eslint-plugin-import
eslint-plugin-import copied to clipboard
import-js
[import/no-restricted-paths]: Correctly resolve symlinks in file path
I have been using this rule for a while in a monorepo and turns out it never worked because it wasn't resolving the symlinks inside node_modules correctly. I am not sure if this is the right location and if I should use the sync or async version of this function but this does solve my problem. Would like to hear what you think of this solution.
Coverage decreased (-0.5%) to 93.651% when pulling 2506f1a64f7ada9563ea9b2c8e6476ca08e27f68 on bensampaio:patch-1 into bdc05aa1d029b70125ae415e5ca5dca22250858b on benmosher:master.
This seems great, but now it needs a test :-)
I'm trying to figure out where to place the test but I don't really get how to test this based on what's already there. I guess I should place my test in tests/src/core/resolve.js. However, in this file I can only find tests that use different types of resolvers (whatever those are 🤔).
Do I need to add a test to every it or create a new it? And what should that test do exactly?
How can I test symlinks and make sure they are resolved correctly? I guess, I could add symlinks to the tests/files folder but what then?
Sorry, but I am not that familiar with all concepts in this repo, and can't really figure out what these tests are doing 😩
Hmm - I believe you can git commit an actual symlink, and use that to ensure it's resolved properly?
@ljharb I added a test as you requested. There are lots of tests failing, is this normal?
Thanks! it's not normal - although there is a few failures on master, they're not the same ones here.
Ok, I thought the tests failing in master were the same as in my branch but I noticed now that I was wrong. I was really looking forward to having this fixed but this is taking me too much time and I am having quite some difficulties understanding the code base. I don't get why these other tests stopped working.
If you still think my initial solution is not ideal then I would say someone has to continue this or you can close it since I cannot invest more time into trying to fix this. Maybe an issue should be open instead in that case?