• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

99designs / gqlgen / 37056503075
0%

Build:
DEFAULT BRANCH: master
Ran 02 Oct 2026 07:51PM UTC
Jobs 1
Files 0
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

02 Oct 2026 07:48PM UTC coverage: 0.0%. Remained the same
37056503075

push

github

web-flow
refactor: move concurrent-field boilerplate into the graphql runtime (#4341)

* refactor: move concurrent-field boilerplate into the graphql runtime

object.gotpl inlined the panic guard, the Invalids bookkeeping and the
whole @defer accounting into every concurrent field of every object: 22
lines and three nested closures per root field, emitted once per field
per object in every generated executor.

Replace it with two runtime helpers, FieldSet.ResolveConcurrently and
FieldSet.ResolveRootConcurrently, plus NewDeferredGroup for the group the
template used to assemble from two locals at the end of the function.
Generated executors shrink by 13,710 lines here (-10% on the largest),
and the logic is now covered by Go tests instead of by diffing generated
output.

Also skip ProcessDeferredGroup, and the atomic add of zero that preceded
it, when the operation has no @defer.

No behaviour change: sentinel choice per nullability, deferred-field
routing, per-label view reuse and RootResolverMiddleware wrapping are all
preserved. The non-atomic fs.Invalids++ fallback was unreachable, since
Object.IsConcurrent() is true whenever any of its fields is, and is
dropped.

* graphql: correct what the deferred branch says about m

The comment claimed a deferred field is "kept out of the initial response".
It is not. m.Values[i] is set to Null, and marshalFieldSet writes every
entry in m.fields, so the field ships in the initial response as null and
the incremental payload replaces it later. A field set holding one plain
and one deferred field marshals {"kept":"v","later":null}.

Say that instead, and record why the slot gets Null rather than being left
unset: marshalFieldSet dereferences every entry, so nil would panic.

Behaviour is unchanged. Only the description was wrong.

* graphql: flatten the concurrent-field tests and pin the panic path

Seven declarations stood between a reader and one case: a shared table, a
scheduleFunc type, two scheduling adapters and a ... (continued)

0 of 0 relevant lines covered (NaN%)

0.0 hits per line

Jobs
ID Job ID Ran Files Coverage
1 37056503075.1 02 Oct 2026 07:51PM UTC 0
0.0
GitHub Action Run
Source Files on build 37056503075
Detailed source file information is not available for this build.
  • Back to Repo
  • f3223264 on github
  • Prev Build on master (#37031144063)
  • Next Build on master (#37082526963)
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE TRIAL · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc