# Flickr Mail

Tool lives here: <https://edwardbetts.com/flickr_mail/>

Flickr Mail is a Flask app that helps find Flickr photos for Wikipedia articles
and contact photographers to request Wikipedia-compatible licensing.

## What It Does

- Searches Flickr from a Wikipedia article title/URL
- Shows license status for each result (free vs non-free CC variants)
- Builds a ready-to-send Flickr message for non-free licenses
- Finds image-less articles in a Wikipedia category
- Shows recent Commons uploads that came from Flickr mail outreach

## Project Layout

- `main.py`: Flask app routes and core logic
- `templates/`: UI templates
- `download_sent_mail.py`: sync Flickr sent messages into DB
- `download_commons_contributions.py`: sync Commons contributions into DB
- `update_flickr_uploads.py`: derive `flickr_uploads` from contributions/sent mail
- `flickr_mail.db`: SQLite database

## Database Pipeline

The recent uploads section depends on a 3-step pipeline:

1. `./download_sent_mail.py` updates `sent_messages`
2. `./download_commons_contributions.py` updates `contributions`
3. `./update_flickr_uploads.py` builds/updates `flickr_uploads`

`main.py` only reads `flickr_uploads`; it does not populate it.

## UploadWizard Detection

`update_flickr_uploads.py` supports both Commons UploadWizard comment styles:

- `User created page with UploadWizard` (older)
- `Uploaded a work by ... with UploadWizard` (newer)

It first tries to extract a Flickr URL directly from the contribution comment.
If absent, it falls back to Commons `extmetadata.Credit`.

## Local Run

Install dependencies (example):

```bash
pip install flask requests beautifulsoup4 sqlalchemy
```

Start the app:

```bash
python3 main.py
```

Then open:

- `http://localhost:5000/`

## Refresh Data

Run in this order:

```bash
./download_sent_mail.py
./download_commons_contributions.py
./update_flickr_uploads.py
```

Before running `./download_sent_mail.py`, create local auth config:

```bash
cp download_sent_mail.example.json download_sent_mail.local.json
```

Then edit `download_sent_mail.local.json` and set `cookies_str` to your full
Flickr `Cookie` header value.

## Notes

- `download_commons_contributions.py` uses an overlap window of known-only
  batches before stopping to avoid full-history scans while still catching
  shallow gaps.
- If a known Commons upload is missing from `flickr_uploads`, re-run the full
  3-step pipeline above.