From 0a870dfa1136bbc179e65101be7fdea25ef6c1d3 Mon Sep 17 00:00:00 2001 From: Jon Michael Aanes Date: Sat, 26 Oct 2024 01:12:30 +0200 Subject: [PATCH] Moved most code into __init__ --- libpurple_to_markdown/__init__.py | 108 ++++++++++++++++++++++++++++++ libpurple_to_markdown/__main__.py | 92 +++---------------------- 2 files changed, 119 insertions(+), 81 deletions(-) diff --git a/libpurple_to_markdown/__init__.py b/libpurple_to_markdown/__init__.py index e69de29..2aa37dd 100644 --- a/libpurple_to_markdown/__init__.py +++ b/libpurple_to_markdown/__init__.py @@ -0,0 +1,108 @@ +import dataclasses +import re +import datetime +from pathlib import Path +import bs4 +import logging + +logger = logging.getLogger(__name__) + +@dataclasses.dataclass(frozen=True, order=True) +class Message: + sent_at: datetime.datetime + sender: str + text: str + + def __post_init__(self): + assert self.sent_at is not None + assert self.sender is not None + assert self.text is not None + +def datetime_sent(chat_start: datetime.datetime, message_sent: datetime.time) -> datetime.datetime: + naive = datetime.datetime.combine(chat_start.date(), message_sent, chat_start.tzinfo) + if chat_start.time() > message_sent: + naive = naive - datetime.timedelta(days=1) + return naive + + +def parse_messages_in_chat_file(path: Path) -> list[Message]: + logger.info('Parsing %s', path) + chat_start = datetime.datetime.fromisoformat(path.stem.removesuffix('CEST')) + + with open(path) as f: + soup = bs4.BeautifulSoup(f) + + messages = [] + + cur_sent_at: datetime.datetime | None = None + cur_sender: str = 'NOT DEFINED' + cur_text: str = '' + + for c in soup.body.children: + if c.name == 'font': + # Get timestamp + m = re.match(r'\((\d+):(\d+):(\d+)\)', c.font.get_text()) + time_sent = datetime.time(int(m.group(1)), int(m.group(2)), int(m.group(3))) + cur_sent_at = datetime_sent(chat_start, time_sent) + + # Get sender + cur_sender = c.b.get_text().strip().removesuffix(':') + + elif c.name is None: + cur_text = c.get_text() + + elif c.name == 'a': + cur_text = cur_text + '<' + c['href'] + '>' + + elif c.name == 'br': + messages.append(Message(cur_sent_at, cur_sender, cur_text.strip())) + cur_sent_at = None + cur_sender = 'NOT DEFINED' + cur_text = '' + + elif c.name == 'h3': + pass # Ignore log header + + + else: + assert False, c + + return messages + + + + +def parse_messages_in_chat_folder(chat_folder_path: Path) -> list[Message]: + messages = [] + for file_path in chat_folder_path.iterdir(): + messages.extend(parse_messages_in_chat_file(file_path)) + + messages.sort() + return messages + + +def format_messages(messages: list[Message]) -> str: + out = [ + '# Chat 2018' + '\n\n' + ] + + for msg in messages: + out.append(f'[[{msg.sent_at.date()}]] {msg.sent_at.time()} [[{msg.sender}]]:\n> {msg.text}\n') + del msg + + return ''.join(out) + +MSG_ADJACENTCY_DIST = datetime.timedelta(minutes=2) + +def is_adjacent_messages(first, second): + return first.sender == second.sender and second.sent_at - first.sent_at <= MSG_ADJACENTCY_DIST + +def merge_adjacent_messages(messages: list[Message]) -> list[Message]: + out = [] + for msg in messages: + if out and is_adjacent_messages(out[-1], msg): + out[-1] = dataclasses.replace(out[-1], text=out[-1].text + '\n\n' + msg.text) + else: + out.append(msg) + return out diff --git a/libpurple_to_markdown/__main__.py b/libpurple_to_markdown/__main__.py index e5549c8..b1035bb 100644 --- a/libpurple_to_markdown/__main__.py +++ b/libpurple_to_markdown/__main__.py @@ -1,90 +1,20 @@ - -import dataclasses -import re -import datetime +import argparse from pathlib import Path -import bs4 - -@dataclasses.dataclass(frozen=True, order=True) -class Message: - sent_at: datetime.datetime - sender: str - text: str - - def __post_init__(self): - assert self.sent_at is not None - assert self.sender is not None - assert self.text is not None - -def datetime_sent(chat_start: datetime.datetime, message_sent: datetime.time) -> datetime.datetime: - naive = datetime.datetime.combine(chat_start.date(), message_sent, chat_start.tzinfo) - if chat_start.time() > message_sent: - naive = naive - datetime.timedelta(days=1) - return naive - - -def parse_messages_in_chat_file(path: Path) -> list[Message]: - chat_start = datetime.datetime.fromisoformat(path.stem.removesuffix('CEST')) - - with open(path) as f: - soup = bs4.BeautifulSoup(f) - - messages = [] - - cur_sent_at: datetime.datetime | None = None - cur_sender: str | None = None - cur_text: str | None = None - - for c in soup.body.children: - if c.name == 'font': - # Get timestamp - m = re.match(r'\((\d+):(\d+):(\d+)\)', c.font.get_text()) - time_sent = datetime.time(int(m.group(1)), int(m.group(2)), int(m.group(3))) - cur_sent_at = datetime_sent(chat_start, time_sent) - - # Get sender - cur_sender = c.b.get_text().strip().removesuffix(':') - - elif c.name is None: - cur_text = c.get_text() - - elif c.name == 'a': - cur_text = '<' + c['href'] + '>' - - elif c.name == 'br': - messages.append(Message(cur_sent_at, cur_sender, cur_text)) - cur_sent_at = None - cur_sender = None - cur_text = None - - elif c.name == 'h3': - pass # Ignore log header - - - else: - assert False, c - - return messages - - - - -def parse_messages_in_chat_folder(chat_folder_path: Path) -> list[Message]: - messages = [] - for file_path in chat_folder_path.iterdir(): - messages.extend(parse_messages_in_chat_file(file_path)) - - messages.sort() - return messages - +from . import (parse_messages_in_chat_folder, merge_adjacent_messages, + format_messages) +def parse_args(): + parser = argparse.ArgumentParser() + parser.add_argument('path', type=Path) + return parser.parse_args() def main(): - path = Path() + args = parse_args() - for message in parse_messages_in_chat_folder(path): - print(f'({message.sent_at}) {message.sender} : {message.text}') + messages = parse_messages_in_chat_folder(args.path) + messages = merge_adjacent_messages(messages) + print(format_messages(messages)) if __name__ == '__main__': main()