Atlassian CLI

This program is an easy-to-use, scriptable API for interacting with JIRA (and eventually other Atlassian products) via the REST API.

Whenever possible, this CLI will implement a "pure JSON" option for input and output, allowing maximum machine-parsability. RecordStream is a huge potential candidate for using with the atlas-cli.

Usage

SYNOPSIS

atlas-jira-cli --help # for full help page

atlas-jira-cli jql "assignee = someuser and project = QA and created > -1d"
atlas-jira-cli view QA-1234
atlas-jira-cli transition QA-1234 --editState close
atlas-jira-cli edit QA-1234 --fix-versions='+1.1.1' --fix-versions='-1.1.0' --comment "bumping to a later release"
atlas-jira-cli comment QA-1234 --comment "This is harder than I thought!"
atlas-jira-cli create --project-key QA --summary "File all the tickets!" \
        --description "Meme of the year, amirite?" \
        --priority p1 \
        --components regression \
        --affects-versions 1.0.0 \
        --affects-versions 1.1.0 \
        --fix-versions 1.1.1

Setting up with RVM

Since atlassian-cli requries a custom version of the terminal-table gem, it must always be executed with bundle exec.

To set this up, it is advised you do the following:

rvm install ruby # if not already done, will grab newest c-ruby
rvm use ruby@atlascli --create
gem install bundler # may be needed
bundle install
# create a bash script in your path, obviously adjust the paths to point to
# your rvm directory and your checkout of atlassian-cli

#!/bin/bash
ATLAS_CLI_DIR=/home/$USER/gitrepos/atlassian-cli
RVM_DIR=/home/$USER/.rvm
GEMSET=ruby-2.0.0-p195@atlascli
cd $ATLAS_CLI_DIR && $RVM_DIR/bin/$GEMSET -S bundle exec $ATLAS_CLI_DIR/bin/atlas-jira-cli "$@"

# you may also have to run this once for each ruby to generate the stubs, depending on your rvm version:
rvm wrapper ruby@atlascli

Authentication

Initially, the script will prompt for a username and a password. If you don't want to type it in every time you can store credentials in your ~/.netrc file using the standard format (see man netrc for details).

DEV GUIDE

To run an instance of jira for testing using the atlassian plugin sdk

atlas-run-standalone --server localhost --product jira --version 5.2.11

Note that you have to create the project and some issues first... intial db is empty.

To hit an endpoint using curl, for example:

curl --user admin:admin 'http://localhost:2990/jira/rest/api/2/search?jql=project%3Dfoo'

TODO

There are many things I'd eventually like to implement. For now, they are broken up by product.

BUGS

• Create new issue with no component/type, try to edit it, profit?

JIRA

• fix transition to also update fields, etc, current ignores all but comment
• Create Components
• Support converting regular issues to sub-tasks (blocked by https://jira.atlassian.com/browse/JRA-27893)
• Move issue (also not supported by the rest API: https://answers.atlassian.com/questions/132846/how-move-issue-to-another-project-via-rest-api)
• JQL Query Pagination
• list components, priorities, assignees, etc.
• issue labels
• Selectively get only fields we need (PERF)
• Clean up regex based things (allow exact match, --dry-run, specify by ID, etc)
• Extract more hashifiers (for comments, etc) for better code reuse
• A way to query required fields

Confluence

• Fetch raw page, upload raw page
• vim/emacs integration

Stash

• Search for projects/repositories
• Search for pull requests
• Create pull request
• Permissions admin?

Generic

• Support using a session cookie to cache credentials (p0, otherwise the CLI puts undue load on crowd/ldap)
• TEST COVERAGE
• Implement oauth support?
• Support mac OSX keychain?

Bamboo

• Search / list builds
• Trigger builds
• Fetch artifacts / logs

Crowd

• Test auth credentials
• Administration?

Fisheye

• lol, jk