walkDeps performs a pre-order DFS (unlike visitDepsDepthFirst which is
a post-order DFS). The visit function takes in both a parent and child
node and returns a bool indicating if the child node should be
traversed.
This removes the need to use $OLDPWD when running tests, which means
that the builddir may be an absolute or relative directory. It also
filters out the "PASS" message on successful test runs to clean up our
output.
Change-Id: I4ab937c7a87b74fe997a47cc0311e2f357f9f7e9
It's difficult for wrapping scripts to handle -b properly. Just pass
BUILDDIR instead, which is easier to handle. This still accepts -b, so
that incremental builds work across this change.
* commit 'ae9aa9216e1b08df5c3a6cfec97b3dae9a94afbc':
Implement plugins for bootstrap go modules
Fix bootstrap and separate buildDir ninja cleanup
context: Changed singletons to execute in order.
bootstrap: Added phony rule for bootstrapManifest.
Add new line to fatal error messages in bootstrap
bpdoc: Cache property struct docs by full package path
Now that we have multi-stage bootstrapping, we can make the primary
builder build more dynamic. Add the concept of plugins that will be
linked and loaded into bootstrap_go_binary or bootstrap_go_package
modules. It's expected that the plugin's init() functions will do
whatever registration is necessary.
Example Blueprint definition:
bootstrap_go_binary {
name: "builder",
...
}
bootstrap_go_package {
name: "plugin1",
pluginFor: ["builder"],
}
A package may specify more than one plugin if it will be inserted into
more than one go module.
Change-Id: I109835f444196b66fc4018c3fa36ba0875823184
With the introduction of $buildDir in bootstrapDir, the ninja cleanup
step hasn't been able to cleanup during the bootstrap stages. The main
stage was unaffected, as long as you were using "." as your buildDir.
Change-Id: I277dd7864989f9052d96cab9ce377548a1391a80
The full package path is no longer stored in the name variable, pass the
pkg variable into getDocs and putDocs so that types with the same name
in different packages don't collide.
Change-Id: I7ecd98815487467dbaadc8a318fa43b123fc83c1
* commit '150e16eab079358d95097f078c975d1b5933918f': (162 commits)
Support go 1.5
Fix bootstrap package build stage default
Support a separate build directory
Switch bootstrap.BinDir to a StaticVariable
Use three stage builds
Enhance bootstrap stage selection
Add test suite for bootstrapping
Switch to custom-built ninja v1.6.0
Migrate to container-based travis infrastructure
Fix running travis on forked repos
Require ninja 1.6.0, remove workaround
Don't print module header if nothing is defined
Fix a sub-Blueprints deps bug.
Add deep Copy() for parser types
Refactor parsing to allow reuse
Add convenience methods to Comment objects
Add self-documenting support
Support filtering nested structures by tag
Add option to build and run tests during bootstrap
Create internal bootstrap Config struct
...
The go compiler and linker changed in v1.5 -- to 'go tool compile' and
'go tool link' instead of 6g and 6l. Move the selection logic to
bootstrap.bash, and have it use compile/link if 6g/6l are missing. This
way the build.ninja.in will continue working with either go 1.4 or 1.5.
Travis and the test suite will fail under 1.5, since the build.ninja.in
is still generated with 1.4, and the function names in the comments
differ between 1.4 and 1.5.
The Go race detector found a race condition in the parser, which
highlighted a few related bugs. A variable could be defined but
not referenced in a Blueprints file, then appended to in multiple
subdirs= Blueprints files. The race detector caught the multiple
writes to assignment.Referenced from the parsers for the subdirs
Blueprints files, but multiple appends would be much more serious.
To fix this, keep local and inherited variables separate in the
Scope object and export that info to the parser. Disallow
appending to non-local variables, which was already the intended
behavior. Only update the referenced boolean for local variables.
Together, this should prevent all writes to Assignment objects
from parsers other than the one that created them.
Also improves the error handling code and some error messages.
Change-Id: Idb4f7d2e61bbe28d90b93074764e64e60d1eba8f
To provide a consistent __FILE__ behavior with cpp, we want to be able
to run with SRCDIR="." and the outputs be saved elsewhere. Other tools
within android also expect to be run from $TOP.
Change-Id: I572bce5c9086b0c3310b42065ae98cbf5a1c6399
API Change -- will require changes to any code using bootstrap.BinDir
This way we can put references to other variables in BinDir, and still
be used properly by other packages.
Change-Id: I497424cb254b3a170401ac9420fa0adbf8d11d1e
This splits the current bootstrap stage into two stages:
A bootstrap stage, which like today, a reference is checked into the
tree. It just builds the "core" blueprint binaries -- minibp,
gotestmain, and choosestage. Just enough to build the next stage's ninja
file.
A primary builder stage. This builds the primary builder, the main ninja
file, and any other bootstrap binaries (bpfmt, etc).
The main advantage here is that the checked in file really only contains
references to blueprint -- not the primary builder. This will allow us
to make the primary builder more dynamic, by loading more module types
that may or may not exist in all trees.
It's even possible to reuse the build.ninja.in in the blueprint repo
directly now. We don't currently do that, since we still want to turn on
tests.
Change-Id: I18683891ed7348b0d7af93084e3a68a04fbd5dbc
This simplifies the bootstrap process while making it more flexible by
moving the stage selection into a go binary(choosestage). It will now be
possible to have more than two build stages.
Now each stage has a ninja template(main.ninja.in) and a timestamp
file(main.ninja.in.timestamp). The timestamp file may be updated by any
build stage that wishes to regenerate the ninja template. If the
choosestage binaries sees that the timestamp is newer than the template,
it will choose the prior stage.
The main stage no longer writes to the source tree to update the
build.ninja.in file. This was a problem for read-only source trees.
Instead, the choosestage binary first checks to see if that file is
newer than the last bootstrap.ninja.in, copies it in place, and starts
the boostrap stage.
The bootstrap stage regenerates it's own ninja template, but that
required a loop through the main stage to actually run it. The
choosestage binary now detects if the template has changed for the
current stage, and will restart the stage.
One change is that if dependencies do get messed up, instead of silently
failing, there's a higher chance that the bootstrap step will just
continue looping, doing nothing. This can happen if the main stage
has a dependency that triggers the bootstrap stage, but the bootstrap
stage doesn't see anything required to rebuild the main ninja file. A
side effect of this requirement is that changes to test code will now
rebuild the main ninja file.
Change-Id: I9965cfba79dc0dbbd3af05f5944f7653054455a2
This only checks to make sure that for a given updated file, the
bootstrap stage is properly run. It doesn't actually check to make sure
that anything was rebuilt.
Change-Id: I9cb6ff1d483264da30e43d5580361d93b148f42c
The workaround no longer works with the new stage selector. We may run
bootstrap.ninja.in twice before running the next stage, but we can't
encode whether to run another ninja instance in the checked in
build.ninja.in.
This can likely be solved, but now that there's an official release with
support for multiple passes, just push up the required version.
Change-Id: I76e321912e323d60e462aabec61bdfcc7118cd5e
Modules may decide not to output any build definitions in some cases.
Clean up the ninja file by not adding headers for empty sections.
One particular usecase is my upcoming multi-stage bootstrapping -
bootstrap_go_* will not output any rules in the first stage unless it's
required to build the primary builder.
Change-Id: I6a6b54da7e1702c63bfa736bcf8daf16956f9449
This change fixes a bug where directories that could contain a Blueprints file
but currently do not were not being added as a dependency. This meant that if
a Blueprints file were added to that directory it would not automatically get
picked up by the build.
