minutes-rss.py

   1 #!/usr/bin/python
   2
   3 # UCC Minutes Arse (RSS) generator
   4
   5 import subprocess
   6 import shlex
   7 import os
   8 import time
   9 from cgi import escape
  10
  11 MINUTESURL = 'http://www.ucc.asn.au/infobase/minutes/'
  12 MINUTESPATH = '/services/http/infobase/minutes/'
  13
  14 def RFCTime(pubtime):
  15         return time.strftime("%a, %d %b %Y %H:%M:%S +0800", pubtime)
  16
  17 class UCCMinutes:
  18         '''Loads relevant information out of UCC minutes'''
  19         def __init__(self, path):
  20                 if os.path.exists(path):
  21                         self.stattime = os.stat(path)[-2]
  22                         self.modtime = RFCTime(time.localtime(self.stattime))
  23                         f = open(path,'r')
  24                         self.body = escape(f.read())
  25                         f.close()
  26
  27                         self.title = self.body.split('\n')[0].strip()
  28
  29                         webpath = path.split('/')[-2:]
  30                         self.url = MINUTESURL + '/'.join(webpath)
  31
  32 def RSS(items,pubtime=''):
  33         pubtime = pubtime or RFCTime(time.localtime())
  34         output = '''<?xml version='1.0' encoding='utf-8' ?>
  35 <rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  36 <channel>
  37  <atom:link href="http://www.ucc.asn.au/infobase/minutes/minutes.xml" rel="self" type="application/rss+xml" />
  38  <title>UCC Meeting Minutes</title>
  39  <link>http://www.ucc.asn.au/infobase/minutes/</link>
  40  <description>Meeting Minutes of the University Computer Club</description>
  41 '''
  42         output += ' <lastBuildDate>%s</lastBuildDate>\n'%pubtime
  43         output += ''' <generator>The giant hack that is Python by JCF</generator>
  44  <copyright>The University Computer Club &#x3C;[email protected]&#x3E;</copyright>
  45
  46 '''
  47         output += items
  48         output += '</channel>\n</rss>'
  49
  50         return output
  51
  52 def itemWriter(minutes):
  53         output = '<item>\n'
  54         output += '<guid>%s</guid>\n'%minutes.url
  55         output += '<title>%s</title>\n'%minutes.title
  56         output += '<pubDate>%s</pubDate>\n'%minutes.modtime
  57         output += '<link>%s</link>\n'%minutes.url
  58         output += '<description><![CDATA[<pre>%s</pre>]]></description>\n'%minutes.body
  59         output += '</item>\n'
  60         return output
  61
  62
  63 if __name__ == '__main__':
  64         xmlpath = os.path.join(MINUTESPATH, 'minutes.xml')
  65         try:
  66                 lastupdate = os.stat(xmlpath)[-2]
  67         except OSError:
  68                 lastupdate = 0
  69
  70         # BEWARE THE GIANT HACK - This should be replaced before it burns someone
  71         args = shlex.split('/usr/bin/find %s'%MINUTESPATH + ' -maxdepth 2 -iname "*txt" -type f -printf "%TY-%Tm-%Td %TT %p\n"')
  72         sub = subprocess.Popen(args, stdout=subprocess.PIPE)
  73         items = sub.communicate()[0].split('\n')
  74         items.sort()
  75
  76         items = [os.path.abspath(x.split(' ')[-1]) for x in items[-15:]]
  77         items.reverse()
  78
  79         latestItem = UCCMinutes(items[0])
  80
  81         if latestItem.stattime > lastupdate:
  82                 minutes = [latestItem] + [UCCMinutes(x) for x in items[1:]]
  83                 output = RSS(''.join([itemWriter(x) for x in minutes]), minutes[0].modtime)
  84                 open(xmlpath, 'w').write(output)