From cbb6f77fa2a1d46ff6052baeefd26a44d878f2a3 Mon Sep 17 00:00:00 2001
From: Yuchen Pei <hi@ypei.me>
Date: Thu, 26 Aug 2021 21:18:22 +1000
Subject: started working on erc type logs.

---
 gentree.py | 58 +++++++++++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 51 insertions(+), 7 deletions(-)

diff --git a/gentree.py b/gentree.py
index 220edf3..0a472bc 100644
--- a/gentree.py
+++ b/gentree.py
@@ -1,6 +1,7 @@
 from enum import Enum
 import sys
 import pprint
+import textwrap
 
 class LogType(Enum):
     QUASSEL = 1
@@ -44,7 +45,7 @@ def parse_log_quassel(log):
     [hh:mm:ss] <sender-nick> blahblah.
     or
     [hh:mm:ss] -*- sender-nick does something.
-    [06:05:28] *** Mode #fsf +o zoe by ChanServ
+    [hh:mm:ss] *** Mode #abc +o def by ChanServ
     args:
       log: a string of quassel log.
     return:
@@ -55,13 +56,12 @@ def parse_log_quassel(log):
     messages = []
     for line in log.splitlines():
         message = Message()
-        is_me = False
+        line = line.lstrip()
         [_, nick, payload] = line.split(' ', maxsplit=2)
         if nick == '***':
             continue
         if nick == '-*-':
             [_, _, nick, payload] = line.split(' ', maxsplit=3)
-            is_me = True
         else:
             nick = nick[1:-1] # removes <>
         message.message = line
@@ -85,13 +85,51 @@ def parse_log_erc(log):
     An erc message is in the form of
     <nick> blahblah
            blahblah blah  [hh:mm]
+    ** nick does something
+    **** old-nick is now known as new-nick
     And the timestamp is optional
     args:
       log: a string of quassel log.
     return:
       a list of Messages
     """
-    return []
+    last_messages = dict()
+    index = 0
+    messages = []
+    log_lines = log.splitlines()
+    log_lines.append('')
+    nick = ''
+    for line, next_line in zip(log_lines, log_lines[1:]):
+        line = line.lstrip()
+        if line[0] == '<':
+            [nick, payload] = line.split(' ', maxsplit=1)
+        elif line[:3] = '** ':
+            [_, nick, payload] = line.split(' ', maxsplit=2)
+        elif line[:5] = '**** ':
+            continue
+        else:
+            payload += f'\n{line}'
+        if nick == '***':
+            continue
+        if nick == '-*-':
+            [_, _, nick, payload] = line.split(' ', maxsplit=3)
+        else:
+            nick = nick[1:-1] # removes <>
+        message.message = line
+        message.id = index
+        # Remove special symbol for op
+        if nick[0] in ['@', '%']:
+            nick = nick[1:]
+        message.sender = nick
+        if nick in last_messages:
+            message.previous_msg = last_messages[nick]
+        last_messages[nick] = index
+        for participant, msg_id in last_messages.items():
+            if payload.find(participant) != -1:
+                message.reply_to_msgs.append(msg_id)
+        messages.append(message)
+        index += 1
+    return messages
 
 def parse_log(log):
     """parse log.
@@ -111,18 +149,24 @@ def parse_log(log):
 def gen_graph(meeting):
     dot = """
 digraph meeting {
+  graph [
+    width = 20
+  ];
+
   node [
     shape = box,
-    width = 40
+    margin = 0,
+    pad = 0
   ];
 """
     for msg in meeting:
-        quoted_message = msg.message.replace('"', '\\"')
+        quoted_message = textwrap.fill(msg.message.replace('"', '\\"'))
         dot += f'\nm{str(msg.id)} [label="{quoted_message}"];'
         if msg.id > 0:
             dot += f'\nm{str(msg.id - 1)} -> m{str(msg.id)} [style="invis"];'
+    for msg in meeting:
         if msg.previous_msg > -1:
-            dot += f'\nm{str(msg.previous_msg)} -> m{str(msg.id)};'
+            dot += f'\nm{str(msg.previous_msg)} -> m{str(msg.id)} [style="dashed"];'
         for id in msg.reply_to_msgs:
             dot += f'\nm{str(id)} -> m{str(msg.id)};'
     dot += '\n}'
-- 
cgit v1.2.3