Project History

Work on PyOxidizer started in November 2018 by Gregory Szorc.

Blog Posts

Version History

0.8.0

Released October 12, 2020.

Backwards Compatibility Notes

  • The default Python distributions have been upgraded to CPython 3.8.6 (from 3.7.7) and support for Python 3.7 has been removed.

  • On Windows, the default_python_distribution() Starlark function now defaults to returning a standalone_dynamic distribution variant, meaning that it picks a distribution that can load standalone .pyd Python extension modules by default.

  • The standalone Python distributions are now validated to be at least version 5 of the distribution format. If you are using the default Python distributions, this change should not affect you.

  • Support for packaging the official Windows embeddable Python distributions has been removed. This support was experimental. The official Windows embeddable distributions are missing critical support files that make them difficult to integrate with PyOxidizer.

  • The pyembed crate now defines a new OxidizedPythonInterpreterConfig type to configure Python interpreters. The legacy PythonConfig type has been removed.

  • Various run_* functions on pyembed::MainPythonInterpreter have been moved to standalone functions in the pyembed crate. The run_as_main() function (which is called by the default Rust program that is generated) will always call Py_RunMain() and finalize the interpreter. See the extensive crate docs for move.

  • Python resources data in the pyembed crate is no longer annotated with the 'static lifetime. Instances of PythonConfig and OxidizedPythonInterpreterConfig must now be annotated with a lifetime for the resources data they hold such that Rust lifetimes can be enforced.

  • The type of the custom Python importer has been renamed from PyOxidizerFinder to OxidizedFinder.

  • The name of the module providing our custom importer has been renamed from _pyoxidizer_importer to oxidized_importer.

  • Minimum Rust version changed from 1.36 to 1.40 to allow for upgrading various dependencies to modern versions.

  • Windows static extension building is possibly broken due to changes to distutils. However, since we changed the default configuration to not use this build mode, we’ve deemed this potential regression acceptable for the 0.8 release. If it exists, it will hopefully be fixed in the 0.9 release.

  • The pip_install(), read_package_root(), read_virtualenv() and setup_py_install() methods of the PythonDistribution Starlark type have been moved to the PythonExecutable type. Existing Starlark config files will need to change references accordingly (often by replacing dist. with exe.).

  • The PythonDistribution.extension_modules() Starlark function no longer accepts arguments filter and preferred_variants. The function now returns every extension in the distribution. The reasons for this change were to make code simpler and the justification for removing it was rather weak. Please file an issue if this feature loss affects you.

  • The PythonInterpreterConfig Starlark type now interally has most of its fields defined to None by default instead of their default values.

  • The following Starlark methods have been renamed: PythonExecutable.add_module_source() -> PythonExecutable.add_python_module_source(); PythonExecutable.add_module_bytecode() -> PythonExecutable.add_python_module_bytecode(); PythonExecutable.add_package_resource() -> PythonExecutable.add_python_package_resource(); PythonExecutable.add_package_distribution_resource() -> PythonExecutable.add_python_package_distribution_resource(); PythonExecutable.add_extension_module() -> PythonExecutable.add_python_extension_module().

  • The location-specific Starlark methods for adding Python resources have been removed. The functionality can be duplicated by modifying the add_location and add_location_fallback attributes on Python resource types. The following methods were removed: PythonExecutable.add_in_memory_module_source(); PythonExecutable.add_filesystem_relative_module_source(), PythonExecutable.add_in_memory_module_bytecode(); PythonExecutable.add_filesystem_relative_module_bytecode(); PythonExecutable.add_in_memory_package_resource(); PythonExecutable.add_filesystem_relative_package_resource(); PythonExecutable.add_in_memory_package_distribution_resource() PythonExecutable.add_filesystem_relative_package_distribution_resource(); PythonExecutable.add_in_memory_extension_module(); PythonExecutable.add_filesystem_relative_extension_module(); PythonExecutable.add_in_memory_python_resource(); PythonExecutable.add_filesystem_relative_python_resource(); PythonExecutable.add_in_memory_python_resources(); PythonExecutable.add_filesystem_relative_python_resources().

  • The Starlark PythonDistribution.to_python_executable() method no longer accepts the arguments extension_module_filter, preferred_extension_module_variants, include_sources, include_resources, and include_test. All of this functionality has been replaced by the optional packaging_policy, which accepts a PythonPackagingPolicy instance. The new type represents all settings influencing executable building and control over resources added to the executable.

  • The Starlark type PythonBytecodeModule has been removed. Previously, this type was internally a request to convert Python module source into bytecode. The introduction of PythonPackagingPolicy and underlying abilities to derive bytecode from a Python source module instance when adding that resource type rendered this Starlark type redundant. There may still be the need for a Starlark type to represent actual Python module bytecode (not derived from source code at build/packaging time). However, this functionality did not exist before so the loss of this type is not a loss in functionality.

  • The Starlark methods PythonExecutable.add_python_resource() and PythonExecutable.add_python_resources() no longer accept the arguments add_source_module, add_bytecode_module, and optimize_level. Instead, set various add_* attributes on resource instances being passed into the methods to influence what happens when they are added.

  • The Starlark methods PythonExecutable.add_python_module_source(), PythonExecutable.add_python_module_bytecode(), PythonExecutable.add_python_package_resource(), PythonExecutable.add_python_package_distribution_resource(), and PythonExecutable.add_python_extension_module() have been removed. The remaining PythonExecutable.add_python_resource() and PythonExecutable.add_python_resources() methods are capable of handling all resource types and should be used. Previous functionality available via argument passing on these methods can be accomplished by setting add_* attributes on individual Python resource objects.

  • The Starlark type PythonSourceModule has been renamed to PythonModuleSource.

  • Serialized Python resources no longer rely on the flavor field to influence how they are loaded at run-time. Instead, the new is_* fields expressing individual type affinity are used. The flavor attributes from the OxidizedResource Python type has been removed since it does nothing.

  • The packed resources data format version has been changed from 1 to 2. The parser has dropped support for reading version 1 files. Packed resources blobs will need to be written and read by the same version of the Rust crate to be compatible.

  • The autogenerated Rust file containing the Python interpreter configuration now emits a pyembed::OxidizedPythonInterpreterConfig instance instead of pyembed::PythonConfig. The new type is more powerful and what is actually used to initialize an embedded Python interpreter.

  • The concept of a resources policy in Starlark has now largely been replaced by attributes denoting valid locations for resources.

  • oxidized_importer.OxidizedResourceCollector.__init__() now

    accepts an allowed_locations argument instead of policy.

  • The PythonInterpreterConfig() constructor has been removed. Instances of this Starlark type are now created via PythonDistribution.make_python_interpreter_config(). In addition, instances are mutated by setting attributes rather than passing perhaps dozens of arguments to a constructor function.

  • The default build configuration for Windows no longer forces extension modules to be loaded from memory and materializes some extension modules as standalone files. This was done because some some extension modules weren’t working when loaded from memory and the configuration caused lots of problems in the wild. The new default should be much more user friendly. To use the old settings, construct a custom PythonPackagingPolicy and set allow_in_memory_shared_library_loading = True and resources_location_fallback = None.

New Features

  • Python distributions upgraded to CPython 3.8.6.

  • CPython 3.9 distributions are now supported by passing python_version="3.9" to the default_python_distribution() Starlark function. CPython 3.8 is the default distribution version.

  • Embedded Python interpreters are now managed via the new apis defined by PEP-587. This gives us much more control over the configuration of interpreters.

  • A FileManifest Starlark instance will now have its default pyoxidizer run executable set to the last added Python executable. Previously, it would only have a run target if there was a single executable file in the FileManifest. If there were multiple executables or executable files (such as Python extension modules) a run target would not be available and pyoxidizer run would do nothing.

  • Default Python distributions upgraded to version 5 of the standalone distribution format. This new format advertises much more metadata about the distribution, enabling PyOxidizer to take fewer guesses about how the distribution works and will help enable more features over time.

  • The pyembed crate now exposes a new OxidizedPythonInterpreterConfig type (and associated types) allowing configuration of every field supported by Python’s interpreter configuration API.

  • Resources data loaded by the pyembed crate can now have a non-'static lifetime. This means that resources data can be more dynamically obtained (e.g. by reading a file). PyOxidizer does not yet support such mechanisms, however.

  • OxidizedFinder instances can now be constructed from Python code. This means that a Python application can instantiate and install its own oxidized module importer.

  • The resources indexed by OxidizedFinder instances are now representable to Python code as OxidizedResource instances. These types can be created, queried, and mutated by Python code. See OxidizedResource for the API.

  • OxidizedFinder instances can now have custom OxidizedResource instances registered against them. This means Python code can collect its own Python modules and register them with the importer. See add_resource(self, resource: OxidizedResource) for more.

  • OxidizedFinder instances can now serialize indexed resources out to a bytes. The serialized data can be loaded into a separate OxidizedFinder instance, perhaps in a different process. This facility enables the creation and reuse of packed resources data structures without having to use pyoxidizer to collect Python resources data.

  • The types returned by OxidizedFinder.find_distributions() now implement entry_points, allowing entry points to be discovered.

  • The types returned by OxidizedFinder.find_distributions() now implement requires, allowing package requirements to be discovered.

  • OxidizedFinder is now able to load Python modules when only source code is provided. Previously, it required that bytecode be available.

  • OxidizedFinder now implements iter_modules(). This enables pkgutil.iter_modules() to return modules serviced by OxidizedFinder.

  • The PythonModuleSource Starlark type now exposes module source code via the source attribute.

  • The PythonExecutable Starlark type now has a make_python_module_source() method to allow construction of PythonModuleSource instances.

  • The PythonModuleSource Starlark type now has attributes add_include, add_location, add_location_fallback, add_source, add_bytecode_optimization_level_zero, add_bytecode_optimization_level_one, and add_bytecode_optimization_level_two to influence what happens when instances are added to to binaries.

  • The Starlark methods for adding Python resources now accept an optional location argument for controlling the load location of the resource. This functionality replaces the prior functionality provided by location-specific APIs such as PythonExecutable.add_in_memory_python_resource(). The following methods gained this argument: PythonExecutable.add_python_module_source(); PythonExecutable.add_python_module_bytecode(); PythonExecutable.add_python_package_resource(); PythonExecutable.add_python_package_distribution_resource(); PythonExecutable.add_python_extension_module(); PythonExecutable.add_python_resource(); PythonExecutable.add_python_resources().

  • Starlark now has a PythonPackagingPolicy type to represent the collection of settings influencing how Python resources are packaged into binaries.

  • The PythonDistribution Starlark type has gained a make_packaging_policy() method for obtaining the default PythonPackagingPolicy for that distribution.

  • The PythonPackagingPolicy.register_resource_callback() method can be used to register a Starlark function that will be called whenever resources are created. The callback allows a single function to inspect and manipulate resources as they are created.

  • Starlark types representing Python resources now expose an is_stdlib attribute denoting whether they came from the Python distribution.

  • The new PythonExecutable.pip_download() method will run pip download to obtain Python wheels for the requested package(s). Those wheels will then be parsed for Python resources, which can be added to the executable.

  • The Starlark function default_python_distribution() now accepts a python_version argument to control the X.Y version of Python to use.

  • The PythonPackagingPolicy Starlark type now exposes a flag to control whether shared libraries can be loaded from memory.

  • The PythonDistribution Starlark type now has a make_python_interpreter_config() method to obtain instances of PythonInterpreterConfig that are appropriate for that distribution.

  • PythonInterpreterConfig Starlark types now expose attributes to query and mutate state. Nearly every setting exposed by Python’s initialization API can be set.

Bug Fixes

  • Fixed potential process crash due to illegal memory access when loading Python bytecode modules from the filesystem.

  • Detection of Python bytecode files based on registered suffixes and cache tags is now more robust. Before, it was possible for modules to get picked up having the cache tag (e.g. cpython-38) in the module name.

  • In the custom Python importer, read_text() of distributions returned from find_distributions() now returns None on unknown file instead of raising IOError. This matches the behavior of importlib.metadata.

  • The pyembed Rust project build script now reruns when the source Starlark file changes.

  • Some Python resource types were improperly installed in the wrong relative directory. The buggy behavior has been fixed.

  • Python extension modules and their shared library dependencies loaded from the filesystem should no longer have the library file suffix stripped when materialized on the filesystem.

  • On Windows, the sqlite module can now be imported. Before, the system for serializing resources thought that sqlite was a shared library and not a Python module.

  • The build script of the pyoxidizer crate now uses the git2 crate to try to resolve the Git commit instead of relying on the git command. This should result in fewer cases where the commit was being identified as unknown.

  • $ORIGIN is properly expanded in sys.path. (This was a regression during the development of version 0.8 and is not a regression from the 0.7 release.)

Other Relevant Changes

  • The registration of the custom Python importer during interpreter initialization no longer relies on running custom frozen bytecode for the importlib._bootstrap_external Python module. This simplifies packaging and interpreter configuration a bit.

  • Packaging documentation now gives more examples on how to use available Starlark packaging methods.

  • The modified distutils files used when building statically linked extensions have been upgraded to those based on Python 3.8.3.

  • The default pyoxidizer.bzl now has comments for the packaging_policy argument to PythonDistribution.to_python_executable().

  • The default pyoxidizer.bzl now uses add_python_resources() instead of add_in_memory_python_resources().

  • The Rust Starlark crate was upgraded from version 0.2 to 0.3. There were numerous changes as part of this upgrade. While we think behavior should be mostly backwards compatible, there may be some slight changes in behavior. Please file issues if any odd behavior or regressions are observed.

  • The configuration documentation was reorganized. The unified document for the complete API document (which was the largest single document) has been split into multiple documents.

  • The serialized data structure for representing Python resources metadata and its data now allows resources to identify as multiple types. For example, a single resource can contain both Python module source/bytecode and a shared library.

  • pyoxidizer --version now prints verbose information about where PyOxidizer was installed, what Git commit was used, and how the pyembed crate will be referenced. This should make it easier to help debug installation issues.

  • The autogenerated/default Starlark configuration file now uses the install target as the default build/run target. This allows extra files required by generated binaries to be available and for built binaries to be usable.

0.7.0

Released April 9, 2020.

Backwards Compatibility Notes

  • Packages imported from memory using PyOxidizer now set __path__ with a value formed by joining the current executable’s path with the package name. This mimics the behavior of zipimport.

  • Resolved Python resource names have changed behavior. See the note in the bug fixes section below.

  • The PythonDistribution.to_python_executable() Starlark method has added a packaging_policy named argument as its 2nd argument / 1st named argument. If you were affected by this, you should add argument names to all arguments passed to this method.

  • The default Rust project for built executables now builds executables such that dynamic symbols are exported from the executable. This change is necessary in order to support executables loading Python extension modules, which are shared libraries which need access to Python symbols defined in executables.

  • The PythonResourceData Starlark type has been renamed to PythonPackageResource.

  • The PythonDistribution.resources_data() Starlark method has been renamed to PythonDistribution.package_resources().

  • The PythonExecutable.to_embedded_data() Starlark method has been renamed to PythonExecutable.to_embedded_resources().

  • The PythonEmbeddedData Starlark type has been renamed to PythonEmbeddedResources.

  • The format of Python resource data embedded in binaries has been completely rewritten. The separate modules and resource data structures have been merged into a single data structure. Embedded resources data can now express more primitives such as package distribution metadata and different bytecode optimization levels.

  • The pyembed crate now has a dev dependency on the pyoxidizer crate in order to run tests.

Bug Fixes

  • PyOxidizer’s importer now always sets __path__ on imported packages in accordance with Python’s stated behavior (#51).

  • The mechanism for resolving Python resource files from the filesystem has been rewritten. Before, it was possible for files like package/resources/foo.txt to be normalized to a (package, resource_name) tuple of (package, resources.foo.txt), which was weird and not compatible with Python’s resource loading mechanism. Resources in sub-directories should no longer encounter munging of directory separators to .. In the above example, the resource path will now be expressed as (package, resources/foo.txt).

  • Certain packaging actions are only performed once during building instead of twice. The user-visible impact of this change is that some duplicate log messages no longer appear.

  • Added a missing ) for add_python_resources() in auto-generated pyoxidizer.bzl files.

New Features

  • Python resource scanning now recognizes *.dist-info and *.egg-info directories as package distribution metadata. Files within these directories are exposed to Starlark as PythonPackageDistributionResource instances. These resources can be added to the embedded resources payload and made available for loading from memory or the filesystem, just like any other resource. The custom Python importer implements get_distributions() and returns objects that expose package distribution files. However, functionality of the returned distribution objects is not yet complete. See importlib.metadata Compatibility for details.

  • The custom Python importer now implements get_data(path), allowing loading of resources from filesystem paths (#139).

  • The PythonDistribution.to_python_executable() Starlark method now accepts a packaging_policy argument to control a policy and default behavior for resources on the produced executable. Using this argument, it is possible to control how resources should be materialized. For example, you can specify that resources should be loaded from memory if supported and from the filesystem if not. The argument can also be used to materialize the Python standard library on the filesystem, like how Python distributions typically work.

  • Python resources can now be installed next to built binaries using the new Starlark functions PythonExecutable.add_filesystem_relative_module_source(), PythonExecutable.add_filesystem_relative_module_bytecode(), PythonExecutable.add_filesystem_relative_package_resource(), PythonExecutable.add_filesystem_relative_extension_module(), PythonExecutable.add_filesystem_relative_python_resource(), PythonExecutable.add_filesystem_relative_package_distribution_resource(), and PythonExecutable.add_filesystem_relative_python_resources(). Unlike adding Python resources to FileManifest instances, Python resources added this way have their metadata serialized into the built executable. This allows the special Python module importer present in built binaries to service the import request without going through Python’s default filesystem-based importer. Because metadata for the file-based Python resources is frozen into the application, Python has to do far less work at run-time to load resources, making operations faster. Resources loaded from the filesystem in this manner have attributes like __file__, __cached__, and __path__ set, emulating behavior of the default Python importer. The custom import now also implements the importlib.abc.ExecutionLoader interface.

  • Windows binaries can now import extension modules defined as shared libraries (e.g. .pyd files) from memory. PyOxidizer will detect .pyd files during packaging and embed them into the binary as resources. When the module is imported, the extension module/shared library is loaded from memory and initialized. This feature enables PyOxidizer to package pre-built extension modules (e.g. from Windows binary wheels published on PyPI) while still maintaining the property of a (mostly) self-contained executable.

  • Multiple bytecode optimization levels can now be embedded in binaries. Previously, it was only possible to embed bytecode for a given module at a single optimization level.

  • The default_python_distribution() Starlark function now accepts values standalone_static and standalone_dynamic to specify a standalone distribution that is either statically or dynamically linked.

  • Support for parsing version 4 of the PYTHON.json distribution descriptor present in standalone Python distribution archives.

  • Default Python distributions upgraded to CPython 3.7.7.

Other Relevant Changes

  • The directory for downloaded Python distributions in the build directory now uses a truncated SHA-256 hash instead of the full hash to help avoid path length limit issues (#224).

  • The documentation for the pyembed crate has been moved out of the Sphinx documentation and into the Rust crate itself. Rendered docs can be seen by following the Documentation link at https://crates.io/crates/pyembed or by running cargo doc from a source checkout.

0.6.0

Released February 12, 2020.

Backwards Compatibility Notes

  • The default_python_distribution() Starlark function now accepts a flavor argument denoting the distribution flavor.

  • The pyembed crate no longer includes the auto-generated default configuration file. Instead, it is consumed by the application that instantiates a Python interpreter.

  • Rust projects for the main executable now utilize and require a Cargo build script so metadata can be passed from pyembed to the project that is consuming it.

  • The pyembed crate is no longer added to created Rust projects. Instead, the generated Cargo.toml will reference a version of the pyembed crate identical to the PyOxidizer version currently running. Or if pyoxidizer is running from a Git checkout of the canonical PyOxidizer Git repository, a local filesystem path will be used.

  • The fields of EmbeddedPythonConfig and pyembed::PythonConfig have been renamed and reordered to align with Python 3.8’s config API naming. This was done for the Starlark type in version 0.5. We have made similar changes to 0.6 so naming is consistent across the various types.

Bug Fixes

  • Module names without a . are now properly recognized when scanning the filesystem for Python resources and a package allow list is used (#223). Previously, if filtering scanned resources through an explicit list of allowed packages, the top-level module/package without a dot in its full name would not be passed through the filter.

New Features

  • The PythonDistribution() Starlark function now accepts a flavor argument to denote the distribution type. This allows construction of alternate distribution types.

  • The default_python_distribution() Starlark function now accepts a flavor argument which can be set to windows_embeddable to return a distribution based on the zip file distributions published by the official CPython project.

  • The pyembed crate and generated Rust projects now have various build-mode-* feature flags to control how build artifacts are built. See Rust Projects for more.

  • The pyembed crate can now be built standalone, without being bound to a specific PyOxidizer configuration.

  • The register_target() Starlark function now accepts an optional default_build_script argument to define the default target when evaluating in Rust build script mode.

  • The pyembed crate now builds against published cpython and python3-sys crates instead of a a specific Git commit.

  • Embedded Python interpreters can now be configured to run a file specified by a filename. See the run_file argument of PythonInterpreterConfig.

Other Relevant Changes

  • Rust internals have been overhauled to use traits to represent various types, namely Python distributions. The goal is to allow different Python distribution flavors to implement different logic for building binaries.

  • The pyembed crate’s build.rs has been tweaked so it can support calling out to pyoxidizer. It also no longer has a build dependency on pyoxidizer.

0.5.1

Released January 26, 2020.

Bug Fixes

  • Fixed bad Starlark example for building black in docs.

  • Remove resources attached to packages that don’t exist. (This was a regression in 0.5.)

  • Warn on failure to annotate a package. (This was a regression in 0.5.)

  • Building embedded Python resources now emits warnings when __file__ is seen. (This was a regression in 0.5.)

  • Missing parent packages are now automatically added when constructing embedded resources. (This was a regression in 0.5.)

0.5.0

Released January 26, 2020.

General Notes

This release of PyOxidizer is significant rewrite of the previous version. The impetus for the rewrite is to transition from TOML to Starlark configuration files. The new configuration file format should allow vastly greater flexibility for building applications and will unlock a world of new possibilities.

The transition to Starlark configuration files represented a shift from static configuration to something more dynamic. This required refactoring a ton of code.

As part of refactoring code, we took the opportunity to shore up lots of the code base. PyOxidizer was the project author’s first real Rust project and a lot of bad practices (such as use of .unwrap()/panics) were prevalent. The code mostly now has proper error handling. Another new addition to the code is unit tests. While coverage still isn’t great, we now have tests performing meaningful packaging activities. So regressions should hopefully be less common going forward.

Because of the scale of the rewritten code in this release, it is expected that there are tons of bugs of regressions. This will likely be a transitional release with a more robust release to follow.

Backwards Compatibility Notes

  • Support for building distributions/installers has been temporarily dropped.

  • Support for installing license files has been temporarily dropped.

  • Python interpreter configuration setting names have been changed to reflect names from Python 3.8’s interpreter initialization API.

  • .egg-info directories are now ignored when scanning for Python resources on the filesystem (matching the behavior for .dist-info directories).

  • The pyoxidizer init sub-command has been renamed to init-rust-project.

  • The pyoxidizer app-path sub-command has been removed.

  • Support for building distributions has been removed.

  • The minimum Rust version to build has been increased from 1.31 to 1.36. This is mainly due to requirements from the starlark crate. We could potentially reduce the minimum version requirements again with minimal changes to 3rd party crates.

  • PyOxidizer configuration files are now Starlark instead of TOML files. The default file name is pyoxidizer.bzl instead of pyoxidizer.toml. All existing configuration files will need to be ported to the new format.

Bug Fixes

  • The repl run mode now properly exits with a non-zero exit code if an error occurs.

  • Compiled C extensions now properly honor the ext_package argument passed to setup(), resulting in extensions which properly have the package name in their extension name (#26).

New Features

  • A glob() function has been added to config files to allow referencing existing files on the filesystem.

  • The in-memory MetaPathFinder now implements find_module().

  • A pyoxidizer init-config-file command has been implemented to create just a pyoxidizer.bzl configuration file.

  • A pyoxidizer python-distribution-info command has been implemented to print information about a Python distribution archive.

  • The EmbeddedPythonConfig() config function now accepts a legacy_windows_stdio argument to control the value of Py_LegacyWindowsStdioFlag (#190).

  • The EmbeddedPythonConfig() config function now accepts a legacy_windows_fs_encoding argument to control the value of Py_LegacyWindowsFSEncodingFlag.

  • The EmbeddedPythonConfig() config function now accepts an isolated argument to control the value of Py_IsolatedFlag.

  • The EmbeddedPythonConfig() config function now accepts a use_hash_seed argument to control the value of Py_HashRandomizationFlag.

  • The EmbeddedPythonConfig() config function now accepts an inspect argument to control the value of Py_InspectFlag.

  • The EmbeddedPythonConfig() config function now accepts an interactive argument to control the value of Py_InteractiveFlag.

  • The EmbeddedPythonConfig() config function now accepts a quiet argument to control the value of Py_QuietFlag.

  • The EmbeddedPythonConfig() config function now accepts a verbose argument to control the value of Py_VerboseFlag.

  • The EmbeddedPythonConfig() config function now accepts a parser_debug argument to control the value of Py_DebugFlag.

  • The EmbeddedPythonConfig() config function now accepts a bytes_warning argument to control the value of Py_BytesWarningFlag.

  • The Stdlib() packaging rule now now accepts an optional excludes list of modules to ignore. This is useful for removing unnecessary Python packages such as distutils, pip, and ensurepip.

  • The PipRequirementsFile() and PipInstallSimple() packaging rules now accept an optional extra_env dict of extra environment variables to set when invoking pip install.

  • The PipRequirementsFile() packaging rule now accepts an optional extra_args list of extra command line arguments to pass to pip install.

Other Relevant Changes

  • PyOxidizer no longer requires a forked version of the rust-cpython project (the python3-sys and cpython crates. All changes required by PyOxidizer are now present in the official project.

0.4.0

Released October 27, 2019.

Backwards Compatibility Notes

  • The setup-py-install packaging rule now has its package_path evaluated relative to the PyOxidizer config file path rather than the current working directory.

Bug Fixes

  • Windows now explicitly requires dynamic linking against msvcrt. Previously, this wasn’t explicit. And sometimes linking the final executable would result in unresolved symbol errors because the Windows Python distributions used external linkage of CRT symbols and for some reason Cargo wasn’t dynamically linking the CRT.

  • Read-only files in Python distributions are now made writable to avoid future permissions errors (#123).

  • In-memory InspectLoader.get_source() implementation no longer errors due to passing a memoryview to a function that can’t handle it (#134).

  • In-memory ResourceReader now properly handles multiple resources (#128).

New Features

  • Added an app-path command that prints the path to a packaged application. This command can be useful for tools calling PyOxidizer, as it will emit the path containing the packaged files without forcing the caller to parse command output.

  • The setup-py-install packaging rule now has an excludes option that allows ignoring specific packages or modules.

  • .py files installed into app-relative locations now have corresponding .pyc bytecode files written.

  • The setup-py-install packaging rule now has an extra_global_arguments option to allow passing additional command line arguments to the setup.py invocation.

  • Packaging rules that invoke pip or setup.py will now set a PYOXIDIZER=1 environment variable so Python code knows at packaging time whether it is running in the context of PyOxidizer.

  • The setup-py-install packaging rule now has an extra_env option to allow passing additional environment variables to setup.py invocations.

  • [[embedded_python_config]] now supports a sys_frozen flag to control setting sys.frozen = True.

  • [[embedded_python_config]] now supports a sys_meipass flag to control setting sys._MEIPASS = <exe directory>.

  • Default Python distribution upgraded to 3.7.5 (from 3.7.4). Various dependency packages also upgraded to latest versions.

All Other Relevant Changes

  • Built extension modules marked as app-relative are now embedded in the final binary rather than being ignored.

0.3.0

Released on August 16, 2019.

Backwards Compatibility Notes

  • The pyembed::PythonConfig struct now has an additional extra_extension_modules field.

  • The default musl Python distribution now uses LibreSSL instead of OpenSSL. This should hopefully be an invisible change.

  • Default Python distributions now use CPython 3.7.4 instead of 3.7.3.

  • Applications are now built into directories named apps/<app_name>/<target>/<build_type> rather than apps/<app_name>/<build_type>. This enables builds for multiple targets to coexist in an application’s output directory.

  • The program_name field from the [[embedded_python_config]] config section has been removed. At run-time, the current executable’s path is always used when calling Py_SetProgramName().

  • The format of embedded Python module data has changed. The pyembed crate and pyoxidizer versions must match exactly or else the pyembed crate will likely crash at run-time when parsing module data.

Bug Fixes

  • The libedit extension variant for the readline extension should now link on Linux. Before, attempting to link a binary using this extension variant would result in missing symbol errors.

  • The setup-py-install [[packaging_rule]] now performs actions to appease setuptools, thus allowing installation of packages using setuptools to (hopefully) work without issue (#70).

  • The virtualenv [[packaging_rule]] now properly finds the site-packages directory on Windows (#83).

  • The filter-include [[packaging_rule]] no longer requires both files and glob_files be defined (#88).

  • import ctypes now works on Windows (#61).

  • The in-memory module importer now implements get_resource_reader() instead of get_resource_loader(). (The CPython documentation steered us in the wrong direction - https://bugs.python.org/issue37459.)

  • The in-memory module importer now correctly populates __package__ in more cases than it did previously. Before, whether a module was a package was derived from the presence of a foo.bar module. Now, a module will be identified as a package if the file providing it is named __init__. This more closely matches the behavior of Python’s filesystem based importer. (#53)

New Features

  • The default Python distributions have been updated. Archives are generally about half the size from before. Tcl/tk is included in the Linux and macOS distributions (but PyOxidizer doesn’t yet package the Tcl files).

  • Extra extension modules can now be registered with PythonConfig instances. This can be useful for having the application embedding Python provide its own extension modules without having to go through Python build mechanisms to integrate those extension modules into the Python executable parts.

  • Built applications now have the ability to detect and use terminfo databases on the execution machine. This allows applications to interact with terminals properly. (e.g. the backspace key will now work in interactive pdb sessions). By default, applications on non-Windows platforms will look for terminfo databases at well-known locations and attempt to load them.

  • Default Python distributions now use CPython 3.7.4 instead of 3.7.3.

  • A warning is now emitted when a Python source file contains __file__. This should help trace down modules using __file__.

  • Added 32-bit Windows distribution.

  • New pyoxidizer distribution command for producing distributable artifacts of applications. Currently supports building tar archives and .msi and .exe installers using the WiX Toolset.

  • Libraries required by C extensions are now passed into the linker as library dependencies. This should allow C extensions linked against libraries to be embedded into produced executables.

  • pyoxidizer --verbose will now pass verbose to invoked pip and setup.py scripts. This can help debug what Python packaging tools are doing.

All Other Relevant Changes

  • The list of modules being added by the Python standard library is no longer printed during rule execution unless --verbose is used. The output was excessive and usually not very informative.

0.2.0

Released on June 30, 2019.

Backwards Compatibility Notes

  • Applications are now built into an apps/<appname>/(debug|release) directory instead of apps/<appname>. This allows debug and release builds to exist side-by-side.

Bug Fixes

  • Extracted .egg directories in Python package directories should now have their resources detected properly and not as Python packages with the name *.egg.

  • site-packages directories are now recognized as Python resource package roots and no longer have their contents packaged under a site-packages Python package.

New Features

  • Support for building and embedding C extensions on Windows, Linux, and macOS in many circumstances. See Native Extension Modules for support status.

  • pyoxidizer init now accepts a --pip-install option to pre-configure generated pyoxidizer.toml files with packages to install via pip. Combined with the --python-code option, it is now possible to create pyoxidizer.toml files for a ready-to-use Python application!

  • pyoxidizer now accepts a --verbose flag to make operations more verbose. Various low-level output is no longer printed by default and requires --verbose to see.

All Other Relevant Changes

  • Packaging now automatically creates empty modules for missing parent packages. This prevents a module from being packaged without its parent. This could occur with namespace packages, for example.

  • pip-install-simple rule now passes --no-binary :all: to pip.

  • Cargo packages updated to latest versions.

0.1.3

Released on June 29, 2019.

Bug Fixes

  • Fix Python refcounting bug involving call to PyImport_AddModule() when mode = module evaluation mode is used. The bug would likely lead to a segfault when destroying the Python interpreter. (#31)

  • Various functionality will no longer fail when running pyoxidizer from a Git repository that isn’t the canonical PyOxidizer repository. (#34)

New Features

  • pyoxidizer init now accepts a --python-code option to control which Python code is evaluated in the produced executable. This can be used to create applications that do not run a Python REPL by default.

  • pip-install-simple packaging rule now supports excludes for excluding resources from packaging. (#21)

  • pip-install-simple packaging rule now supports extra_args for adding parameters to the pip install command. (#42)

All Relevant Changes

  • Minimum Rust version decreased to 1.31 (the first Rust 2018 release). (#24)

  • Added CI powered by Azure Pipelines. (#45)

  • Comments in auto-generated pyoxidizer.toml have been tweaked to improve understanding. (#29)

0.1.2

Released on June 25, 2019.

Bug Fixes

  • Honor HTTP_PROXY and HTTPS_PROXY environment variables when downloading Python distributions. (#15)

  • Handle BOM when compiling Python source files to bytecode. (#13)

All Relevant Changes

  • pyoxidizer now verifies the minimum Rust version meets requirements before building.

0.1.1

Released on June 24, 2019.

Bug Fixes

  • pyoxidizer binaries built from crates should now properly refer to an appropriate commit/tag in PyOxidizer’s canonical Git repository in auto-generated Cargo.toml files. (#11)

0.1

Released on June 24, 2019. This is the initial formal release of PyOxidizer. The first pyoxidizer crate was published to crates.io.

New Features

  • Support for building standalone, single file executables embedding Python for 64-bit Windows, macOS, and Linux.

  • Support for importing Python modules from memory using zero-copy.

  • Basic Python packaging support.

  • Support for jemalloc as Python’s memory allocator.

  • pyoxidizer CLI command with basic support for managing project lifecycle.