The well-known types generate C code into wkt.inc, and this C code was
not testing isset($msg->submsg_field) like the generated code does:
```php
// PHP generated getter: checks isset().
public function getOptions()
{
return isset($this->options) ? $this->options : null;
}
```
```c
// C generated getter, does not check upb_msg_has()
static PHP_METHOD(google_protobuf_Value, getListValue) {
Message* intern = (Message*)Z_OBJ_P(getThis());
const upb_fielddef *f = upb_msgdef_ntofz(intern->desc->msgdef,
"list_value");
zval ret;
Message_get(intern, f, &ret);
RETURN_COPY_VALUE(&ret);
}
```
This led to an error where we wnuld try to get a sub-message field from upb
when it `upb_msg_has(msg, field) == false`, which is an error according to upb.
There are two possible fixes for this bug. A guiding principle is that we want
the generated C code in wkt.inc to have the same behavior as PHP generated
code. Following this principle, the two possible fixes are:
1. Change the code generator for wkt.inc to check upb_msg_has(f) before
calling Message_get(). This would match the isset() check that the
The PHP generated code does, and we would leave the PHP code unchanged.
2. Change Message_get() to itself perform the upb_msg_has(f) check for
sub-message fields. This means that generated code would no longer need
to perform an isset() check, so we would want to remove this check from
the PHP generated code also to avoid a redundant check.
Both of these are reasonable fixes, and it is not immediately obvious which is
better. (1) has the benefit of resolving this case when we are in more
specialized code (a getter function that already knows this is a sub-message
field), and therefore avoids performing the check later in more generic code
that would have to test the type again. On the other hand, the isset() check is
not needed for the pure PHP implementation, as an unset PHP variable will
return `null` anyway. And for the C extension, we'd rather check upb_msg_has()
at the C level instead of PHP.
So this change implements (2). The generated code in wkt.inc remains unchanged,
and the PHP generated code for sub-message fields is changed to remove the
isset() check.
* Some more updates to PHP testing infrastructure (#8576)
* WIP.
* Added build config for all of the tests.
* Use ../src/protoc if it is available, for cases where Bazel isn't available.
* Added test_php.sh.
* Fix for the broken macOS tests.
* Move all jobs to use php80 instead of lots of separate jobs.
* Only pass -t flag if we are running in a terminal.
* Updated php_all job to use new Docker stuff.
* Fixed PHP memory leaks and arginfo errors (#8614)
* Fixed a bunch of incorrect arginfo and a few incorrect error messages.
* Passes mem check test with no leaks!
* WIP.
* Fix build warning that was causing Bazel build to fail.
* Added compatibility code for PHP <8.0.
* Added test_valgrind target and made tests Valgrind-clean.
* Updated Valgrind test to fail if memory leaks are detected.
* Removed intermediate shell script so commands are easier to cut, paste, and modify.
* Passing all Valgrind tests!
* Hoist addref into ObjCache_Get().
* Removed special case of map descriptors by keying object map on upb_msgdef.
* Removed all remaining RETURN_ZVAL() macros.
* Removed all explicit reference add/del operations.
* Added REFCOUNTING.md to Makefile.am.
* Updated upb version and fixed PHP to not get unset message field. (#8621)
* Updated upb version and fixed PHP to not get unset message field.
* Updated changelog.
* Fixed preproc test to handle old versions of Clang withot __has_attribute().
* A second try at fixing __has_attribute().
* Copy __has_attribute() fix to cc file also.
* Updated failure list for PHP for fixed test.
* Updated version of upb for Ruby (#8624)
* Updated upb.
* Preserve legacy behavior for unset messages.
* Updated failure list.
* Updated CHANGES.txt.
* Added erroneously-deleted test file.
* Fixed condition on compatibility code.
* Re-introduced deleted file again, and fixed Rakefile to not delete it.
* Fix generation of test protos.
* Updated upb version and fixed PHP to not get unset message field.
* Updated changelog.
* Fixed preproc test to handle old versions of Clang withot __has_attribute().
* A second try at fixing __has_attribute().
* Copy __has_attribute() fix to cc file also.
* Updated failure list for PHP for fixed test.
* [PHP] Fixed $msg->setMessage(null) to properly clear the message.
Fixes: https://github.com/protocolbuffers/protobuf/issues/8457
* Changed pure PHP to throw TypeError, and added a test for null.
* Added more tests and fixed null in setter for oneof.
* Fixed a bunch of incorrect arginfo and a few incorrect error messages.
* Passes mem check test with no leaks!
* WIP.
* Fix build warning that was causing Bazel build to fail.
* Added compatibility code for PHP <8.0.
* Added test_valgrind target and made tests Valgrind-clean.
* Updated Valgrind test to fail if memory leaks are detected.
* Removed intermediate shell script so commands are easier to cut, paste, and modify.
* Passing all Valgrind tests!
* Hoist addref into ObjCache_Get().
* Removed special case of map descriptors by keying object map on upb_msgdef.
* Removed all remaining RETURN_ZVAL() macros.
* Removed all explicit reference add/del operations.
* Added REFCOUNTING.md to Makefile.am.
* WIP.
* Added build config for all of the tests.
* Use ../src/protoc if it is available, for cases where Bazel isn't available.
* Added test_php.sh.
* Fix for the broken macOS tests.
* Move all jobs to use php80 instead of lots of separate jobs.
* Only pass -t flag if we are running in a terminal.
* Updated php_all job to use new Docker stuff.
* Simplified PHP testing setup.
- Consolidated on a single autoloader, created by composer.
- Consolidated on a single phpunit invocation strategy: we run
phpunit on a directory, which will run all tests matching *Test.php
in that directory.
- We now rely on autoloading to import all test protos. require_once()
calls for test protos are removed.
- For now the valgrind tests are removed. A follow-up PR will re-enable
them in a more robust way.
* More improvements to PHP testing.
1. Replace custom PHPUnit-selection logic in test.sh with generic
composer version selection.
2. Optimized both test proto generation and the custom extension
build to avoid unnecessary work when the files are already up
to date.
* Added assertions to verify that the C test doesn't use PHP sources.
* Updated tests.sh for the new PHP testing commands.
* Removed obsolete rules from tests.sh.
* Fixed generate_test_protos.sh for when tmp does not exist.
Also removed undefined_test.php and fixed Makefile.am.
* Added php8.0_all again which is still used.
* Added missing file to Makefile.am.
* Re-added php_all_32 rule which is also still used.
* Updated testing commands for macOS and download composer.
* Use /usr/local/bin on mac instead of /usr/bin, since the latter is not writable.
Message accessors will return null when when the field is not
set, so this should be reflected in the PhpDoc.
Also updated the code generator for the well-known types to reflect
the edits made in https://github.com/protocolbuffers/protobuf/pull/8105.
Also explicitly check for upb_msg_has() in the oneof accessor, so
we are not implicitly relying on unset message fields returning NULL
at the upb level.
* Port for php8
* Port php c extension for php8
* Update composer.json
* Drop php7.0 support
* Update phpunit for php7.1 in c extension test
* Add back support for php7.0
* Add badge for php8 continuous build
The param was typed as `object` which is invalid as when items are set in the container the key passes through `checkKey` where it enforces the key must be various classes of integer, bool or string. Thus I have replaced it with a union type (in future this can be a proper PHP level typehint, but for now only doctype)
This commit updates protobuf.c to refer to the PHP version macro so that
it stays up to date. I also deleted the notes element near the top of
package.xml. This seems to be a global notes field rather than a
per-release field, so it was showing up on new releases where it doesn't
make sense.
Fixes#8017.
* Only ported c extension to php8.
* Didn't fixed the issue of throwing warnings for missing arginfo in bundled files.
* Tests not fixed, because syntax of phpunit (<7 vs >9.3) are not compatible.
* In next release, needs to drop php5 and php7.0 support (in order to use phpunit > 7)
* Fix a typo
* Fix lots of spelling errors
* Fix a few more spelling mistakes
* s/parsable/parseable/
* Don't touch the third party files
* Cloneable is the preferred C# term
* Copyable is the preferred C++ term
* Revert "s/parsable/parseable/"
This reverts commit 534ecf7675.
* Revert unparseable->unparsable corrections
* Updated upb to pick up several bugfixes. This fixes most conformance failures.
* Updated golden file, now that we are serializing as packed by default.
* Set "packed" properly: it is on by default for packable fields in proto3.
* Updated failure list for PHP now that we properly respect "packed".
* Temporarily disable encode_decode_test
Co-authored-by: Joshua Haberman <jhaberman@gmail.com>
* Removed references to legacy class names from generated code.
* Removed old PHP extension in favor of new extension. This drops PHP5 compat.
* Updated failure lists for things fixed by the new extension.
* Updated Makefile.am for new file list.
* Fix distcheck.
* Added code for the new PHP extension.
* Removed a bunch of unused variables shown by compiler warnings.
* Test protobuf2 in the tests.
* Updated upb version to fix a goldenfile test.
* Added pure-PHP changes. Extension now passes all tests.
* Enabled protobuf2 for all C extension tests.
* Fixed pure=PHP lib: full names no longer start with '.'.
* Added files for new extension to Makefile.am.
* Downgraded make-preload.py to python 2, since python3 isn't available in the php_all Kokoro image.
* Disable tests of new C extension with PHP 5.x.
* Also do not compile the extension for PHP5.
* Accept version 5.*.*, and use /usr/bin/python.
* Addressed PR comments.
* Addressed PR comments.
* Added "const" to a parameter and fixed a memory leak seen in Valgrind.
* Stop testing the C extension for PHP5.
The next release of protobuf will deprecate the C extension
for PHP5, see:
https://github.com/protocolbuffers/protobuf/issues/7525
* Made the PHP5.6 Mac test only exercise pure-PHP.
* Build protoc for PHP 5.6 test.
* Rewrote bundling script in PHP to avoid dependency on Python.
* A few more fixes.
* Fixed int32/int64 behavior for 32-bit builds.
* Match more PHP versions in testing script.
* Use phpize --clean before building extension.
* Force-delete configure.in to avoid phpize problems cross-version.
* Delete both configure.ac and configure.in.
PHP does not have a long type, and IDEs don't necessarily understand that long is the same as integer.
E.g. PhpStorm will give an error like this: `Expected parameter of type '\Google\Protobuf\Internal\long', 'int' provided`
* De-duplicated code to generate PHP protos and install phpunit.
* Removed all references to generate_php_test_proto.
* Replaced with internal_build_cpp.
* Make Timestamp::__construct() static to avoid conflicts with MongoDB.
* Replicated PHPUnit versions and added new script to Makefile.am.
* Fixed filename in Makefile.am.
* Disabled test that SEGV's on macOS.
* Removed extraneous "set -e".
* Make sure generate_protos.sh happens on every test path.
* Removed stray '$' chars.
* Added proper support for aggregate_metadata tests. But now I get a stack overflow.
Stack overflow:
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
/Users/haberman/code/protobuf/php/tests/generated/GPBMetadata/Proto/TestDescriptors.php:16
namespace GPBMetadata\Proto;
class TestDescriptors
{
public static $is_initialized = false;
public static function initOnce() {
$pool = \Google\Protobuf\Internal\DescriptorPool::getGeneratedPool();
if (static::$is_initialized == true) {
return;
}
\GPBMetadata\Proto\TestDescriptors::initOnce();
$pool->internalAddGeneratedFile(hex2bin(
""
), true);
static::$is_initialized = true;
}
}
* Fixed and verified metadata aggregation testing.
* Simplified PHP testing scripts.
- allow them to be run from any directory.
- remove "VERSION" arg, we can get PHP from $PATH.
* Added a few places I missed.
* Removed redundant "cd `dirname $0`".
Also replaced all backticks with $(), for consistency.
* Set execute bit on files if and only if they begin with (#!).
Git only tracks the 'x' (executable) bit on each file. Prior to this
CL, our files were a random mix of executable and non-executable.
This change imposes some order by making files executable if and only
if they have shebang (#!) lines at the beginning.
We don't have any executable binaries checked into the repo, so
we shouldn't need to worry about that case.
* Added fix_permissions.sh script to set +x iff a file begins with (#!).
Instead of calling initOnce of dependencies, initialize metadata of dependencies in the same file.
Needs to pass aggregate_metadata option to protoc to trigger, e.g.:
--php_out=aggregate_metadata=foo#bar:generated_dir
For each input file, transitive dependencies (including itself), whose package name has the prefix of foo or bar, will be aggregated, in which their metadata string will be aggregated in the same internalAddGeneratedFile call. For other dependencies, initOnce is called as before.
This feature is EXPERIMENTAL. DO NOT USE!!!
Instead of calling initOnce of dependencies, initialize metadata of dependencies in the same file.
Needs to pass aggregate_metadata option to protoc to trigger, e.g.:
--php_out=aggregate_metadata=foo#bar:generated_dir
For each input file, transitive dependencies (including itself), whose package name has the prefix of foo or bar, will be aggregated, in which their metadata string will be aggregated in the same internalAddGeneratedFile call. For other dependencies, initOnce is called as before.
This feature is EXPERIMENTAL. DO NOT USE!!!
