Last active
September 2, 2022 07:55
-
-
Save daniele-athome/ea10806fa3d37f4b2285fff235b5555b to your computer and use it in GitHub Desktop.
Convert a Kontalk database (v4) to HTML files
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python3 | |
# Convert a Kontalk conversation to HTML | |
# Usage: ./convert-threads.py messages.db | |
# HTML goes to files in current directory, one per thread | |
import sys | |
import hashlib | |
import sqlite3 | |
from datetime import datetime | |
DISPLAYNAME_ME = 'Me' | |
COLOR_ME = 'black' | |
COLOR_PEER = '#a94138' | |
def sha1(text): | |
hashed = hashlib.sha1(text) | |
return hashed.hexdigest() | |
def dict_factory(cursor, row): | |
d = {} | |
for idx, col in enumerate(cursor.description): | |
d[col[0]] = row[idx] | |
return d | |
def get_group_name(conn, threadId): | |
g = conn.cursor() | |
g.execute('SELECT * FROM groups WHERE thread_id = ?', (threadId, )) | |
val = g.fetchone() | |
g.close() | |
return val['subject'] if val else None | |
dbfile = sys.argv[1] | |
conn = sqlite3.connect(dbfile) | |
conn.row_factory = dict_factory | |
t = conn.cursor() | |
t.execute('SELECT * from threads ORDER by timestamp') | |
thread = t.fetchone() | |
while thread: | |
c = conn.cursor() | |
c.execute('SELECT * FROM messages WHERE thread_id = ? ORDER BY timestamp', (thread['_id'], )) | |
groupName = get_group_name(conn, thread['_id']) | |
if groupName: | |
displayName = groupName | |
else: | |
displayName = thread['peer'] | |
output_file = open(thread['peer'] + '.html', mode='w+') | |
print(""" | |
<!DOCTYPE html> | |
<html lang="en"> | |
<head> | |
<title>Kontalk messages with %s</title> | |
<meta charset="UTF-8"> | |
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1"> | |
<style type="text/css"> | |
* { | |
padding: 0; | |
margin: 0; | |
} | |
body { | |
font-family: arial, helvetica, serif; | |
} | |
h1 { | |
margin: 12px; | |
} | |
#conversation { | |
width: 100%%; | |
} | |
.stamp { | |
background-color: #f2f2f2; | |
color: #20435c; | |
font-weight: bold; | |
text-align: center; | |
padding: 8px; | |
} | |
.chat { | |
padding: 4px; | |
} | |
.chat .incoming { | |
color: %s; | |
} | |
.chat .outgoing { | |
color: %s; | |
} | |
.peer { | |
font-weight: bold; | |
} | |
</style> | |
</head> | |
<body> | |
<h1>Conversation with %s</h1> | |
<div id="conversation"> | |
""" % (displayName, COLOR_PEER, COLOR_ME, displayName), file=output_file) | |
cur_date = None | |
row = c.fetchone() | |
while row: | |
date = datetime.fromtimestamp(row['timestamp']/1000) | |
date_fmt = date.strftime('%Y-%m-%d') | |
if not row['body_mime'] and not row['att_mime']: | |
content = "(encrypted)" | |
elif row['body_content']: | |
try: | |
content = row['body_content'].decode('utf-8') | |
except UnicodeDecodeError: | |
# damn Windows!! | |
content = row['body_content'].decode('iso-8859-1') | |
elif row['att_mime']: | |
content = "Media: %s" % (row['att_mime'], ) | |
else: | |
raise ValueError("Unsupported message") | |
if not cur_date or date_fmt != cur_date.strftime('%Y-%m-%d'): | |
cur_date = date | |
print(""" | |
<div class="stamp">%s</div> | |
""" % (date_fmt,), file=output_file) | |
if row['direction'] == 0: | |
name = row['peer'] | |
css = 'incoming' | |
else: | |
name = DISPLAYNAME_ME | |
css = 'outgoing' | |
print(""" | |
<div class="chat"> | |
[%s] <span class="peer %s">%s</span>: %s | |
</div> | |
""" % (date.strftime('%H:%M:%S'), css, name, content), file=output_file) | |
row = c.fetchone() | |
print(""" | |
</div> | |
</body> | |
</html> | |
""", file=output_file) | |
thread = t.fetchone() | |
conn.close() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment