Project Standards¶
Development Environment¶
If you haven’t already, please visit “(b) Development Environment” to set up your local development environment.
Version Control (VC)¶
The repository uses a fork-based Git workflow with tag releases.
Guidelines for VC¶
main
must always be deployableAll changes are made through support branches on forks
Rebase with
main
to avoid/resolve conflictsMake sure
pre-commit
checks pass when committing (enforced in CI/CD build)Open a pull-request (PR) early for discussion
Once the CI/CD build passes and PR is approved, squash and rebase your commits
Merge PR into
main
and delete the branch
Things to Avoid in VC¶
Don’t merge in broken or commented out code
Don’t commit onto
main
directlyDon’t merge with conflicts (handle conflicts upon rebasing)
Pre-commit¶
The repository uses the pre-commit
package to manage pre-commit hooks.
These hooks help enforce quality assurance standards and identify simple issues
at the commit level before submitting code reviews.
Helpful pre-commit
Commands¶
Install into your cloned repo
conda activate zppy_dev
pre-commit install
Automatically run all pre-commit hooks (just commit)
# Tip: If there is an issue with pre-commit, you can bypass with the `--no-verify` flag. Please do NOT use this on a regular basis.
git commit -m '...'
Manually run all pre-commit hooks
pre-commit run --all-files
Run individual hook
# Available hook ids: trailing-whitespace, end-of-file-fixer, check-yaml, black, isort, flake8, mypy
pre-commit run <hook_id>
Squash and Rebase Commits¶
Before you merge a support branch back into main
, the branch is typically
squashed down to a single* buildable commit, and then rebased on top of the main repo’s main
branch.
* In some cases, it might be logical to have multiple squashed commits, as long as each commit passes the CI/CD build
Why squash and rebase commits?
Ensures build passes from the commit
Cleans up Git history for easy navigation
Makes collaboration and review process more efficient
Makes handling conflicts from rebasing simple since there are fewer commits to fix conflicts on
Makes
git bisect
easier and more effective to use. For example, it will show the exact commit that introduced a bug since the commit contains a relatively small changeset. On the otherhand, a merge commit bringing in many commits makes it much harder since it includes a large changeset. A changeset is a collection of commits that are logically grouped together.
How to squash and rebase commits¶
Assuming that you followed “(b) Development Environment”:
Sync
main
with the main repo’smain
git checkout main git rebase <upstream-origin>/main git push -f <fork-origin> main
Get the SHA of the commit OR number of commits to rebase to
git log --graph --decorate --pretty=oneline --abbrev-commit
Squash commits:
git rebase -i [SHA] # OR git rebase -i HEAD~[NUMBER OF COMMITS]
Make sure your squashed commit messages are refined
Rebase branch onto
main
git checkout <branch-name> git rebase main git push -f <fork-origin> <branch-name>
Force push to remote branch
# You have to force push because the rebase rewrites the commit SHAs git push -f <fork-origin> <branch-name>
Source: https://blog.carbonfive.com/always-squash-and-rebase-your-git-commits/
Code Quality Assurance¶
This project uses several tools for code formatting, linting, and type checking listed below.
You can run them as hooks manually/automatically when committing using pre-commit
, or manually through the terminal or IDE/text editor.
Helpful Commands¶
- Run a tool
# Available tool names: black, flake8, isort, mypy <tool_name> .
Continuous Integration / Continuous Delivery (CI/CD)¶
This project uses GitHub Actions to run two CI/CD workflows.
CI/CD Build Workflow
This workflow is triggered by Git
pull_request
andpush
(merging PRs) events to the the main repo’smain
.Jobs:
Run
pre-commit
for formatting, linting, and type checkingRun test suite in a conda environment
Publish latest
main
docs (only onpush
)
CI/CD Release Workflow
This workflow is triggered by the Git
publish
event, which occurs when a new release is tagged.Jobs:
Publish new release docs
Publish Anaconda package
API Documentation¶
In most cases, code should be self-documenting.
If necessary, documentation should explain why something is done, its purpose, and its goal. The code shows how it is done, so commenting on this can be redundant.
Guidelines For Documenting¶
Embrace documentation as an integral part of the overall development process
Treat documentation as code and follow principles such as Don’t Repeat Yourself and Easier to Change
Use comments and docstrings to explain ambiguity, complexity, or to avoid confusion
Co-locate API documentation with related code
Use Python type annotations and type comments where helpful
Things to Avoid with Documenting¶
Don’t write comments as a crutch for poor code
Don’t comment every function, data structure, type declaration