ReleaseNotes.load does not parse SemVer correctly

[Freymaurer]

Description

I used ReleaseNotes.load to read in my markdown release notes and tried to add the github commit to the SemVer metadata. As "#" is not a allowed symbol in metadata the version looks like this 0.1.1+09da94a.

let release = ReleaseNotes.load "RELEASE_NOTES.md"

release.SemVer.BuildMetaData returns in this case 09
release.SemVer.Original returns in this case 0.1.1+09

if i change the SemVer to 0.1.1+test it returns nothing for meta data.

Repro steps

1. Create RELEASE_NOTES.md with

### 0.1.1+09da94a (Released 2020-11-18)
* TestDes.:
    * TestSubDesc

2. Parse the markdown file with ReleaseNotes.load "RELEASE_NOTES.md"

3. Check SemVer.BuildMetaData/SemVer.Original of the ReleaseNotes record type.

Expected behavior

As far as i understand the specification it should allow both numbers and letters for
metadata.

Identifiers MUST comprise only ASCII alphanumerics and hyphens [0-9A-Za-z-].

Examples: 1.0.0-alpha+001, 1.0.0+20130313144700, 1.0.0-beta+exp.sha.5114f85, 1.0.0+21AF26D3—-117B344092BD.

Actual behavior

It stops parsing the metadata at the first letter.

[github-actions]

Welcome to the FAKE community! Thank you so much for creating your first issue and therefore improving the project!

[github-actions]

There has not been any activity in this issue for the last 3 months so it will be closed in 14 days if there is no activity.

[Freymaurer]

Okay, so i looked a bit into this:

The problem seems to be that SemVer is parsed from the nuget version

static member New(assemblyVersion,nugetVersion,date,notes) = { 
  AssemblyVersion = assemblyVersion
  NugetVersion = nugetVersion
  SemVer = SemVer.parse nugetVersion
  Date = date
  Notes = notes }

and nuget seems to get parsed with this Regex pattern: ([0-9]+.)+[0-9]+(-[a-zA-Z]+\d*)?(.[0-9]+)?, which does actually not hit any metadata and is faulty as it will parse wrong patterns.

My suggestion would be to use the offcially recommended Regex to parse SemVer. Either by parsing the full header for SemVer again or by using it to also parse the nuget version.

In the following i removed starting and ending requirement to confidently pick the complete SemVer from the start line.

(?<major>0|[1-9]\d*)\.(?<minor>0|[1-9]\d*)\.(?<patch>0|[1-9]\d*)(?:-(?<prerelease>(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?(?:\+(?<buildmetadata>[0-9a-zA-Z-]+(?:\.[0-9a-zA-Z-]+)*))?

You can check this regex against some examples:

### 0.1.1+7c567fd (Released 2020-11-18)

### 0.2.0+899b535 (Released 2021-1-11)

### 0.3.0+0d31c43 (Released 2021-2-11)

### 0.3.1+cbc655c (Released 2021-2-12)

1.0.0-alpha+001
1.0.0+20130313144700, 
1.0.0-beta+exp.sha.5114f85
1.0.0+21AF26D3—-117B344092BD

[Freymaurer]

Would it help if i did a PR for this? There are some open PRs and i don't really know if this would be useful.

[aggieben]

I think this may still be an issue and should be reopened.

[Freymaurer]

Yes sadly it is 😢 I am also having an eye on this issue