1
0
mirror of https://github.com/qTox/qTox.git synced 2024-03-22 14:00:36 +08:00
qTox/MAINTAINING.md
Mick Sayson 8abe8320d2 chore(CI): Use docker for CI scripts
Motiviation:
* Reproducing issues in CI is currently difficult
* Predicting issues in CI is currently difficult if you are not on
  ubuntu 18.04
* Reproducing issues submitted from other distros is currently done by
  creating a VM of that distro and building qtox for it locally
* Documentation for how to build on different distros is out of date
* Issues on non-ubuntu distributions are not caught by CI
* Cross compiling for windows locally is not trivial
* Iterating when working with custom build scripts is slow, scripts
  don't necessarily support re-running without re-starting the docker
  container and re-building qtox again
* Updating dependencies is a pain

Changes:
* docker-compose file has been added to the root of our repo.
  After `docker compose run --rm ubuntu` (or other supported distros),
  you are ready to compile and run qtox
* Dependencies are owned by dependency install scripts in buildscripts/.
  This allows us to use the same exact dependencies in our
  OSX/windows/linux scripts
* New docker images have been added for a variety of distributions.
  These are now run in CI in a variety of configurations
  * Docker images are cached in CI so rebuild time for the majority of
    jobs is quite quick
* Build scripts have been trimmed to leverage state of docker
  containers.
  * Windows build script no longer installs anything, dependencies are
    now managed by the windows_builder docker images
  * Build scripts should now be easily re-runnable. Usage is now `docker
    compose run --rm <image>` and then run the scripts
* All artifacts are now uploaded to github after build, this means we
  can take an appimage/flatpak/exe/dmg for any given PR and try it out
  without having to build it ourselves

Notes:
* Docker image size is quite important. We have a maximum of 5GB cache
  space on github actions. The majority of the linux distro docker
  images cache at ~300-400MB, which gives us room to test ~6 distros
  after accounting for the sizes of flatpak/windows docker images
* Docker layer ordering is relatively intentional. Approximate order
  should be that large dependencies that change infrequently should be
  farther up. This lowers the amount of rebuilding we have to do when
  dependencies are updated
* download_xxx.sh scripts are the cleanest way I could find to implement
  a shared dependency map between osx scripts and docker containers.
  Although it would be nice to have a single dependency mapping file,
  splitting it into individual scripts allows us to only rebuild some
  docker layers when dependencies are updated.
* Github actions are split between docker image building and docker
  image use. This allows us to re-use the same docker images for
  multiple jobs, but only build it once
  * Unfortunately I could not find a way to de-duplicate the stitching
    between jobs, so we have a lot of copy pasta in that area
2021-12-19 14:56:05 -08:00

11 KiB
Raw Blame History

Guidelines, overview of maintenance process, etc.

“Thou shall GPG-sign.”

Git config

GPG signing

While contributors are suggested to use GPG, as a maintainer you are required to use GPG to sign commits & merges.

If you don't have GPG signing set up yet, now is the moment to do it.

Config, etc.

SSH

Preferably use SSH.

There are quite a few articles about that: https://help.github.com/categories/ssh/

Useful aliases

Check whether commits are GPG-signed with git logs

git config --global alias.logs 'log --show-signature'

Commits

  • always use commit message format

  • always GPG-sign your commits.

  • it's preferable to make a PR with changes that you're about to commit.

    Yes, there might be a situation where something has to be fixed "right now" on master..

    Perhaps a security fix, who knows what future holds. If it's not that important, you're still better off making a PR. Even when you'll just fast-forward commits from PR onto the master branch.

    Reasoning for it is that it's always hard to catch bugs/mistakes that you create, while someone else who just briefly looked at the changeset possibly can see a problem :)

Pull requests

  • do not push any Merge, Squash & Merge, etc. buttons on the website! The only allowed way of merging is locally, since otherwise merge will not be signed, and websites can fairly well mess things up.

  • always test PR that is being merged.

  • always GPG-sign PR that you're merging.

    Commits that are about to be merged don't have to be signed, but the merge-commit must be signed. To simplify the process, and ensure that things are done "right", it's preferable to use the merge-pr.sh script, which does that for you automatically.

  • use merge-pr.sh script to merge PRs. First checkout the target branch, usually either master or a release dev branch e.g. v1.17-dev, make sure it's up to date with qTox/qTox, then e.g. ./merge-pr.sh 1234.

    You don't have to use it, but then you're running into risk of breaking travis build of master & other PRs, since it verifies all commit messages, indlucing merge messages.

    Risk, that can be avoided when one doesn't type manually merge message 😉

  • might want to use test-pr.sh.

  • give a PR some "breathing space" right after it's created i.e. merging something right away can lead to bugs & regressions suddenly popping up, thus it's preferable to wait at least a day or so, to let people test & comment on the PR before merging.

    • with trivial changes, like fixing typos or something along those lines, feel free to merge right away.
  • if a PR requires some changes, comment what parts need to be adjusted, preferably by using the Reviewable button on the first comment.

  • if PR doesn't apply properly on top of current master (when using merge-pr.sh script), request a rebase

  • if a PR requires changes but there has been no activity from the PR submitter for more than 2 months, close the PR.

Continous Integration

qTox nightly builds can be found in qTox-nightly-release. Should one build fail, it is important to restart the whole Travis CI build and not just a single job. The tool managing the nightly builds deletes all build artifacts on any job failure, so all need to be rebuilt.

Issues

Tagging Issues

  • When you request more info to be provided in the issue, tag it with O-need-info. Remove tag once the needed info has been provided.
    • If the needed information is not provided after 30 days, add the O-stale tag and a comment requesting the information again.
      • If the O-stale tag is present for more than 30 day, the issue should be closed.
  • If you're going to fix the issue, assign yourself to it.
  • when closing an issue, preferably state the reason why it was closed, unless it was closed automatically by commit message.
  • When issue is a duplicate, close the issue with less useful information and comment the link to the other issue.

Determining Priority

The priority of an issue should be determined by taking into account the user impact and how hard it is to fix the issue.

User impact

We have two labels to rate user impact U-high and U-low.

Use U-high if

  • Many users have reported this issue
  • The problem is triggered often during typical use of qTox
  • The problem causes data loss
  • There is no workaround

Use U-low if

  • Few users reported this problem
  • The problem occurs very sporadically
  • The problem needs a very specific set of conditions to appear
  • There is a workaround
  • The problem appears only after multiple days of usage

Difficulty to fix

We have two labels to estimate the difficulty of a fix, D-easy and D-hard.

Use D-easy if you think that:

  • The issue is well described
  • The issue can be consistently reproduced
  • The issue needs no specific equipment to fix, e.g. specific OS, webcam,...
  • The code that causes the problem is known

Use D-hard if you think that:

  • The issue is described only vaguely
  • The exact way to reproduce the issue is not known
  • The issue happens only on a specifc OS
  • The issue is not only caused by code from qTox

Determining initial priority

After assesing the user impact and the difficulty to fix the issue you look up the initial priority for the issue in the following table:

U-high U-low
E-easy P-high P-medium
E-hard P-medium P-low

Possible security issues should be tagged with P-high initally. If they are confirmed security issues, the tag should be changed to P-very-high, else apply the normal rating process.

Translations from Weblate

Weblate provides an easy way for people to translate qTox. On one hand, it does require a bit more attention & regular checking whether there are new translations, on the other, it lessened problems that were happening with "manual" way of providing translations.

To get translations into qTox:

  1. Go to https://hosted.weblate.org/projects/tox/qtox/#repository and lock the repository for translations.
  2. Make sure you have git setup to automatically gpg sign commits.
  3. To update translated strings from Weblate, in the root of the qTox repository execute the script tools/update-weblate.sh
  4. If a new translation language has been added, update the following files:
    • translations/CMakeLists.txt
    • src/widget/form/settings/generalform.cpp
    • translations/README.md
    • translations/i18n.pri
    • translations/translations.qrc
  5. To update translatable strings from qTox for Weblate, run ./tools/update-translation-files.sh ALL
  6. Checkout a new branch with e.g. git checkout -b update_weblate and open a Pull Request for it on Github.
  7. After the Pull Request has been merged, reset Weblate to master and unlock it.

Releases

Tagging scheme

  • tag versions that are to be released, make sure that they are GPG-signed, i.e. git tag -s v1.8.0
  • use semantic versions for tags: vMAJOR.MINOR.PATCH
    • MAJOR bump version when there are breaking changes to video, audio, text chats, groupchats, file transfers, and any other basic functionality. For other things, MINOR and PATCH are to be bumped.
    • MINOR bump version when there are:
      • new features added
      • UI/feature breaking changes
      • other non-breaking changes
    • PATCH bump when there have been only fixes added. If changes include something more than just bugfixes, bump MAJOR or MINOR version accordingly.
  • bumping a higher-level version "resets" lower-version numbers, e.g. v1.7.1 → v2.0.0

Steps for release

Before tagging

  • Format all code using the ./tools/format-code.sh script
  • Update the Flatpak manifest of our Flathub repository with the script in flatpak/update_flathub_descriptor_dependencies.py
    • Make sure to check if new dependencies need to be added, add them if necessary
  • Update version number for windows/osx packages using the ./tools/update-versions.sh script, e.g. ./tools/update-versions.sh 1.11.0
  • Update toxcore version number to the latest tag using [`./tools/update-toxcore-version.sh]
  • Update the bootstrap nodelist at ./res/nodes.json from https://nodes.tox.chat/json. This can be done by running ./tools/update-nodes.sh
  • Generate changelog with clog.
    • In a MAJOR/MINOR release tag should include information that changelog is located in the CHANGELOG.md file, e.g. For details see CHANGELOG.md
  • To release a PATCH version after non-fix changes have landed on master branch, checkout latest MAJOR/MINOR version and git cherry-pick -x commits from master that you want PATCH release to include. Once cherry-picking has been done, tag HEAD of the branch.
    • When making a PATCH tag, include in tag message short summary of what the tag release fixes, and to whom it's interesting (often only some OSes/distributions would find given PATCH release interesting).

After tagging

  • Create and GPG-sign the tar.lz and tar.gz archives using [./tools/create-tarballs.sh] script, and upload both archives plus both signature files to the github release that was created by a Travis OSX release job.
  • Download the binaries that are part of the draft release, sign them in in detached and ascii armored mode, e.g. gpg -a -b <artifact>, and upload the signatures to the draft release.
  • Add a title and description to the draft release, then publish the release.
  • Update download links on https://tox.chat to point to the new release.
  • Write a short blog post for https://github.com/qTox/blog/ and advertise the post on Tox IRC channels, popular Tox groups, reddit, or whatever other platforms.
  • Merge ./flatpak/io.github.qtox.qTox.json into the manifest of our Flathub repository. Keep the Flathub repository's version of "sources" for qTox.
  • After the build passed for qTox on all architectures on the Flathub build bot, merge the PR into the master branch of our Flathub repository.

How to become a maintainer?

Contribute, review & test pull requests, be active, oh and don't forget to mention that you would want to become a maintainer :P

Aside from contents of CONTRIBUTING.md you should also know the contents of this file.

Once you're confident about your knowledge and you've been around the project helping for a while, ask to be added to the qTox organization on GitHub.