import pyblog # (http://code.google.com/p/python-blogger/)
import re,pickle,traceback
from pyblog import *

wp = MetaWeblog('http://blog.jonudell.net/xmlrpc.php',
  '{USER}','{PASS}', appkey='{KEY}')

posts = wp.get_recent_posts(530) # a number larger than the total

f = open('metabydate.pickle','rb') # created in a previous step,
                                   # from old archive. e.g.:
                                   # { '2006/11/01' : 
                                   #   {'title' : '...'},
                                   #   {'etc' : '...'},
                                   # }
metabydate = pickle.load(f)

def make_fname(title,date):
  fname = title.replace('\n','')
  fname = re.sub('[&\/:]+','-',fname)
  fname = re.sub('[^\w\s\-]+','',fname)
  fname = re.sub('\s+','-',fname)
  fname = re.sub('\-+','-',fname)
  fname = fname.lower() + '.html'
  fname = date.replace('/','-') + '-' + fname
  return fname

def make_new_url(fname):
  return 'http://jonudell.net/udell/%s' % fname

def subrepl(matchobj):
  oldurl = matchobj.group(1)
  date = re.findall('(\d+/\d+/\d+)',oldurl)[0]
  title = metabydate[date]['title']
  fname = make_fname(title,date)
  newurl = make_new_url(fname)
  print "from %s\n to %s\n" % (oldurl,newurl)
  return newurl

targets = []

for post in posts:       # first collect edit targets

 desc = post['description']
 if ( len(re.findall('weblog.infoworld.com/udell', desc)) > 0 ):
   targets.append(post)

for post in targets[2:]: # skipping the most recent 2, which really 
                         # do refer to the old namespace

 desc = post['description']
 id = post['postid']
 print id, post['dateCreated'],post['title']

 try:

   desc = re.sub('(http://weblog\.infoworld\.com/udell/\d+/\d+/\d+\.html#a\d+)', 
     subrepl, desc) # some urls end like 2006/01/04.html#a527

   desc = re.sub('(http://weblog\.infoworld\.com/udell/\d+/\d+/\d+\.html)', 
     subrepl, desc) # others just like 2006/01/05.html

   post['description'] = desc

   wp.edit_post(id,post)

 except:
   traceback.print_exc()

# sample output
# 50 20070206T15:06:17 Search strategies, part 2
# from http://weblog.infoworld.com/udell/2006/12/04.html#a1571
# to http://jonudell.net/udell/2006-12-04-hunting-the-elusive-search-strategy.html