emoji works

README.md

@@ -37,6 +37,8 @@
 ```
 
 ### TODO
-+ Parse custom emoji, and other message types
++ Parse uncommon message types
 + Show name of emoji in text
 + Search by uid/username..
++ Fix emoji sizes..
++ Sometimes beautiful soup failed on links..

lib/msg.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python2
 # -*- coding: UTF-8 -*-
 # File: msg.py
-# Date: Sat Dec 20 15:12:34 2014 +0800
+# Date: Sat Dec 20 18:38:59 2014 +0800
 # Author: Yuxin Wu <[email protected]>
 
 from datetime import datetime
@@ -77,6 +77,9 @@ class WeChatMsg(object):
             if not name:
                 name = msg.get('alias', '')
             return u"NAMECARD: {}".format(name)
+        elif self.type == TYPE_EMOJI:
+            # TODO add emoji name
+            return self.content
         else:
             return self.content
 
@@ -95,3 +98,11 @@ class WeChatMsg(object):
 
     def __lt__(self, r):
         return self.createTime < r.createTime
+
+    def get_emoji_product_id(self):
+        assert self.type == TYPE_EMOJI, "Wrong call to get_emoji_product_id()!"
+        soup = BeautifulSoup(self.content)
+        emoji = soup.find('emoji')
+        if emoji is None:
+            return None
+        return emoji.get('productid', None)

lib/parser.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python2
 # -*- coding: UTF-8 -*-
 # File: parser.py
-# Date: Sat Dec 20 15:58:16 2014 +0800
+# Date: Sat Dec 20 19:40:31 2014 +0800
 # Author: Yuxin Wu <[email protected]>
 
 import sqlite3
@@ -30,6 +30,7 @@ class WeChatDBParser(object):
         self.cc = self.db_conn.cursor()
         self.contacts = {}
         self.msgs_by_talker = defaultdict(list)
+        self.emojis = {}
         self.parse()
 
     def _parse_contact(self):
@@ -86,8 +87,17 @@ SELECT {} FROM message
                 ret.append(msg)
         return sorted(ret)
 
+    def _parse_emoji(self):
+        # wechat provided emojis
+        emojiinfo_q = self.cc.execute(
+""" SELECT md5, desc, groupid FROM EmojiInfoDesc """)
+        for row in emojiinfo_q:
+            md5, desc, group = row
+            self.emojis[md5] = (group, desc)
+
     def parse(self):
         self._parse_userinfo()
         self._parse_contact()
         self._parse_msg()
         self._parse_imginfo()
+        self._parse_emoji()

lib/render.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python2
 # -*- coding: UTF-8 -*-
 # File: render.py
-# Date: Sat Dec 20 17:48:48 2014 +0800
+# Date: Sat Dec 20 20:15:47 2014 +0800
 # Author: Yuxin Wu <[email protected]>
 
 import os
@@ -24,7 +24,8 @@ from smiley import SmileyProvider
 
 TEMPLATES_FILES = {TYPE_MSG: "TP_MSG",
                    TYPE_IMG: "TP_IMG",
-                   TYPE_SPEAK: "TP_SPEAK"}
+                   TYPE_SPEAK: "TP_SPEAK",
+                   TYPE_EMOJI: "TP_EMOJI"}
 TEMPLATES = dict([(k, open(os.path.join(
     LIB_PATH, 'static/{}.html'.format(v))).read())
     for k, v in TEMPLATES_FILES.iteritems()])
@@ -96,10 +97,15 @@ class HTMLRender(object):
                                    big_img=bigimg)
         elif msg.type == TYPE_EMOJI:
             imgpath = msg.imgPath
-            print imgpath
-            return fallback()
-            # custom emoji
 
+            if imgpath in self.parser.emojis:
+                group, _ = self.parser.emojis[imgpath]
+            else:
+                group = None
+            emoji_img, format = self.res.get_emoji(imgpath, group)
+            return template.format(sender_label=sender,
+                                  emoji_format=format,
+                                  emoji_img=emoji_img)
         return fallback()
 
     def render_msgs(self, msgs):

lib/res.py

@@ -1,22 +1,26 @@
 #!/usr/bin/env python2
 # -*- coding: UTF-8 -*-
 # File: res.py
-# Date: Sat Dec 20 16:26:51 2014 +0800
+# Date: Sat Dec 20 20:14:40 2014 +0800
 # Author: Yuxin Wu <[email protected]>
 
 import glob
 import os
+import re
 import Image
 import cStringIO
 import base64
 import logging
-import eyed3
+import imghdr
 logger = logging.getLogger(__name__)
 
+import eyed3
+
 from lib.avatar import AvatarReader
 
 VOICE_DIRNAME = 'voice2'
 IMG_DIRNAME = 'image2'
+EMOJI_DIRNAME = 'emoji'
 JPEG_QUALITY = 50
 
 class Resource(object):
@@ -25,6 +29,7 @@ class Resource(object):
         assert os.path.isdir(res_dir), "No such directory: {}".format(res_dir)
         self.res_dir = res_dir
         self.img_dir = os.path.join(res_dir, IMG_DIRNAME)
+        self.emoji_dir = os.path.join(res_dir, EMOJI_DIRNAME)
         assert os.path.isdir(self.img_dir), \
                      "No such directory: {}".format(self.img_dir)
         self.avt_reader = AvatarReader(self.res_dir)
@@ -78,7 +83,7 @@ class Resource(object):
         jpeg_str = buf.getvalue()
         return base64.b64encode(jpeg_str)
 
-    def get_img_file(self, fnames):
+    def _get_img_file(self, fnames):
         """ fnames: a list of filename to search for
             return (filename, filename) of (big, small) image.
             could be empty string.
@@ -120,13 +125,13 @@ class Resource(object):
 
     def get_img(self, fnames):
         """ return two base64 jpg string"""
-        big_file, small_file = self.get_img_file(fnames)
+        big_file, small_file = self._get_img_file(fnames)
 
         def get_jpg_b64(img_file):
             if not img_file:
                 return None
-            if not img_file.endswith('jpg'):
-                # possibly not jpg
+            if not img_file.endswith('jpg') and \
+               imghdr.what(img_file) != 'jpeg':
                 im = Image.open(open(img_file, 'rb'))
                 buf = cStringIO.StringIO()
                 im.save(buf, 'JPEG', quality=JPEG_QUALITY)
@@ -140,3 +145,17 @@ class Resource(object):
             pass
         return (big_file, small_file)
 
+    def get_emoji(self, md5, pack_id):
+        path = self.emoji_dir
+        if pack_id:
+            path = os.path.join(path, pack_id)
+        candidates = glob.glob(os.path.join(path, '{}*'.format(md5)))
+        candidates = [k for k in candidates if \
+                      not k.endswith('_thumb') and not k.endswith('_cover')]
+        if len(candidates) > 1:
+            # annimation
+            candidates = [k for k in candidates if not re.match('.*_[0-9]+$', k)]
+            assert len(candidates) == 1
+        fname = candidates[0]
+        return Resource.get_file_b64(fname), imghdr.what(fname)
+

lib/static/wx.css

@@ -5242,7 +5242,6 @@ a.btnBlue:active .btnBluePanel {
 }
 .cloudImg .cloudContent {
   min-height: 30px;
-  max-height: 200px;
 }
 .cloudImg .cloudContent img {
   min-width: 50px;
@@ -5447,8 +5446,6 @@ a.btnBlue:active .btnBluePanel {
 }
 .you .cloudImg .img_wrap{
     display: block;
-    max-width: 280px;
-    max-height: 150px;
     overflow: hidden;
 }
 .you .cloudImg .cloudArrow {