Make the packages publishable, and verify them against a real consumer - #74
Merged
Conversation
Publishing was blocked by a package that could not be built and metadata that had
never been filled in. Both are fixed; the publishing steps themselves are
documented in RELEASING.md rather than run, since they take credentials and cannot
be undone.
Settex.Build did not pack at all. It declared the task assemblies into
lib/netstandard2.0 with a plain ItemGroup, whose $(OutputPath) evaluates too early
— they actually landed under contentFiles/ — and NU5100 then failed the build,
because TreatWarningsAsErrors is on. The assemblies now go to tasks/netstandard2.0
through a pack-time target, which is where a task assembly belongs: anything in
lib/ becomes a compile-time reference for every consuming project, which is what
NU5100 was warning about all along. NU5100 and NU5128 are suppressed with that
reasoning written down, and the targets file loads the task from its new home.
Metadata was still the template's: PackageProjectUrl and RepositoryUrl pointed at
github.com/yourusername/settex. Settex.Cli pinned its own Version of 1.0.0 while
everything else derived 0.1.0-alpha, so the three packages would have shipped
looking unrelated; there is one version for all of them now. Every package carries
the README, so its NuGet page is not blank. Settex.LanguageServer is marked not
packable — it ships inside the editor extensions.
Packing cleanly proves nothing about whether a package works, so all three were
tested against a throwaway consumer on a local feed. A fresh class library with a
PackageReference to Settex.Build compiles appsettings.settex into appsettings.json
plus a Production delta of {"App":{"Name":"Prod"}}. The tool installs and both its
commands run, import included, reporting the round trip verified exact. The VS
Code extension packages to a 2.5 MB VSIX with the language server bundled.
RELEASING.md records the verified commands, including the packageSourceMapping
section a local-feed test needs on a machine where mapping is enabled — without it
the local source is silently ignored and the test appears to pass while testing
nothing.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
La publication était bloquée par deux vrais défauts. Ils sont corrigés ; les étapes de publication elles-mêmes sont documentées dans
RELEASING.mdplutôt qu'exécutées — elles demandent des identifiants et sont irréversibles.Settex.Buildne se construisait pas du toutIl déclarait les assemblys de tâche dans
lib/netstandard2.0via unItemGroupsimple, dont le$(OutputPath)s'évalue trop tôt — ils atterrissaient en réalité souscontentFiles/. NU5100 faisait alors échouer le build,TreatWarningsAsErrorsétant actif.Les assemblys vont maintenant dans
tasks/netstandard2.0via une cible exécutée au moment du pack. C'est leur place : tout ce qui est danslib/devient une référence de compilation pour chaque projet consommateur — ce que NU5100 signalait depuis le début.Métadonnées de gabarit
PackageProjectUrletRepositoryUrlpointaient encore versgithub.qkg1.top/yourusername/settex. EtSettex.Cliépinglait sa propreVersion1.0.0 pendant que le reste dérivait 0.1.0-alpha : les trois paquets seraient partis comme s'ils n'avaient aucun rapport. Une seule version pour tous, README embarqué dans chacun, etSettex.LanguageServermarqué non-packable (il vit dans les extensions).Packer proprement ne prouve rien
Les trois paquets ont donc été testés contre un consommateur jetable sur un flux local :
PackageReference Settex.Buildappsettings.json+ delta{"App":{"Name":"Prod"}}dotnet tool install Settex.Clibuildetimportfonctionnentimportsur la sortievsce packageRELEASING.mdconsigne les commandes vérifiées — dont la sectionpackageSourceMappingindispensable au test sur flux local : sans elle la source locale est silencieusement ignorée et le test semble passer sans rien tester.🤖 Generated with Claude Code