deprecation(semver): deprecate ltr() and gtr()

[timreichen]

• renames ltr() to lessThanRange() and deprecates the old function
• renames gtr() to greaterThanRange() and deprecates the old function
• deprecates ltr()
• deprecates gtr()

[iuioiua]

I agree with the changes. Nice.

[kt3k]

I wonder what are the use cases of these functions. I guess most users are only interested in testRange for comparing a version and a range. In what cases, would people try checking these conditions?

[timreichen]

I wonder what are the use cases of these functions. I guess most users are only interested in testRange for comparing a version and a range. In what cases, would people try checking these conditions?

I have no idea. I think it is pretty trivial to get the min/max version of a range and compare it to another version. But I guess there must be a reason why it was ported over.

[kt3k]

But I guess there must be a reason why it was ported over.

I feel we blindly ported everything from npm:semver and these wasn't a discussion about utilities of these APIs.

I'd suggest we should just deprecate these without replacement APIs because they don't seem common operation and also the alternative ways to perform these are trivial as you point out. The deprecation note would be something like use greaterThan(version, rangeMax(range)) instead

[timreichen]

But I guess there must be a reason why it was ported over.

I feel we blindly ported everything from npm:semver and these wasn't a discussion about utilities of these APIs.

I'd suggest we should just deprecate these without replacement APIs because they don't seem common operation and also the alternative ways to perform these are trivial as you point out. The deprecation note would be something like use greaterThan(version, rangeMax(range)) instead

done

[iuioiua]

Towards #4228

[iuioiua]

LGTM. I've made some modifications and opened a new issue (#4273) to track the deprecation. Does this new way of deprecation look good to us?

[kt3k]

LGTM. I've made some modifications and opened a new issue (#4273) to track the deprecation. Does this new way of deprecation look good to us?

Thanks! #4273 Looks good to me.

[kt3k]

LGTM