On Fri, 2024-11-01 at 07:36 +0100, Tobias Burnus wrote: > Am 31.10.24 um 21:57 schrieb David Malcolm: > > On Thu, 2024-10-31 at 11:36 -0700, Andrew Pinski wrote: > > > This is recorded as > > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=117361 already. > > > I think one idea is to set LANG=C before doing the self-check. > > > But I > > > will allow David and other think of better ways of fixing it. > > > > Sorry about this. For some reason this isn't reproducing for me. > > > > That said, I think setting LANG=C when running selftests is > > probably > > the best solution. How does the attached (untested) patch look? > > Or you do the same as for https://gcc.gnu.org/PR115203 ; namely, > your commit > https://gcc.gnu.org/r15-866-g2dbb1c124c1e585dc413132d7a8d4be62c6b7baa > > added: > > test_diagnostic_path (pretty_printer *event_pp) > : simple_diagnostic_path (event_pp) > { > + disable_event_localization (); > } > > and then disabled the _(...) localization for those: > > - text_info ti (_(fmt), &ap, 0, nullptr, &rich_loc); > + text_info ti (m_localize_events ? _(fmt) : fmt, > + &ap, 0, nullptr, &rich_loc); > > > In any case, I think it would be good to either use LC_ALL or > code like disable_event_localization but not both, depending > on the code path.
Doing so for this case would require implementing a new feature in the diagnostic_context to suppress message translation, so I went for the simpler option (bigger hammer?) of using LC_ALL=C. Dave
