Added script to update the errorcodes module.

1 files changed, 110 insertions, 0 deletions

diff --git a/scripts/make_errorcodes.py b/scripts/make_errorcodes.py
new file mode 100755
index 0000000..3dad76a
--- /dev/null
+++ b/scripts/make_errorcodes.py
@@ -0,0 +1,110 @@
+#!/usr/bin/env python
+"""Generate the errorcodes module starting from PostgreSQL documentation.
+
+The script can be run at a new PostgreSQL release to refresh the module.
+"""
+
+# Copyright (C) 2010 Daniele Varrazzo  <daniele.varrazzo@gmail.com>
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by the
+# Free Software Foundation; either version 2, or (at your option) any later
+# version.
+#
+# This program is distributed in the hope that it will be useful, but
+# WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTIBILITY
+# or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
+# for more details.
+
+import sys
+import urllib2
+from collections import defaultdict
+
+from BeautifulSoup import BeautifulSoup as BS
+
+def main():
+    if len(sys.argv) != 2:
+        print >>sys.stderr, "usage: %s /path/to/errorcodes.py" % sys.argv[0]
+        return 2
+
+    filename = sys.argv[1]
+
+    file_start = read_base_file(filename)
+    classes, errors = fetch_errors(['8.1', '8.2', '8.3', '8.4'])
+
+    f = open(filename, "w")
+    for line in file_start:
+        print >>f, line
+    for line in generate_module_data(classes, errors):
+        print >>f, line
+
+def read_base_file(filename):
+    rv = []
+    for line in open(filename):
+        rv.append(line.rstrip("\n"))
+        if line.startswith("# autogenerated"):
+            return rv
+
+    raise ValueError("can't find the separator. Is this the right file?")
+
+def parse_errors(url):
+    page = urllib2.urlopen(url).read()
+    page = page.replace( # make things easier
+        '<SPAN CLASS="PRODUCTNAME">PostgreSQL</SPAN>',
+        'PostgreSQL')
+    page = BS(page)
+    table = page('table')[1]('tbody')[0]
+
+    classes = {}
+    errors = defaultdict(dict)
+
+    for tr in table('tr'):
+        if tr.td.get('colspan'): # it's a class
+            label = tr.b.string.encode("ascii")
+            assert label.startswith('Class')
+            label = label.replace("&mdash;", "-")
+            class_ = label.split()[1]
+            assert len(class_) == 2
+            classes[class_] = label
+
+        else: # it's an error
+            errcode = tr.tt.string.encode("ascii")
+            assert len(errcode) == 5
+            errlabel = tr('td')[1].string.replace(" ", "_").encode("ascii")
+            errors[class_][errcode] = errlabel
+
+    return classes, errors
+
+errors_url="http://www.postgresql.org/docs/%s/static/errcodes-appendix.html"
+
+def fetch_errors(versions):
+    classes = {}
+    errors = defaultdict(dict)
+
+    for version in versions:
+        c1, e1 = parse_errors(errors_url % version)
+        classes.update(c1)
+        for c, cerrs in e1.iteritems():
+            errors[c].update(cerrs)
+
+    return classes, errors
+
+def generate_module_data(classes, errors):
+    yield ""
+    yield "# Error classes"
+    for clscode, clslabel in sorted(classes.items()):
+        err = clslabel.split(" - ")[1].split("(")[0] \
+                .strip().replace(" ", "_").replace('/', "_").upper()
+        yield "CLASS_%s = %r" % (err, clscode)
+    
+    for clscode, clslabel in sorted(classes.items()):
+        yield ""
+        yield "# %s" % clslabel
+
+        for errcode, errlabel in sorted(errors[clscode].items()):
+            yield "%s = %r" % (errlabel, errcode)
+
+if __name__ == '__main__':
+    sys.exit(main())
+
+