RFC: reduce the time it takes to clone the repository

[aminya]

Checklist

Please ensure the following tasks are completed before submitting a feature request.

• Read and understood the Code of Conduct.
• Searched for existing issues and pull requests.
• The issue name begins with RFC:.

Description

Description of the feature request.

I noticed that this repository is very huge, and it takes a long time to clone, which makes it hard to contribute.

There are multiple ways to reduce the size:

• Generate the duplicated files instead of copying them
• Moving some of the files to other links or repositories
• Using git lfs
• Removing the files that are not used anymore from the history.

Example script for cleaning history

PowerShell

# clone:
# git clone https://github.com/stdlib-js/stdlib.git
# cd "stdlib"

$paths = "some/folder",
         "some/file"

foreach ($path in $paths) {
   # using https://github.com/newren/git-filter-repo
   git filter-repo --invert-paths --path $path
   if (test-path $path) {
      rm $path -Recurse -Force
   }
   git reflog expire --expire=now --all && git gc --prune=now --aggressive
}

git remote add origin git@github.com:stdlib-js/stdlib.git
git push --set-upstream origin master -f
git push --force --all origin 

Related Issues

Does this feature request have any related issues?

Questions

Any questions for reviewers?

No.

Other

Any other information relevant to this feature request? This may include screenshots, references, sample output, and/or implementation notes.

No.

[kgryte]

@aminya Thanks for filing this issue. We are certainly interested in exploring approaches for making this project easier to contribute to.

Re: clone times. One thing we suggest in our development guide, and which is not included in your suggestions above, is to limit the clone depth. We do this on our CI environments and seem to have decent success.

[kgryte]

Another possibility for reducing clone time is performing a shallow clone + sparse checkout. This is discussed on SO.