| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
| |
This was actually not breaking the build, just outputing an error
--
MOS_MIGRATED_REVID=117481375
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Bazel expects to be told the width of the terminal; during
bootstrap, we do not have a good mechanism of determining it,
so we have to resort to a hard-coded value. However, 97 is a
not very well motivated value (and actually not a very useful
value either) which is hard to guess that this is a default value.
Resorting to Bazel's built-in default of 80 is only slightly
more useful, but, at least, it is easily recognizable as a default
value, as 80 is the width of the famous VT100, the standard punch
card, etc.
--
Change-Id: I62403ca37ee74a090067a5a4248028e3d624b7c6
Reviewed-on: https://bazel-review.googlesource.com/#/c/3082
MOS_MIGRATED_REVID=117346018
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For the process wrapper the value '-' has a special meaning
as file name to which stdout/stderr are to be redirected: do
not redirect. However, the simple shell script that serves
as process wrapper during the bootstrap phase was not aware
of this special meaning, resulting in loss of useful process
output, which is especially annoying during bootstrap. Fix this.
--
Change-Id: Ifcf84e9000d74dafc69b675f192c1fc1cce484e8
Reviewed-on: https://bazel-review.googlesource.com/#/c/3081
MOS_MIGRATED_REVID=117338558
|
|
|
|
|
|
|
|
|
|
|
| |
All Bazel should now contain the latest version with a date
modifier to ensure we can track all Bazel builds out there.
It is especially important for #1014.
--
Change-Id: I8d17d9e1660b104c2b8623d5a8a06d010d044ea7
Reviewed-on: https://bazel-review.googlesource.com/3070
MOS_MIGRATED_REVID=117120584
|
|
|
|
|
|
|
| |
Fixes #739.
--
MOS_MIGRATED_REVID=116257685
|
|
|
|
|
|
|
|
| |
This is a temporary workaround to enable the use
of a cc_configure() rules to auto-configure C++ toolchain.
--
MOS_MIGRATED_REVID=116140726
|
|
|
|
|
|
|
|
| |
This target include all non tests targets of Bazel to do integration tests of
bootstrapping.
--
MOS_MIGRATED_REVID=115830741
|
|
|
|
|
|
|
|
|
|
|
|
| |
Rename run_silent to run and add a global VERBOSE variable that tunes
whether the run function prints its output or not.
This is for better debugging possibilities of Bazel's self-build, though
compile.sh remains silent as before and only displays the command's output
in case of an error.
--
MOS_MIGRATED_REVID=115599355
|
|
|
|
|
|
|
|
|
|
| |
As part of this, change atexit to receive the name of a function instead of
just a chunk of code. This way, we can also control the return value of
each individual atexit function and report it to the user, should these
routines fail when the program apparently succeeded.
--
MOS_MIGRATED_REVID=115575895
|
|
|
|
|
|
|
|
|
| |
explicitly create it.
Turns out, mktemp doesn't like being passed a non-existent directory.
--
MOS_MIGRATED_REVID=115454169
|
|
|
|
|
|
|
|
|
|
|
| |
It seems like the change introduced in pull request #785 does not
function universally. On some releases of Debian and Ubuntu, it looks
like packages are declared as providing java{7,8}-sdk.
Fixes #961.
--
MOS_MIGRATED_REVID=115338913
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fix three issues, one cosmetic, with setup-intellij.sh: running bazel
query "deps(${target}, 1)" on generated java files with a visibility
specification was returning two targets, one of which was the
visibility. Filtering for rules fixes that.
Bazel uses the javac langtools in third_party/ in preference to the
SDK's, but that wasn't reflected in the order of dependencies in the
module. This requires a small hack, since the SDK should appear before
some other jars in the dependencies.
Finally, remove some apparently dead code related to ijar that was just
printing errors when the script was run.
--
Change-Id: Iba5ae5a46c720282d99f193cd098ea14cc5eef52
Reviewed-on: https://bazel-review.googlesource.com/2990
MOS_MIGRATED_REVID=115246698
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=115207057
|
|
|
|
|
|
|
|
|
| |
Jenkins setup is not done yet, but it appears doable.
Work towards #927.
--
MOS_MIGRATED_REVID=115059680
|
|
|
|
|
|
|
| |
This should fix the current ci.bazel.io breakage.
--
MOS_MIGRATED_REVID=114892993
|
|
|
|
|
|
|
| |
Needed for #276.
--
MOS_MIGRATED_REVID=114693711
|
|
|
|
|
|
|
|
|
| |
So that bazel can read it.
Fixes #894.
--
MOS_MIGRATED_REVID=114689351
|
|
|
|
|
|
|
| |
Fixes #887.
--
MOS_MIGRATED_REVID=114681512
|
|
|
|
|
|
|
|
| |
The recent change in the bootstrap process was causing a warning
about sandboxing not being activated.
--
MOS_MIGRATED_REVID=114106185
|
|
|
|
|
|
|
| |
--
Change-Id: Ideb2011d23da470276420afcc8b5a71f1470f391
Reviewed-on: https://bazel-review.googlesource.com/#/c/2771
MOS_MIGRATED_REVID=113639747
|
|
|
|
|
|
|
|
|
|
|
|
| |
This will fix some issue with loading skylark rules from @bazel_tools.
Tested with ./compile.sh all and also a project that use pkg_tar from
@bazel_tools (which is broken at HEAD).
--
Change-Id: Iffbb7134ef5fee497890c4d01b85084973d45ad0
Reviewed-on: https://bazel-review.googlesource.com/2800
MOS_MIGRATED_REVID=113508661
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=113356411
|
|
|
|
|
|
|
|
|
|
|
| |
On Windows, MSYS will mangle all arguments that resemble Unix paths
when executing (exec*()) non-msys executables (in an attempt to convert
them to Windows paths). This affects ``//src:bazel`` (it becomes
``/src:bazel``) but not ``src:bazel``. This CL converts to the latter in
bootstrapping shell scripts to work around this issue.
--
MOS_MIGRATED_REVID=113349821
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Issue that caused the rollback (#819) is resolved
*** Original change description ***
Automated [] rollback of [].
*** Reason for rollback ***
Prerequisite for rolling back j2objc base workspace change
*** Original change description ***
Remove base_workspace from bazel setup
I also removed a couple places the documentation referred to it incorrectly.
There are still a couple of blog posts that mention it, but that seemed okay.
RELNOTES: A bazelrc with --package_path set is no longer required for Bazel to
find its tools. This also means that building //......
--
MOS_MIGRATED_REVID=113259357
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Prerequisite for rolling back j2objc base workspace change
*** Original change description ***
Remove base_workspace from bazel setup
I also removed a couple places the documentation referred to it incorrectly.
There are still a couple of blog posts that mention it, but that seemed okay.
RELNOTES: A bazelrc with --package_path set is no longer required for Bazel to
find its tools. This also means that building //... should work to build
everything in the workspace without including Bazel's own targets.
--
MOS_MIGRATED_REVID=113164089
|
|
|
|
|
|
|
|
| |
RELNOTES: JDK-related targets are now available via @local_jdk (instead of
@local-jdk).
--
MOS_MIGRATED_REVID=113067049
|
|
|
|
|
|
|
|
| |
Update buildenv.sh to detect aarch64 as on Nvidias Jetson TX1 as ARM architecture. Final goal is to build tensorflow on jetson tx1
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/814
MOS_MIGRATED_REVID=112967980
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
terminal, not the bazel server.
This fixes two user annoyances:
1) the command can now read from stdin
2) the bazel lock is not held when the command is invoked
The script use's bazel's `--script_path` flag to write out the shell script that runs the command.
--
MOS_MIGRATED_REVID=112815712
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=112685830
|
|
|
|
|
|
|
|
|
|
|
|
| |
I also removed a couple places the documentation referred to it incorrectly.
There are still a couple of blog posts that mention it, but that seemed okay.
RELNOTES: A bazelrc with --package_path set is no longer required for Bazel to
find its tools. This also means that building //... should work to build
everything in the workspace without including Bazel's own targets.
--
MOS_MIGRATED_REVID=112604261
|
|
|
|
|
|
|
| |
The generated project is not perfect, but this does unblock Manuel for the time being.
--
MOS_MIGRATED_REVID=112571724
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This executable in the workspace can be another Bazel binary whose
version will change with the code it's next to, or a shell script which
downloads a fixed version from some location.
RELNOTES: A tools/bazel script in the workspace will be executed
as an opportunity to use a fixed version of Bazel (not implemented for
the homebrew recipe yet).
Fixes #521
--
Change-Id: Id06177d9c2b259cd9d6fd62edb5abe541342dd05
Reviewed-on: https://bazel-review.googlesource.com/2620
MOS_MIGRATED_REVID=112477232
|
|
|
|
|
|
|
|
|
| |
Somewhat. The generated IntelliJ project is still somewhat broken and you need to manually exclude a few directories to use it and there is a mystical error about a missing method on truth.MapSubject, but at least now it fails in a less spectacular way.
Fixes #796.
--
MOS_MIGRATED_REVID=112458948
|
|
|
|
|
|
|
|
|
|
|
| |
With latest change to the bootstrap compilation, some options
were wrongly moved around.
Tested with `source scripts/ci/build.sh; bazel_build` for JAVA_VERSION
1.7 and 1.8.
--
MOS_MIGRATED_REVID=112409496
|
|
|
|
|
|
|
| |
This remove all C++ compilation in bootstrapping itself.
--
MOS_MIGRATED_REVID=112407516
|
|
|
|
|
|
|
| |
Another step towards no C++ compilation outside of Bazel for bootstrapping.
--
MOS_MIGRATED_REVID=112399835
|
|
|
|
|
|
|
|
| |
process-wrapper is a C++ tool, replacing by a dummy shell version to
reduce the number of C++ compiler calls during the bootstrap process.
--
MOS_MIGRATED_REVID=112398152
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=112397181
|
|
|
|
|
|
|
| |
This method is used only once now so inline.
--
MOS_MIGRATED_REVID=112396086
|
|
|
|
|
|
|
|
| |
This simplify the bootstrap process and remove a C++ tool from
the bootstrap binary.
--
MOS_MIGRATED_REVID=112394555
|
|
|
|
|
|
|
|
|
| |
Instead use the current tools from the repository. This
simplifies the bootstrap binary and is a step towards
no C++ compilation for the bootstrap binary.
--
MOS_MIGRATED_REVID=112240524
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=112239696
|
|
|
|
|
|
|
|
|
| |
Using java8-jdk will match both openjdk-8-jdk and oracle-java8-jdk similarly with java7-jdk. If
either of them are installed on the system allowing bazel to be installed in either case. Also the jdk source is not needed to create the bazel deb pacakge.
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/785
MOS_MIGRATED_REVID=112132285
|
|
|
|
|
|
|
|
|
|
| |
0.1.3 was transformed into 0.1.~3 instead of 0.1.3
Tested the regexp manually with:
echo 0.1.3 | sed ...
echo 0.1.3rc3 | sed ...
--
MOS_MIGRATED_REVID=111884245
|
|
|
|
|
|
|
|
|
| |
The installer was missing the StdRedirect.dylib after a recent change.
This was breaking the Tutorial job on ci.bazel.io (and iOS builds using
that installer).
--
MOS_MIGRATED_REVID=111855462
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Roll-forward with fix
*** Original change description ***
Automated [] rollback of [].
*** Reason for rollback ***
Broke tests on Mac: https://google.com/url?sa=D&q=http%3A%2F%2Fci.bazel.io%2Fjob%2FBazel%2FJAVA_VERSION%3D1.8%2CPLATFORM_NAME%3Ddarwin-x86_64%2F269%2Fconsole
*** Original change description ***
Speed-up bootstrap on OS X by removing tool compilation.
--
MOS_MIGRATED_REVID=111833617
|
|
|
|
|
|
|
| |
For bazel on non-darwin architectures, this will simply be a stub, and should never be invoked. On darwin arcitectures, the tool will map xcode version to xcode path on the host system.
--
MOS_MIGRATED_REVID=111651147
|
|
|
|
|
|
|
| |
bazel server exits abruptly.
--
MOS_MIGRATED_REVID=111641619
|
|
|
|
|
|
|
| |
Fixes #752.
--
MOS_MIGRATED_REVID=111614541
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Broke tests on Mac: https://google.com/url?sa=D&q=http%3A%2F%2Fci.bazel.io%2Fjob%2FBazel%2FJAVA_VERSION%3D1.8%2CPLATFORM_NAME%3Ddarwin-x86_64%2F269%2Fconsole
*** Original change description ***
Speed-up bootstrap on OS X by removing tool compilation.
--
MOS_MIGRATED_REVID=110785493
|