diff options
author | Sebastian Thiel <byronimo@gmail.com> | 2010-06-02 00:48:16 +0200 |
---|---|---|
committer | Sebastian Thiel <byronimo@gmail.com> | 2010-06-02 01:04:16 +0200 |
commit | df0892351a394d768489b5647d47b73c24d3ef5f (patch) | |
tree | fc2cbca826f2b728fe58aa22a09e3a65c12543ae | |
parent | 6d9b1f4f9fa8c9f030e3207e7deacc5d5f8bba4e (diff) | |
download | gitpython-df0892351a394d768489b5647d47b73c24d3ef5f.tar.gz |
commit: initial version of commit_from_tree which could create commit objects if it could serialize itself
-rw-r--r-- | CHANGES | 6 | ||||
-rw-r--r-- | lib/git/objects/base.py | 1 | ||||
-rw-r--r-- | lib/git/objects/commit.py | 777 | ||||
-rw-r--r-- | lib/git/objects/utils.py | 411 | ||||
-rw-r--r-- | test/git/test_utils.py | 233 |
5 files changed, 815 insertions, 613 deletions
@@ -1,7 +1,11 @@ ======= CHANGES ======= - + +0.2 Beta 2 +=========== + * Commit objects now carry the 'encoding' information of their message. It wasn't parsed previously, and defaults to UTF-8 + 0.2 ===== General diff --git a/lib/git/objects/base.py b/lib/git/objects/base.py index 6a51eed3..bb15192d 100644 --- a/lib/git/objects/base.py +++ b/lib/git/objects/base.py @@ -141,6 +141,7 @@ class Object(LazyMixin): self.repo.git.cat_file(self.type, self.sha, output_stream=ostream) return self + class IndexObject(Object): """ Base for all objects that can be part of the index file , namely Tree, Blob and diff --git a/lib/git/objects/commit.py b/lib/git/objects/commit.py index 826f684c..87eed49b 100644 --- a/lib/git/objects/commit.py +++ b/lib/git/objects/commit.py @@ -7,372 +7,425 @@ from git.utils import Iterable import git.diff as diff import git.stats as stats +from git.actor import Actor from tree import Tree import base import utils -import tempfile +import time import os class Commit(base.Object, Iterable, diff.Diffable, utils.Traversable): - """ - Wraps a git Commit object. - - This class will act lazily on some of its attributes and will query the - value on demand only if it involves calling the git binary. - """ - - # object configuration - type = "commit" - __slots__ = ("tree", - "author", "authored_date", "author_tz_offset", - "committer", "committed_date", "committer_tz_offset", - "message", "parents") - _id_attribute_ = "sha" - - def __init__(self, repo, sha, tree=None, author=None, authored_date=None, author_tz_offset=None, - committer=None, committed_date=None, committer_tz_offset=None, message=None, parents=None): - """ - Instantiate a new Commit. All keyword arguments taking None as default will - be implicitly set if id names a valid sha. - - The parameter documentation indicates the type of the argument after a colon ':'. - - ``sha`` - is the sha id of the commit or a ref - - ``parents`` : tuple( Commit, ... ) - is a tuple of commit ids or actual Commits - - ``tree`` : Tree - is the corresponding tree id or an actual Tree - - ``author`` : Actor - is the author string ( will be implicitly converted into an Actor object ) - - ``authored_date`` : int_seconds_since_epoch - is the authored DateTime - use time.gmtime() to convert it into a - different format - - ``author_tz_offset``: int_seconds_west_of_utc - is the timezone that the authored_date is in - - ``committer`` : Actor - is the committer string - - ``committed_date`` : int_seconds_since_epoch - is the committed DateTime - use time.gmtime() to convert it into a - different format - - ``committer_tz_offset``: int_seconds_west_of_utc - is the timezone that the authored_date is in - - ``message`` : string - is the commit message - - Returns - git.Commit - """ - super(Commit,self).__init__(repo, sha) - self._set_self_from_args_(locals()) - - if parents is not None: - self.parents = tuple( self.__class__(repo, p) for p in parents ) - # END for each parent to convert - - if self.sha and tree is not None: - self.tree = Tree(repo, tree, path='') - # END id to tree conversion - - @classmethod - def _get_intermediate_items(cls, commit): - return commit.parents - - def _set_cache_(self, attr): - """ - Called by LazyMixin superclass when the given uninitialized member needs - to be set. - We set all values at once. - """ - if attr in Commit.__slots__: - # prepare our data lines to match rev-list - data_lines = self.data.splitlines() - data_lines.insert(0, "commit %s" % self.sha) - temp = self._iter_from_process_or_stream(self.repo, iter(data_lines), False).next() - self.parents = temp.parents - self.tree = temp.tree - self.author = temp.author - self.authored_date = temp.authored_date - self.author_tz_offset = temp.author_tz_offset - self.committer = temp.committer - self.committed_date = temp.committed_date - self.committer_tz_offset = temp.committer_tz_offset - self.message = temp.message - else: - super(Commit, self)._set_cache_(attr) - - @property - def summary(self): - """ - Returns - First line of the commit message. - """ - return self.message.split('\n', 1)[0] - - def count(self, paths='', **kwargs): - """ - Count the number of commits reachable from this commit - - ``paths`` - is an optinal path or a list of paths restricting the return value - to commits actually containing the paths - - ``kwargs`` - Additional options to be passed to git-rev-list. They must not alter - the ouput style of the command, or parsing will yield incorrect results - Returns - int - """ - # yes, it makes a difference whether empty paths are given or not in our case - # as the empty paths version will ignore merge commits for some reason. - if paths: - return len(self.repo.git.rev_list(self.sha, '--', paths, **kwargs).splitlines()) - else: - return len(self.repo.git.rev_list(self.sha, **kwargs).splitlines()) - - - @property - def name_rev(self): - """ - Returns - String describing the commits hex sha based on the closest Reference. - Mostly useful for UI purposes - """ - return self.repo.git.name_rev(self) - - @classmethod - def iter_items(cls, repo, rev, paths='', **kwargs): - """ - Find all commits matching the given criteria. - - ``repo`` - is the Repo - - ``rev`` - revision specifier, see git-rev-parse for viable options - - ``paths`` - is an optinal path or list of paths, if set only Commits that include the path - or paths will be considered - - ``kwargs`` - optional keyword arguments to git rev-list where - ``max_count`` is the maximum number of commits to fetch - ``skip`` is the number of commits to skip - ``since`` all commits since i.e. '1970-01-01' - - Returns - iterator yielding Commit items - """ - options = {'pretty': 'raw', 'as_process' : True } - options.update(kwargs) - - args = list() - if paths: - args.extend(('--', paths)) - # END if paths - - proc = repo.git.rev_list(rev, args, **options) - return cls._iter_from_process_or_stream(repo, proc, True) - - def iter_parents(self, paths='', **kwargs): - """ - Iterate _all_ parents of this commit. - - ``paths`` - Optional path or list of paths limiting the Commits to those that - contain at least one of the paths - - ``kwargs`` - All arguments allowed by git-rev-list - - Return: - Iterator yielding Commit objects which are parents of self - """ - # skip ourselves - skip = kwargs.get("skip", 1) - if skip == 0: # skip ourselves - skip = 1 - kwargs['skip'] = skip - - return self.iter_items( self.repo, self, paths, **kwargs ) - - @property - def stats(self): - """ - Create a git stat from changes between this commit and its first parent - or from all changes done if this is the very first commit. - - Return - git.Stats - """ - if not self.parents: - text = self.repo.git.diff_tree(self.sha, '--', numstat=True, root=True) - text2 = "" - for line in text.splitlines()[1:]: - (insertions, deletions, filename) = line.split("\t") - text2 += "%s\t%s\t%s\n" % (insertions, deletions, filename) - text = text2 - else: - text = self.repo.git.diff(self.parents[0].sha, self.sha, '--', numstat=True) - return stats.Stats._list_from_string(self.repo, text) - - @classmethod - def _iter_from_process_or_stream(cls, repo, proc_or_stream, from_rev_list): - """ - Parse out commit information into a list of Commit objects - - ``repo`` - is the Repo - - ``proc`` - git-rev-list process instance (raw format) - - ``from_rev_list`` - If True, the stream was created by rev-list in which case we parse - the message differently - Returns - iterator returning Commit objects - """ - stream = proc_or_stream - if not hasattr(stream,'next'): - stream = proc_or_stream.stdout - - for line in stream: - commit_tokens = line.split() - id = commit_tokens[1] - assert commit_tokens[0] == "commit" - tree = stream.next().split()[1] - - parents = [] - next_line = None - for parent_line in stream: - if not parent_line.startswith('parent'): - next_line = parent_line - break - # END abort reading parents - parents.append(parent_line.split()[-1]) - # END for each parent line - - author, authored_date, author_tz_offset = utils.parse_actor_and_date(next_line) - committer, committed_date, committer_tz_offset = utils.parse_actor_and_date(stream.next()) - - # empty line - stream.next() - - message_lines = [] - if from_rev_list: - for msg_line in stream: - if not msg_line.startswith(' '): - # and forget about this empty marker - break - # END abort message reading - # strip leading 4 spaces - message_lines.append(msg_line[4:]) - # END while there are message lines - else: - # a stream from our data simply gives us the plain message - for msg_line in stream: - message_lines.append(msg_line) - # END message parsing - message = '\n'.join(message_lines) - - yield Commit(repo, id, parents=tuple(parents), tree=tree, - author=author, authored_date=authored_date, author_tz_offset=author_tz_offset, - committer=committer, committed_date=committed_date, committer_tz_offset=committer_tz_offset, - message=message) - # END for each line in stream - - - @classmethod - def create_from_tree(cls, repo, tree, message, parent_commits=None, head=False): - """ - Commit the given tree, creating a commit object. - - ``repo`` - is the Repo - - ``tree`` - Sha of a tree or a tree object to become the tree of the new commit - - ``message`` - Commit message. It may be an empty string if no message is provided. - It will be converted to a string in any case. - - ``parent_commits`` - Optional Commit objects to use as parents for the new commit. - If empty list, the commit will have no parents at all and become - a root commit. - If None , the current head commit will be the parent of the - new commit object - - ``head`` - If True, the HEAD will be advanced to the new commit automatically. - Else the HEAD will remain pointing on the previous commit. This could - lead to undesired results when diffing files. - - Returns - Commit object representing the new commit - - Note: - Additional information about hte committer and Author are taken from the - environment or from the git configuration, see git-commit-tree for - more information - """ - parents = parent_commits - if parent_commits is None: - try: - parent_commits = [ repo.head.commit ] - except ValueError: - # empty repositories have no head commit - parent_commits = list() - # END handle parent commits - # END if parent commits are unset - - parent_args = [ ("-p", str(commit)) for commit in parent_commits ] - - # create message stream - tmp_file_path = tempfile.mktemp() - fp = open(tmp_file_path,"wb") - fp.write(str(message)) - fp.close() - fp = open(tmp_file_path,"rb") - fp.seek(0) - - try: - # write the current index as tree - commit_sha = repo.git.commit_tree(tree, parent_args, istream=fp) - new_commit = cls(repo, commit_sha) - - if head: - try: - repo.head.commit = new_commit - except ValueError: - # head is not yet set to the ref our HEAD points to. - import git.refs - master = git.refs.Head.create(repo, repo.head.ref, commit=new_commit) - repo.head.reference = master - # END handle empty repositories - # END advance head handling - - return new_commit - finally: - fp.close() - os.remove(tmp_file_path) - - def __str__(self): - """ Convert commit to string which is SHA1 """ - return self.sha - - def __repr__(self): - return '<git.Commit "%s">' % self.sha + """ + Wraps a git Commit object. + + This class will act lazily on some of its attributes and will query the + value on demand only if it involves calling the git binary. + """ + + # ENVIRONMENT VARIABLES + # read when creating new commits + env_author_name = "GIT_AUTHOR_NAME" + env_author_email = "GIT_AUTHOR_EMAIL" + env_author_date = "GIT_AUTHOR_DATE" + env_committer_name = "GIT_COMMITTER_NAME" + env_committer_email = "GIT_COMMITTER_EMAIL" + env_committer_date = "GIT_COMMITTER_DATE" + env_email = "EMAIL" + + # CONFIGURATION KEYS + conf_email = 'email' + conf_name = 'name' + conf_encoding = 'i18n.commitencoding' + + # INVARIANTS + default_encoding = "UTF-8" + + + # object configuration + type = "commit" + __slots__ = ("tree", + "author", "authored_date", "author_tz_offset", + "committer", "committed_date", "committer_tz_offset", + "message", "parents", "encoding") + _id_attribute_ = "sha" + + def __init__(self, repo, sha, tree=None, author=None, authored_date=None, author_tz_offset=None, + committer=None, committed_date=None, committer_tz_offset=None, + message=None, parents=None, encoding=None): + """ + Instantiate a new Commit. All keyword arguments taking None as default will + be implicitly set if id names a valid sha. + + The parameter documentation indicates the type of the argument after a colon ':'. + + :param sha: is the sha id of the commit or a ref + :param parents: tuple( Commit, ... ) + is a tuple of commit ids or actual Commits + :param tree: Tree + is the corresponding tree id or an actual Tree + :param author: Actor + is the author string ( will be implicitly converted into an Actor object ) + :param authored_date: int_seconds_since_epoch + is the authored DateTime - use time.gmtime() to convert it into a + different format + :param author_tz_offset: int_seconds_west_of_utc + is the timezone that the authored_date is in + :param committer: Actor + is the committer string + :param committed_date: int_seconds_since_epoch + is the committed DateTime - use time.gmtime() to convert it into a + different format + :param committer_tz_offset: int_seconds_west_of_utc + is the timezone that the authored_date is in + :param message: string + is the commit message + :param encoding: string + encoding of the message, defaults to UTF-8 + :return: git.Commit + + :note: Timezone information is in the same format and in the same sign + as what time.altzone returns. The sign is inverted compared to git's + UTC timezone. + """ + super(Commit,self).__init__(repo, sha) + self._set_self_from_args_(locals()) + + if parents is not None: + self.parents = tuple( self.__class__(repo, p) for p in parents ) + # END for each parent to convert + + if self.sha and tree is not None: + self.tree = Tree(repo, tree, path='') + # END id to tree conversion + + @classmethod + def _get_intermediate_items(cls, commit): + return commit.parents + + def _set_cache_(self, attr): + """ + Called by LazyMixin superclass when the given uninitialized member needs + to be set. + We set all values at once. + """ + if attr in Commit.__slots__: + # prepare our data lines to match rev-list + data_lines = self.data.splitlines() + data_lines.insert(0, "commit %s" % self.sha) + temp = self._iter_from_process_or_stream(self.repo, iter(data_lines), False).next() + self.parents = temp.parents + self.tree = temp.tree + self.author = temp.author + self.authored_date = temp.authored_date + self.author_tz_offset = temp.author_tz_offset + self.committer = temp.committer + self.committed_date = temp.committed_date + self.committer_tz_offset = temp.committer_tz_offset + self.message = temp.message + self.encoding = temp.encoding + else: + super(Commit, self)._set_cache_(attr) + + @property + def summary(self): + """ + Returns + First line of the commit message. + """ + return self.message.split('\n', 1)[0] + + def count(self, paths='', **kwargs): + """ + Count the number of commits reachable from this commit + + ``paths`` + is an optinal path or a list of paths restricting the return value + to commits actually containing the paths + + ``kwargs`` + Additional options to be passed to git-rev-list. They must not alter + the ouput style of the command, or parsing will yield incorrect results + Returns + int + """ + # yes, it makes a difference whether empty paths are given or not in our case + # as the empty paths version will ignore merge commits for some reason. + if paths: + return len(self.repo.git.rev_list(self.sha, '--', paths, **kwargs).splitlines()) + else: + return len(self.repo.git.rev_list(self.sha, **kwargs).splitlines()) + + + @property + def name_rev(self): + """ + Returns + String describing the commits hex sha based on the closest Reference. + Mostly useful for UI purposes + """ + return self.repo.git.name_rev(self) + + @classmethod + def iter_items(cls, repo, rev, paths='', **kwargs): + """ + Find all commits matching the given criteria. + + ``repo`` + is the Repo + + ``rev`` + revision specifier, see git-rev-parse for viable options + + ``paths`` + is an optinal path or list of paths, if set only Commits that include the path + or paths will be considered + + ``kwargs`` + optional keyword arguments to git rev-list where + ``max_count`` is the maximum number of commits to fetch + ``skip`` is the number of commits to skip + ``since`` all commits since i.e. '1970-01-01' + + Returns + iterator yielding Commit items + """ + options = {'pretty': 'raw', 'as_process' : True } + options.update(kwargs) + + args = list() + if paths: + args.extend(('--', paths)) + # END if paths + + proc = repo.git.rev_list(rev, args, **options) + return cls._iter_from_process_or_stream(repo, proc, True) + + def iter_parents(self, paths='', **kwargs): + """ + Iterate _all_ parents of this commit. + + ``paths`` + Optional path or list of paths limiting the Commits to those that + contain at least one of the paths + + ``kwargs`` + All arguments allowed by git-rev-list + + Return: + Iterator yielding Commit objects which are parents of self + """ + # skip ourselves + skip = kwargs.get("skip", 1) + if skip == 0: # skip ourselves + skip = 1 + kwargs['skip'] = skip + + return self.iter_items( self.repo, self, paths, **kwargs ) + + @property + def stats(self): + """ + Create a git stat from changes between this commit and its first parent + or from all changes done if this is the very first commit. + + Return + git.Stats + """ + if not self.parents: + text = self.repo.git.diff_tree(self.sha, '--', numstat=True, root=True) + text2 = "" + for line in text.splitlines()[1:]: + (insertions, deletions, filename) = line.split("\t") + text2 += "%s\t%s\t%s\n" % (insertions, deletions, filename) + text = text2 + else: + text = self.repo.git.diff(self.parents[0].sha, self.sha, '--', numstat=True) + return stats.Stats._list_from_string(self.repo, text) + + @classmethod + def _iter_from_process_or_stream(cls, repo, proc_or_stream, from_rev_list): + """ + Parse out commit information into a list of Commit objects + + ``repo`` + is the Repo + + ``proc`` + git-rev-list process instance (raw format) + + ``from_rev_list`` + If True, the stream was created by rev-list in which case we parse + the message differently + Returns + iterator returning Commit objects + """ + stream = proc_or_stream + if not hasattr(stream,'next'): + stream = proc_or_stream.stdout + + for line in stream: + commit_tokens = line.split() + id = commit_tokens[1] + assert commit_tokens[0] == "commit" + tree = stream.next().split()[1] + + parents = [] + next_line = None + for parent_line in stream: + if not parent_line.startswith('parent'): + next_line = parent_line + break + # END abort reading parents + parents.append(parent_line.split()[-1]) + # END for each parent line + + author, authored_date, author_tz_offset = utils.parse_actor_and_date(next_line) + committer, committed_date, committer_tz_offset = utils.parse_actor_and_date(stream.next()) + + + # empty line + encoding = stream.next() + encoding.strip() + if encoding: + encoding = encoding[encoding.find(' ')+1:] + # END parse encoding + + message_lines = list() + if from_rev_list: + for msg_line in stream: + if not msg_line.startswith(' '): + # and forget about this empty marker + break + # END abort message reading + # strip leading 4 spaces + message_lines.append(msg_line[4:]) + # END while there are message lines + else: + # a stream from our data simply gives us the plain message + for msg_line in stream: + message_lines.append(msg_line) + # END message parsing + message = '\n'.join(message_lines) + + + yield Commit(repo, id, tree, + author, authored_date, author_tz_offset, + committer, committed_date, committer_tz_offset, + message, tuple(parents), + encoding or cls.default_encoding) + # END for each line in stream + + + @classmethod + def create_from_tree(cls, repo, tree, message, parent_commits=None, head=False): + """Commit the given tree, creating a commit object. + + :param repo: Repo object the commit should be part of + :param tree: Sha of a tree or a tree object to become the tree of the new commit + :param message: Commit message. It may be an empty string if no message is provided. + It will be converted to a string in any case. + :param parent_commits: + Optional Commit objects to use as parents for the new commit. + If empty list, the commit will have no parents at all and become + a root commit. + If None , the current head commit will be the parent of the + new commit object + :param head: + If True, the HEAD will be advanced to the new commit automatically. + Else the HEAD will remain pointing on the previous commit. This could + lead to undesired results when diffing files. + + :return: Commit object representing the new commit + + :note: + Additional information about the committer and Author are taken from the + environment or from the git configuration, see git-commit-tree for + more information + """ + parents = parent_commits + if parent_commits is None: + try: + parent_commits = [ repo.head.commit ] + except ValueError: + # empty repositories have no head commit + parent_commits = list() + # END handle parent commits + # END if parent commits are unset + + # retrieve all additional information, create a commit object, and + # serialize it + # Generally: + # * Environment variables override configuration values + # * Sensible defaults are set according to the git documentation + + # COMMITER AND AUTHOR INFO + cr = repo.config_reader() + env = os.environ + default_email = utils.get_user_id() + default_name = default_email.split('@')[0] + + conf_name = cr.get_value('user', cls.conf_name, default_name) + conf_email = cr.get_value('user', cls.conf_email, default_email) + + author_name = env.get(cls.env_author_name, conf_name) + author_email = env.get(cls.env_author_email, default_email) + + committer_name = env.get(cls.env_committer_name, conf_name) + committer_email = env.get(cls.env_committer_email, conf_email) + + # PARSE THE DATES + unix_time = int(time.time()) + offset = time.altzone + + author_date_str = env.get(cls.env_author_date, '') + if author_date_str: + author_time, author_offset = utils.parse_date(author_date_str) + else: + author_time, author_offset = unix_time, offset + # END set author time + + committer_date_str = env.get(cls.env_committer_date, '') + if committer_date_str: + committer_time, committer_offset = utils.parse_date(committer_date_str) + else: + committer_time, committer_offset = unix_time, offset + # END set committer time + + # assume utf8 encoding + enc_section, enc_option = cls.conf_encoding.split('.') + conf_encoding = cr.get_value(enc_section, enc_option, default_encoding) + + author = Actor(author_name, author_email) + committer = Actor(committer_name, committer_email) + + + # CREATE NEW COMMIT + new_commit = cls(repo, cls.NULL_HEX_SHA, tree, + author, author_time, author_offset, + committer, committer_time, committer_offset, + message, parent_commits, conf_encoding) + + # serialize ! + + if head: + try: + repo.head.commit = new_commit + except ValueError: + # head is not yet set to the ref our HEAD points to + # Happens on first commit + import git.refs + master = git.refs.Head.create(repo, repo.head.ref, commit=new_commit) + repo.head.reference = master + # END handle empty repositories + # END advance head handling + + return new_commit + + + def __str__(self): + """ Convert commit to string which is SHA1 """ + return self.sha + + def __repr__(self): + return '<git.Commit "%s">' % self.sha diff --git a/lib/git/objects/utils.py b/lib/git/objects/utils.py index 4f17b652..7060e293 100644 --- a/lib/git/objects/utils.py +++ b/lib/git/objects/utils.py @@ -9,159 +9,274 @@ Module for general utility functions import re from collections import deque as Deque from git.actor import Actor +import platform + +from string import digits +import time +import os + +__all__ = ('get_object_type_by_name', 'get_user_id', 'parse_date', 'parse_actor_and_date', + 'ProcessStreamAdapter', 'Traversable') def get_object_type_by_name(object_type_name): - """ - Returns - type suitable to handle the given object type name. - Use the type to create new instances. - - ``object_type_name`` - Member of TYPES - - Raises - ValueError: In case object_type_name is unknown - """ - if object_type_name == "commit": - import commit - return commit.Commit - elif object_type_name == "tag": - import tag - return tag.TagObject - elif object_type_name == "blob": - import blob - return blob.Blob - elif object_type_name == "tree": - import tree - return tree.Tree - else: - raise ValueError("Cannot handle unknown object type: %s" % object_type_name) - - + """ + Returns + type suitable to handle the given object type name. + Use the type to create new instances. + + ``object_type_name`` + Member of TYPES + + Raises + ValueError: In case object_type_name is unknown + """ + if object_type_name == "commit": + import commit + return commit.Commit + elif object_type_name == "tag": + import tag + return tag.TagObject + elif object_type_name == "blob": + import blob + return blob.Blob + elif object_type_name == "tree": + import tree + return tree.Tree + else: + raise ValueError("Cannot handle unknown object type: %s" % object_type_name) + + +def get_user_id(): + """:return: string identifying the currently active system user as name@node + :note: user can be set with the 'USER' environment variable, usually set on windows""" + ukn = 'UNKNOWN' + username = os.environ.get('USER', ukn) + if username == ukn and hasattr(os, 'getlogin'): + username = os.getlogin() + # END get username from login + return "%s@%s" % (username, platform.node()) + + +def _utc_tz_to_altz(utctz): + """we convert utctz to the timezone in seconds, it is the format time.altzone + returns. Git stores it as UTC timezon which has the opposite sign as well, + which explains the -1 * ( that was made explicit here ) + :param utctz: git utc timezone string, i.e. +0200""" + return -1 * int(float(utctz)/100*3600) + +def _verify_utctz(offset): + """:raise ValueError: if offset is incorrect + :return: offset""" + fmt_exc = ValueError("Invalid timezone offset format: %s" % offset) + if len(offset) != 5: + raise fmt_exc + if offset[0] not in "+-": + raise fmt_exc + if offset[1] not in digits or \ + offset[2] not in digits or \ + offset[3] not in digits or \ + offset[4] not in digits: + raise fmt_exc + # END for each char + return offset + +def parse_date(string_date): + """ + Parse the given date as one of the following + * Git internal format: timestamp offset + * RFC 2822: Thu, 07 Apr 2005 22:13:13 +0200. + * ISO 8601 2005-04-07T22:13:13 + The T can be a space as well + + :return: Tuple(int(timestamp), int(offset), both in seconds since epoch + :raise ValueError: If the format could not be understood + :note: Date can also be YYYY.MM.DD, MM/DD/YYYY and DD.MM.YYYY + """ + # git time + try: + if string_date.count(' ') == 1 and string_date.rfind(':') == -1: + timestamp, offset = string_date.split() + timestamp = int(timestamp) + return timestamp, _utc_tz_to_altz(_verify_utctz(offset)) + else: + offset = "+0000" # local time by default + if string_date[-5] in '-+': + offset = _verify_utctz(string_date[-5:]) + string_date = string_date[:-6] # skip space as well + # END split timezone info + + # now figure out the date and time portion - split time + date_formats = list() + splitter = -1 + if ',' in string_date: + date_formats.append("%a, %d %b %Y") + splitter = string_date.rfind(' ') + else: + # iso plus additional + date_formats.append("%Y-%m-%d") + date_formats.append("%Y.%m.%d") + date_formats.append("%m/%d/%Y") + date_formats.append("%d.%m.%Y") + + splitter = string_date.rfind('T') + if splitter == -1: + splitter = string_date.rfind(' ') + # END handle 'T' and ' ' + # END handle rfc or iso + + assert splitter > -1 + + # split date and time + time_part = string_date[splitter+1:] # skip space + date_part = string_date[:splitter] + + # parse time + tstruct = time.strptime(time_part, "%H:%M:%S") + + for fmt in date_formats: + try: + dtstruct = time.strptime(date_part, fmt) + fstruct = time.struct_time((dtstruct.tm_year, dtstruct.tm_mon, dtstruct.tm_mday, + tstruct.tm_hour, tstruct.tm_min, tstruct.tm_sec, + dtstruct.tm_wday, dtstruct.tm_yday, tstruct.tm_isdst)) + return int(time.mktime(fstruct)), _utc_tz_to_altz(offset) + except ValueError: + continue + # END exception handling + # END for each fmt + + # still here ? fail + raise ValueError("no format matched") + # END handle format + except Exception: + raise ValueError("Unsupported date format: %s" % string_date) + # END handle exceptions + + # precompiled regex _re_actor_epoch = re.compile(r'^.+? (.*) (\d+) ([+-]\d+).*$') def parse_actor_and_date(line): - """ - Parse out the actor (author or committer) info from a line like:: - - author Tom Preston-Werner <tom@mojombo.com> 1191999972 -0700 - - Returns - [Actor, int_seconds_since_epoch, int_timezone_offset] - """ - m = _re_actor_epoch.search(line) - actor, epoch, offset = m.groups() - return (Actor._from_string(actor), int(epoch), -int(float(offset)/100*3600)) - - - + """ + Parse out the actor (author or committer) info from a line like:: + + author Tom Preston-Werner <tom@mojombo.com> 1191999972 -0700 + + Returns + [Actor, int_seconds_since_epoch, int_timezone_offset] + """ + m = _re_actor_epoch.search(line) + actor, epoch, offset = m.groups() + return (Actor._from_string(actor), int(epoch), _utc_tz_to_altz(offset)) + + + class ProcessStreamAdapter(object): - """ - Class wireing all calls to the contained Process instance. - - Use this type to hide the underlying process to provide access only to a specified - stream. The process is usually wrapped into an AutoInterrupt class to kill - it if the instance goes out of scope. - """ - __slots__ = ("_proc", "_stream") - def __init__(self, process, stream_name): - self._proc = process - self._stream = getattr(process, stream_name) - - def __getattr__(self, attr): - return getattr(self._stream, attr) - - + """ + Class wireing all calls to the contained Process instance. + + Use this type to hide the underlying process to provide access only to a specified + stream. The process is usually wrapped into an AutoInterrupt class to kill + it if the instance goes out of scope. + """ + __slots__ = ("_proc", "_stream") + def __init__(self, process, stream_name): + self._proc = process + self._stream = getattr(process, stream_name) + + def __getattr__(self, attr): + return getattr(self._stream, attr) + + class Traversable(object): - """Simple interface to perforam depth-first or breadth-first traversals - into one direction. - Subclasses only need to implement one function. - Instances of the Subclass must be hashable""" - __slots__ = tuple() - - @classmethod - def _get_intermediate_items(cls, item): - """ - Returns: - List of items connected to the given item. - Must be implemented in subclass - """ - raise NotImplementedError("To be implemented in subclass") - - - def traverse( self, predicate = lambda i,d: True, - prune = lambda i,d: False, depth = -1, branch_first=True, - visit_once = True, ignore_self=1, as_edge = False ): - """ - ``Returns`` - iterator yieling of items found when traversing self - - ``predicate`` - f(i,d) returns False if item i at depth d should not be included in the result - - ``prune`` - f(i,d) return True if the search should stop at item i at depth d. - Item i will not be returned. - - ``depth`` - define at which level the iteration should not go deeper - if -1, there is no limit - if 0, you would effectively only get self, the root of the iteration - i.e. if 1, you would only get the first level of predessessors/successors - - ``branch_first`` - if True, items will be returned branch first, otherwise depth first - - ``visit_once`` - if True, items will only be returned once, although they might be encountered - several times. Loops are prevented that way. - - ``ignore_self`` - if True, self will be ignored and automatically pruned from - the result. Otherwise it will be the first item to be returned. - If as_edge is True, the source of the first edge is None - - ``as_edge`` - if True, return a pair of items, first being the source, second the - destinatination, i.e. tuple(src, dest) with the edge spanning from - source to destination""" - visited = set() - stack = Deque() - stack.append( ( 0 ,self, None ) ) # self is always depth level 0 - - def addToStack( stack, item, branch_first, depth ): - lst = self._get_intermediate_items( item ) - if not lst: - return - if branch_first: - stack.extendleft( ( depth , i, item ) for i in lst ) - else: - reviter = ( ( depth , lst[i], item ) for i in range( len( lst )-1,-1,-1) ) - stack.extend( reviter ) - # END addToStack local method - - while stack: - d, item, src = stack.pop() # depth of item, item, item_source - - if visit_once and item in visited: - continue - - if visit_once: - visited.add(item) - - rval = ( as_edge and (src, item) ) or item - if prune( rval, d ): - continue - - skipStartItem = ignore_self and ( item == self ) - if not skipStartItem and predicate( rval, d ): - yield rval - - # only continue to next level if this is appropriate ! - nd = d + 1 - if depth > -1 and nd > depth: - continue - - addToStack( stack, item, branch_first, nd ) - # END for each item on work stack + """Simple interface to perforam depth-first or breadth-first traversals + into one direction. + Subclasses only need to implement one function. + Instances of the Subclass must be hashable""" + __slots__ = tuple() + + @classmethod + def _get_intermediate_items(cls, item): + """ + Returns: + List of items connected to the given item. + Must be implemented in subclass + """ + raise NotImplementedError("To be implemented in subclass") + + + def traverse( self, predicate = lambda i,d: True, + prune = lambda i,d: False, depth = -1, branch_first=True, + visit_once = True, ignore_self=1, as_edge = False ): + """ + ``Returns`` + iterator yieling of items found when traversing self + + ``predicate`` + f(i,d) returns False if item i at depth d should not be included in the result + + ``prune`` + f(i,d) return True if the search should stop at item i at depth d. + Item i will not be returned. + + ``depth`` + define at which level the iteration should not go deeper + if -1, there is no limit + if 0, you would effectively only get self, the root of the iteration + i.e. if 1, you would only get the first level of predessessors/successors + + ``branch_first`` + if True, items will be returned branch first, otherwise depth first + + ``visit_once`` + if True, items will only be returned once, although they might be encountered + several times. Loops are prevented that way. + + ``ignore_self`` + if True, self will be ignored and automatically pruned from + the result. Otherwise it will be the first item to be returned. + If as_edge is True, the source of the first edge is None + + ``as_edge`` + if True, return a pair of items, first being the source, second the + destinatination, i.e. tuple(src, dest) with the edge spanning from + source to destination""" + visited = set() + stack = Deque() + stack.append( ( 0 ,self, None ) ) # self is always depth level 0 + + def addToStack( stack, item, branch_first, depth ): + lst = self._get_intermediate_items( item ) + if not lst: + return + if branch_first: + stack.extendleft( ( depth , i, item ) for i in lst ) + else: + reviter = ( ( depth , lst[i], item ) for i in range( len( lst )-1,-1,-1) ) + stack.extend( reviter ) + # END addToStack local method + + while stack: + d, item, src = stack.pop() # depth of item, item, item_source + + if visit_once and item in visited: + continue + + if visit_once: + visited.add(item) + + rval = ( as_edge and (src, item) ) or item + if prune( rval, d ): + continue + + skipStartItem = ignore_self and ( item == self ) + if not skipStartItem and predicate( rval, d ): + yield rval + + # only continue to next level if this is appropriate ! + nd = d + 1 + if depth > -1 and nd > depth: + continue + + addToStack( stack, item, branch_first, nd ) + # END for each item on work stack diff --git a/test/git/test_utils.py b/test/git/test_utils.py index f843c12e..2c3c392b 100644 --- a/test/git/test_utils.py +++ b/test/git/test_utils.py @@ -9,112 +9,141 @@ import tempfile from test.testlib import * from git.utils import * +from git.objects.utils import * from git import * from git.cmd import dashify import time class TestUtils(TestCase): - def setup(self): - self.testdict = { - "string": "42", - "int": 42, - "array": [ 42 ], - } + def setup(self): + self.testdict = { + "string": "42", + "int": 42, + "array": [ 42 ], + } - def test_it_should_dashify(self): - assert_equal('this-is-my-argument', dashify('this_is_my_argument')) - assert_equal('foo', dashify('foo')) - - - def test_lock_file(self): - my_file = tempfile.mktemp() - lock_file = LockFile(my_file) - assert not lock_file._has_lock() - # release lock we don't have - fine - lock_file._release_lock() - - # get lock - lock_file._obtain_lock_or_raise() - assert lock_file._has_lock() - - # concurrent access - other_lock_file = LockFile(my_file) - assert not other_lock_file._has_lock() - self.failUnlessRaises(IOError, other_lock_file._obtain_lock_or_raise) - - lock_file._release_lock() - assert not lock_file._has_lock() - - other_lock_file._obtain_lock_or_raise() - self.failUnlessRaises(IOError, lock_file._obtain_lock_or_raise) - - # auto-release on destruction - del(other_lock_file) - lock_file._obtain_lock_or_raise() - lock_file._release_lock() - - def test_blocking_lock_file(self): - my_file = tempfile.mktemp() - lock_file = BlockingLockFile(my_file) - lock_file._obtain_lock() - - # next one waits for the lock - start = time.time() - wait_time = 0.1 - wait_lock = BlockingLockFile(my_file, 0.05, wait_time) - self.failUnlessRaises(IOError, wait_lock._obtain_lock) - elapsed = time.time() - start - assert elapsed <= wait_time + 0.02 # some extra time it may cost - - def _cmp_contents(self, file_path, data): - # raise if data from file at file_path - # does not match data string - fp = open(file_path, "rb") - try: - assert fp.read() == data - finally: - fp.close() - - def test_safe_operation(self): - my_file = tempfile.mktemp() - orig_data = "hello" - new_data = "world" - my_file_fp = open(my_file, "wb") - my_file_fp.write(orig_data) - my_file_fp.close() - - try: - cwrite = ConcurrentWriteOperation(my_file) - - # didn't start writing, doesnt matter - cwrite._end_writing(False) - cwrite._end_writing(True) - assert not cwrite._is_writing() - - # write data and fail - stream = cwrite._begin_writing() - assert cwrite._is_writing() - stream.write(new_data) - cwrite._end_writing(successful=False) - self._cmp_contents(my_file, orig_data) - assert not os.path.exists(stream.name) - - # write data - concurrently - ocwrite = ConcurrentWriteOperation(my_file) - stream = cwrite._begin_writing() - self.failUnlessRaises(IOError, ocwrite._begin_writing) - - stream.write("world") - cwrite._end_writing(successful=True) - self._cmp_contents(my_file, new_data) - assert not os.path.exists(stream.name) - - # could test automatic _end_writing on destruction - finally: - os.remove(my_file) - # END final cleanup - - - - + def test_it_should_dashify(self): + assert_equal('this-is-my-argument', dashify('this_is_my_argument')) + assert_equal('foo', dashify('foo')) + + + def test_lock_file(self): + my_file = tempfile.mktemp() + lock_file = LockFile(my_file) + assert not lock_file._has_lock() + # release lock we don't have - fine + lock_file._release_lock() + + # get lock + lock_file._obtain_lock_or_raise() + assert lock_file._has_lock() + + # concurrent access + other_lock_file = LockFile(my_file) + assert not other_lock_file._has_lock() + self.failUnlessRaises(IOError, other_lock_file._obtain_lock_or_raise) + + lock_file._release_lock() + assert not lock_file._has_lock() + + other_lock_file._obtain_lock_or_raise() + self.failUnlessRaises(IOError, lock_file._obtain_lock_or_raise) + + # auto-release on destruction + del(other_lock_file) + lock_file._obtain_lock_or_raise() + lock_file._release_lock() + + def test_blocking_lock_file(self): + my_file = tempfile.mktemp() + lock_file = BlockingLockFile(my_file) + lock_file._obtain_lock() + + # next one waits for the lock + start = time.time() + wait_time = 0.1 + wait_lock = BlockingLockFile(my_file, 0.05, wait_time) + self.failUnlessRaises(IOError, wait_lock._obtain_lock) + elapsed = time.time() - start + assert elapsed <= wait_time + 0.02 # some extra time it may cost + + def _cmp_contents(self, file_path, data): + # raise if data from file at file_path + # does not match data string + fp = open(file_path, "rb") + try: + assert fp.read() == data + finally: + fp.close() + + def test_safe_operation(self): + my_file = tempfile.mktemp() + orig_data = "hello" + new_data = "world" + my_file_fp = open(my_file, "wb") + my_file_fp.write(orig_data) + my_file_fp.close() + + try: + cwrite = ConcurrentWriteOperation(my_file) + + # didn't start writing, doesnt matter + cwrite._end_writing(False) + cwrite._end_writing(True) + assert not cwrite._is_writing() + + # write data and fail + stream = cwrite._begin_writing() + assert cwrite._is_writing() + stream.write(new_data) + cwrite._end_writing(successful=False) + self._cmp_contents(my_file, orig_data) + assert not os.path.exists(stream.name) + + # write data - concurrently + ocwrite = ConcurrentWriteOperation(my_file) + stream = cwrite._begin_writing() + self.failUnlessRaises(IOError, ocwrite._begin_writing) + + stream.write("world") + cwrite._end_writing(successful=True) + self._cmp_contents(my_file, new_data) + assert not os.path.exists(stream.name) + + # could test automatic _end_writing on destruction + finally: + os.remove(my_file) + # END final cleanup + + + + def test_user_id(self): + assert '@' in get_user_id() + + def test_parse_date(self): + # test all supported formats + def assert_rval(rval, veri_time, offset=0): + assert len(rval) == 2 + assert isinstance(rval[0], int) and isinstance(rval[1], int) + assert rval[0] == veri_time + assert rval[1] == offset + # END assert rval utility + + rfc = ("Thu, 07 Apr 2005 22:13:11 +0000", 0) + iso = ("2005-04-07T22:13:11 -0200", 7200) + iso2 = ("2005-04-07 22:13:11 +0400", -14400) + iso3 = ("2005.04.07 22:13:11 -0000", 0) + alt = ("04/07/2005 22:13:11", 0) + alt2 = ("07.04.2005 22:13:11", 0) + veri_time = 1112904791 # the time this represents + for date, offset in (rfc, iso, iso2, iso3, alt, alt2): + assert_rval(parse_date(date), veri_time, offset) + # END for each date type + + # and failure + self.failUnlessRaises(ValueError, parse_date, 'invalid format') + self.failUnlessRaises(ValueError, parse_date, '123456789 -02000') + self.failUnlessRaises(ValueError, parse_date, ' 123456789 -0200') + + |