Alex Crichton [Thu, 14 Apr 2016 06:37:57 +0000 (23:37 -0700)]
Stream build script output to the console
This commit alters Cargo's behavior when the `-vv` option is passed (two verbose
flags) to stream output of all build scripts to the console. Cargo makes not
attempt to prevent interleaving or indicate *which* build script is producing
output, rather it simply forwards all output to one to the console.
Cargo still acts as a middle-man, capturing the output, to parse build script
output and interpret the results. The parsing is still deferred to completion
but the stream output happens while the build script is running.
On Unix this is implemented via `select` and on Windows this is implemented via
IOCP.
Closes #1106
Alex Crichton [Tue, 12 Apr 2016 21:55:19 +0000 (14:55 -0700)]
Allow specifying -v multiple times
This commit modifies the CLI interface to allow the verbose (-v) flag to be
specified multiple times. This'll be used soon to have `-vv` indicate that more
output should be generated than `-v` during a normal build.
Currently this commit changes the behavior of whether warnings are printed to
print warnings for the **entire DAG of dependencies** if `-vv` is specified.
That is, `--cap-lints` is never passed nor is any warning from build scripts if
`-vv` is specified.
Alex Crichton [Tue, 12 Apr 2016 21:28:52 +0000 (14:28 -0700)]
Forward warnings from build scripts
This adds support for forwarding warnings from build scripts to the main console
for diagnosis. A new `warning` metadata key is recognized by Cargo and is
printed after a build script has finished executing.
The purpose of this key is for build dependencies to try to produce useful
warnings for local crates being developed. For example a parser generator could
emit warnings about ambiguous grammars or the gcc crate could print warnings
about the C/C++ code that's compiled.
Warnings are only emitted for path dependencies by default (like lints) so
warnings printed in crates.io crates are suppressed.
cc #1106
bors [Sun, 12 Jun 2016 15:18:50 +0000 (08:18 -0700)]
Auto merge of #2720 - alexcrichton:ignore-nfs-more-often, r=brson
Ignore file locks on OSX NFS mounts
We already ignore NFS on Linux so let's do the same on OSX as well.
Alex Crichton [Thu, 19 May 2016 19:09:00 +0000 (12:09 -0700)]
Ignore file locks on OSX NFS mounts
We already ignore NFS on Linux so let's do the same on OSX as well.
bors [Sun, 12 Jun 2016 13:03:28 +0000 (06:03 -0700)]
Auto merge of #2741 - alexcrichton:cdylib, r=brson
Add support for cdylib crate types
And while we're at it this also extends support for all future crate types
assuming that they aren't linkable by default.
bors [Sun, 12 Jun 2016 12:21:23 +0000 (05:21 -0700)]
Auto merge of #2680 - alexcrichton:update-toml, r=brson
Update TOML parser to pick up a bugfix
Cargo has previously accepted invalid TOML as valid, but this bugfix should fix
the problem. In order to prevent breaking all crates immediately toml-rs has a
compatibility mode which emulates the bug that was fixed. Cargo will issue a
warning if this compatibility is required to parse a crate.
bors [Sun, 12 Jun 2016 11:46:54 +0000 (04:46 -0700)]
Auto merge of #2668 - alexcrichton:dont-warn-about-metadata, r=brson
Don't warn about `metadata` keys in the manifest
External tools may want to store metadata in `Cargo.toml` that they read but
Cargo itself doesn't read. For example `cargo-apk` uses this for pieces of
configuration. Cargo unfortunately, however, warns about these keys as "unused
keys in the manifest"
This commit instead whitelists the `package.metadata` key to not warn about any
data inside.
Alex Crichton [Thu, 12 May 2016 18:44:00 +0000 (11:44 -0700)]
Update TOML parser to pick up a bugfix
Cargo has previously accepted invalid TOML as valid, but this bugfix should fix
the problem. In order to prevent breaking all crates immediately toml-rs has a
compatibility mode which emulates the bug that was fixed. Cargo will issue a
warning if this compatibility is required to parse a crate.
Alex Crichton [Tue, 10 May 2016 19:55:19 +0000 (12:55 -0700)]
Don't warn about `metadata` keys in the manifest
External tools may want to store metadata in `Cargo.toml` that they read but
Cargo itself doesn't read. For example `cargo-apk` uses this for pieces of
configuration. Cargo unfortunately, however, warns about these keys as "unused
keys in the manifest"
This commit instead whitelists the `package.metadata` key to not warn about any
data inside.
Alex Crichton [Fri, 20 May 2016 23:34:50 +0000 (16:34 -0700)]
Add support for cdylib crate types
And while we're at it this also extends support for all future crate types
assuming that they aren't linkable by default.
bors [Sat, 11 Jun 2016 02:09:39 +0000 (19:09 -0700)]
Auto merge of #2781 - alexcrichton:block-if-dirty, r=brson
Prevent packaging a crate if any files are dirty
This commit alters Cargo's behavior to prevent publishing a crate by default if
any files in that crate are determined to be dirty, that is either modified or
not part of the working tree.
This can prevent common mistakes like many listed in #2063 and enables features like https://github.com/rust-lang/cargo/issues/841.
Closes https://github.com/rust-lang/cargo/issues/1597
Closes #2063
bors [Sat, 11 Jun 2016 01:32:24 +0000 (18:32 -0700)]
Auto merge of #2757 - alexcrichton:panic-profile-dox, r=brson
Document the `panic` profile option
Closes #2750
bors [Sat, 11 Jun 2016 00:34:29 +0000 (17:34 -0700)]
Auto merge of #2704 - alexcrichton:clean-no-fetch, r=brson
Don't fetch all crates on clean
Only fetch those that are needed by avoiding `ops::fetch`
bors [Fri, 10 Jun 2016 14:23:32 +0000 (07:23 -0700)]
Auto merge of #2779 - matklad:propagete-color-config, r=alexcrichton
Propagate --color option to rustc
closes #2740
Will try to add a test for this soon (and fix failing tests if any, compiling/running tests locally is slow :( ).
I am not sure what is the right place to add `--color` option to the command line. I use [`build_base_args`]. [`process`] also looks like a good candidate, because it is more general, but if we look at the [`CommandType`] we see that only `rustc` command supports `--color`.
[`build_base_args`]: https://github.com/matklad/cargo/blob/
1f7504397ce7c40ff708e2d31da164822e88ed37/src/cargo/ops/cargo_rustc/mod.rs#L449
[`process`]: https://github.com/matklad/cargo/blob/
1f7504397ce7c40ff708e2d31da164822e88ed37/src/cargo/ops/cargo_rustc/mod.rs#L608
[`CommandType`]: https://github.com/matklad/cargo/blob/
1f7504397ce7c40ff708e2d31da164822e88ed37/src/cargo/ops/cargo_rustc/engine.rs#L102
Alex Crichton [Fri, 10 Jun 2016 13:43:34 +0000 (06:43 -0700)]
Prevent packaging a crate if any files are dirty
This commit alters Cargo's behavior to prevent publishing a crate by default if
any files in that crate are determined to be dirty, that is either modified or
not part of the working tree.
This can prevent common mistakes like many listed in #2063, and subsequently...
Closes #2063
bors [Fri, 10 Jun 2016 13:48:41 +0000 (06:48 -0700)]
Auto merge of #2780 - Boddlnagg:patch-1, r=alexcrichton
Fix `cargo doc --open` on Windows
This fixes #2446. Note that I have not built cargo with this change, but I have tested the functionality in isolation (on Windows 10).
As to the issue itself, I don't know why the previous version didn't work, but `start` is redundant when `cmd /C` is used.
Patrick Reisert [Fri, 10 Jun 2016 13:07:55 +0000 (15:07 +0200)]
Fix `cargo doc --open` on Windows
This fixes #2446. Note that I have not built cargo with this change, but I have tested the functionality in isolation (on Windows 10).
As to the issue itself, I don't know why the previous version didn't work, but `start` is redundant when `cmd /C` is used.
Aleksey Kladov [Thu, 9 Jun 2016 20:55:17 +0000 (23:55 +0300)]
Propagate --color option to rustc
bors [Fri, 10 Jun 2016 01:52:51 +0000 (18:52 -0700)]
Auto merge of #2778 - alexcrichton:dont-print-on-q-when-run, r=brson
Don't print extra info on -q + `cargo run`
If a process fails then it's probably printing out why and Cargo doesn't need to
do it all over again.
Closes #2735
Alex Crichton [Thu, 9 Jun 2016 11:14:41 +0000 (04:14 -0700)]
Don't print extra info on -q + `cargo run`
If a process fails then it's probably printing out why and Cargo doesn't need to
do it all over again.
Closes #2735
bors [Thu, 9 Jun 2016 10:03:54 +0000 (03:03 -0700)]
Auto merge of #2777 - matklad:harness-help, r=alexcrichton
Explain how to get help for the test harness
TIL that the default test harness has a non-trivial number of options. Guess it's better to share this knowledge.
Aleksey Kladov [Wed, 8 Jun 2016 18:01:00 +0000 (21:01 +0300)]
Explain how to get help for the test harness
bors [Tue, 7 Jun 2016 23:32:25 +0000 (16:32 -0700)]
Auto merge of #2771 - alexcrichton:update-replace, r=brson
Don't lock overrides if we're updating them
There was already a function for this, `keep`, it was just forgotten to be
called.
Closes #2766
bors [Tue, 7 Jun 2016 22:52:14 +0000 (15:52 -0700)]
Auto merge of #2770 - alexcrichton:new-curl, r=brson
Update to curl 0.3
bors [Tue, 7 Jun 2016 08:05:44 +0000 (01:05 -0700)]
Auto merge of #2763 - mbrubeck:native-dirs, r=alexcrichton
Correctly record multiple native dirs per package
This fixes a bug when a package's build script outputs multiple library search
paths. Because Compilation::native_dirs is a `HashMap<PackageId, PathBuf>` it
can only store one path per package. Currently if there are multiple paths,
all but the last will be inserted and then overwritten.
The key from this map is never used anyway, so this fixes the bug by changing
it from a HashMap to a HashSet.
Matt Brubeck [Mon, 6 Jun 2016 17:09:33 +0000 (10:09 -0700)]
Test for correct library search paths during cargo run
Matt Brubeck [Thu, 2 Jun 2016 20:53:31 +0000 (13:53 -0700)]
Strip native= prefix from -L arg before additing it to LD_LIBRARY_PATH
Fixes #2765.
Matt Brubeck [Thu, 2 Jun 2016 18:37:32 +0000 (11:37 -0700)]
Correctly record multiple native dirs per package
This fixes a bug when a package's build script outputs multiple library search
paths. Because Compilation::native_dirs is a `HashMap<PackageId, PathBuf>` it
can only store one path per package. Currently if there are multiple paths,
all but the last will be inserted and then overwritten.
The key from this map is never used anyway, so this fixes the bug by changing
it from a HashMap to a HashSet.
bors [Mon, 6 Jun 2016 13:24:15 +0000 (06:24 -0700)]
Auto merge of #2774 - andrehjr:patch-doc-2754, r=alexcrichton
Document file destination of build scripts.
Close #2754
André Luis Leal Cardoso Junior [Mon, 6 Jun 2016 03:57:15 +0000 (00:57 -0300)]
Document file destination of build scripts.
Close #2754
bors [Mon, 6 Jun 2016 09:16:38 +0000 (02:16 -0700)]
Auto merge of #2772 - knight42:patch, r=alexcrichton
Improve autocompletion
1. Add path completion to manifest-path options
2. Support subcommand metadata
Knight [Sun, 5 Jun 2016 10:08:20 +0000 (18:08 +0800)]
Add zsh autocompletion to subcommand metadata
Knight [Sun, 5 Jun 2016 09:45:24 +0000 (17:45 +0800)]
zsh: add path completion to manifest-path options
Alex Crichton [Sun, 5 Jun 2016 07:19:55 +0000 (00:19 -0700)]
Don't lock overrides if we're updating them
There was already a function for this, `keep`, it was just forgotten to be
called.
Closes #2766
Alex Crichton [Wed, 18 May 2016 17:01:40 +0000 (10:01 -0700)]
Update to curl 0.3
bors [Sun, 5 Jun 2016 06:06:58 +0000 (23:06 -0700)]
Auto merge of #2769 - alexcrichton:no-oom-travis, r=alexcrichton
Pass -j1 on Travis
Looks like if we do things concurrently we hit the OOM killer, so let's not do
that!
Alex Crichton [Tue, 31 May 2016 23:44:56 +0000 (16:44 -0700)]
Pass -j1 on Travis
Looks like if we do things concurrently we hit the OOM killer, so let's not do
that!
Alex Crichton [Tue, 31 May 2016 20:30:54 +0000 (13:30 -0700)]
Document the `panic` profile option
Closes #2750
bors [Tue, 31 May 2016 16:26:55 +0000 (09:26 -0700)]
Auto merge of #2747 - WiSaGaN:feature/rename-main-thread, r=alexcrichton
Rename main thread from '<main>' to 'main'.
This pull request resolves the test failure in rust pull request 33803
https://github.com/rust-lang/rust/pull/33803
Wangshan Lu [Tue, 31 May 2016 11:43:44 +0000 (19:43 +0800)]
Use place holder for main thread name in tests
bors [Tue, 31 May 2016 05:03:58 +0000 (22:03 -0700)]
Auto merge of #2755 - matklad:remove-useless-formats, r=alexcrichton
Remove useless format calls from tests
bors [Tue, 31 May 2016 03:32:21 +0000 (20:32 -0700)]
Auto merge of #2752 - wesleywiser:patch-1, r=alexcrichton
Fix example in cargo manpage
Fixes #2731
Aleksey Kladov [Mon, 30 May 2016 22:41:36 +0000 (01:41 +0300)]
Remove useless format calls from tests
Wesley Wiser [Sun, 29 May 2016 23:18:59 +0000 (19:18 -0400)]
Fix example in cargo manpage
Fixes #2731
Wangshan Lu [Sat, 28 May 2016 10:05:58 +0000 (18:05 +0800)]
Rename main thread from '<main>' to 'main'.
This pull request resolves the test failure in rust pull request 33803
bors [Fri, 27 May 2016 00:14:54 +0000 (17:14 -0700)]
Auto merge of #2742 - alexcrichton:new-error-format, r=wycats
Fix tests for RUST_NEW_ERROR_FORMAT
A few tests were a bit too strict in the output they were looking for, this
instead relaxes some assertions to just what we're interested in.
bors [Thu, 26 May 2016 16:37:11 +0000 (09:37 -0700)]
Auto merge of #2744 - alexcrichton:bump, r=alexcrichton
Bump version numbers
Working on 0.12.0 now
Alex Crichton [Thu, 26 May 2016 16:36:37 +0000 (09:36 -0700)]
Bump version numbers
Working on 0.12.0 now
Alex Crichton [Wed, 25 May 2016 17:04:39 +0000 (10:04 -0700)]
Fix tests for RUST_NEW_ERROR_FORMAT
A few tests were a bit too strict in the output they were looking for, this
instead relaxes some assertions to just what we're interested in.
bors [Thu, 26 May 2016 04:54:45 +0000 (21:54 -0700)]
Auto merge of #2743 - alexcrichton:better-test-suite, r=alexcrichton
Split the test suite into multiple binaries
Helps reduce compile time of tests, more ergonomic names to type, more conventional configuration, and makes it basically easier to work with.
The previous `support` module is now a separate `cargotest` crate.
Alex Crichton [Thu, 26 May 2016 00:06:25 +0000 (17:06 -0700)]
Shared the test suite into multiple binaries
Compiling everything in one binary was getting annoying as it just took forever
to build, instead shard it all up so we can build just particular test suites at
a time.
Alex Crichton [Wed, 25 May 2016 20:55:42 +0000 (13:55 -0700)]
Remove the test! macro
This isn't really necessary and we can do setup manually in a few cases and
lazily do it in all the others.
bors [Tue, 24 May 2016 20:03:39 +0000 (13:03 -0700)]
Auto merge of #2739 - alexcrichton:configure-colors, r=alexcrichton
Configure colors of both stdout/stderr
Previously color configuration only affected stdout, not stder as well.
Closes #2734
Alex Crichton [Tue, 24 May 2016 18:52:31 +0000 (11:52 -0700)]
Configure colors of both stdout/stderr
Previously color configuration only affected stdout, not stder as well.
Closes #2734
bors [Tue, 24 May 2016 18:49:48 +0000 (11:49 -0700)]
Auto merge of #2737 - aidanhs:aphs-hash-url-serialization, r=alexcrichton
Hash url serialization to ensure hash stability
`as_str` is documented to return the serialization of a url (http://servo.github.io/rust-url/url/struct.Url.html#method.as_str).
The serialization of a url has a spec (https://url.spec.whatwg.org/#url-serializing).
Hashing the serialization of a url insulates cargo against possible decisions to alter how rust-url does hashing (I've observed it happening twice). Note that rust-url happens to do this internally (but is not guaranteed to in future) and so this change doesn't actually have any effect on hash values.
r? @alexcrichton
Closes #1710
Aidan Hobson Sayers [Tue, 24 May 2016 13:29:52 +0000 (14:29 +0100)]
Hash url serialization to ensure hash stability
Closes #1710
bors [Mon, 23 May 2016 20:49:01 +0000 (13:49 -0700)]
Auto merge of #2730 - alexcrichton:fix-panic-abort-build-script, r=alexcrichton
Fix build scripts and panic=abort
Build scripts were apparently always compiled with the "dev" profile rather than
the standard "match whatever the normal build was" profile, which meant that if
dev/release disagreed on panic=abort you'd get compile errors. Seems bad!
This commit fixes this by just having build scripts always compile with the same
profile as libraries (for now), as this was the original intention anyway.
Closes #2726
Alex Crichton [Mon, 23 May 2016 15:44:27 +0000 (08:44 -0700)]
Fix build scripts and panic=abort
Build scripts were apparently always compiled with the "dev" profile rather than
the standard "match whatever the normal build was" profile, which meant that if
dev/release disagreed on panic=abort you'd get compile errors. Seems bad!
This commit fixes this by just having build scripts always compile with the same
profile as libraries (for now), as this was the original intention anyway.
Closes #2726
bors [Sat, 21 May 2016 01:52:38 +0000 (18:52 -0700)]
Auto merge of #2687 - alexcrichton:panic-abort, r=wycats
Implement the `panic` profile option
This is the Cargo half of the implementation of [RFC 1513] which adds a new
`profile.*.panic` option to customize the `-C panic` argument to the compiler.
This is not passed by default and can otherwise be specified as `abort` or
`unwind` on the nightly compiler.
[RFC 1513]: https://github.com/rust-lang/rfcs/pull/1513
The `profile.*.panic` option is *only* used from the top-level crate, not each
crate individually. This means that applications should customize this value as
they see fit, and libraries will only use their own value when they're being
tested.
Cargo also has specific knowledge that when *testing* a crate it can't pass
`-C panic=abort` for now as the default test harness requires `panic=unwind`.
This essentially just means that `cargo test` will continue to work for crates
that specify `panic=abort` in Cargo.toml.
Alex Crichton [Fri, 13 May 2016 18:35:52 +0000 (11:35 -0700)]
Implement the `panic` profile option
This is the Cargo half of the implementation of [RFC 1513] which adds a new
`profile.*.panic` option to customize the `-C panic` argument to the compiler.
This is not passed by default and can otherwise be specified as `abort` or
`unwind` on the nightly compiler.
[RFC 1513]: https://github.com/rust-lang/rfcs/pull/1513
The `profile.*.panic` option is *only* used from the top-level crate, not each
crate individually. This means that applications should customize this value as
they see fit, and libraries will only use their own value when they're being
tested.
Cargo also has specific knowledge that when *testing* a crate it can't pass
`-C panic=abort` for now as the default test harness requires `panic=unwind`.
This essentially just means that `cargo test` will continue to work for crates
that specify `panic=abort` in Cargo.toml.
bors [Fri, 20 May 2016 23:56:13 +0000 (16:56 -0700)]
Auto merge of #2723 - alexcrichton:override-errors, r=wycats
Don't throw away errors with `-p` arguments
This was unfortunately ignoring errors which would helpfully tell you how to
rerun a command with a more precise specification.
Closes #2641
Alex Crichton [Fri, 20 May 2016 21:25:43 +0000 (14:25 -0700)]
Don't throw away errors with `-p` arguments
This was unfortunately ignoring errors which would helpfully tell you how to
rerun a command with a more precise specification.
Closes #2641
bors [Fri, 20 May 2016 21:09:55 +0000 (14:09 -0700)]
Auto merge of #2722 - alexcrichton:stderr-not-stdout, r=alexcrichton
Report status to stderr -- last few tests
Fixup of #2693 for the last few tests.
Alex Crichton [Fri, 20 May 2016 16:23:50 +0000 (09:23 -0700)]
Fix up last few remaining tests status-to-stderr
Alex Crichton [Fri, 20 May 2016 16:08:48 +0000 (09:08 -0700)]
Merge branch 'status-2-stderr' of https://github.com/matklad/cargo into stdout-not-stderr
Aleksey Kladov [Fri, 20 May 2016 14:19:49 +0000 (17:19 +0300)]
fix cargo_compile_path_deps
Aleksey Kladov [Fri, 20 May 2016 14:00:15 +0000 (17:00 +0300)]
fix test_cargo_publish
Aleksey Kladov [Fri, 20 May 2016 13:54:01 +0000 (16:54 +0300)]
fix test_cargo_freshness
Aleksey Kladov [Fri, 20 May 2016 13:39:58 +0000 (16:39 +0300)]
fix test_compile_git_deps
Aleksey Kladov [Fri, 20 May 2016 13:22:58 +0000 (16:22 +0300)]
fix some cargo_compile tests
Aleksey Kladov [Fri, 20 May 2016 12:02:09 +0000 (15:02 +0300)]
fix test_cargo_bench
Aleksey Kladov [Fri, 20 May 2016 11:47:39 +0000 (14:47 +0300)]
fix test_cargo_overrides
Aleksey Kladov [Fri, 20 May 2016 01:07:54 +0000 (04:07 +0300)]
one more regexp
Aleksey Kladov [Fri, 20 May 2016 00:52:13 +0000 (03:52 +0300)]
fix test_cargo_package
Aleksey Kladov [Fri, 20 May 2016 00:21:55 +0000 (03:21 +0300)]
fix test_cargo_registry
bors [Thu, 19 May 2016 22:00:03 +0000 (15:00 -0700)]
Auto merge of #2719 - semarie:test-git-vcs, r=alexcrichton
PR #1124 have replaced `git` with `vcs`
changes test_cargo_new::author_prefers_cargo to testing configuration
with `vcs`
Sébastien Marie [Thu, 19 May 2016 19:42:32 +0000 (21:42 +0200)]
PR #1124 have replaced `git` with `vcs`
changes test_cargo_new::author_prefers_cargo to testing configuration
with `vcs`
Aleksey Kladov [Thu, 19 May 2016 00:51:07 +0000 (03:51 +0300)]
more auto refactoring fixes
Aleksey Kladov [Thu, 19 May 2016 00:02:41 +0000 (03:02 +0300)]
fix more tests with auto refactoring
Aleksey Kladov [Wed, 18 May 2016 23:41:30 +0000 (02:41 +0300)]
more regexp fixes
Aleksey Kladov [Sun, 15 May 2016 22:16:54 +0000 (01:16 +0300)]
fix test_cargo_install
Aleksey Kladov [Sun, 15 May 2016 22:03:35 +0000 (01:03 +0300)]
fix exit_code tests
Aleksey Kladov [Sun, 15 May 2016 21:48:11 +0000 (00:48 +0300)]
fix more tests
Aleksey Kladov [Sun, 15 May 2016 21:26:24 +0000 (00:26 +0300)]
fix more test with insane regexes
Aleksey Kladov [Sun, 15 May 2016 21:17:56 +0000 (00:17 +0300)]
fix more tests with regex
Aleksey Kladov [Sun, 15 May 2016 21:07:04 +0000 (00:07 +0300)]
fix test_cargo_test
Aleksey Kladov [Sun, 15 May 2016 01:11:00 +0000 (04:11 +0300)]
fix some tests with IntelliJ-Rust automated refactoring
Aleksey Kladov [Sat, 14 May 2016 21:44:18 +0000 (00:44 +0300)]
fix some more tests with regexp
Aleksey Kladov [Sat, 14 May 2016 21:15:22 +0000 (00:15 +0300)]
fix a bunch of tests with regexp
Aleksey Kladov [Sat, 14 May 2016 21:00:00 +0000 (00:00 +0300)]
fix test_cargo_clean
Aleksey Kladov [Sat, 14 May 2016 20:53:49 +0000 (23:53 +0300)]
fix test_cargo_build_lib
Aleksey Kladov [Sat, 14 May 2016 20:49:39 +0000 (23:49 +0300)]
fix test_cargo_build_auth
Aleksey Kladov [Sat, 14 May 2016 20:31:11 +0000 (23:31 +0300)]
report status to stderr instead of stdout
bors [Wed, 18 May 2016 22:15:03 +0000 (15:15 -0700)]
Auto merge of #2710 - sbeckeriv:the-missing-lib, r=alexcrichton
The missing lib
Dearest Reviewer,
This pull request allows the value of project.links to be passed into
the build script via an environment variable. This closes #1220 . I have
added a test that makes sure that the environment variable is set. Also
note I am using CARGO_LIB not LIB because it appears to be used for
windows in appveyor. I think CARGO_LIB fits better.
I have also updated the documentation to reflect the new variable.
Thanks!
Becker
Stephen Becker IV [Tue, 17 May 2016 18:54:43 +0000 (11:54 -0700)]
The missing lib
Dearest Reviewer,
This pull request allows the value of project.links to be passed into
the build script via an environment variable. This closes #1220 . I have
added a test that makes sure that the environment variable is set. Also
note I am using CARGO_LIB not LIB because it appears to be used for
windows in appveyor. I think CARGO_LIB fits better.
I have also updated the documentation to reflect the new variable.
Thanks!
Becker
[Updated]
Github comments moved to CARGO_MANIFEST_LINKS
Fix if statement
bors [Wed, 18 May 2016 18:18:39 +0000 (11:18 -0700)]
Auto merge of #2703 - alexcrichton:target-dir-abs-path, r=alexcrichton
Fix relative `build.target-dir` configuration
Closes #2700
bors [Wed, 18 May 2016 17:08:25 +0000 (10:08 -0700)]
Auto merge of #2706 - alx741:extend_manpages, r=alexcrichton
Add man pages for cargo commands
So, the markdown port (#2672) is in idle, I'll be working in a way to make it without depending on external tools, but that is going to take a bit longer.
These are the **troff** pages for the rest of the cargo commands; previously the *cargo-build(1)* was added, so here are the rest.
No extra build dependencies here, just the new man pages.
bors [Wed, 18 May 2016 03:54:03 +0000 (20:54 -0700)]
Auto merge of #2688 - carols10cents:docs-refresh, r=alexcrichton
Docs refresh
Hi! Sooooo I've been thinking about the cargo docs lately and today I decided to try some of the things I've been thinking about.
There are some smaller changes in here, typos and clarifications and stuff, that I'm totally willing to break out into a separate PR(s) if yinz want.
My goals for the larger changes were:
- **Make one obvious place to go for info on every which way to specify a dependency.** I've found myself clicking around the different doc pages a lot to find the semver syntax, the different keys that can go with a git location, the `path` override syntax, etc. So I made one top-level "Specifying Dependencies" page and moved relevant info from the guide, the manifest page, and the crates.io page.
- **Make The Guide a bit more focused.** There were a few topics in there that didn't really fit in the story arc of trying to get someone up and running with cargo-- it veered off into some more advanced topics that I've moved to places I think they fit better.
- **Make the crates.io page only about publishing, not using and publishing.** The info about adding dependencies got folded into either the guide or the new specifying dependencies page, so this should close https://github.com/rust-lang/cargo/issues/1035.
I'm also thinking about having someone new to rust/cargo try following the guide, but wanted to get a first pass from experienced people :)
bors [Wed, 18 May 2016 02:09:43 +0000 (19:09 -0700)]
Auto merge of #2713 - sbeckeriv:fix-name-ordering, r=alexcrichton
Correct author order
Dearest Reviewer,
PR #2696 added some new checks for environment variables for author and
email. After the pull request was merged a comment was left about gits
ordering for looking at the variables. This pull request closes #2705 by
reordering the look up and also addeds CARGO_NAME and CARGO_EMAIL to the
top of the order.
Thanks
Becker