Repo should be initialized either through git init
template or through script.
For the git init
approach, check git-scm.com/docs/git-init#git-init---template.
- use
.gitignore
for the languages found in that repo, ignoring:- compiled files (eg:
.bin
,.pyc
,.class
etc.); - data files (with the exception of small
sqlite
files); - big binaries (eg:
.exe
); - libraries (eg:
node_modules
,bower_components
etc.).
- compiled files (eg:
- use a common
.gitignore
like the one found in this repo (language and project-agnostic): - if there are important root folders, use one level of nesting for
.gitignore
(a main.gitignore
for the main problems, and shallow nested.gitignores
for specific files or folders) - this allows for cleaner root gitignore, and easier tracking of files that should be removed from.gitignore
TODO
: some Git workflow described here ?
- commit wording (reflect this in
.gitignore
):- fix - something is broken
- patch - something is added
- ...
- branch naming
- policy for features - eg: 'feature/dice-roller'
Proposal
: refactor passes:- open issue with targeted feature - eg: javadoc
- mark with emoji in commit and add reference to issue (without closing)
- no emoji == no refactor passes
- evaluating quality of code by the number of emojis (eg: 2 emojis - slightly better than 1 ...)