bugzilla.py

   1 #!/usr/bin/env python
   2 # -*- Mode: Python -*-
   3 # vi:si:et:sw=4:sts=4:ts=4
   4
   5
   6 # parse HTML from bugzilla.gnome.org to create a list of bugs for a given
   7 # product, component and target_milestone
   8
   9 import re
  10 import os
  11 import sys
  12 import codecs
  13 import urllib
  14 import HTMLParser
  15
  16 # a sample bug line we parse for future reference:
  17 #<TR VALIGN=TOP ALIGN=LEFT CLASS="Nor" ><TD><A HREF="show_bug.cgi?id=78267">78267</A> <td class=severity><nobr>min</nobr></td><td class=priority><nobr>Nor</nobr></td><td class=owner><nobr>thomas@apestaart.org</nobr></td><td class=status><nobr>RESO</nobr></td><td class=resolution><nobr>FIXE</nobr></td><td class=summary>autogen.sh doesn't take --prefix and similar to configure</td></TR>
  18
  19 # a sample bug section after olav's updating of bugzilla
  20 #    <td class="first-child">
  21 #      <a href="show_bug.cgi?id=147641">147641</a>
  22 #      <span style="display: none"></span>
  23 #    </td>
  24 #
  25 #    <td style="white-space: nowrap">nor
  26 #    </td>
  27 #    <td style="white-space: nowrap">Nor
  28 #    </td>
  29 #    <td style="white-space: nowrap">Linu
  30 #    </td>
  31 #    <td style="white-space: nowrap">GStreamer
  32 #    </td>
  33 #    <td style="white-space: nowrap">RESO
  34 #    </td>
  35 #    <td style="white-space: nowrap">FIXE
  36 #    </td>
  37 #    <td >[docs] pydoc segfaults when viewing gst package doc
  38 #    </td>
  39 #
  40 #  </tr>
  41
  42
  43 URL = 'http://bugzilla.gnome.org/buglist.cgi?product=%s&component=%s&target_milestone=%s'
  44
  45 # reg = re.compile('<TR.*id=(\d+)".*summary>(.*)<\/td')
  46
  47 HEADER = ' Changes\n'
  48 ITEM = ' o %s : %s'
  49 FOOTER = '\n Contributors\n'
  50
  51 default_product = "gtk-doc"
  52
  53 TD_ID = 1
  54 TD_SUMMARY = 7
  55 # after Olav's changes, it's now number 8
  56 #TD_SUMMARY = 8
  57
  58 # Horrible, don't look here
  59 class HP(HTMLParser.HTMLParser):
  60     def __init__(self):
  61         HTMLParser.HTMLParser.__init__(self)
  62         self.tr = 0
  63         self.td = 0
  64         self.bugs = []
  65         self.bugno = 0
  66         self.descr = ""
  67
  68     def handle_starttag(self, tag, data):
  69         if tag == 'tr':
  70             self.tr = 1
  71             return
  72         # count td's
  73         elif self.tr and tag.startswith('td'):
  74             self.td += 1
  75
  76     # all &gt; refs are handled through this method; append them to self.descr
  77     def handle_entityref(self, name):
  78         self.descr += " &%s; " % name
  79
  80     # can be called more than once for one td
  81     def handle_data(self, data):
  82         if not self.tr:
  83             return
  84         data = data.strip()
  85         if not data:
  86             return
  87
  88         #print self.td, self.tr, repr(data)
  89
  90         # check what td it is in
  91         if self.td == TD_ID:
  92             try:
  93                 self.bugno = int(data)
  94                 #print "got id: ", self.bugno
  95             except ValueError:
  96                 self.bugno = 0
  97         elif self.td == TD_SUMMARY:
  98             # the summary td
  99             self.descr += data
 100             #print "got descr: ", self.descr
 101
 102     def handle_endtag(self, tag):
 103         if tag == 'tr':
 104             self.tr = 0
 105             self.td = 0
 106             #print "end tag: ", self.bugno, self.descr
 107             if self.bugno != 0:
 108                 self.bugs.append((self.bugno, self.descr))
 109                 self.bugno = 0
 110                 self.descr = ""
 111
 112 def main(args):
 113     if len(args) < 3:
 114         print 'Usage: %s component milestone [product] [file]' % args[0]
 115         return 2
 116
 117     component = args[1]
 118     milestone = args[2]
 119
 120     if len(args) <= 3:
 121         product = default_product
 122     else:
 123         product = args[3]
 124
 125     if len(args) <= 4:
 126         output = None
 127     else:
 128         output = args[4]
 129
 130     url = URL % (product, urllib.quote(component), milestone)
 131     fd = urllib.urlopen(url)
 132
 133     hp = HP()
 134     hp.feed(fd.read())
 135
 136     lines = ["\n", ]
 137     lines.append(HEADER)
 138     for bug_id, summary in hp.bugs:
 139         lines.append(ITEM % (bug_id, summary))
 140     lines.append(FOOTER)
 141     bugs = "\n".join(lines)
 142
 143     if not output:
 144         print bugs
 145     else:
 146         # get original
 147         #doc = codecs.open(output, "r", encoding='utf-8').read()
 148         doc = open(output, "r").read()
 149         matcher = re.compile('(.*)<bugs>.*</bugs>(.*)',
 150             re.DOTALL)
 151         match = matcher.search(doc)
 152         pre = match.expand('\\1')
 153         post = match.expand('\\2')
 154
 155         backup = output + ".bugs.bak"
 156         os.rename(output, backup)
 157         handle = open(output, "w")
 158         handle.write(pre + bugs + post)
 159
 160 if __name__ == '__main__':
 161     sys.exit(main(sys.argv))