msg can work

dump.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python2
 # -*- coding: UTF-8 -*-
 # File: dump.py
-# Date: Fri Nov 21 13:07:58 2014 +0800
+# Date: Fri Nov 21 14:08:45 2014 +0800
 # Author: Yuxin Wu <[email protected]>
 
 import sqlite3
@@ -10,6 +10,7 @@ from pprint import PrettyPrinter
 pp = PrettyPrinter()
 def log(x): print repr(x).decode('unicode-escape')
 from lib.Msg import WeChatMsg
+from lib.utils import ensure_unicode
 
 """ tables in concern:
 emojiinfo
@@ -28,42 +29,48 @@ class WeChatDBParser(object):
         self.db_conn = sqlite3.connect(self.db_fname)
         self.cc = self.db_conn.cursor()
         self.contacts = {}
+        self.msgs_by_talker = defaultdict(list)
 
     def _parse_contact(self):
         contacts = self.cc.execute(
 """
-SELECT username,conRemark,nickname
-FROM rcontact
+SELECT username,conRemark,nickname FROM rcontact
 """)
         for row in contacts:
             username, remark, nickname = row
             if remark:
-                self.contacts[username] = remark
+                self.contacts[username] = ensure_unicode(remark)
             else:
-                self.contacts[username] = nickname
+                self.contacts[username] = ensure_unicode(nickname)
 
     def _parse_msg(self):
-        msgs_by_talker = defaultdict(list)
         db_msgs = self.cc.execute(
 """
-SELECT {}
-FROM message ORDER BY createTime
+SELECT {} FROM message
 """.format(','.join(WeChatMsg.FIELDS)))
         for row in db_msgs:
             msg = WeChatMsg(row)
-            msgs_by_talker[msg.talker].append(msg)
-        msgs_by_talker = dict([
+            if msg.type not in WeChatMsg.FILTER_TYPES:
+                self.msgs_by_talker[msg.talker].append(msg)
+        self.msgs_by_talker = dict([
             (self.contacts[k], sorted(v, key=lambda x: x.createTime))
-                           for k, v in msgs_by_talker.iteritems()])
-        for k, v in msgs_by_talker.iteritems():
+                           for k, v in self.msgs_by_talker.iteritems()])
+        for k, v in self.msgs_by_talker.iteritems():
             for msg in v:
                 msg.talker = k
-        pp.pprint(msgs_by_talker.items()[0])
+
+    def _find_msg_by_type(self):
+        ret = []
+        for v in self.msgs_by_talker.itervalues():
+            for msg in v:
+                if msg.type == 34:
+                    print msg
+                    print
+        return ret
 
     def parse(self):
         self._parse_contact()
-        msg = self._parse_msg()
-        pass
+        self._parse_msg()
 
 if __name__ == '__main__':
     parser = WeChatDBParser('./decoded_database.db')