Content value now extracted differently

+ remove useless script

million/analyze/message_evaluation.py

@@ -7,12 +7,14 @@ import million.analyze.dns_solver as dns
 _memoization: Dict[Message, int] = {}
 _dns_solver: dns.DNS_solver = dns.DNS_solver()
 
+
 def get(msg: Message) -> int:
     """
     Returns the estimated value counted in this message
     """
     return _memoization.get(msg, _compute(msg))
 
+
 def reset(msg: Message) -> None:
     """
     Drop memorized value of this Message
@@ -20,6 +22,7 @@ def reset(msg: Message) -> None:
     if msg in _memoization:
         _memoization.pop(msg)
 
+
 def reset() -> None:
     """
     Drop every memorized message value
@@ -28,20 +31,23 @@ def reset() -> None:
 
 
 def _compute(msg: Message) -> int:
-    value = _dns_solver.solve(msg) or \
-        _computeContent(msg) or \
-        None
+    value = _dns_solver.solve(msg) or _computeContent(msg) or None
 
     _memoization[msg] = value
     return value
 
+
+# 1🍁 pour 1420 ?
+# @Elias Cheddar pour 69 ?
+
 def _computeContent(msg: Message) -> int:
-    # TODO parse potential math expressions in content
-    match = msg.content and re.search(r"\d+", msg.content)
+    if not msg.content:
+        return
     
+    s = re.sub(r"[^\s\d]","", msg.content)
+    match = re.search(r"\d+", s)
+
     if match:
-        value = int(match.group())
-    else:
-        value = None
-    
-    return value
+        return int(match.group())
+
+    return None

scripts/find_holes_v2.py

@@ -1,77 +0,0 @@
-from million.analyze.word_finder import retain_counts
-import million.analyze.message_evaluation as msg_val
-from million.model.sequence import Sequence
-import million.parse.fb_exports as fb
-
-
-def _find_value_around_index(messages, value, idx, amplitude) -> int:
-    check_value = lambda x: msg_val.get(messages[x]) == value
-
-    if check_value(idx):
-        return idx
-
-    for offset in range(1, amplitude):
-        o_idx = idx + offset * +1
-        if check_value(o_idx):
-            return o_idx
-
-        o_idx = idx + offset * -1
-        if check_value(o_idx):
-            return o_idx
-
-    return -1
-
-
-def _open_sequence(sequences, msg):
-    sequence = Sequence(
-        start=msg_val.get(msg), start_message=msg, end=-1, end_message=msg
-    )
-
-    sequences.append(sequence)
-
-
-def _close_sequence(sequences):
-    if len(sequences) == 0:
-        return
-
-    sequences[-1].end = msg_val.get(sequences[-1].end_message)
-
-
-def _opened_sequence(sequences):
-    return len(sequences) > 0 and sequences[-1].end == -1
-
-
-export = fb.parse_dirfiles("./data/")
-messages = retain_counts(export.messages)
-
-current = 1
-base_idx = 0
-amplitude = 200
-
-sequences = []
-
-while base_idx < len(messages):
-    curr_idx = _find_value_around_index(messages, current, base_idx, amplitude)
-    print(
-        f"searching {current} from [{messages[base_idx]}]\t-> {'Not found' if curr_idx == -1 else 'Itself' if curr_idx == base_idx else messages[curr_idx]}"
-    )
-
-    if curr_idx != -1:  # trouvé
-
-        if not _opened_sequence(sequences):
-            _open_sequence(sequences, messages[curr_idx])
-        else:
-            sequences[-1].end_message = messages[curr_idx]
-
-        base_idx = curr_idx + 1
-        current += 1
-    else:  # pas trouvé
-
-        # fermer la sequence si ouverte
-        if _opened_sequence(sequences):
-            _close_sequence(sequences)
-
-        if msg_val.get(messages[base_idx]) < current:
-            base_idx += 1
-        else:
-            current += 1