Implement beta publishing on development branch by VikramGoyal23 · Pull Request #85 · git-mastery/app

VikramGoyal23 · 2026-03-23T19:13:36Z

Should be merged after the actions PR and after gitmastery-beta is registered on AUR and Winget.

RFC: https://docs.google.com/document/d/17ZhVua_eH_mWQIDc90ojq6EkiolRXQa5asBBvZhgfAU/edit?tab=t.0

jovnc · 2026-04-11T09:56:24Z

@VikramGoyal23 Can we check on status of this? Hopefully we can get this merged soon before the semester ends.

While the version parsing could be trivially solved using regex matching, I opted to build on the current logic for now. Should maybe come back to it in the future.

Copilot

Pull request overview

Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-15T06:30:59Z

+        if ("beta" in version and len(parts) != 4) or ("beta" not in version and len(parts) != 3):
            raise ValueError(
-                f"Invalid version string (expected 'MAJOR.MINOR.PATCH'): {version!r}"
+                f"Invalid version string (expected 'MAJOR.MINOR.PATCH[-beta.PRERELEASE]'): {version!r}"
            )
        try:
-            major, minor, patch = (int(part) for part in parts)
+            if "beta" in version:
+                version_part, prerelease_str = version.split("-beta.")
+                prerelease = int(prerelease_str)
+                major, minor, patch = (int(part) for part in version_part.split("."))


parse uses "beta" in version to decide beta parsing/validation. This can incorrectly route versions that merely contain the substring "beta" and it doesn't assert the presence of the -beta. separator before calling version.split("-beta."). Use a stricter check like "-beta." in version (and validate the split yields exactly 2 parts) to avoid false positives and confusing exceptions.

Copilot · 2026-04-15T06:30:59Z

        """Returns if the current version is behind the other version based on major and minor versions."""
+        if self.prerelease is not None and other.prerelease is not None:
+            return (other.major, other.minor, other.patch, other.prerelease) > (
+                self.major,
+                self.minor,
+                self.patch,
+                self.prerelease,
+            )
        return (other.major, other.minor) > (self.major, self.minor)



is_behind now compares (major, minor, patch, prerelease) only when both versions are prereleases, but falls back to comparing only (major, minor) for stable versions and for mixed stable/prerelease comparisons. This breaks SemVer ordering in common cases (e.g., 1.2.3-beta.1 should be behind 1.2.3, and 1.2.3 should be behind 1.2.4), and the docstring no longer matches behavior. Consider comparing (major, minor, patch) first in all cases, then applying prerelease ordering when base versions are equal (stable > prerelease; if both prerelease, compare prerelease number).

Suggested change

"""Returns if the current version is behind the other version based on major and minor versions."""

if self.prerelease is not None and other.prerelease is not None:

return (other.major, other.minor, other.patch, other.prerelease) > (

self.major,

self.minor,

self.patch,

self.prerelease,

)

return (other.major, other.minor) > (self.major, self.minor)

"""Return whether the current version is behind the other version using SemVer ordering."""

self_base = (self.major, self.minor, self.patch)

other_base = (other.major, other.minor, other.patch)

if self_base != other_base:

return self_base < other_base

if self.prerelease is None and other.prerelease is None:

return False

if self.prerelease is None:

return False

if other.prerelease is None:

return True

return self.prerelease < other.prerelease

Copilot · 2026-04-15T06:30:59Z

+  winget-publish:
+    needs: [prepare, windows]
+    if: needs.prepare.outputs.should_publish == 'true' && !contains(needs.prepare.outputs.ref_name, 'beta') # temporarily disable winget publishing 
+
+    runs-on: windows-latest
+
+    permissions:
+      contents: read
+
+    steps:
+      - name: Submit to WinGet
+        uses: vedantmgoyal9/winget-releaser@v2
+        with:
+          identifier: GitMastery.GitMasteryBeta
+          version: ${{ needs.prepare.outputs.version_number }}
+          release-tag: ${{ needs.prepare.outputs.ref_name }}
+          token: ${{ secrets.ORG_PAT }}
+          installers-regex: '\.exe$'


The winget-publish job is effectively unreachable in this workflow: the workflow triggers on v*.*.*-beta.* tags and is named/purposed for beta publishing, but the job additionally requires !contains(..., 'beta'), so it will never run for the intended beta tags. If WinGet publishing is intentionally disabled, consider removing the job or gating it behind an explicit workflow input so the condition is not contradictory/misleading.

Copilot · 2026-04-15T06:31:00Z

+on:
+  workflow_dispatch:
+    secrets:
+      ORG_PAT:
+        required: true
+


workflow_dispatch does not support a secrets: schema (only inputs:). As written, this won't actually enforce that ORG_PAT is provided, and the workflow may run and fail later with a confusing auth error. Consider removing this block and instead adding an early step that validates ${{ secrets.ORG_PAT }} is set (or use a protected environment with required reviewers/required secrets).

Copilot · 2026-04-15T06:31:00Z

        only_version = version[1:]
+        if "beta" in only_version:
+            version_part, prerelease = only_version.split("-beta.")
+            [major, minor, patch] = version_part.split(".")
+            return Version(int(major), int(minor), int(patch), int(prerelease))


parse_version_string checks "beta" in only_version and then blindly splits on -beta.. This will mis-detect unrelated substrings (e.g., "alphabet") and will raise an unhelpful ValueError if the string contains "beta" but not the expected -beta. delimiter. Prefer checking for the exact -beta. marker and raising a clear ValueError (or falling back to stable parsing) when the beta format is malformed.

jovnc · 2026-04-16T11:47:44Z

@VikramGoyal23 feel free to merge and test once ur ready, let us know when it is ready so we can approve it

VikramGoyal23 · 2026-04-16T14:04:19Z

@jovnc It's ready, can be approved

jovnc · 2026-04-16T14:18:39Z

@VikramGoyal23 I have disabled require approval as of now, so feel free to merge the PRs and debug as necessary

To unblock

VikramGoyal23 temporarily deployed to e2e-test March 23, 2026 19:13 — with GitHub Actions Inactive

VikramGoyal23 requested a review from jiaxinnns March 23, 2026 19:13

VikramGoyal23 marked this pull request as draft March 23, 2026 19:17

VikramGoyal23 force-pushed the beta-publishing branch from 1707c2d to a7c9451 Compare March 30, 2026 05:50

VikramGoyal23 requested a deployment to e2e-test March 30, 2026 05:50 — with GitHub Actions Waiting

VikramGoyal23 force-pushed the beta-publishing branch from a7c9451 to 275576b Compare April 1, 2026 02:35

VikramGoyal23 requested a deployment to e2e-test April 1, 2026 02:36 — with GitHub Actions Waiting

VikramGoyal23 requested a deployment to e2e-test April 1, 2026 06:27 — with GitHub Actions Waiting

VikramGoyal23 force-pushed the beta-publishing branch from 8c7e033 to bb5aba5 Compare April 4, 2026 11:09

VikramGoyal23 requested a deployment to e2e-test April 4, 2026 11:09 — with GitHub Actions Waiting

VikramGoyal23 requested a deployment to e2e-test April 8, 2026 05:34 — with GitHub Actions Waiting

VikramGoyal23 requested a deployment to e2e-test April 10, 2026 07:54 — with GitHub Actions Waiting

VikramGoyal23 force-pushed the beta-publishing branch from 70ffcf3 to 03d3902 Compare April 13, 2026 05:42

VikramGoyal23 requested a deployment to e2e-test April 13, 2026 05:42 — with GitHub Actions Waiting

VikramGoyal23 requested a deployment to e2e-test April 15, 2026 04:03 — with GitHub Actions Waiting

Revert unnecessary changes and fix version logic

53fc315

While the version parsing could be trivially solved using regex matching, I opted to build on the current logic for now. Should maybe come back to it in the future.

VikramGoyal23 requested a deployment to e2e-test April 15, 2026 05:41 — with GitHub Actions Waiting

Fix mypy error

01c9519

VikramGoyal23 requested a deployment to e2e-test April 15, 2026 05:43 — with GitHub Actions Waiting

Fix AUR installation directory

e75ec6f

VikramGoyal23 requested a deployment to e2e-test April 15, 2026 05:45 — with GitHub Actions Waiting

VikramGoyal23 added 2 commits April 15, 2026 14:10

Fix changelog name

f5d8dcc

Shift bump-version-beta logic to app repo

b5ae1bc

VikramGoyal23 requested a deployment to e2e-test April 15, 2026 06:13 — with GitHub Actions Waiting

jovnc requested a review from Copilot April 15, 2026 06:25

Copilot started reviewing on behalf of jovnc April 15, 2026 06:25 View session

Copilot AI reviewed Apr 15, 2026

View reviewed changes

Remove incorrect secrets syntax

2fc8a46

VikramGoyal23 deployed to e2e-test April 15, 2026 06:36 — with GitHub Actions Active

VikramGoyal23 temporarily deployed to e2e-test April 15, 2026 06:36 — with GitHub Actions Inactive

jovnc approved these changes Apr 16, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implement beta publishing on development branch#85

Implement beta publishing on development branch#85
VikramGoyal23 wants to merge 11 commits intogit-mastery:mainfrom
VikramGoyal23:beta-publishing

VikramGoyal23 commented Mar 23, 2026

Uh oh!

jovnc commented Apr 11, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

Copilot AI Apr 15, 2026

Uh oh!

jovnc commented Apr 16, 2026

Uh oh!

VikramGoyal23 commented Apr 16, 2026

Uh oh!

jovnc commented Apr 16, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

-        """Returns if the current version is behind the other version based on major and minor versions."""
-        if self.prerelease is not None and other.prerelease is not None:
-            return (other.major, other.minor, other.patch, other.prerelease) > (
-                self.major,
-                self.minor,
-                self.patch,
-                self.prerelease,
-            )
-        return (other.major, other.minor) > (self.major, self.minor)
+        """Return whether the current version is behind the other version using SemVer ordering."""
+        self_base = (self.major, self.minor, self.patch)
+        other_base = (other.major, other.minor, other.patch)
+        if self_base != other_base:
+            return self_base < other_base
+        if self.prerelease is None and other.prerelease is None:
+            return False
+        if self.prerelease is None:
+            return False
+        if other.prerelease is None:
+            return True
+        return self.prerelease < other.prerelease

Conversation

VikramGoyal23 commented Mar 23, 2026

Uh oh!

jovnc commented Apr 11, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 15, 2026

Choose a reason for hiding this comment

Uh oh!

jovnc commented Apr 16, 2026

Uh oh!

VikramGoyal23 commented Apr 16, 2026

Uh oh!

jovnc commented Apr 16, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants