Merge WhatsApp 'Message Yourself' exports into one sorted, deduplicated txt file.

script.py

```
#!/usr/bin/env python3
"""
Merge WhatsApp 'Message Yourself' exports into one sorted, deduplicated .txt.

Usage:
    python merge_whatsapp.py <folder> [output.txt]

- Reads every .txt in <folder>.
- Parses messages (multi-line messages preserved as one entry).
- Deduplicates messages by exact text — duplicate files and overlapping date
  ranges across exports both collapse automatically.
- Sorts oldest -> newest.
- Writes to output.txt (default: <folder>/merged.txt).
"""

import re
import sys
from datetime import datetime
from pathlib import Path

# Header of a new message: "20/01/2026, 6:30 pm - "
# Tolerates invisible unicode chars (LRM/RLM/BOM) WhatsApp sometimes inserts,
# and the narrow-no-break-space WhatsApp uses around am/pm (matched by \s).
MSG_START = re.compile(
    r'^[\u200e\u200f\ufeff]*'
    r'(\d{1,2}/\d{1,2}/\d{4}),\s+'   # date
    r'(\d{1,2}:\d{2})\s*'             # time
    r'(am|pm|AM|PM)\s+-\s',           # period + " - "
)


def parse_timestamp(date: str, time: str, ampm: str) -> datetime:
    return datetime.strptime(f"{date} {time} {ampm.upper()}", "%d/%m/%Y %I:%M %p")


def parse_messages(path: Path):
    """Return list of (timestamp, full_text) for every message in a chat export."""
    out = []
    buf_ts, buf_lines = None, []
    with open(path, encoding='utf-8', errors='replace') as f:
        for line in f:
            line = line.rstrip('\n')
            m = MSG_START.match(line)
            if m:
                if buf_ts is not None:
                    out.append((buf_ts, '\n'.join(buf_lines)))
                buf_ts = parse_timestamp(m.group(1), m.group(2), m.group(3))
                buf_lines = [line]
            elif buf_ts is not None:
                # continuation of a multi-line message
                buf_lines.append(line)
            # lines before the first message (encryption notice) are skipped
        if buf_ts is not None:
            out.append((buf_ts, '\n'.join(buf_lines)))
    return out


def main():
    if len(sys.argv) < 2:
        print(__doc__)
        sys.exit(1)

    folder = Path(sys.argv[1])
    output = Path(sys.argv[2]) if len(sys.argv) > 2 else folder / 'merged.txt'

    txts = sorted(p for p in folder.glob('*.txt') if p.resolve() != output.resolve())
    if not txts:
        print(f"no .txt files found in {folder}")
        sys.exit(1)

    # 1. parse every file
    all_msgs = []
    for p in txts:
        msgs = parse_messages(p)
        print(f"  {p.name:40s} {len(msgs):>6} messages")
        all_msgs.extend(msgs)

    # 2. dedupe by exact message text
    seen, deduped = set(), []
    for ts, text in all_msgs:
        if text not in seen:
            seen.add(text)
            deduped.append((ts, text))

    # 3. sort oldest -> newest (stable)
    deduped.sort(key=lambda x: x[0])

    # 4. write
    with open(output, 'w', encoding='utf-8') as f:
        for _, text in deduped:
            f.write(text + '\n')

    print(f"\n{len(all_msgs)} parsed  ->  {len(deduped)} unique  ->  {output}")


if __name__ == '__main__':
    main()
```