Skip to content

Latest commit

 

History

History
54 lines (38 loc) · 2.32 KB

CONTRIBUTING.md

File metadata and controls

54 lines (38 loc) · 2.32 KB

Contribution Guidelines

Rough guidelines:

  1. Please ensure that the build passes.
  2. Packages in other languages are welcome.

Bank Names Guidelines

The banknames.json file is committed to the repo and is used to maintain a list of human-readable bank names in India. If you are touching that file, here are a few guidelines:

  1. Do not include ltd, Ltd at the end.
  2. Ensure that the case is consistent with the official name.
  3. Do not prefix the bank name with "The".
  4. The canonical way to spell Coop is Co-operative.
  5. Do not include a city name in the bank name, unless it is part of the IFSC code.
  6. Include the city name in brackets at the end if it is necessary for disambiguation.
  7. Try not to leave any unexpanded abbreviations in the name
  8. Also ensure that the 4 character code is committed to Bank.php
  9. No period after Co-operative
  10. Grameen is spelled three ways: Grameen/Gramin/Grameena. Check the RBI List for the corresponding bank here.
  11. Sahakari, not sahkari.

Code Guidelines

We use prettier for the javascript and rubocop for the Ruby code for style fixes

Releases

Releases are partially automated. To draft a new release, follow these steps:

  1. Create a new release/{version} branch
  2. Copy the artifacts from the build pipeline and commit it (IFSC.json and sublet.json)
  3. Make sure that the tests are passing
  4. Download the release.md file from the release pipeline
  5. Bump the versions in the following places: package.json, ifsc.gemspec and commit it
  6. Merge the PR
  7. Tag the merge commit (don't use a prefix, just X.Y.Z)
  8. Push the tag to GitHub.
  9. Create a release on GitHub for the tag. Use release.md from Step 4 as the template. Replace TODO as applicable for the release.

Patches

Sometimes, when you know a certain information to be incorrect in the dataset (temporarily or permanently), you might want to override what the official dataset says. For such cases, the library maintains patches in YAML format in the patches/ directory. Each patch has 2 components:

  1. Diff that must be applied (patch)
  2. List of IFSC that it must be applied to (ifsc)

Using this, you can selectively correct data for various IFSC, including turning "NEFT"/"IMPS" properties on or off, or setting the right state name.