Skip to content

karlicoss/grasp

Repository files navigation

Grasp is a browser extension for Chrome and Firefox, which adds a button/keybinding to capture current page title and url, possibly selected text, additional comments or tags and adds it into your Org Mode file.

Screenshot, short demo.

Running

In the simplest setup, the server runs locally, and you can use 'localhost' version of the extension. If you have to work on a computer where you can't run python scripts, or your target capture file is just not there, you can selfhost the server part elsewhere. Don't forget to set the endpoint in extension settings!

Setup

  • install grasp_backend package: pip3 install --user grasp-backend

  • install systemd/launchd service to autorun grasp

    python3 -m grasp_backend setup --path /path/to/your/capture.org [--port <custom port>] [--template <custom org-capture template>]

    Or alternatively, just run it directly if you don't want to autostart python3 -m grasp_backend serve --path /path/to/your/capture.org [--port <custom port>] [--template <custom org-capture template>]

  • install chrome extension and configure hotkeys

That's it! If you're using custom port make sure it's the same as in the extension settings (default is 12212).

Configuration

Here you can find some references for the --template syntax.

If you are looking for more flexible formatting that's not supported by template syntax, see config.py.example. You can modify it to your liking and pass as --config to grasp_backend setup command.

Motivation

Why use org-capture? Well, it's hard to explain, maybe some other time... However, if you do know you want to use it instead of/alongside your browser bookmarks, by default you don't have much choice and have to copy everything manually. For an experienced enough org-mode user it's no less than a torture.

This tool:

  • + shows a notification when capturing fails/succeeds, so you won't lose your notes

  • + doesn't require always running Emacs, simply appends an org-mode text entry to a file

  • + can capture things that org-protocol can't handle (e.g. extra comment or tags)

  • + can potentially use any plaintext format as a storage.

    E.g. you might be more of a Markdown or Todo.txt fan (let me know if you are interested in that!).

  • - doesn't talk to Emacs, so can't benefit from Emacs capture templates

    E.g. currently you can't point at a specific header in an org file, it would just append at the end.

  • - requires running a small HTTP server

    However, there are no dependencies apart from python3, so in many ways, it's even more portable than Emacs.

Comparison with similar tools:

  • - relies on org-protocol and MIME handler: flaky for many people and has no feedback whether capture failed or succeeded

    Losing few days of captured stuff due to MIME handler mysteriously not working was the main motivator for me to develop grasp.

  • - requires always running Emacs, which might not be the case for some people

  • + relies on org-protocol, so can potentially be better integrated with Emacs and your org-mode files

Same pros/cons as org-capture-extension as it's relying on org-protocol.

In addition:

  • + using a bookmarklet, hence browser-agnostic
  • + capable of on the fly HTML to org-mode markup conversion

Potential improvements

Permissions used

  • http://localhost/capture for talking with the backend

    If you want to use an external URL as an endpoint, you will be prompted for a permission dynamically.

  • storage for settings

  • notifications for showing notification

  • activeTab for requesting page info

Building & developing

The most up-to-date instructions should be in CI config.

You need npm for building the extension.

npm install
./build --target <browser> # e.g. ./build --target chrome or ./build --target firefox

After that you can find the extension in dist directory and 'Load unpacked** if necessary. There is also Flow and Eslint set up.

testing and linting

Check CI config to figure out all the checks I'm doing.

There are some end2end tests which check both web extension and the browser, but require GUI, so they can't run on github actions. You can run them manually though.

  • pytest -s --pyargs tests.test_end2end

publishing

  • run ./publish to generate extension zip files

  • firefox: ./build --firefox --release --lint --sign

    After than, upload the signed xpi file on AMO

  • chrome: ./build --chrome --release --lint

    After that, upload the zip (generated by publish script) on Web store

Credits