| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Needs one backport to fix a test regression when not building with
debug.
I've also added an experimental dedup thing. One of the major issues
with rust is that cargo locks and requires all crates to be downloaded
offline, even if you're not using most of them because the crate
dependencies are Windows-specific. I haven't solved that, but I did
manage to remove about 40 of them by removing some workspace members
from the ruff monorepo that aren't actually part of the ruff binary in
any configuration (which means their dependencies don't need to be
locked, either). This has to be done before running pycargoebuild *and*
also done in the ebuild itself.
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Tidyups:
- $(cargo_crate_uris)
- Rust BDEPENDS come from the eclass except in very rare
circumstances (RUST_OPTIONAL=1)
- RUST_M{AX,IN}_VER where required.
- Suboptimal crate separator (`-` -> `@`)
Signed-off-by: Matt Jolly <kangie@gentoo.org>
Closes: https://github.com/gentoo/gentoo/pull/39218
Signed-off-by: Matt Jolly <kangie@gentoo.org>
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Arthur Zamarin <arthurzam@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Jakov Smolić <jsmolic@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
|
|
|
| |
New feature pulls in the zstd code that was formerly testsuite-only, so
it is now moved to RDEPEND.
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
I have recently noticed that running cargo test causes the resulting
tested binary to be configured without color support. In general this is
a huge flaw of the cargo "build" system -- it cannot handle building
something, then testing it, so instead it rebuilds the original thing.
It makes me incredibly nervous and clearly it is going wrong somehow --
and this is "expected functionality" in rust land, so plausibly the
canonical solution is to simply run tests as part of a completely
separate build.
It is not like reusing the same build directory will make anything go
faster. cargo's caching is very primitive here and basically all crates
will get recompiled anyway, for one reason or another.
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
| |
Restores logic introduced in 4.10.0 via commit
29c7ab715e11ae45f090a887c23c5c2bb3c9776d and overwritten by updating
0.4.8 to 0.5.0 in commit b4fba2fce4058ab102be23b0828f3d51744b8092.
No revbump needed as it doesn't affect the installed package.
Signed-off-by: Eli Schwartz <eschwartz93@gmail.com>
Signed-off-by: Sam James <sam@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Includes a backport for a test failure that I originally hit when trying
to package this version, which only manifests because we build with
release mode.
Thanks, debug asserts with side effects. :)
Also the testsuite has introduced the use of the zstd-sys crate. It's
not used at runtime though. Either way, depend on it so that the
testsuite builds faster.
Signed-off-by: Eli Schwartz <eschwartz93@gmail.com>
Signed-off-by: Sam James <sam@gentoo.org>
|
|
|
|
| |
Signed-off-by: James Le Cuirot <chewi@gentoo.org>
|
|
|
|
| |
Signed-off-by: Patrick McLean <chutzpah@gentoo.org>
|
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz93@gmail.com>
Signed-off-by: Sam James <sam@gentoo.org>
|
|
|
|
|
| |
Signed-off-by: Eli Schwartz <eschwartz93@gmail.com>
Signed-off-by: Sam James <sam@gentoo.org>
|
|
|
|
|
|
|
|
| |
The github project is now under a company organization, rather than the
lead developer's personal account.
Signed-off-by: Eli Schwartz <eschwartz93@gmail.com>
Signed-off-by: Sam James <sam@gentoo.org>
|