init

README.md

@@ -0,0 +1 @@
+## Dump Wechat Messages from Android

decrypt_db.sh

@@ -0,0 +1,23 @@
+#!/bin/bash -e
+# File: decrypt_db.sh
+# Date: Fri Nov 21 13:07:33 2014 +0800
+# Author: Yuxin Wu <[email protected]>
+
+MMSGDB=$1
+imei=$2
+uin=$3
+
+if [[ -z "$1" || -z "$2" || -z "$3" ]]; then
+	echo "Usage: $0 <path to EnMicroMsg.db> <imei> <uin>"
+	exit
+fi
+
+KEY=$(echo -n "$imei$uin" | md5sum | cut -b 1-7)
+
+LD_LIBRARY_PATH=./lib ./lib/sqlcipher $MMSGDB << EOF
+PRAGMA key='$KEY';
+PRAGMA cipher_use_hmac = off;
+ATTACH DATABASE "decrypted_database.db" AS decrypted_database KEY "";
+SELECT sqlcipher_export("decrypted_database");
+DETACH DATABASE decrypted_database;
+EOF

dump.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python2
+# -*- coding: UTF-8 -*-
+# File: dump.py
+# Date: Fri Nov 21 13:07:58 2014 +0800
+# Author: Yuxin Wu <[email protected]>
+
+import sqlite3
+from collections import defaultdict
+from pprint import PrettyPrinter
+pp = PrettyPrinter()
+def log(x): print repr(x).decode('unicode-escape')
+from lib.Msg import WeChatMsg
+
+""" tables in concern:
+emojiinfo
+imginfo2
+addr_upload2
+chatroom
+message
+rcontact
+
+"""
+
+class WeChatDBParser(object):
+    def __init__(self, db_fname):
+        """ db_fname: EnMicroMsg.db"""
+        self.db_fname = db_fname
+        self.db_conn = sqlite3.connect(self.db_fname)
+        self.cc = self.db_conn.cursor()
+        self.contacts = {}
+
+    def _parse_contact(self):
+        contacts = self.cc.execute(
+"""
+SELECT username,conRemark,nickname
+FROM rcontact
+""")
+        for row in contacts:
+            username, remark, nickname = row
+            if remark:
+                self.contacts[username] = remark
+            else:
+                self.contacts[username] = nickname
+
+    def _parse_msg(self):
+        msgs_by_talker = defaultdict(list)
+        db_msgs = self.cc.execute(
+"""
+SELECT {}
+FROM message ORDER BY createTime
+""".format(','.join(WeChatMsg.FIELDS)))
+        for row in db_msgs:
+            msg = WeChatMsg(row)
+            msgs_by_talker[msg.talker].append(msg)
+        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 msg in v:
+                msg.talker = k
+        pp.pprint(msgs_by_talker.items()[0])
+
+    def parse(self):
+        self._parse_contact()
+        msg = self._parse_msg()
+        pass
+
+if __name__ == '__main__':
+    parser = WeChatDBParser('./decrypted_database.db')
+    parser.parse()

lib/Msg.py

@@ -0,0 +1,23 @@
+#!/usr/bin/env python2
+# -*- coding: UTF-8 -*-
+# File: Msg.py
+# Date: Fri Nov 21 12:15:27 2014 +0800
+# Author: Yuxin Wu <[email protected]>
+
+from datetime import datetime
+
+class WeChatMsg(object):
+    """ fields in concern"""
+    FIELDS = ["msgSvrId","type","isSend","createTime","talker","content","imgPath"]
+
+    def __init__(self, row):
+        """ row: a tuple corresponding to FIELDS"""
+        assert len(row) == len(WeChatMsg.FIELDS)
+        for f, v in zip(WeChatMsg.FIELDS, row):
+            setattr(self, f, v)
+        self.createTime = datetime.fromtimestamp(self.createTime / 1000)
+        if self.content:
+            self.content = self.content.encode('utf-8')
+
+    def __repr__(self):
+        return "{}:{}:{}".format(self.talker.encode('utf-8') if not self.isSend else 'me', self.createTime, self.content)