-
Notifications
You must be signed in to change notification settings - Fork 6.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[vcpkg] Miscellaneous internal improvements extracted from #15424 #15677
[vcpkg] Miscellaneous internal improvements extracted from #15424 #15677
Conversation
LGTM but there is a regression about
|
From the cartographer:x86-windows failure:
However, |
From the stdout:
stderr:
Looks like a linker crash. GDAL doesn't appear to add |
Depends on #15802 |
…roschuma/tool-ports-infra
…roschuma/tool-ports-infra
Somehow |
From
And in the config log:
So another config variation between release and debug that doesn't repro locally. |
…roschuma/tool-ports-infra
From opencascade:x86-windows:
|
…roschuma/tool-ports-infra
From armadillo:x64-uwp:
I have a fix prepared in https://github.com/microsoft/vcpkg/pull/15808/files#diff-20fb8471fdc2c697e520789ae25d430898df4b1a99ea13f97d571d22e12d4732 which changes armadillo to use |
7d53e10
to
a6cdaa4
Compare
…15424 (microsoft#15677) * [vcpkg] Miscellaneous internal improvements extracted from microsoft#15424 * [vcpkg] CR comments * [armadillo] Use vcpkg_from_git() to workaround gitlab missing archive Co-authored-by: Robert Schumacher <[email protected]> Co-authored-by: Billy Robert O'Neal III <[email protected]>
While waiting on #15424, I've extracted a few improvements:
g_debugging
in unit tests to an environment variableVCPKG_DEBUG
which is enabled in CI. This makes it reasonable to run unit tests as part of every build when developing by cutting out noise (ninja && ./vcpkg-test
).ExportPlanAction::dependencies
had a totally unused argumentOptional::map()
+ testsBinaryParagraph
and it causes database corruption if we add this to the status database unchanged. There's an open question about whether differently-named triplets with identical contents should hash the same, but with the current implementation they are not substitutable.