Features

• Catches and logs workflow errors for easier development and support

• “Magic” arguments to help development/debugging

• Auto-saves settings

• Super-simple data caching

• Fuzzy, Alfred-like search/filtering with diacritic folding

• Keychain support for secure storage (and syncing) of passwords, API keys etc.

• Simple generation of Alfred feedback (XML output)

• Input/output decoding for handling non-ASCII text

• Lightweight web API with modelled on requests

• Pre-configured logging

• Painlessly add directories to sys.path

• Easily launch background tasks (daemons) to keep your workflow responsive

• Check for new versions and update workflows hosted on GitHub.

• Post notifications via Notification Center.

Quick Example

Here’s how to show recent Pinboard.in posts in Alfred.

Create a new workflow in Alfred’s preferences. Add a Script Filter with Language /usr/bin/python and paste the following into the Script field (changing API_KEY):

import sys
from workflow import Workflow, ICON_WEB, web

API_KEY = 'your-pinboard-api-key'

def main(wf):
    url = 'https://api.pinboard.in/v1/posts/recent'
    params = dict(auth_token=API_KEY, count=20, format='json')
    r = web.get(url, params)
    r.raise_for_status()
    for post in r.json()['posts']:
        wf.add_item(post['description'], post['href'], arg=post['href'],
                    uid=post['hash'], valid=True, icon=ICON_WEB)
    wf.send_feedback()


if __name__ == u"__main__":
    wf = Workflow()
    sys.exit(wf.run(main))

Add an Open URL action to your workflow with {query} as the URL, connect your Script Filter to it, and you can now hit ENTER on a Pinboard item in Alfred to open it in your browser.

Installation

Note: If you intend to distribute your workflow to other users, you should include Alfred-Workflow (and other Python libraries your workflow requires) within your workflow’s directory as described below. Do not ask users to install anything into their system Python. Python installations cannot support multiple versions of the same library, so if you rely on globally-installed libraries, the chances are very good that your workflow will sooner or later break—or be broken by—some other software doing the same naughty thing.

pip install --target=/path/to/my/workflow Alfred-Workflow

Your Workflow/
    info.plist
    icon.png
    workflow/
        __init__.py
        background.py
        notify.py
        Notify.tgz
        update.py
        version
        web.py
        workflow.py
    yourscript.py
    etc.

Your Workflow/
    info.plist
    icon.png
    workflow-1.X.X.zip
    yourscript.py
    etc.

Documentation

Detailed documentation, including a tutorial, is available at http://www.deanishe.net/alfred-workflow/.