Add constraints to IoAwaitable - #383
Merged
Merged
Conversation
This add constraints to IoAwaitabe, so that the concept reflects what is needed for a type to be used in Capy-coroutines: * `await_ready`, * `await_resume`, * move construction.
Contributor
Author
|
This addresses #356. |
|
An automated preview of the documentation is available at https://383.capy.prtest3.cppalliance.org/index.html If more commits are pushed to the pull request, the docs will rebuild at the same URL. 2026-08-23 23:43:19 UTC |
|
GCOVR code coverage report https://383.capy.prtest3.cppalliance.org/gcovr/index.html Build time: 2026-08-23 23:56:36 UTC |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## develop #383 +/- ##
========================================
Coverage 98.09% 98.09%
========================================
Files 130 130
Lines 6291 6291
========================================
Hits 6171 6171
Misses 120 120
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report in Codecov by Harness.
🚀 New features to boost your workflow:
|
Member
|
Typo in "funciton" Should it be convertible to bool instead of The return value excludes typed handles (e.g. The Doxygen comment example drops the old semantic-requirements list about stop tokens and frame allocator propagation. Consider restoring this guidance in some form. |
Contributor
Author
Fixed.
C++ awaiters are not allowed to return "convertible to
My bad. I changed the concept to allow this.
They seem not to apply to the awaitable, but rather to the bodies of the resumed coroutines. Maybe instead, I will mention this in a (new) chapter which explains how to define your own coroutine? |
Contributor
Author
|
Added an issue for restoring the removed part about the discipline of using the passed execution context: |
mvandeberg
added a commit
to mvandeberg/capy
that referenced
this pull request
Aug 25, 2026
Maintainer decision: the Documentation workflow should not go red on doc findings. Drops --strict from the gate step, so the job exits 0 while the gated findings are still called out and annotated on the diff. This required decoupling two things that were conflated. Attention (which findings get called out and annotated) now keys on the --gate spec; failure keys on --strict alone. Previously attention keyed on --strict, so simply dropping it would have silently removed the annotations too -- the opposite of the intent. --strict still works, so promoting the gate back to blocking is a one-word change. Also re-anchors the reseed's gate-spec extractor. It located the spec by grepping the workflow for `check-no-new-violations.mjs --strict`, which this commit removes; it now anchors on the step name. Left unfixed the reseed would have aborted with "could not extract any --gate spec" -- it fails closed, but it would have failed. Verified the extractor still recovers all six specs. The gate step's comment claimed the expected state was two when_any.hpp C2 refusals. Those no longer exist -- they were resolved before cppalliance#383, and the live gated findings are cppalliance#383's own. Corrected. Trade-off, recorded: new gated violations no longer block, so they can accumulate until a reseed grandfathers them. That is the mechanism behind the existing backlog. The annotations and the job summary are now the only thing prompting a fix. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
mvandeberg
added a commit
to mvandeberg/capy
that referenced
this pull request
Aug 25, 2026
… the stale "blocking" claims The pre-authorised reseed exception named two when_any.hpp C2 refusals from the doc-improvement branch. Those findings no longer exist -- later docstring work resolved them before cppalliance#383 landed, and the hard C2 slice measured 0 at 64e6c7e^. A reseed run in that window would have reported `none gated` and needed no exception at all. Anyone following the old text would have been told to treat the real findings as regressions. The exception is now cppalliance#383's eight, listed verbatim so the report can be checked against them rather than trusted: seven C2 sentences over the limit (one of 64 words) across concept/io_awaitable.hpp, 2b.syntax.adoc and 4d.io-awaitable.adoc, plus one Capy.NoFluff "In order to". Bisected to 64e6c7e -- hard C2 was 0 at its parent, 7 at the commit. Recorded as what it is: accepting real findings against merged prose as backlog rather than blocking on an already-merged PR, not a judgement that the prose is fine. They surface in --show-baseline from the moment the reseed lands. Also retires the section that described the slice as those two refusals, and corrects four places still calling the gate "blocking" -- 4d78475 made it report-only and did not update the docs. Adds an explicit note under "How the gate works" saying the gate reports rather than blocks, what that costs, and that restoring it is one --strict away. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
mvandeberg
added a commit
to mvandeberg/capy
that referenced
this pull request
Aug 25, 2026
Retires 3778 stale fingerprints: 2356 vale_adoc, 1152 vale_docstrings, 193 mrdocs_warnings, 64 doc_lint A1 (+1 D2), 12 a11y. The vale removals are dominated by rules deliberately demoted in .vale.ini (584 Google.Headings, 59 Google.Will, 56 Google.EmDash) plus the vocabulary work; the gated rules retire in proportion to the Phase-4 pass (61 Capy.SimpleTense, 17 Capy.Terminology, 9 Capy.NoFluff). doc_lint's 64 A1 are the :page-mode: sweep. All six per-check rows reconcile as committed - removed + added = candidate, so nothing was silently truncated. Adds 124. Eight match the gate spec, and they are the one hand-granted exception this reseed exists for: - sentence_length :: C2:lint/.docstrings/concept/io_awaitable.hpp.adoc:cppalliance#1..cppalliance#4 - sentence_length :: C2:modules/ROOT/pages/2.cpp20-coroutines/2b.syntax.adoc:cppalliance#1,cppalliance#2 - sentence_length :: C2:modules/ROOT/pages/4.coroutines/4d.io-awaitable.adoc:cppalliance#1 - vale_adoc :: modules/ROOT/pages/4.coroutines/4d.io-awaitable.adoc:cppalliance#2:Capy.NoFluff All eight arrived with 64e6c7e, "Add constraints to IoAwaitable (cppalliance#383)": the hard C2 slice measured 0 at its parent and 7 at the commit, and the Capy.NoFluff addition is in the same commit's edit to 4d.io-awaitable.adoc. Seven over-length sentences (one of 64 words) and one "In order to". They are real findings against merged prose. Accepting them is a maintainer decision to carry them as backlog rather than block an already-merged PR -- not a judgement that the prose is correct. They appear in --show-baseline output from here on, which is where they should be picked up. Of the remaining 116 additions, 58 are advisory-C2 design-essay findings (the deliberate Part C2 carve-out), and most of the rest are new corpus rather than new defects: the buffers chapter was consolidated from six pages into 5a.buffers.adoc, and detail/io_result_combinators.hpp arrived with the io_result combinator rewrite. 21 are a11y; note only 6 are color-contrast, so 15 sit outside the rationale STYLE_GUIDE.md E4 records for demoting that check -- they look generator-produced rather than hand-authored, but E4's wording does not cover them and they are now permanent. Also removes the one-shot reseed blockquote, as it instructed, and the two statements that depended on it: the "one pre-authorised exception" clause in the never-reseed-to-go-green rule, and the section asserting sentence_length has no baseline entry. It has one now -- 65 fingerprints, 7 C2 and 58 advisory-C2. Candidate authored by CI (generatedAt 2026-08-25T18:03:26Z); a local run differs by ~54 fingerprints on this machine alone, which is why it was not. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
mvandeberg
added a commit
to mvandeberg/capy
that referenced
this pull request
Aug 25, 2026
The gate printed its comparison as JSON -- ~250 lines of which 8 mattered, none carrying a line number or a quote, so you still had to grep the corpus to find what broke. It now prints one block per finding: severity, file:line, rule, and an excerpt. --show-baseline adds the grandfathered findings still present, as warnings; --json restores the old dump. Under Actions each gated finding is also a ::warning annotation on the diff. Line numbers cannot live in the fingerprint -- keying on them once renamed 27 grandfathered findings from a comment-only insertion (df68f9b) -- so baseline.mjs gains --details, a sidecar the comparator requests and the reseed does not. Fingerprints are byte-identical either way. Docstring findings resolve back to the .hpp through a line map from extract-docstrings.mjs, then narrow within the block by locating the excerpt, since a doc comment can be 100 lines long. The gate is report-only by maintainer decision. Attention keys on --gate and failure on --strict, so restoring the block is one word. The Vale steps treat exit 1 (alerts found, the steady state) as success and propagate exit 2, so an annotation from them now means Vale actually broke. selftest.mjs runs in CI, non-blocking -- it is the only thing between a silent linter regression and a green run. Step names are standardised on "Docs:", "Lint:" and "Reseed:"; the reseed locates its --gate spec by matching the gate step's name, so that anchor moves too. Stops linting detail/ prose, which mrdocs.yml marks implementation-defined: 11 of 73 extracted files, plus 33 doc comments in `namespace detail` blocks inside public headers. The extractor now clears its output directory -- it never did, so a header that stopped producing output left its last .adoc linted forever. Vale vocabulary gains the domain prose it was missing and the coroutine keywords. The keyword split is the point: io_result has a reference page, so bare in prose it is a missing cpp: link and B1 says fix it; co_await has no page, so that alert can never be actioned. One real misspelling is fixed rather than whitelisted, and STYLE_GUIDE E4's a11y carve-out is extended to the three rule classes it did not name. Reseeds doc/lint/baseline.json from CI: retires 3845 stale fingerprints, adds 107. Eight match the gate spec and are PR cppalliance#383's, accepted as backlog rather than blocking an already-merged PR. The one-shot reseed blockquote is removed as it instructed. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
mvandeberg
added a commit
to mvandeberg/capy
that referenced
this pull request
Aug 25, 2026
The gate printed its comparison as JSON -- ~250 lines of which 8 mattered, none carrying a line number or a quote, so you still had to grep the corpus to find what broke. It now prints one block per finding: severity, file:line, rule, and an excerpt. --show-baseline adds the grandfathered findings still present, as warnings; --json restores the old dump. Under Actions each gated finding is also a ::warning annotation on the diff. Line numbers cannot live in the fingerprint -- keying on them once renamed 27 grandfathered findings from a comment-only insertion (df68f9b) -- so baseline.mjs gains --details, a sidecar the comparator requests and the reseed does not. Fingerprints are byte-identical either way. Docstring findings resolve back to the .hpp through a line map from extract-docstrings.mjs, then narrow within the block by locating the excerpt, since a doc comment can be 100 lines long. The gate is report-only by maintainer decision. Attention keys on --gate and failure on --strict, so restoring the block is one word. The Vale steps treat exit 1 (alerts found, the steady state) as success and propagate exit 2, so an annotation from them now means Vale actually broke. selftest.mjs runs in CI, non-blocking -- it is the only thing between a silent linter regression and a green run. Step names are standardised on "Docs:", "Lint:" and "Reseed:"; the reseed locates its --gate spec by matching the gate step's name, so that anchor moves too. Stops linting detail/ prose, which mrdocs.yml marks implementation-defined: 11 of 73 extracted files, plus 33 doc comments in `namespace detail` blocks inside public headers. The extractor now clears its output directory -- it never did, so a header that stopped producing output left its last .adoc linted forever. Vale vocabulary gains the domain prose it was missing and the coroutine keywords. The keyword split is the point: io_result has a reference page, so bare in prose it is a missing cpp: link and B1 says fix it; co_await has no page, so that alert can never be actioned. One real misspelling is fixed rather than whitelisted, and STYLE_GUIDE E4's a11y carve-out is extended to the three rule classes it did not name. Reseeds doc/lint/baseline.json from CI: retires 3845 stale fingerprints, adds 107. Eight match the gate spec and are PR cppalliance#383's, accepted as backlog rather than blocking an already-merged PR. The one-shot reseed blockquote is removed as it instructed.
mvandeberg
added a commit
that referenced
this pull request
Aug 25, 2026
The gate printed its comparison as JSON -- ~250 lines of which 8 mattered, none carrying a line number or a quote, so you still had to grep the corpus to find what broke. It now prints one block per finding: severity, file:line, rule, and an excerpt. --show-baseline adds the grandfathered findings still present, as warnings; --json restores the old dump. Under Actions each gated finding is also a ::warning annotation on the diff. Line numbers cannot live in the fingerprint -- keying on them once renamed 27 grandfathered findings from a comment-only insertion (df68f9b) -- so baseline.mjs gains --details, a sidecar the comparator requests and the reseed does not. Fingerprints are byte-identical either way. Docstring findings resolve back to the .hpp through a line map from extract-docstrings.mjs, then narrow within the block by locating the excerpt, since a doc comment can be 100 lines long. The gate is report-only by maintainer decision. Attention keys on --gate and failure on --strict, so restoring the block is one word. The Vale steps treat exit 1 (alerts found, the steady state) as success and propagate exit 2, so an annotation from them now means Vale actually broke. selftest.mjs runs in CI, non-blocking -- it is the only thing between a silent linter regression and a green run. Step names are standardised on "Docs:", "Lint:" and "Reseed:"; the reseed locates its --gate spec by matching the gate step's name, so that anchor moves too. Stops linting detail/ prose, which mrdocs.yml marks implementation-defined: 11 of 73 extracted files, plus 33 doc comments in `namespace detail` blocks inside public headers. The extractor now clears its output directory -- it never did, so a header that stopped producing output left its last .adoc linted forever. Vale vocabulary gains the domain prose it was missing and the coroutine keywords. The keyword split is the point: io_result has a reference page, so bare in prose it is a missing cpp: link and B1 says fix it; co_await has no page, so that alert can never be actioned. One real misspelling is fixed rather than whitelisted, and STYLE_GUIDE E4's a11y carve-out is extended to the three rule classes it did not name. Reseeds doc/lint/baseline.json from CI: retires 3845 stale fingerprints, adds 107. Eight match the gate spec and are PR #383's, accepted as backlog rather than blocking an already-merged PR. The one-shot reseed blockquote is removed as it instructed.
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
3 participants
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.
This add constraints to IoAwaitabe, so that the concept reflects what is needed for a type to be used in Capy-coroutines:
await_ready,await_resume,