This project writes American English. In 47 files it does not, in 80 places:
cancelling, 53 times, across the engine, the payment, VBO and UI submodules and their tests;honour,honouredandhonours, 19 times, most of them new in this release cycle;modelledandmodelling, 6 times;favour, twice.
Nineteen of them sit on the documented API surface: OrchestraClientInterface::start() and WireAccount both describe a name being honoured, and docs/distributed.md and docs/subprocesses.md carry it into the published documentation. One is a string a person reads, Cancelling selected instances, the batch title on the instance listing, so the French catalog moves with it.
Why the lint lane does not catch it
The cspell job is green on 1.x and stays green. Its en-US dictionary accepts both forms of these four words, so the lane that enforces spelling cannot see the forms this module actually uses. That is why they accumulated.
Proposed resolution
Sweep all 80, and move the one translated string's French entry with it. No test comes with this: tests are for code logic, and the tests that read source files to assert things about the words in them are removed in #3624520: Drop the tests that scan source files instead of exercising code, and retranslate two French strings. A spelling problem is fixed in the spelling.
AI-Generated: Yes (Claude Code was used to help draft this issue summary and to write the code on the merge request. I reviewed and ran the work myself before posting it.)
Issue fork orchestra-3624521
Show commands
Start within a Git clone of the project using the version control instructions.
Or, if you do not have SSH keys set up on git.drupalcode.org:
Comments
Comment #3
mably commentedComment #4
mably commentedComment #5
mably commentedComment #7
mably commented