Skip to content

Latest commit

 

History

History
224 lines (153 loc) · 7.05 KB

BUILD.rst

File metadata and controls

224 lines (153 loc) · 7.05 KB

Creating Pygment Lexer releases

These instructions cover steps needed to create new releases of Robot Framework Pygments lexer. Many individual steps are automated, but we don't want to automate the whole procedure because it would be hard to react if something goes terribly wrong. When applicable, the steps are listed as commands that can be copied and executed on the command line.

Contents

Preconditions

Operating system and Python requirements

Generating releases has only been tested on Linux, but it ought to work the same way also on OSX and other unixes. Generating releases on Windows may work but is not tested, supported, or recommended.

Creating releases is only supported with Python 3.6 or newer. If you are using Ubuntu or one of its derivatives and don't have Python 3.6 in the official package repository, you may consider using the Dead Snakes PPA.

The pip and invoke commands below are also expected to run on Python 3.6+. Alternatively, it's possible to use the python3.6 -m pip approach to run these commands.

Required Python modules

Many steps are automated using Invoke and also other tools/modules such as twine are needed. A pre-condition is installing all these tools, and that's easiest done using pip and the requirements-build.txt file:

pip install -r requirements-build.txt

Using Invoke

Invoke tasks are defined in tasks.py file and they are executed from the command line like:

inv[oke] task [options]

Run invoke without arguments for help. All tasks can be listed using invoke --list and each task's usage with invoke --help task.

Different Git workflows

Git commands used below always expect that origin is the project main repository. If that's not the case, and instead origin is your personal fork, you probably still want to push to the main repository. In that case you need to add upstream or similar to git push commands before running them.

Preparation

  1. Check that you are on the master branch and have nothing left to commit, pull, or push:

    git branch
git status
git pull --rebase
git push

  2. Clean up:

    invoke clean

  3. Set version information to a shell variable to ease copy-pasting further commands. Add aN, bN or rcN postfix if creating a pre-release:

    VERSION=<version>

    For example, VERSION=3.0.1 or VERSION=3.1a2.

Release notes

  1. Generate a template for the release notes. Either first create shell variables with GitHub login info or just replace them in the command to actually generate the notes:

    GITHUB_USERNAME=<username>
GITHUB_PASSWORD=<password>

invoke release-notes -w -v $VERSION -u $GITHUB_USERNAME -p $GITHUB_PASSWORD

    The -v $VERSION option can be omitted if version is already set. Omit the -w option if you just want to get release notes printed to the console, not written to a file.

    When generating release notes for a preview release like 3.0.2rc1, the list of issues is only going to contain issues with that label (e.g. rc1) or with a label of an earlier preview release (e.g. alpha1, beta2).

  2. Fill the missing details in the generated release notes template.

  3. Make sure that issues have correct information:

    • All issues should have type (bug, enhancement or task) and priority set. Notice that issues with the task type are automatically excluded from the release notes.
    • Issue priorities should be consistent.
    • Issue titles should be informative. Consistency is good here too, but no need to overdo it.

    If information needs to be added or edited, its better to edit it in the issue tracker than in the generated release notes. This allows re-generating the list of issues later if more issues are added.

  4. Add, commit and push:

    git add docs/robotframeworklexer-$VERSION.rst
git commit -m "Release notes for $VERSION" docs/robotframeworklexer-$VERSION.rst
git push

  5. Update later if necessary. Writing release notes is typically the biggest task when generating releases, and getting everything done in one go is often impossible.

Set version

  1. Set version information in robotframeworklexer.py:

    invoke set-version $VERSION

  2. Commit and push changes:

    git commit -m "Updated version to $VERSION" robotframeworklexer.py
git push

Tagging

  1. Create an annotated tag and push it:

    git tag -a v$VERSION -m "Release $VERSION"
git push --tags

  2. Add short release notes to GitHub's releases page with a link to the full release notes.

Creating distributions

  1. Checkout the earlier created tag if necessary:

    git checkout v$VERSION

  2. Cleanup. This removes temporary files as well as build and dist directories:

    invoke clean

  3. Create and validate source distribution in zip format and universal (i.e. Python 2 and 3 compatible) wheel:

    python3 setup.py sdist bdist_wheel
ls -l dist
twine check dist/*

    Distributions can be tested locally if needed.

  4. Upload to PyPI:

    twine upload dist/*

  5. Verify that project page at PyPI looks good.

  6. Test installation:

    pip install --pre --upgrade robotframeworklexer

Post actions

  1. Set dev version based on the previous version:

    invoke set-version dev
git commit -m "Back to dev version" robotframeworklexer.py
git push

    For example, 1.2.3 is changed to 1.2.4.dev1 and 2.0.1a1 to 2.0.1a2.dev1.

  2. Close the issue tracker milestone.

  3. Create new milestone for the next release unless one exists already.

Announcements

  1. robotframework-users and robotframework-announce lists. The latter is not needed with preview releases but should be used at least with major updates. Notice that sending to it requires admin rights.

  2. Twitter. Either Tweet something yourself and make sure it's re-tweeted by @robotframework, or send the message directly as @robotframework. This makes the note appear also at http://robotframework.org.

    Should include a link to more information. Possibly a link to the full release notes or an email to the aforementioned mailing lists.

  3. #devel and #general channels on Slack.

  4. Robot Framework LinkedIn group.