Merge branch 'feature/find_gromots' of DemiSel/Messenger_1Million_Stats into master

million/analyze/word_finder.py

@@ -0,0 +1,11 @@
+import re
+from typing import List
+from million.model.message import Message
+
+
+def _wordFilter(msg: Message, words: List[str]) -> bool:
+    rgx = r"(\b"+ r'\b|\b'.join(words) + r"\b)"
+    return msg.content and re.search(rgx, msg.content, re.I)
+
+def findWords(messages: List[Message], words: List[str]) -> List[Message]:
+    return filter(lambda m: _wordFilter(m, words), messages)

scripts/find_gromots.py

@@ -0,0 +1,44 @@
+from datetime import datetime
+from million.analyze.word_finder import findWords
+from million.parse.fb_exports import FacebookExportParser
+
+
+DATA_PATH = './data/'
+
+parser = FacebookExportParser()
+
+export = parser.parse(DATA_PATH)
+
+gros_mots = [
+    '.*merde.*',
+    'sexe',
+    'pute',
+    'pé?dé?',
+    'putain',
+    'bite',
+    'encul.*',
+    'cul',
+    'nichon',
+    'gueule',
+    'con(ne)?',
+    'chatte',
+    'niqu.*',
+    'chi(é|e).*',
+    'bais.*'
+    ]
+
+msg_gros_mots = findWords(export.messages, gros_mots)
+
+msg_gros_mots_grp = {}
+
+for msg in msg_gros_mots:
+    if msg.sender_name not in msg_gros_mots_grp: msg_gros_mots_grp[msg.sender_name] = []
+    msg_gros_mots_grp[msg.sender_name].append(msg)
+
+for name in sorted(msg_gros_mots_grp, key = lambda k: len(msg_gros_mots_grp[k])):
+    print(name)
+
+    for msg in msg_gros_mots_grp[name]:
+        time = datetime.fromtimestamp(msg.timestamp_ms / 1000)
+        time_str = time.strftime("%d/%m/%Y %H:%M:%S")
+        print(f"\t{time_str} : {msg.content}")