c0de-fox / simple-bookmarking-service

This is a small Python3 API and website that will store bookmarks

A simple bookmarking engine

This is a small python3 API and website that allows saving bookmarks into an SQLite3 database.

The production environment will eventually be hosted at save.c0de.link

Usage

This project uses Pipenv to manage the dependencies, so install that first after cloning this repository.

• Install dependencies - pipenv install
• Create a Telegram bot (see Creating a Telegram Bot)
• Enter the environment - pipenv shell
• Start the server - bash run.sh

The server will be running on http://localhost:8080/

Creating a Telegram bot

1. Login to my.telegram.org/apps and create a Web App - You will need the api_id and api_hash
2. Message the BotFather to get an api_token
3. Copy sample.env to .env and fill in the details from the previous steps

Available Requests

Now that you have a running server, you can issue requests via a web browser, curl, wget, etc to the following endpoints:

• / - Currently returns the string "This is the index"
• /save/<title>/<uri> - Creates a new bookmark entry for the provided uri. title is a human readable word/set of words to recognize saved bookmarks quickly. Returns a JSON object containing the uuid
• /get/all or /getall - Returns a JSON object containing all bookmarks saved, or a 404 error if there are none
• /get/<uuid> - Returns a JSON object containing the bookmark. uuid is a UUID that is returned from /save, /get/all and /get/<uuid>
• /delete/<uuid> - Returns a JSON object containing the uuid and deletion status. Deletes the bookmark. uuid is a UUID that is returned from /save, /get/all and /get/<uuid>
• /update/title/<uuid>/<title> - Changes the bookmark's title with the newly provided one. Returns the same bookmark object as /get/<uuid>. uuid is a UUID that is returned from /save, /get/all and /get/<uuid>
• /update/uri/<uuid>/<uri> - Changes the bookmark's uri with the newly provided one. Returns the same bookmark object as /get/<uuid>. uuid is a UUID that is returned from /save, /get/all and /get/<uuid>. By design, the bookmark's uuid will be updated. If the updated uuid is already in the database, an HTTP 409 Conflict will be raised.

The Bookmark Object

The following fields will be returned with /get and /update requests:

• uuid - Required (generated automatically) - This is used for /get, /update and /delete requests
  • This is regnerated when updating the URI due to the fact that it is based on the URI of the bookmark
  • This is generated with the following criteria:

  Generate a UUID based on the SHA-1 hash of the python URL namespace identifier
  (which is 6ba7b811-9dad-11d1-80b4-00c04fd430c8) and a uri (which is a string)
  

• uri - Required (user provided) - This is a full link. Currently the API does not verify that it is reachable, so it can be any form of URI, including UNC Paths, or even a bare string
• title - Required (user provided) - A human readable word/set of words to recognize saved bookmarks quickly
• date_created - Required (generated automatically) - This is the date and time of the server when the request was received to save a bookmark
• date_updated - Optional (generated automatically) - This is the date and time of the server when the request was received to update a bookmark

Completed Features:

• Database - Capable of any user-defined database file (multiple instances); Has Create/Read/Update/Delete methods for bookmarks
• API - All of the features in the database have been implemented in the API (currently unauthenticated and unsanitized and everything is GET Requests) - Uses Bottle

Features In Progress:

• Authentication - Uses Telethon to provide me with an API key

Upcoming Features (TODO):

• Authentication - I'm thinking about using Telethon which will send me auth tokens, which would then be sent along with the request
• Website UI - I want the website to be extremely easy to use. I plan on using Bootstrap 4 with a style very similar to lob.li
• API - Able to accept various HTTP methods (POST, GET for sure, possibly PUT and DELETE). This will interface with the database, and provide data sanitization. I'm planning on using Bottle to accomplish this
  • Fetch various meta-data from the URI if it is reachable
    • HTML title, favicon, etc
  • Determine if a link is reachable
• Chrome Extension - This will allow exporting bookmarks from Chrome into this service, as well as quickly adding pages, most likely through a right click menu and by clicking on the icon
• Firefox Extension - Everything the Chrome extension can do