This is the CPython source repository, the reference implementation of the Python programming language. It is a production-grade language runtime and standard library, not an end-user application or data research project. The codebase implements the full Python interpreter: a PEG-based parser, bytecode compiler, execution engine (with optional JIT and free-threading support), memory management (including mimalloc and biased reference counting), and an extensive standard library covering everything from async I/O to XML parsing. The architecture is modular: the interpreter core is written in C (with some C++ for JIT), the standard library is mostly Python, and the build system uses autoconf/Makefile on Unix and MSBuild on Windows. Key entry points include `Programs/python.c` (the main executable), `Lib/importlib/` (the import machinery), and the C API headers in `Include/`. External dependencies are minimal and vendored (e.g., Expat, SQLite, HACL* crypto). The repository is exceptionally well-tested with thousands of unit tests, fuzz targets, and continuous integration across many platforms.
JaCoCo is a production-grade Java code coverage library that instruments JVM bytecode at runtime or offline to collect execution data, then analyzes and reports coverage metrics (lines, branches, instructions, methods, cyclomatic complexity). It is not a standalone application but a multi-module toolset: a Java agent (`jacocoagent.jar`) for on-the-fly instrumentation, a CLI (`jacococli.jar`) for offline operations, a Maven plugin (`jacoco-maven-plugin`) with goals like `prepare-agent`, `report`, `check`, and Ant tasks. The core library (`org.jacoco.core`) handles bytecode manipulation via ASM, probe insertion, and coverage analysis; the report module (`org.jacoco.report`) generates HTML, CSV, XML, and rule-based check outputs. The architecture follows a layered design: instrumentation → runtime data collection → analysis → reporting, with a visitor pattern for report generation and a filter framework to ignore compiler-generated bytecode (e.g., synthetic methods, Kotlin coroutines, try-with-resources). External dependencies include ASM for bytecode engineering, args4j for CLI parsing, and Google Code Prettify for HTML syntax highlighting. The project supports Java versions 5 through 28 (including preview features) and JVM languages like Kotlin, Groovy, and Scala through extensive compiler-specific filters.
Gravity is an embeddable, dynamically typed scripting language written in portable C99 with no external dependencies beyond the C standard library. The repository implements a full language toolchain: a multi-pass compiler (lexer, Pratt parser, AST, IR, optimizer, bytecode codegen) and a register-based virtual machine with mark-and-sweep garbage collection. It provides a CLI (`gravity` executable) for running scripts and a well-defined C embedding API. The architecture is modular, with clear separation between compiler (`src/compiler`), runtime (`src/runtime`), shared utilities (`src/shared`), and optional modules (Math, File, JSON, ENV). The codebase is extensively tested with unit tests (embedded `#unittest` blocks) and a large fuzzing corpus for crash resistance. Despite some monolithic files (`gravity_core.c` >4000 lines) and heavy macro usage, the implementation demonstrates strong software engineering craftsmanship, especially in the hash table, dynamic array macros, and Pratt parser. The project appears to be under active maintenance (last commit 2026-08-05, 34 open issues, 4.5k stars) but shows some technical drift such as outdated CI configurations, platform-specific ifdefs, and a fuzz corpus from 2017.
This repository implements `cuda_tile`, an MLIR-based compiler dialect and toolchain for GPU tile programming, primarily targeting NVIDIA CUDA architectures. It defines a rich set of operations for tile-level computation (arithmetic, matrix multiply, memory views, atomics, control flow) with explicit memory ordering and synchronization. The project includes a custom bytecode serialization format (.tileirbc) with versioned backward compatibility, allowing efficient IR transport and JIT compilation. Key tools are `cuda-tile-opt` (dialect testing), `cuda-tile-translate` (MLIR-to-bytecode conversion), `cuda-tile-optimize` (bytecode-to-bytecode optimization passes), and `cuda-tile-tblgen` (TableGen-based code generation for bytecode readers/writers). Python bindings enable programmatic IR construction. The architecture is a set of modular libraries (dialect, bytecode, optimizer, C API) that can be integrated into larger compiler pipelines. External dependencies are LLVM/MLIR at a pinned commit and CUDA. The codebase demonstrates high engineering rigor with extensive test coverage (bytecode stability, version compatibility, round-trip, transformation passes) and careful version management for evolving the IR format. While primarily a compiler infrastructure application, the project's focus on low-precision matrix operations (FP8, FP4) and tensor core optimizations touches on ML research areas.
Bob is a Scheme language implementation consisting of a Python interpreter, compiler, and virtual machine (VM), plus an optional C++ VM implementation (BareVM). The system compiles Scheme source code to custom bytecode, which can be executed by either the Python VM or the C++ BareVM. It includes features like bytecode serialization/deserialization and WebAssembly compilation. The architecture is modular with clear separation of concerns: parser, compiler, bytecode handling, VM execution, built-in functions, expression representation, and environment management. Key entry points are in bob/cmd.py (command-line interface) and example scripts demonstrating interpretation, compilation, and execution workflows.