Generate the adapter capability matrix instead of maintaining three by hand
Les mainteneurs répondent en général sous 1 jour
Personne n'a encore pris cette issue.
Évaluation
- Difficulté
- 5/5
- Temps estimé
- Plus d'une semaine
- Accessibilité débutants
- 42/100
- Type d'issue
- Fonctionnalité
- Clarté
- Plutôt claire
- Activité
- Active
- Stack technique
- typescript
- Domaine
- build-system, documentation, tooling
Piste de recherche
Read the Capability table in README.md, CrossFrameworkSupport.md, MobileTesting.md, and CLAUDE.md’s Known debt to inventory the current rows, adapters, and reasons before locating the repository’s generation or build entry point. Done means one declared source generates all three tables, requires reasons for exemptions, rejects undeclared adapters, and visibly distinguishes proven from declared cells.
Rédigé par le modèle d'indexation à partir du texte de l'issue.
Description
Part of #278
Three capability matrices are maintained by hand today: the Capability table in
this repo's README.md, CrossFrameworkSupport.md on the docs site, and the
adapter table on MobileTesting.md. The first two carry three adapter columns.
The mobile one carries four.
The Python adapter published to PyPI on 2026-09-29 and has yet to appear in the
other two, which is the whole problem in one observation: a markdown table has no
way to notice that a fourth adapter now exists. It is the epic's complaint about
capture logic, in prose rather than code.
The tables are otherwise good. They already do the thing worth keeping, which is
naming the capability, marking it per adapter, and footnoting why an adapter
falls short rather than leaving a bare gap. That part should survive.
What to change
Make the matrix an output rather than a document. One declared source listing
every capability and, per adapter, either support or an exemption carrying a
required reason. Render it to the README and to the docs pages.
That buys three things a table cannot:
- Adding an adapter turns every unlisted capability into an error, rather than
into a column nobody updated. - A reason becomes mandatory. Today a warning marker with no footnote is valid
markdown; in a declared source, an exemption without a reason does not build.
Most of the reasons already exist inCLAUDE.md's Known debt in more detail
than the current footnotes carry, so this is largely relocation. - The README and both docs tables render from one source and cannot disagree.
Relationship to #300
Complementary, and deliberately separable. Where a capability is testable over
the emitted frame stream, its cell can be proven by the conformance suite
rather than declared. Where it is not testable from a stream (inline Allure
attach, Cucumber nesting, live-dashboard overlay), the cell stays a declared
claim.
The two states must be visibly distinct in the rendered table. Most cells will
start declared, and a matrix that reads as conformance-backed when only a few
rows are is worse than an honest hand-maintained one.
This issue is useful whether or not #300 is built, which is why it is not folded
into it.
Acceptance criteria
One declared source of capabilities and per-adapter support. An exemption
requires a reason. The README table and the docs tables are generated from it,
with no adapter column missing. Adding an adapter without declaring its support
fails the build. Proven and declared cells are distinguishable in the output.
Size: M.
- Langage dominant
- TypeScript
- Étoiles
- 10
- Forks
- 2
- Merge moyen
- 1 j 2 h
- PR mergées (30 j)
- 25
Préparer son environnement
- Aucun Dockerfile ni fichier Docker Compose
- Propose un modèle de pull request
- Lire le guide de contribution
Par où commencer
- Lisez l'issue en entier, puis le guide de contribution du projet.
- Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
- Forkez le dépôt et travaillez sur une branche.
- Ouvrez une pull request qui référence le numéro de l'issue.
Autres issues de webdriverio/devtools
-
Cross-adapter conformance suite for capture invariantsPeut-être à nouveau libre @vishnuv688 l’a pris il y a 57 jours, et aucune pull request n’est ouverte. Ouvertecapture-architecture enhancement
webdriverio/devtools#300 · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
-
Spike: how to avoid re-implementing capture per languagePeut-être à nouveau libre @vishnuv688 l’a pris il y a 57 jours, et aucune pull request n’est ouverte. Ouvertecapture-architecture
webdriverio/devtools#298 · 3 commentaires · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
-
Epic: Cross-language capture architecturePeut-être à nouveau libre @vishnuv688 l’a pris il y a 57 jours, et aucune pull request n’est ouverte. Ouvertecapture-architecture
webdriverio/devtools#278 · 1 commentaire · 1 personne assignée ·
Les mainteneurs répondent en général sous 1 jour
Toutes les issues de webdriverio/devtools
Issues similaires
-
Difficulté 1/5 1-3 heures Accessibilité débutants 84/100
Les mainteneurs répondent en général sous 1 jour
-
core
Difficulté 2/5 1-3 heures Accessibilité débutants 70/100
vectorize-io/hindsight#5457 ·
Les mainteneurs répondent en général sous 1 jour
-
beginner friendly community contributions-welcome good first issue hacktoberfest help wanted testing up-for-grabs
Difficulté 2/5 1-3 heures Accessibilité débutants 72/100
Les mainteneurs répondent en général sous 1 jour
-
Difficulté 1/5 Moins d'une heure Accessibilité débutants 85/100
lukilabs/beautiful-mermaid#160 ·
-
Difficulté 2/5 1-3 heures Accessibilité débutants 66/100
rescript-lang/rescript-lang.org#1420 ·
Les mainteneurs répondent en général sous 2 jours