fix filename generation for platforms that don't quite handle UTF-8 filenames

[oss/ljdump.git] / ljdump.py

diff --git a/ljdump.py b/ljdump.py
index 751dfffca02771f93405f44dbd755e1fbfd3cb85..0ab61ec5c79ba6a480479afbca76a245b66a3e24 100755 (executable)
--- a/ljdump.py
+++ b/ljdump.py
@@ -2,7 +2,7 @@
 #
 # ljdump.py - livejournal archiver
 # Greg Hewgill <greg@hewgill.com> http://hewgill.com
-# Version 1.1
+# Version 1.3.1
 #
 # $Id$
 #
@@ -26,6 +26,9 @@
 # This program may be run as often as needed to bring the backup copy up
 # to date. Both new and updated items are downloaded.
 #
+# The community http://ljdump.livejournal.com has been set up for questions
+# or comments.
+#
 # LICENSE
 #
 # This software is provided 'as-is', without any express or implied
@@ -44,11 +47,17 @@
 #    misrepresented as being the original software.
 # 3. This notice may not be removed or altered from any source distribution.
 #
-# Copyright (c) 2005-2006 Greg Hewgill
+# Copyright (c) 2005-2009 Greg Hewgill
 
-import codecs, md5, os, pickle, pprint, sys, urllib2, xml.dom.minidom, xmlrpclib
+import codecs, md5, os, pickle, pprint, re, shutil, sys, urllib2, xml.dom.minidom, xmlrpclib
 from xml.sax import saxutils
 
+MimeExtensions = {
+    "image/gif": ".gif",
+    "image/jpeg": ".jpg",
+    "image/png": ".png",
+}
+
 def calcchallenge(challenge, password):
     return md5.new(challenge+md5.new(password).hexdigest()).hexdigest()
 
@@ -67,10 +76,10 @@ def flatresponse(response):
     return r
 
 def getljsession(username, password):
-    r = urllib2.urlopen("http://livejournal.com/interface/flat", "mode=getchallenge")
+    r = urllib2.urlopen(Server+"/interface/flat", "mode=getchallenge")
     response = flatresponse(r)
     r.close()
-    r = urllib2.urlopen("http://livejournal.com/interface/flat", "mode=sessiongenerate&user=%s&auth_method=challenge&auth_challenge=%s&auth_response=%s" % (username, response['challenge'], calcchallenge(response['challenge'], password)))
+    r = urllib2.urlopen(Server+"/interface/flat", "mode=sessiongenerate&user=%s&auth_method=challenge&auth_challenge=%s&auth_response=%s" % (username, response['challenge'], calcchallenge(response['challenge'], password)))
     response = flatresponse(r)
     r.close()
     return response['ljsession']
@@ -100,6 +109,12 @@ def writedump(fn, event):
     dumpelement(f, "event", event)
     f.close()
 
+def writelast():
+    f = open("%s/.last" % Username, "w")
+    f.write("%s\n" % lastsync)
+    f.write("%s\n" % lastmaxid)
+    f.close()
+
 def createxml(doc, name, map):
     e = doc.createElement(name)
     for k in map.keys():
@@ -118,6 +133,10 @@ Server = config.documentElement.getElementsByTagName("server")[0].childNodes[0].
 Username = config.documentElement.getElementsByTagName("username")[0].childNodes[0].data
 Password = config.documentElement.getElementsByTagName("password")[0].childNodes[0].data
 
+m = re.search("(.*)/interface/xmlrpc", Server)
+if m:
+    Server = m.group(1)
+
 print "Fetching journal entries for: %s" % Username
 try:
     os.mkdir(Username)
@@ -127,7 +146,7 @@ except:
 
 ljsession = getljsession(Username, Password)
 
-server = xmlrpclib.ServerProxy(Server)
+server = xmlrpclib.ServerProxy(Server+"/interface/xmlrpc")
 
 newentries = 0
 newcomments = 0
@@ -152,6 +171,15 @@ except:
     pass
 origlastsync = lastsync
 
+r = server.LJ.XMLRPC.login(dochallenge({
+    'username': Username,
+    'ver': 1,
+    'getpickws': 1,
+    'getpickwurls': 1,
+}, Password))
+userpics = dict(zip(map(str, r['pickws']), r['pickwurls']))
+userpics['*'] = r['defaultpicurl']
+
 while True:
     r = server.LJ.XMLRPC.syncitems(dochallenge({
         'username': Username,
@@ -171,13 +199,18 @@ while True:
                     'selecttype': "one",
                     'itemid': item['item'][2:],
                 }, Password))
-                writedump("%s/%s" % (Username, item['item']), e['events'][0])
-                newentries += 1
+                if e['events']:
+                    writedump("%s/%s" % (Username, item['item']), e['events'][0])
+                    newentries += 1
+                else:
+                    print "Unexpected empty item: %s" % item['item']
+                    errors += 1
             except xmlrpclib.Fault, x:
                 print "Error getting item: %s" % item['item']
                 pprint.pprint(x)
                 errors += 1
         lastsync = item['time']
+        writelast()
 
 # The following code doesn't work because the server rejects our repeated calls.
 # http://www.livejournal.com/doc/server/ljp.csp.xml-rpc.getevents.html
@@ -219,7 +252,7 @@ except:
 
 maxid = lastmaxid
 while True:
-    r = urllib2.urlopen(urllib2.Request("http://livejournal.com/export_comments.bml?get=comment_meta&startid=%d" % (maxid+1), headers = {'Cookie': "ljsession="+ljsession}))
+    r = urllib2.urlopen(urllib2.Request(Server+"/export_comments.bml?get=comment_meta&startid=%d" % (maxid+1), headers = {'Cookie': "ljsession="+ljsession}))
     meta = xml.dom.minidom.parse(r)
     r.close()
     for c in meta.getElementsByTagName("comment"):
@@ -243,10 +276,31 @@ f = open("%s/user.map" % Username, "w")
 pickle.dump(usermap, f)
 f.close()
 
+print "Fetching userpics for: %s" % Username
+f = open("%s/userpics.xml" % Username, "w")
+print >>f, """<?xml version="1.0"?>"""
+print >>f, "<userpics>"
+for p in userpics:
+    print >>f, """<userpic keyword="%s" url="%s" />""" % (p, userpics[p])
+    pic = urllib2.urlopen(userpics[p])
+    ext = MimeExtensions.get(pic.info()["Content-Type"], "")
+    try:
+        picfn = codecs.utf_8_decode(p)[0]
+        picf = open("%s/%s%s" % (Username, picfn, ext), "wb")
+    except:
+        # for installations where the above utf_8_decode doesn't work
+        picfn = "".join([ord(x) < 128 and x or "?" for x in p])
+        picf = open("%s/%s%s" % (Username, picfn, ext), "wb")
+    shutil.copyfileobj(pic, picf)
+    pic.close()
+    picf.close()
+print >>f, "</userpics>"
+f.close()
+
 newmaxid = maxid
 maxid = lastmaxid
 while True:
-    r = urllib2.urlopen(urllib2.Request("http://livejournal.com/export_comments.bml?get=comment_body&startid=%d" % (maxid+1), headers = {'Cookie': "ljsession="+ljsession}))
+    r = urllib2.urlopen(urllib2.Request(Server+"/export_comments.bml?get=comment_body&startid=%d" % (maxid+1), headers = {'Cookie': "ljsession="+ljsession}))
     meta = xml.dom.minidom.parse(r)
     r.close()
     for c in meta.getElementsByTagName("comment"):
@@ -286,10 +340,7 @@ while True:
 
 lastmaxid = maxid
 
-f = open("%s/.last" % Username, "w")
-f.write("%s\n" % lastsync)
-f.write("%s\n" % lastmaxid)
-f.close()
+writelast()
 
 if origlastsync:
     print "%d new entries, %d new comments (since %s)" % (newentries, newcomments, origlastsync)