAMESPACE) + writer.startPrefixMapping(u"", SPARQL_XML_NAMESPACE) + writer.startPrefixMapping(u"xml", XML_NAMESPACE) writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'sparql'), - u'sparql', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"sparql"), u"sparql", AttributesNSImpl({}, {}) + ) self.writer = writer self._output = output self._encoding = encoding @@ -149,102 +142,99 @@ class SPARQLXMLWriter: def write_header(self, allvarsL): self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'head'), - u'head', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"head"), u"head", AttributesNSImpl({}, {}) + ) for i in range(0, len(allvarsL)): attr_vals = { - (None, u'name'): str(allvarsL[i]), + (None, u"name"): str(allvarsL[i]), } attr_qnames = { - (None, u'name'): u'name', + (None, u"name"): u"name", } self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'variable'), - u'variable', AttributesNSImpl(attr_vals, attr_qnames)) - self.writer.endElementNS((SPARQL_XML_NAMESPACE, - u'variable'), u'variable') - self.writer.endElementNS((SPARQL_XML_NAMESPACE, u'head'), u'head') + (SPARQL_XML_NAMESPACE, u"variable"), + u"variable", + AttributesNSImpl(attr_vals, attr_qnames), + ) + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"variable"), u"variable") + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"head"), u"head") def write_ask(self, val): self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'boolean'), - u'boolean', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"boolean"), u"boolean", AttributesNSImpl({}, {}) + ) self.writer.characters(str(val).lower()) - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'boolean'), u'boolean') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"boolean"), u"boolean") def write_results_header(self): self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'results'), - u'results', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"results"), u"results", AttributesNSImpl({}, {}) + ) self._results = True def write_start_result(self): self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'result'), - u'result', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"result"), u"result", AttributesNSImpl({}, {}) + ) self._resultStarted = True def write_end_result(self): assert self._resultStarted - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'result'), u'result') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"result"), u"result") self._resultStarted = False def write_binding(self, name, val): assert self._resultStarted attr_vals = { - (None, u'name'): str(name), + (None, u"name"): str(name), } attr_qnames = { - (None, u'name'): u'name', + (None, u"name"): u"name", } self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'binding'), - u'binding', AttributesNSImpl(attr_vals, attr_qnames)) + (SPARQL_XML_NAMESPACE, u"binding"), + u"binding", + AttributesNSImpl(attr_vals, attr_qnames), + ) if isinstance(val, URIRef): self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'uri'), - u'uri', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"uri"), u"uri", AttributesNSImpl({}, {}) + ) self.writer.characters(val) - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'uri'), u'uri') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"uri"), u"uri") elif isinstance(val, BNode): self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'bnode'), - u'bnode', AttributesNSImpl({}, {})) + (SPARQL_XML_NAMESPACE, u"bnode"), u"bnode", AttributesNSImpl({}, {}) + ) self.writer.characters(val) - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'bnode'), u'bnode') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"bnode"), u"bnode") elif isinstance(val, Literal): attr_vals = {} attr_qnames = {} if val.language: - attr_vals[(XML_NAMESPACE, u'lang')] = val.language - attr_qnames[(XML_NAMESPACE, u'lang')] = u"xml:lang" + attr_vals[(XML_NAMESPACE, u"lang")] = val.language + attr_qnames[(XML_NAMESPACE, u"lang")] = u"xml:lang" elif val.datatype: - attr_vals[(None, u'datatype')] = val.datatype - attr_qnames[(None, u'datatype')] = u'datatype' + attr_vals[(None, u"datatype")] = val.datatype + attr_qnames[(None, u"datatype")] = u"datatype" self.writer.startElementNS( - (SPARQL_XML_NAMESPACE, u'literal'), - u'literal', AttributesNSImpl(attr_vals, attr_qnames)) + (SPARQL_XML_NAMESPACE, u"literal"), + u"literal", + AttributesNSImpl(attr_vals, attr_qnames), + ) self.writer.characters(val) - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'literal'), u'literal') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"literal"), u"literal") else: raise Exception("Unsupported RDF term: %s" % val) - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'binding'), u'binding') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"binding"), u"binding") def close(self): if self._results: - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'results'), u'results') - self.writer.endElementNS( - (SPARQL_XML_NAMESPACE, u'sparql'), u'sparql') + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"results"), u"results") + self.writer.endElementNS((SPARQL_XML_NAMESPACE, u"sparql"), u"sparql") self.writer.endDocument() diff --git a/rdflib/plugins/sparql/sparql.py b/rdflib/plugins/sparql/sparql.py index d9d781db..5b6eab2e 100644 --- a/rdflib/plugins/sparql/sparql.py +++ b/rdflib/plugins/sparql/sparql.py @@ -129,8 +129,7 @@ class FrozenDict(Mapping): return self._hash def project(self, vars): - return FrozenDict( - (x for x in self.items() if x[0] in vars)) + return FrozenDict((x for x in self.items() if x[0] in vars)) def disjointDomain(self, other): return not bool(set(self).intersection(other)) @@ -146,8 +145,7 @@ class FrozenDict(Mapping): return True def merge(self, other): - res = FrozenDict( - itertools.chain(self.items(), other.items())) + res = FrozenDict(itertools.chain(self.items(), other.items())) return res @@ -159,7 +157,6 @@ class FrozenDict(Mapping): class FrozenBindings(FrozenDict): - def __init__(self, ctx, *args, **kwargs): FrozenDict.__init__(self, *args, **kwargs) self.ctx = ctx @@ -178,12 +175,10 @@ class FrozenBindings(FrozenDict): return self._d[key] def project(self, vars): - return FrozenBindings( - self.ctx, (x for x in self.items() if x[0] in vars)) + return FrozenBindings(self.ctx, (x for x in self.items() if x[0] in vars)) def merge(self, other): - res = FrozenBindings( - self.ctx, itertools.chain(self.items(), other.items())) + res = FrozenBindings(self.ctx, itertools.chain(self.items(), other.items())) return res @@ -210,18 +205,23 @@ class FrozenBindings(FrozenDict): # bindings from initBindings are newer forgotten return FrozenBindings( - self.ctx, ( - x for x in self.items() if ( - x[0] in _except or - x[0] in self.ctx.initBindings or - before[x[0]] is None))) + self.ctx, + ( + x + for x in self.items() + if ( + x[0] in _except + or x[0] in self.ctx.initBindings + or before[x[0]] is None + ) + ), + ) def remember(self, these): """ return a frozen dict only of bindings in these """ - return FrozenBindings( - self.ctx, (x for x in self.items() if x[0] in these)) + return FrozenBindings(self.ctx, (x for x in self.items() if x[0] in these)) class QueryContext(object): @@ -253,7 +253,10 @@ class QueryContext(object): def clone(self, bindings=None): r = QueryContext( - self._dataset if self._dataset is not None else self.graph, bindings or self.bindings, initBindings=self.initBindings) + self._dataset if self._dataset is not None else self.graph, + bindings or self.bindings, + initBindings=self.initBindings, + ) r.prologue = self.prologue r.graph = self.graph r.bnodes = self.bnodes @@ -262,30 +265,30 @@ class QueryContext(object): def _get_dataset(self): if self._dataset is None: raise Exception( - 'You performed a query operation requiring ' + - 'a dataset (i.e. ConjunctiveGraph), but ' + - 'operating currently on a single graph.') + "You performed a query operation requiring " + + "a dataset (i.e. ConjunctiveGraph), but " + + "operating currently on a single graph." + ) return self._dataset dataset = property(_get_dataset, doc="current dataset") def load(self, source, default=False, **kwargs): - def _load(graph, source): try: return graph.load(source, **kwargs) except: pass try: - return graph.load(source, format='n3', **kwargs) + return graph.load(source, format="n3", **kwargs) except: pass try: - return graph.load(source, format='nt', **kwargs) + return graph.load(source, format="nt", **kwargs) except: raise Exception( - "Could not load %s as either RDF/XML, N3 or NTriples" % ( - source)) + "Could not load %s as either RDF/XML, N3 or NTriples" % (source) + ) if not rdflib.plugins.sparql.SPARQL_LOAD_GRAPHS: # we are not loading - if we already know the graph @@ -320,9 +323,8 @@ class QueryContext(object): """ if vars: return FrozenBindings( - self, ((k, v) - for k, v in self.bindings.items() - if k in vars)) + self, ((k, v) for k, v in self.bindings.items() if k in vars) + ) else: return FrozenBindings(self, self.bindings.items()) @@ -366,13 +368,12 @@ class Prologue(object): def __init__(self): self.base = None - self.namespace_manager = NamespaceManager( - Graph()) # ns man needs a store + self.namespace_manager = NamespaceManager(Graph()) # ns man needs a store def resolvePName(self, prefix, localname): ns = self.namespace_manager.store.namespace(prefix or "") if ns is None: - raise Exception('Unknown namespace prefix : %s' % prefix) + raise Exception("Unknown namespace prefix : %s" % prefix) return URIRef(ns + (localname or "")) def bind(self, prefix, uri): @@ -387,13 +388,13 @@ class Prologue(object): """ if isinstance(iri, CompValue): - if iri.name == 'pname': + if iri.name == "pname": return self.resolvePName(iri.prefix, iri.localname) - if iri.name == 'literal': + if iri.name == "literal": return Literal( - iri.string, lang=iri.lang, - datatype=self.absolutize(iri.datatype)) - elif isinstance(iri, URIRef) and not ':' in iri: + iri.string, lang=iri.lang, datatype=self.absolutize(iri.datatype) + ) + elif isinstance(iri, URIRef) and not ":" in iri: return URIRef(iri, base=self.base) return iri diff --git a/rdflib/plugins/sparql/update.py b/rdflib/plugins/sparql/update.py index 44ea40a3..f979c387 100644 --- a/rdflib/plugins/sparql/update.py +++ b/rdflib/plugins/sparql/update.py @@ -11,7 +11,7 @@ from rdflib.plugins.sparql.evaluate import evalBGP, evalPart def _graphOrDefault(ctx, g): - if g == 'DEFAULT': + if g == "DEFAULT": return ctx.graph else: return ctx.dataset.get_context(g) @@ -21,12 +21,13 @@ def _graphAll(ctx, g): """ return a list of graphs """ - if g == 'DEFAULT': + if g == "DEFAULT": return [ctx.graph] - elif g == 'NAMED': - return [c for c in ctx.dataset.contexts() - if c.identifier != ctx.graph.identifier] - elif g == 'ALL': + elif g == "NAMED": + return [ + c for c in ctx.dataset.contexts() if c.identifier != ctx.graph.identifier + ] + elif g == "ALL": return list(ctx.dataset.contexts()) else: return [ctx.dataset.get_context(g)] @@ -280,30 +281,30 @@ def evalUpdate(graph, update, initBindings={}): ctx.prologue = u.prologue try: - if u.name == 'Load': + if u.name == "Load": evalLoad(ctx, u) - elif u.name == 'Clear': + elif u.name == "Clear": evalClear(ctx, u) - elif u.name == 'Drop': + elif u.name == "Drop": evalDrop(ctx, u) - elif u.name == 'Create': + elif u.name == "Create": evalCreate(ctx, u) - elif u.name == 'Add': + elif u.name == "Add": evalAdd(ctx, u) - elif u.name == 'Move': + elif u.name == "Move": evalMove(ctx, u) - elif u.name == 'Copy': + elif u.name == "Copy": evalCopy(ctx, u) - elif u.name == 'InsertData': + elif u.name == "InsertData": evalInsertData(ctx, u) - elif u.name == 'DeleteData': + elif u.name == "DeleteData": evalDeleteData(ctx, u) - elif u.name == 'DeleteWhere': + elif u.name == "DeleteWhere": evalDeleteWhere(ctx, u) - elif u.name == 'Modify': + elif u.name == "Modify": evalModify(ctx, u) else: - raise Exception('Unknown update operation: %s' % (u,)) + raise Exception("Unknown update operation: %s" % (u,)) except: if not u.silent: raise diff --git a/rdflib/plugins/stores/auditable.py b/rdflib/plugins/stores/auditable.py index 7a3492b7..ff21716b 100644 --- a/rdflib/plugins/stores/auditable.py +++ b/rdflib/plugins/stores/auditable.py @@ -20,8 +20,8 @@ from rdflib import Graph, ConjunctiveGraph import threading destructiveOpLocks = { - 'add': None, - 'remove': None, + "add": None, + "remove": None, } @@ -50,59 +50,79 @@ class AuditableStore(Store): def add(self, triple, context, quoted=False): (s, p, o) = triple - lock = destructiveOpLocks['add'] + lock = destructiveOpLocks["add"] lock = lock if lock else threading.RLock() with lock: - context = context.__class__(self.store, context.identifier) if context is not None else None + context = ( + context.__class__(self.store, context.identifier) + if context is not None + else None + ) ctxId = context.identifier if context is not None else None if list(self.store.triples(triple, context)): return # triple already in store, do nothing - self.reverseOps.append((s, p, o, ctxId, 'remove')) + self.reverseOps.append((s, p, o, ctxId, "remove")) try: - self.reverseOps.remove((s, p, o, ctxId, 'add')) + self.reverseOps.remove((s, p, o, ctxId, "add")) except ValueError: pass self.store.add((s, p, o), context, quoted) def remove(self, spo, context=None): subject, predicate, object_ = spo - lock = destructiveOpLocks['remove'] + lock = destructiveOpLocks["remove"] lock = lock if lock else threading.RLock() with lock: # Need to determine which quads will be removed if any term is a # wildcard - context = context.__class__(self.store, context.identifier) if context is not None else None + context = ( + context.__class__(self.store, context.identifier) + if context is not None + else None + ) ctxId = context.identifier if context is not None else None if None in [subject, predicate, object_, context]: if ctxId: for s, p, o in context.triples((subject, predicate, object_)): try: - self.reverseOps.remove((s, p, o, ctxId, 'remove')) + self.reverseOps.remove((s, p, o, ctxId, "remove")) except ValueError: - self.reverseOps.append((s, p, o, ctxId, 'add')) + self.reverseOps.append((s, p, o, ctxId, "add")) else: - for s, p, o, ctx in ConjunctiveGraph(self.store).quads((subject, predicate, object_)): + for s, p, o, ctx in ConjunctiveGraph(self.store).quads( + (subject, predicate, object_) + ): try: - self.reverseOps.remove((s, p, o, ctx.identifier, 'remove')) + self.reverseOps.remove((s, p, o, ctx.identifier, "remove")) except ValueError: - self.reverseOps.append((s, p, o, ctx.identifier, 'add')) + self.reverseOps.append((s, p, o, ctx.identifier, "add")) else: if not list(self.triples((subject, predicate, object_), context)): return # triple not present in store, do nothing try: - self.reverseOps.remove((subject, predicate, object_, ctxId, 'remove')) + self.reverseOps.remove( + (subject, predicate, object_, ctxId, "remove") + ) except ValueError: - self.reverseOps.append((subject, predicate, object_, ctxId, 'add')) + self.reverseOps.append((subject, predicate, object_, ctxId, "add")) self.store.remove((subject, predicate, object_), context) def triples(self, triple, context=None): (su, pr, ob) = triple - context = context.__class__(self.store, context.identifier) if context is not None else None + context = ( + context.__class__(self.store, context.identifier) + if context is not None + else None + ) for (s, p, o), cg in self.store.triples((su, pr, ob), context): yield (s, p, o), cg def __len__(self, context=None): - context = context.__class__(self.store, context.identifier) if context is not None else None + context = ( + context.__class__(self.store, context.identifier) + if context is not None + else None + ) return self.store.__len__(context) def contexts(self, triple=None): @@ -129,11 +149,13 @@ class AuditableStore(Store): # order with self.rollbackLock: for subject, predicate, obj, context, op in self.reverseOps: - if op == 'add': + if op == "add": self.store.add( - (subject, predicate, obj), Graph(self.store, context)) + (subject, predicate, obj), Graph(self.store, context) + ) else: self.store.remove( - (subject, predicate, obj), Graph(self.store, context)) + (subject, predicate, obj), Graph(self.store, context) + ) self.reverseOps = [] diff --git a/rdflib/plugins/stores/concurrent.py b/rdflib/plugins/stores/concurrent.py index 40747fb1..a258e778 100644 --- a/rdflib/plugins/stores/concurrent.py +++ b/rdflib/plugins/stores/concurrent.py @@ -4,7 +4,7 @@ from threading import Lock class ResponsibleGenerator(object): """A generator that will help clean up when it is done being used.""" - __slots__ = ['cleanup', 'gen'] + __slots__ = ["cleanup", "gen"] def __init__(self, gen, cleanup): self.cleanup = cleanup @@ -21,7 +21,6 @@ class ResponsibleGenerator(object): class ConcurrentStore(object): - def __init__(self, store): self.store = store @@ -60,9 +59,11 @@ class ConcurrentStore(object): yield s, p, o for (s, p, o) in self.__pending_adds: - if (su is None or su == s) \ - and (pr is None or pr == p) \ - and (ob is None or ob == o): + if ( + (su is None or su == s) + and (pr is None or pr == p) + and (ob is None or ob == o) + ): yield s, p, o def __len__(self): diff --git a/rdflib/plugins/stores/regexmatching.py b/rdflib/plugins/stores/regexmatching.py index 55c1e671..f890405d 100644 --- a/rdflib/plugins/stores/regexmatching.py +++ b/rdflib/plugins/stores/regexmatching.py @@ -12,7 +12,6 @@ from rdflib.store import Store from rdflib.graph import Graph - import re # Store is capable of doing its own REGEX matching @@ -32,13 +31,14 @@ class REGEXTerm(str): self.compiledExpr = re.compile(expr) def __reduce__(self): - return (REGEXTerm, (str(''),)) + return (REGEXTerm, (str(""),)) def regexCompareQuad(quad, regexQuad): for index in range(4): - if isinstance(regexQuad[index], REGEXTerm) and not \ - regexQuad[index].compiledExpr.match(quad[index]): + if isinstance(regexQuad[index], REGEXTerm) and not regexQuad[ + index + ].compiledExpr.match(quad[index]): return False return True @@ -67,29 +67,36 @@ class REGEXMatching(Store): def remove(self, triple, context=None): (subject, predicate, object_) = triple - if isinstance(subject, REGEXTerm) or \ - isinstance(predicate, REGEXTerm) or \ - isinstance(object_, REGEXTerm) or \ - (context is not None and - isinstance(context.identifier, REGEXTerm)): + if ( + isinstance(subject, REGEXTerm) + or isinstance(predicate, REGEXTerm) + or isinstance(object_, REGEXTerm) + or (context is not None and isinstance(context.identifier, REGEXTerm)) + ): # One or more of the terms is a REGEX expression, so we must # replace it / them with wildcard(s)and match after we query. s = not isinstance(subject, REGEXTerm) and subject or None p = not isinstance(predicate, REGEXTerm) and predicate or None o = not isinstance(object_, REGEXTerm) and object_ or None - c = (context is not None and - not isinstance(context.identifier, REGEXTerm)) \ - and context \ + c = ( + (context is not None and not isinstance(context.identifier, REGEXTerm)) + and context or None + ) removeQuadList = [] for (s1, p1, o1), cg in self.storage.triples((s, p, o), c): for ctx in cg: ctx = ctx.identifier if regexCompareQuad( - (s1, p1, o1, ctx), - (subject, predicate, object_, context - is not None and context.identifier or context)): + (s1, p1, o1, ctx), + ( + subject, + predicate, + object_, + context is not None and context.identifier or context, + ), + ): removeQuadList.append((s1, p1, o1, ctx)) for s, p, o, c in removeQuadList: self.storage.remove((s, p, o), c and Graph(self, c) or c) @@ -98,37 +105,40 @@ class REGEXMatching(Store): def triples(self, triple, context=None): (subject, predicate, object_) = triple - if isinstance(subject, REGEXTerm) or \ - isinstance(predicate, REGEXTerm) or \ - isinstance(object_, REGEXTerm) or \ - (context is not None and - isinstance(context.identifier, REGEXTerm)): + if ( + isinstance(subject, REGEXTerm) + or isinstance(predicate, REGEXTerm) + or isinstance(object_, REGEXTerm) + or (context is not None and isinstance(context.identifier, REGEXTerm)) + ): # One or more of the terms is a REGEX expression, so we must # replace it / them with wildcard(s) and match after we query. s = not isinstance(subject, REGEXTerm) and subject or None p = not isinstance(predicate, REGEXTerm) and predicate or None o = not isinstance(object_, REGEXTerm) and object_ or None - c = (context is not None and - not isinstance(context.identifier, REGEXTerm)) \ - and context \ + c = ( + (context is not None and not isinstance(context.identifier, REGEXTerm)) + and context or None + ) for (s1, p1, o1), cg in self.storage.triples((s, p, o), c): matchingCtxs = [] for ctx in cg: if c is None: - if context is None \ - or context.identifier.compiledExpr.match( - ctx.identifier): + if context is None or context.identifier.compiledExpr.match( + ctx.identifier + ): matchingCtxs.append(ctx) else: matchingCtxs.append(ctx) - if matchingCtxs \ - and regexCompareQuad((s1, p1, o1, None), - (subject, predicate, object_, None)): + if matchingCtxs and regexCompareQuad( + (s1, p1, o1, None), (subject, predicate, object_, None) + ): yield (s1, p1, o1), (c for c in matchingCtxs) else: for (s1, p1, o1), cg in self.storage.triples( - (subject, predicate, object_), context): + (subject, predicate, object_), context + ): yield (s1, p1, o1), cg def __len__(self, context=None): diff --git a/rdflib/plugins/stores/sparqlconnector.py b/rdflib/plugins/stores/sparqlconnector.py index abb69a55..abec85a8 100644 --- a/rdflib/plugins/stores/sparqlconnector.py +++ b/rdflib/plugins/stores/sparqlconnector.py @@ -14,13 +14,14 @@ log = logging.getLogger(__name__) class SPARQLConnectorException(Exception): pass + # TODO: Pull in these from the result implementation plugins? _response_mime_types = { - 'xml': 'application/sparql-results+xml, application/rdf+xml', - 'json': 'application/sparql-results+json', - 'csv': 'text/csv', - 'tsv': 'text/tab-separated-values', - 'application/rdf+xml': 'application/rdf+xml', + "xml": "application/sparql-results+xml, application/rdf+xml", + "json": "application/sparql-results+json", + "csv": "text/csv", + "tsv": "text/tab-separated-values", + "application/rdf+xml": "application/rdf+xml", } @@ -30,7 +31,14 @@ class SPARQLConnector(object): this class deals with nitty gritty details of talking to a SPARQL server """ - def __init__(self, query_endpoint=None, update_endpoint=None, returnFormat='xml', method='GET', **kwargs): + def __init__( + self, + query_endpoint=None, + update_endpoint=None, + returnFormat="xml", + method="GET", + **kwargs + ): """ Any additional keyword arguments will be passed to requests, and can be used to setup timesouts, basic auth, etc. """ @@ -48,9 +56,9 @@ class SPARQLConnector(object): @property def session(self): - k = 'session_%d' % os.getpid() + k = "session_%d" % os.getpid() self._session.__dict__.setdefault(k, requests.Session()) - log.debug('Session %s %s', os.getpid(), id(self._session.__dict__[k])) + log.debug("Session %s %s", os.getpid(), id(self._session.__dict__[k])) return self._session.__dict__[k] @property @@ -59,7 +67,7 @@ class SPARQLConnector(object): @method.setter def method(self, method): - if method not in ('GET', 'POST'): + if method not in ("GET", "POST"): raise SPARQLConnectorException('Method must be "GET" or "POST"') self._method = method @@ -69,26 +77,26 @@ class SPARQLConnector(object): if not self.query_endpoint: raise SPARQLConnectorException("Query endpoint not set!") - params = {'query': query} + params = {"query": query} if default_graph: params["default-graph-uri"] = default_graph - headers = {'Accept': _response_mime_types[self.returnFormat]} + headers = {"Accept": _response_mime_types[self.returnFormat]} args = dict(self.kwargs) args.update(url=self.query_endpoint) # merge params/headers dicts - args.setdefault('params', {}) + args.setdefault("params", {}) - args.setdefault('headers', {}) - args['headers'].update(headers) + args.setdefault("headers", {}) + args["headers"].update(headers) - if self.method == 'GET': - args['params'].update(params) - elif self.method == 'POST': - args['headers'].update({'Content-Type': 'application/sparql-query'}) - args['data'] = params + if self.method == "GET": + args["params"].update(params) + elif self.method == "POST": + args["headers"].update({"Content-Type": "application/sparql-query"}) + args["data"] = params else: raise SPARQLConnectorException("Unknown method %s" % self.method) @@ -96,7 +104,9 @@ class SPARQLConnector(object): res.raise_for_status() - return Result.parse(BytesIO(res.content), content_type=res.headers['Content-type']) + return Result.parse( + BytesIO(res.content), content_type=res.headers["Content-type"] + ) def update(self, update, default_graph=None): if not self.update_endpoint: @@ -108,20 +118,19 @@ class SPARQLConnector(object): params["using-graph-uri"] = default_graph headers = { - 'Accept': _response_mime_types[self.returnFormat], - 'Content-Type': 'application/sparql-update', + "Accept": _response_mime_types[self.returnFormat], + "Content-Type": "application/sparql-update", } args = dict(self.kwargs) - args.update(url=self.update_endpoint, - data=update.encode('utf-8')) + args.update(url=self.update_endpoint, data=update.encode("utf-8")) # merge params/headers dicts - args.setdefault('params', {}) - args['params'].update(params) - args.setdefault('headers', {}) - args['headers'].update(headers) + args.setdefault("params", {}) + args["params"].update(params) + args.setdefault("headers", {}) + args["headers"].update(headers) res = self.session.post(**args) diff --git a/rdflib/plugins/stores/sparqlstore.py b/rdflib/plugins/stores/sparqlstore.py index 989b0126..63e92f54 100644 --- a/rdflib/plugins/stores/sparqlstore.py +++ b/rdflib/plugins/stores/sparqlstore.py @@ -7,9 +7,9 @@ This was first done in layer-cake, and then ported to RDFLib """ # Defines some SPARQL keywords -LIMIT = 'LIMIT' -OFFSET = 'OFFSET' -ORDERBY = 'ORDER BY' +LIMIT = "LIMIT" +OFFSET = "OFFSET" +ORDERBY = "ORDER BY" import re import collections @@ -24,7 +24,7 @@ from rdflib.graph import DATASET_DEFAULT_GRAPH_ID from rdflib.term import Node -BNODE_IDENT_PATTERN = re.compile('(?P<label>_\:[^\s]+)') +BNODE_IDENT_PATTERN = re.compile("(?P<label>_\:[^\s]+)") def _node_to_sparql(node): @@ -87,21 +87,26 @@ class SPARQLStore(SPARQLConnector, Store): will use HTTP basic auth. """ + formula_aware = False transaction_aware = False graph_aware = True regex_matching = NATIVE_REGEX - def __init__(self, - endpoint=None, - sparql11=True, context_aware=True, - node_to_sparql=_node_to_sparql, - returnFormat='xml', - **sparqlconnector_kwargs): + def __init__( + self, + endpoint=None, + sparql11=True, + context_aware=True, + node_to_sparql=_node_to_sparql, + returnFormat="xml", + **sparqlconnector_kwargs + ): """ """ super(SPARQLStore, self).__init__( - endpoint, returnFormat=returnFormat, **sparqlconnector_kwargs) + endpoint, returnFormat=returnFormat, **sparqlconnector_kwargs + ) self.node_to_sparql = node_to_sparql self.nsBindings = {} @@ -112,7 +117,7 @@ class SPARQLStore(SPARQLConnector, Store): # Database Management Methods def create(self, configuration): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def open(self, configuration, create=False): """ @@ -125,23 +130,23 @@ class SPARQLStore(SPARQLConnector, Store): self.query_endpoint = configuration def destroy(self, configuration): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") # Transactional interfaces def commit(self): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def rollback(self): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def add(self, _, context=None, quoted=False): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def addN(self, quads): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def remove(self, _, context): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def _query(self, *args, **kwargs): self._queries += 1 @@ -152,20 +157,18 @@ class SPARQLStore(SPARQLConnector, Store): bindings = list(self.nsBindings.items()) + list(extra_bindings.items()) if not bindings: return query - return '\n'.join([ - '\n'.join(['PREFIX %s: <%s>' % (k, v) for k, v in bindings]), - '', # separate ns_bindings from query with an empty line - query - ]) + return "\n".join( + [ + "\n".join(["PREFIX %s: <%s>" % (k, v) for k, v in bindings]), + "", # separate ns_bindings from query with an empty line + query, + ] + ) def _preprocess_query(self, query): return self._inject_prefixes(query) - def query(self, query, - initNs={}, - initBindings={}, - queryGraph=None, - DEBUG=False): + def query(self, query, initNs={}, initBindings={}, queryGraph=None, DEBUG=False): self.debug = DEBUG assert isinstance(query, str) @@ -173,17 +176,18 @@ class SPARQLStore(SPARQLConnector, Store): if initBindings: if not self.sparql11: - raise Exception( - "initBindings not supported for SPARQL 1.0 Endpoints.") + raise Exception("initBindings not supported for SPARQL 1.0 Endpoints.") v = list(initBindings) # VALUES was added to SPARQL 1.1 on 2012/07/24 - query += "\nVALUES ( %s )\n{ ( %s ) }\n"\ - % (" ".join("?" + str(x) for x in v), - " ".join(self.node_to_sparql(initBindings[x]) for x in v)) + query += "\nVALUES ( %s )\n{ ( %s ) }\n" % ( + " ".join("?" + str(x) for x in v), + " ".join(self.node_to_sparql(initBindings[x]) for x in v), + ) - return self._query(query, - default_graph=queryGraph if self._is_contextual(queryGraph) else None) + return self._query( + query, default_graph=queryGraph if self._is_contextual(queryGraph) else None + ) def triples(self, spo, context=None): """ @@ -225,28 +229,31 @@ class SPARQLStore(SPARQLConnector, Store): vars = [] if not s: - s = Variable('s') + s = Variable("s") vars.append(s) if not p: - p = Variable('p') + p = Variable("p") vars.append(p) if not o: - o = Variable('o') + o = Variable("o") vars.append(o) if vars: - v = ' '.join([term.n3() for term in vars]) - verb = 'SELECT %s ' % v + v = " ".join([term.n3() for term in vars]) + verb = "SELECT %s " % v else: - verb = 'ASK' + verb = "ASK" nts = self.node_to_sparql query = "%s { %s %s %s }" % (verb, nts(s), nts(p), nts(o)) # The ORDER BY is necessary - if hasattr(context, LIMIT) or hasattr(context, OFFSET) \ - or hasattr(context, ORDERBY): + if ( + hasattr(context, LIMIT) + or hasattr(context, OFFSET) + or hasattr(context, ORDERBY) + ): var = None if isinstance(s, Variable): var = s @@ -254,28 +261,33 @@ class SPARQLStore(SPARQLConnector, Store): var = p elif isinstance(o, Variable): var = o - elif hasattr(context, ORDERBY) \ - and isinstance(getattr(context, ORDERBY), Variable): + elif hasattr(context, ORDERBY) and isinstance( + getattr(context, ORDERBY), Variable + ): var = getattr(context, ORDERBY) - query = query + ' %s %s' % (ORDERBY, var.n3()) + query = query + " %s %s" % (ORDERBY, var.n3()) try: - query = query + ' LIMIT %s' % int(getattr(context, LIMIT)) + query = query + " LIMIT %s" % int(getattr(context, LIMIT)) except (ValueError, TypeError, AttributeError): pass try: - query = query + ' OFFSET %s' % int(getattr(context, OFFSET)) + query = query + " OFFSET %s" % int(getattr(context, OFFSET)) except (ValueError, TypeError, AttributeError): pass - result = self._query(query, - default_graph=context.identifier if self._is_contextual(context) else None) + result = self._query( + query, + default_graph=context.identifier if self._is_contextual(context) else None, + ) if vars: for row in result: - yield (row.get(s, s), - row.get(p, p), - row.get(o, o)), None # why is the context here not the passed in graph 'context'? + yield ( + row.get(s, s), + row.get(p, p), + row.get(o, o), + ), None # why is the context here not the passed in graph 'context'? else: if result.askAnswer: yield (s, p, o), None @@ -288,18 +300,23 @@ class SPARQLStore(SPARQLConnector, Store): which will iterate over each term in the list and dispatch to triples. """ - raise NotImplementedError('Triples choices currently not supported') + raise NotImplementedError("Triples choices currently not supported") def __len__(self, context=None): if not self.sparql11: raise NotImplementedError( - "For performance reasons, this is not" + - "supported for sparql1.0 endpoints") + "For performance reasons, this is not" + + "supported for sparql1.0 endpoints" + ) else: q = "SELECT (count(*) as ?c) WHERE {?s ?p ?o .}" - result = self._query(q, - default_graph=context.identifier if self._is_contextual(context) else None) + result = self._query( + q, + default_graph=context.identifier + if self._is_contextual(context) + else None, + ) return int(next(iter(result)).c) @@ -321,12 +338,14 @@ class SPARQLStore(SPARQLConnector, Store): if triple: nts = self.node_to_sparql s, p, o = triple - params = (nts(s if s else Variable('s')), - nts(p if p else Variable('p')), - nts(o if o else Variable('o'))) - q = 'SELECT ?name WHERE { GRAPH ?name { %s %s %s }}' % params + params = ( + nts(s if s else Variable("s")), + nts(p if p else Variable("p")), + nts(o if o else Variable("o")), + ) + q = "SELECT ?name WHERE { GRAPH ?name { %s %s %s }}" % params else: - q = 'SELECT ?name WHERE { GRAPH ?name {} }' + q = "SELECT ?name WHERE { GRAPH ?name {} }" result = self._query(q) @@ -338,9 +357,7 @@ class SPARQLStore(SPARQLConnector, Store): def prefix(self, namespace): """ """ - return dict( - [(v, k) for k, v in self.nsBindings.items()] - ).get(namespace) + return dict([(v, k) for k, v in self.nsBindings.items()]).get(namespace) def namespace(self, prefix): return self.nsBindings.get(prefix) @@ -350,10 +367,10 @@ class SPARQLStore(SPARQLConnector, Store): yield prefix, ns def add_graph(self, graph): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def remove_graph(self, graph): - raise TypeError('The SPARQL store is read only') + raise TypeError("The SPARQL store is read only") def _is_contextual(self, graph): """ Returns `True` if the "GRAPH" keyword must appear @@ -362,7 +379,7 @@ class SPARQLStore(SPARQLConnector, Store): if (not self.context_aware) or (graph is None): return False if isinstance(graph, str): - return graph != '__UNION__' + return graph != "__UNION__" else: return graph.identifier != DATASET_DEFAULT_GRAPH_ID @@ -414,19 +431,27 @@ class SPARQLUpdateStore(SPARQLStore): STRING_LITERAL2 = u'"([^"\\\\]|\\\\.)*"' STRING_LITERAL_LONG1 = u"'''(('|'')?([^'\\\\]|\\\\.))*'''" STRING_LITERAL_LONG2 = u'"""(("|"")?([^"\\\\]|\\\\.))*"""' - String = u'(%s)|(%s)|(%s)|(%s)' % (STRING_LITERAL1, STRING_LITERAL2, - STRING_LITERAL_LONG1, STRING_LITERAL_LONG2) + String = u"(%s)|(%s)|(%s)|(%s)" % ( + STRING_LITERAL1, + STRING_LITERAL2, + STRING_LITERAL_LONG1, + STRING_LITERAL_LONG2, + ) IRIREF = u'<([^<>"{}|^`\\]\\\\\[\\x00-\\x20])*>' - COMMENT = u'#[^\\x0D\\x0A]*([\\x0D\\x0A]|\\Z)' + COMMENT = u"#[^\\x0D\\x0A]*([\\x0D\\x0A]|\\Z)" # Simplified grammar to find { at beginning and } at end of blocks - BLOCK_START = u'{' - BLOCK_END = u'}' - ESCAPED = u'\\\\.' + BLOCK_START = u"{" + BLOCK_END = u"}" + ESCAPED = u"\\\\." # Match anything that doesn't start or end a block: - BlockContent = u'(%s)|(%s)|(%s)|(%s)' % (String, IRIREF, COMMENT, ESCAPED) - BlockFinding = u'(?P<block_start>%s)|(?P<block_end>%s)|(?P<block_content>%s)' % (BLOCK_START, BLOCK_END, BlockContent) + BlockContent = u"(%s)|(%s)|(%s)|(%s)" % (String, IRIREF, COMMENT, ESCAPED) + BlockFinding = u"(?P<block_start>%s)|(?P<block_end>%s)|(?P<block_content>%s)" % ( + BLOCK_START, + BLOCK_END, + BlockContent, + ) BLOCK_FINDING_PATTERN = re.compile(BlockFinding) # Note that BLOCK_FINDING_PATTERN.finditer() will not cover the whole @@ -435,15 +460,17 @@ class SPARQLUpdateStore(SPARQLStore): ################################################################## - def __init__(self, - queryEndpoint=None, update_endpoint=None, - sparql11=True, - context_aware=True, - postAsEncoded=True, - autocommit=True, - dirty_reads=False, - **kwds - ): + def __init__( + self, + queryEndpoint=None, + update_endpoint=None, + sparql11=True, + context_aware=True, + postAsEncoded=True, + autocommit=True, + dirty_reads=False, + **kwds + ): """ :param autocommit if set, the store will commit after every writing operations. If False, we only make queries on the @@ -521,7 +548,7 @@ class SPARQLUpdateStore(SPARQLStore): and reads can degenerate to the original call-per-triple situation that originally existed. """ if self._edits and len(self._edits) > 0: - self._update('\n;\n'.join(self._edits)) + self._update("\n;\n".join(self._edits)) self._edits = None def rollback(self): @@ -539,8 +566,7 @@ class SPARQLUpdateStore(SPARQLStore): nts = self.node_to_sparql triple = "%s %s %s ." % (nts(subject), nts(predicate), nts(obj)) if self._is_contextual(context): - q = "INSERT DATA { GRAPH %s { %s } }" % ( - nts(context.identifier), triple) + q = "INSERT DATA { GRAPH %s { %s } }" % (nts(context.identifier), triple) else: q = "INSERT DATA { %s }" % triple self._transaction().append(q) @@ -559,12 +585,13 @@ class SPARQLUpdateStore(SPARQLStore): nts = self.node_to_sparql for context in contexts: triples = [ - "%s %s %s ." % ( - nts(subject), nts(predicate), nts(obj) - ) for subject, predicate, obj in contexts[context] + "%s %s %s ." % (nts(subject), nts(predicate), nts(obj)) + for subject, predicate, obj in contexts[context] ] - data.append("INSERT DATA { GRAPH %s { %s } }\n" % ( - nts(context.identifier), '\n'.join(triples))) + data.append( + "INSERT DATA { GRAPH %s { %s } }\n" + % (nts(context.identifier), "\n".join(triples)) + ) self._transaction().extend(data) if self.autocommit: self.commit() @@ -586,7 +613,10 @@ class SPARQLUpdateStore(SPARQLStore): triple = "%s %s %s ." % (nts(subject), nts(predicate), nts(obj)) if self._is_contextual(context): cid = nts(context.identifier) - q = "WITH %(graph)s DELETE { %(triple)s } WHERE { %(triple)s }" % { 'graph': cid, 'triple': triple } + q = "WITH %(graph)s DELETE { %(triple)s } WHERE { %(triple)s }" % { + "graph": cid, + "triple": triple, + } else: q = "DELETE { %s } WHERE { %s } " % (triple, triple) self._transaction().append(q) @@ -602,11 +632,7 @@ class SPARQLUpdateStore(SPARQLStore): SPARQLConnector.update(self, update) - def update(self, query, - initNs={}, - initBindings={}, - queryGraph=None, - DEBUG=False): + def update(self, query, initNs={}, initBindings={}, queryGraph=None, DEBUG=False): """ Perform a SPARQL Update Query against the endpoint, INSERT, LOAD, DELETE etc. @@ -656,9 +682,10 @@ class SPARQLUpdateStore(SPARQLStore): # have a WHERE clause. This also works for updates with # more than one INSERT/DELETE. v = list(initBindings) - values = "\nVALUES ( %s )\n{ ( %s ) }\n"\ - % (" ".join("?" + str(x) for x in v), - " ".join(self.node_to_sparql(initBindings[x]) for x in v)) + values = "\nVALUES ( %s )\n{ ( %s ) }\n" % ( + " ".join("?" + str(x) for x in v), + " ".join(self.node_to_sparql(initBindings[x]) for x in v), + ) query = self.where_pattern.sub("WHERE { " + values, query) @@ -677,7 +704,7 @@ class SPARQLUpdateStore(SPARQLStore): if isinstance(query_graph, Node): query_graph = self.node_to_sparql(query_graph) else: - query_graph = '<%s>' % query_graph + query_graph = "<%s>" % query_graph graph_block_open = " GRAPH %s {" % query_graph graph_block_close = "} " @@ -696,16 +723,18 @@ class SPARQLUpdateStore(SPARQLStore): modified_query = [] pos = 0 for match in self.BLOCK_FINDING_PATTERN.finditer(query): - if match.group('block_start') is not None: + if match.group("block_start") is not None: level += 1 if level == 1: - modified_query.append(query[pos:match.end()]) + modified_query.append(query[pos : match.end()]) modified_query.append(graph_block_open) pos = match.end() - elif match.group('block_end') is not None: + elif match.group("block_end") is not None: if level == 1: - since_previous_pos = query[pos:match.start()] - if modified_query[-1] is graph_block_open and (since_previous_pos == "" or since_previous_pos.isspace()): + since_previous_pos = query[pos : match.start()] + if modified_query[-1] is graph_block_open and ( + since_previous_pos == "" or since_previous_pos.isspace() + ): # In this case, adding graph_block_start and # graph_block_end results in an empty GRAPH block. Some # enpoints (e.g. TDB) can not handle this. Therefore @@ -725,8 +754,7 @@ class SPARQLUpdateStore(SPARQLStore): if not self.graph_aware: Store.add_graph(self, graph) elif graph.identifier != DATASET_DEFAULT_GRAPH_ID: - self.update( - "CREATE GRAPH %s" % self.node_to_sparql(graph.identifier)) + self.update("CREATE GRAPH %s" % self.node_to_sparql(graph.identifier)) def close(self, commit_pending_transaction=False): @@ -741,5 +769,4 @@ class SPARQLUpdateStore(SPARQLStore): elif graph.identifier == DATASET_DEFAULT_GRAPH_ID: self.update("DROP DEFAULT") else: - self.update( - "DROP GRAPH %s" % self.node_to_sparql(graph.identifier)) + self.update("DROP GRAPH %s" % self.node_to_sparql(graph.identifier)) |