Archived since my core message stream is Mastodon now. While my current solution for the requirement is to paste the link manually, I plan to build a tool to transform XBEL to RSS/ActivityStream. The new tool should cover the functionality better.
After you save a page as bookmark, post it in the Telegram channel
Config options are passed by either envs or dotenv
| name | description | default value | extra |
|---|---|---|---|
| BOT_TOKEN | Telegram bot token | / | |
| XBEL_URL | URL of the XBEL bookmark file | / | |
| DB_PATH | SQLite database file path | db.sqlite3 |
|
| POLL_INTERVAL | interval of requests to the XBEL URL | 360 | unit: second |
| CHAN_NAME | Telegram channel name sent messages to | / | Either starting with @ or not is OK, like @myl7s or myl7s |
| MSG_TEMPLATE | Message format string to generate sent messages | New bookmark: {title} {url} |
Python .format() style format string, whose {title} and {url} will be replaced with bookmark title and URL. Here string without {title} or {url} will cause exceptions. |
We will take Chrome bookmarks as example. First you need to export your bookmarks as XBEL (XML Bookmark Exchange Language) file and sync the export with your current bookmarks frequently. We can use floccus, which is a free open source software under MPL-2.0 License, to do it. As a browser addon, It can export and sync your bookmarks in Chrome, Firefox and some other browsers, to Google Drive or any other WebDAV service like NextCloud. You can also specify the sync policy, which includes merging, push-only and pull-only. Here we can install the addon in Chrome, add a push-only sync to export bookmarks to Google Drive, share the file from Google Drive, use this tool to convert the share link to direct link, and now you got the required XBEL URL.
Docker image myl7/tgchan-bookmark-bot is available in Docker Hub.
Tag edge is bleeding edge build from CI and tag latest is latest release with a version.
Semantic versioning is available so feel free to use tag 1, 1.1 or 1.1.1.
Dir /db is created in advance and it is suggested to store the database file in it for Docker binding mounting.
Run the image like:
docker run -d --name=bookmark --restart=unless-stopped \
-e BOT_TOKEN=<...> \
-e XBEL_URL=<...> \
-e DB_PATH=/db/db.sqlite3 \
-e CHAN_NAME=<...> \
-v <...>:/db \
myl7/tgchan-bookmark-bot:latestCopyright (c) 2022 myl7
SPDX-License-Identifier: Apache-2.0