Synchronize user stories in Pivotal Tracker with Cucumber features.
If you aren't using Cucumber, you can still use pickler as a Pivotal Tracker command line client, provided you humor it with a features/ directory containing a tracker.yml file.
gem install pickler --source=http://gemcutter.org echo "api_token: ..." > ~/.tracker.yml echo "project_id: ..." > ~/my/app/features/tracker.yml echo "ssl: [true|false]" >> ~/my/app/features/tracker.yml pickler --help
“ssl” defaults to false if not configured in the yml file.
For details about the Pivotal Tracker API, including where to find your API token and project id, see www.pivotaltracker.com/help/api .
The pull and push commands map the story's name into the “Feature: …” line and the story's description with an additional two space indent into the feature's body. Keep this in mind when entering stories into Pivotal Tracker.
Download all well formed stories (basically, any story with “Scenario:” in the body) that are not in the “unstarted” or “unscheduled” state to the features/ directory.
Upload all features with a tracker url in a comment on the first line.
pickler search <query>
List all stories matching the given query.
pickler start <story> <name>
Pull a given feature to features/<name>.feature and change its state to started.
pickler start <story> -
Pull a given feature to a file name based on the title and change its state to started. (I use this more than any other command, and you probably should, too).
pickler finish <story>
Push a given feature and change its state to finished.
pickler install-vim-plugin [<directory>]
Installs the Vim plugin to <directory>, or ~/.vim/plugin. This plugin currently only provides omnicomplete (CTRL-X, CTRL-O) of feature ids in Git commit messages.
Full list of commands.
pickler <command> --help
Further help for a given command.
Pull requests will be ignored if they don't follow the Git convention of a 50 character or less subject and optional paragraphs wrapped to 72 characters. See stopwritingramblingcommitmessages.com/.
If I provide you with feedback on your pull request, generally you should squash your changes into the previous commit when submitting a second request.
No warranties, expressed or implied.
Notably, the push and pull commands are quite happy to blindly clobber features if so instructed. Pivotal Tracker has a history to recover things server side.