feat: allows seperate prefixTag version sequences

[jahead]

Solves #572

[coveralls]

Coverage increased (+0.0005%) to 99.429% when pulling c8016dd on jahead:master into 2f04ac8 on conventional-changelog:master.

[jvitor83]

@bcoe need that PR merged and a new release please.
Solves #702 too.

[jahead]

there's conflicts I'll fix them on the weekend and update

[jvitor83]

@jahead don't forget that please. :)

[codecov-io]

Codecov Report

Merging #573 (1c93d7f) into master (f5bff12) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #573   +/-   ##
=======================================
  Coverage   97.71%   97.71%           
=======================================
  Files          24       24           
  Lines        1051     1052    +1     
=======================================
+ Hits         1027     1028    +1     
  Misses         24       24           

Impacted Files	Coverage Δ
index.js	100.00% <100.00%> (ø)
lib/latest-semver-tag.js	100.00% <100.00%> (ø)
test/git.spec.js	99.07% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f5bff12...1c93d7f. Read the comment docs.

[jahead]

Done. Sorry for taking so long. Let me know if anything else is needed

[bcoe]

@jahead this is looking good to me, but any chance we could add a test that covers the additional functionality?

[jahead]

Yea mate, I"ll do it tonight.

I was thinking how to test this

[jahead]

@bcoe Hey so I added some tests here and tested tagPrefix in general, as there didn't seem to be any tests already that I could see.

I'm not overly happy with is this as git-semver-tags actually is the one filtering the tags, and due to the mock we won't know if git-semver-tags changes will break the logic here. But the tests cover the use case in #702

resolve if you're happy

[jahead]

Oddly, through testing this.
I discovered/re-discovered that

    if (pkg) {
      version = pkg.version
    } else if (args.gitTagFallback) {
      version = await latestSemverTag(args.tagPrefix)
    } else {
      throw new Error('no package file found')
    }

Which means that if there is a packageFiles defined we never get to use the gitSemverTags logic to resolve versions of different prefixes. And there's no option to force resolution by tag if packageFiles are defined.

It will solve #702 but people should take care to configure no packageFiles.

I'm not sure we should solve that problem in this PR, but it did seem like odd to me.
I've outlined it in the tests.

[bcoe]

@jahead released in v9.2.0, and appreciated. Let me know if there are any issues.

I do gradually try to make my way through @ messages on GitHub, so feel free to keep engaging in this way.