Pollution from .mlc_config.json

[rngadam]

1. .mlc_config is getting added to the root of every repo - a file that is used to configure the github
   https://github.com/search?q=org%3Aai-cfia%20.mlc_config&type=code

github-workflows/.github/workflows/workflow-markdown-check.yml

Line 24 in 8521ad2

config-file: '${{ inputs.config-file-path }}'

2. this was created to fix links to LinkedIn where it is returning a non-standard 999 HTTP code for links accessed without a proper account

https://excellentwebcheck.com/blogs/what-is-http-999-request-denied-response-code

3. this really should be a parameter to the github-workflows with default only expecting 200 (we generally shouldn't be pointing to pages that are 403

• modify workflow-markdown-check.yml to take list of acceptable status code and reasonable default
• remove .mlc_config from repo where it was inserted
• fix the documentation to not require that file anymore

[ibrahim-kabir]

@SonOfLope I will remove my assignment to this issue, currently not available to work on a devops issue

[SonOfLope]

I update the workflow due to a different issue where we notice the pipeline wasnt working properly. The updated workflow (once merged) will allow to provide the config file optionally. A default config file is provided in the github-workflow repo so we can now remove .mlc_config.json if needed. These changes resolve all the checks except

this really should be a parameter to the github-workflows with default only expecting 200 (we generally shouldn't be pointing to pages that are 403

I don't know about this. The workflow gives false positive sometimes about 403. Here is an example :
https://github.com/ai-cfia/github-workflows/actions/runs/8573751804/job/23499098138?pr=114
Doing a curl or manually visiting the website gives a 200 response but the workflow resolves into a 403.

[rngadam]

Doing a curl or manually visiting the website gives a 200 response but the workflow resolves into a 403.

hmm, interesting. weird that it only denies checks from markdown. Then I guess we need to keep accepting 403 for now.