Disallowing periods in slugs (Response to issue#2383)

[alinaghale88]

I have updated the regular expression pattern in isSlug.js to disallow periods for strings of type slug. I have also ensured that the modification made does not impact all the correct conditions the existing pattern supports at present.

Checklist

• PR contains only changes related; no stray files, etc.
• README updated (where applicable)
• Tests written (where applicable)
• References provided in PR (where applicable)

[WikiRik]

Just fyi;

This is the basis of the validator: https://gist.github.com/hagemann/382adfc57adbd5af078dc93feef01fe1

In my opinion any changes to isSlug should still conform to that gist.

I'll check later if that's the case. In the meantime; don't forget to add additional tests.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 100.00%. Comparing base (4197b86) to head (5e19e96).
Report is 17 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff            @@
##            master     #2385   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files          107       107           
  Lines         2449      2449           
  Branches       619       619           
=========================================
  Hits          2449      2449           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[alinaghale88]

@WikiRik I have added a new invalid test case and made further changes to ensure that the regular expression complies with all the valid and invalid test cases. Please review and provide your feedback.

[Zhuochengyu]

Should 'a-a' be considered a slug? I ran this example with this regex and it is not highlighting.

[alinaghale88]

Yes, the string 'a-a' should be considered a slug @Zhuochengyu

Thank you for pointing out this exception. I have pushed another update for the regex and also included a new test case.

[alinaghale88]

Just fyi;

This is the basis of the validator: https://gist.github.com/hagemann/382adfc57adbd5af078dc93feef01fe1

In my opinion any changes to isSlug should still conform to that gist.

I'll check later if that's the case. In the meantime; don't forget to add additional tests.

Just fyi;

This is the basis of the validator: https://gist.github.com/hagemann/382adfc57adbd5af078dc93feef01fe1

In my opinion any changes to isSlug should still conform to that gist.

I'll check later if that's the case. In the meantime; don't forget to add additional tests.

@WikiRik The reference you provided seems to be changing any string that cannot be considered a slug to an actual definition of a slug. But the isSlug function's role is to validate whether or not a provided string is of type slug.

[jfaMan]

@alinaghale88 As mentioned here, the original regex effectively required at least 3 characters, which your revision doesn't seem to do when I tested it. Not sure if it's technically required, but just something I thought I'd mention. I use the isLength validator for my use case anyway to catch those. This change otherwise fixes my issue with special characters passing.

[jfaMan]

Just fyi;
This is the basis of the validator: https://gist.github.com/hagemann/382adfc57adbd5af078dc93feef01fe1
In my opinion any changes to isSlug should still conform to that gist.
I'll check later if that's the case. In the meantime; don't forget to add additional tests.

Just fyi;
This is the basis of the validator: https://gist.github.com/hagemann/382adfc57adbd5af078dc93feef01fe1
In my opinion any changes to isSlug should still conform to that gist.
I'll check later if that's the case. In the meantime; don't forget to add additional tests.

@WikiRik The reference you provided seems to be changing any string that cannot be considered a slug to an actual definition of a slug. But the isSlug function's role is to validate whether or not a provided string is of type slug.

I believe he means in the sense that any characters Slugify retains when it 'slugifies' a string are considered valid characters in a slug, and isSlug uses this as the basis of how it validates slugs.

[jfaMan]

Something else I noticed, but is there a reason this allows &, * and \ special characters? The original allows \ though I'm not sure why since Slugify ignores \, and @WikiRik mentioned that Slugify is the basis of the regex for isSlug. Slugify also ignores * and replaces & with 'and', so I would consider all of them to be invalid.