Struct BytecodeCompiler

pub struct BytecodeCompiler { /* private fields */ }

An entity to perform Python bytecode compilation.

Implementations

impl BytecodeCompiler

pub fn new( python: &Path, script_dir: impl AsRef<Path>, ) -> Result<BytecodeCompiler>

Create a bytecode compiler using a Python executable.

A Python process will be started and it will start executing a Python source file embedded in this crate. That process interacts with this object via a pipe, which is used to send bytecode compilation requests and receive the compiled bytecode. The process is terminated when this object is dropped.

A Python script is written to the directory passed. This should ideally be a temporary directory. The file name is deterministic, so it isn’t safe for multiple callers to simultaneously pass the same directory. The temporary file is deleted before this function returns. Ideally this function would use a proper temporary file internally. The reason this isn’t done is to avoid an extra crate dependency.

Trait Implementations

impl Debug for BytecodeCompiler

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Drop for BytecodeCompiler

fn drop(&mut self)

Executes the destructor for this type.

impl PythonBytecodeCompiler for BytecodeCompiler

fn get_magic_number(&self) -> u32

Obtain the magic number to use in the bytecode header.

fn compile( self: &mut BytecodeCompiler, source: &[u8], filename: &str, optimize: BytecodeOptimizationLevel, output_mode: CompileMode, ) -> Result<Vec<u8>>

Compile Python source into bytecode with an optimization level.