How ambersub/minor-project works
ChunkFlow is a Python-facing CSV row utility and parallel chunk-processing library backed by a C++/OpenMP core. The main usage path is a Python caller invoking the CSV facade for parsing, arithmetic, or filtering; an optional path serializes records in ChunkProcessor, delegates chunk work to the core, and reads results or status from SQLite. The excerpts establish the public bindings and delegation, but do not expose the full chunk persistence implementation, so storage links are limited to the core’s stated responsibilities.
Main entry points: chunking.py ChunkProcessor · chunkflow_core.cpp CSV core · __init__.py Public API · csv_math.py CSV facade
Components 14
Python API
- Public API
chunkflow/__init__.py - CSV facade
chunkflow/csv_math.py - ChunkProcessor
chunkflow/chunking.py
CSV Operations
- CSV core
src/chunkflow_core.cpp - Row parsing
src/chunkflow_core.cpp - Row arithmetic
src/chunkflow_core.cpp - Batch arithmetic
src/chunkflow_core.cpp - Row filtering
src/chunkflow_core.cpp
Chunk Processing
- Chunk engine
src/chunkflow_core.cpp - Transform callback
chunkflow/chunking.py
Run State
- SQLite results
chunkflow/chunking.py - Run summary
chunkflow/chunking.py - Run log
src/chunkflow_core.cpp
Other components
- Python caller
Connections 15
- Python caller → Public API: calls
- Public API → CSV facade: imports
chunkflow/__init__.py - CSV facade → CSV core: delegates
chunkflow/csv_math.py - CSV core → Row parsing: provides parsing
src/chunkflow_core.cpp - CSV core → Row arithmetic: provides operations
src/chunkflow_core.cpp - CSV core → Batch arithmetic: provides operations
src/chunkflow_core.cpp - CSV core → Row filtering: provides filters
src/chunkflow_core.cpp - Python caller → ChunkProcessor: uses optionally
- ChunkProcessor → Chunk engine: calls process
chunkflow/chunking.py - Chunk engine → Transform callback: invokes callback
- ChunkProcessor → SQLite results: reads results
chunkflow/chunking.py - ChunkProcessor → SQLite results: reads status
chunkflow/chunking.py - ChunkProcessor → Run summary: returns summary
chunkflow/chunking.py - Chunk engine → SQLite results: writes results
- Chunk engine → Run log: writes log
src/chunkflow_core.cpp