Module data 

Data pipeline for storing verified test cases

This module handles the storage and retrieval of verified (source, target, correctness) tuples in Parquet format.

Features

• Large-scale parallel generation with progress tracking
• Automatic Parquet sharding for large datasets
• Support for all sampling strategies

Re-exports

pub use corpus::CorpusFormat;

pub use corpus::CorpusManager;

pub use corpus::CorpusMetadata;

pub use corpus::TrainingCorpus;

pub use pipeline::DataPipeline;

pub use pipeline::PipelineConfig;

pub use pipeline::PipelineStats;

pub use pipeline::PipelineStrategy;

Modules

corpus

Corpus management for verified tuples

pipeline

Large-scale data generation pipeline

Structs

CodeFeatures

Features extracted from source code for ML

GenerationMetadata

Metadata about how the test case was generated

TestCase

Test case with full metadata

TestCaseBuilder

Builder for test cases with mutations

VerifiedTuple

Verified transpilation tuple for ML training

Enums

TestResult

Test result enum