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

steve-downey / free_value_or / 34069335376
98%

Build:
DEFAULT BRANCH: main
Ran 07 Sep 2026 12:21AM UTC
Jobs 1
Files 7
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

07 Sep 2026 12:17AM UTC coverage: 96.803%. Remained the same
34069335376

push

github

steve-downey
Fix defects found reviewing the paper, and tighten its prose

Fact and citation check of D4270R1, with the technical findings applied to
the reference implementation as well as to the paper.

The concept did not enforce the const-observability rule the paper argues
for. Spelled requires(const T t), the const is discarded when T is deduced
as an lvalue reference, which is the ordinary case, so a type with non-const
operator bool and operator* modelled nullable whenever it was passed as an
lvalue. Stated on const remove_reference_t<T>& the rule holds for both
deduced forms. Every type the paper lists still models the concept, function
pointers included, and the suite is unchanged at 1424 passing tests.

reference_or's second dangling guard was stated on T&, the nullable, rather
than on iter_reference_t<T>, the type *m yields. A reference to the payload
never binds to the holder, so the trait was false for the reason that made
the guard unreachable: it could not fire. The prose claimed it rejected the
symmetric materializing case; now it does.

Also: or_invoke forwards the invocable, in the wording and the listing, as
the implementation always did -- the Effects called it as an lvalue while
the return type was computed from invoke_result_t<I>, which disagree for an
rvalue-only call operator. or_construct's wording gains the engaged-path
Mandates its member and reference-specialization counterparts already have.
or_construct can not dangle only when Ret is defaulted; naming a reference
Ret is now described, and added to the open questions. Corrected: "these
three functions" (there are four), the claim that smart pointers have no
members to add, the test count, and "the current implementation, in full",
which omitted the portability shims.

Wording moves from N5008 (2025-03) to N5054 (2026-07). N5008 predates
optional<T&> reaching the working draft, which this paper adds wording to.
Stale R0-isms and the singular "the author" are gone now that the paper ha... (continued)

1090 of 1126 relevant lines covered (96.8%)

17812.94 hits per line

Jobs
ID Job ID Ran Files Coverage
1 34069335376.1 07 Sep 2026 12:21AM UTC 7
96.8
GitHub Action Run
Source Files on build 34069335376
  • Tree
  • List 7
  • Changed 1
  • Source Changed 1
  • Coverage Changed 1
Coverage ∆ File Lines Relevant Covered Missed Hits/Line
  • Back to Repo
  • Github Actions Build #34069335376
  • 51e4b90e on github
  • Prev Build on main (#34008546503)
  • Next Build on main (#34070152215)
  • Delete
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