fuzzy-search-rs 0.1.0__tar.gz

fuzzy_search_rs-0.1.0/Cargo.lock

@@ -0,0 +1,191 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+version = 4
+
+[[package]]
+name = "crossbeam-deque"
+version = "0.8.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "622f3fc73690be383c7214310406f28a90e6edeadc3cea882f9d71e495b9711a"
+dependencies = [
+ "crossbeam-epoch",
+ "crossbeam-utils",
+]
+
+[[package]]
+name = "crossbeam-epoch"
+version = "0.9.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "dc74980687109a3b14c72fd458107bf0baa1da1a1a805e178d15501ba9b86d9d"
+dependencies = [
+ "crossbeam-utils",
+]
+
+[[package]]
+name = "crossbeam-utils"
+version = "0.8.23"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a31eee39dddec8330830986fcd7625edb5a24ec90ea038215273bbc3adb08ac6"
+
+[[package]]
+name = "either"
+version = "1.19.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0e9c71c2167ca323c882b99918929403426e2373ea17242ff5653e0d5e1058be"
+
+[[package]]
+name = "fuzzy-search-rs"
+version = "0.1.0"
+dependencies = [
+ "pyo3",
+ "rayon",
+ "rustc-hash",
+]
+
+[[package]]
+name = "heck"
+version = "0.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2304e00983f87ffb38b55b444b5e3b60a884b5d30c0fca7d82fe33449bbe55ea"
+
+[[package]]
+name = "libc"
+version = "0.2.191"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "65cd87c9a6d60536580b316ad985726104730a5151c99845f2490428ef91240a"
+
+[[package]]
+name = "once_cell"
+version = "1.21.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9f7c3e4beb33f85d45ae3e3a1792185706c8e16d043238c593331cc7cd313b50"
+
+[[package]]
+name = "portable-atomic"
+version = "1.15.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "05c8b63e8d9609db387f0324918f81d68fe27748f084ef092fb35954d0539a85"
+
+[[package]]
+name = "proc-macro2"
+version = "1.0.107"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "985e7ec9bb745e6ce6535b544d84d6cd6f7ad8bd711c398938ae983b91a766d9"
+dependencies = [
+ "unicode-ident",
+]
+
+[[package]]
+name = "pyo3"
+version = "0.30.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "229ca13b8a15bb0d1487b81ab975bfcd97513416499a8312011b6036e42e0de4"
+dependencies = [
+ "libc",
+ "once_cell",
+ "portable-atomic",
+ "pyo3-build-config",
+ "pyo3-ffi",
+ "pyo3-macros",
+]
+
+[[package]]
+name = "pyo3-build-config"
+version = "0.30.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "70130e2b6dd1ee589a8788026a429561ca5c40fa7a2fa7914c2282eddae24700"
+dependencies = [
+ "target-lexicon",
+]
+
+[[package]]
+name = "pyo3-ffi"
+version = "0.30.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5e7ff73c555c286d212d4ce7e5fbda845a973c4f61bd5f160f5f2c3c2a04d1c4"
+dependencies = [
+ "libc",
+ "pyo3-build-config",
+]
+
+[[package]]
+name = "pyo3-macros"
+version = "0.30.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c46d2e5862c7bb8207e3b639f713dda3c499933444784154c11d013c5a691b60"
+dependencies = [
+ "proc-macro2",
+ "pyo3-macros-backend",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "pyo3-macros-backend"
+version = "0.30.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "38b89f066f2ae72a2aef52d1537ca2fcd104440ebfadee06e97aa90738a25b9b"
+dependencies = [
+ "heck",
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
+name = "quote"
+version = "1.0.47"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1fbf4db142a473a8d80c26bbf18454ed458bf8d26c8219c331daecfdbd079001"
+dependencies = [
+ "proc-macro2",
+]
+
+[[package]]
+name = "rayon"
+version = "1.12.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fb39b166781f92d482534ef4b4b1b2568f42613b53e5b6c160e24cfbfa30926d"
+dependencies = [
+ "either",
+ "rayon-core",
+]
+
+[[package]]
+name = "rayon-core"
+version = "1.13.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "22e18b0f0062d30d4230b2e85ff77fdfe4326feb054b9783a3460d8435c8ab91"
+dependencies = [
+ "crossbeam-deque",
+ "crossbeam-utils",
+]
+
+[[package]]
+name = "rustc-hash"
+version = "2.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6b1e7f9a428571be2dc5bc0505c13fb6bf936822b894ec87abf8a08a4e51742d"
+
+[[package]]
+name = "syn"
+version = "3.0.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4ae9ccbea3a7b0b09443036490c8f34a8b1b760fd345e5c0dae0aac718faea79"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-ident",
+]
+
+[[package]]
+name = "target-lexicon"
+version = "0.13.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "adb6935a6f5c20170eeceb1a3835a49e12e19d792f6dd344ccc76a985ca5a6ca"
+
+[[package]]
+name = "unicode-ident"
+version = "1.0.27"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a2c754d6c33795a1c324727428e5a7dedb5b06195f9890bdbcba760d3e246563"

fuzzy_search_rs-0.1.0/Cargo.toml

@@ -0,0 +1,20 @@
+[package]
+name = "fuzzy-search-rs"
+version = "0.1.0"
+edition = "2021"
+description = "Rust backend for the skipgram scan and candidate building of fuzzy-search"
+license = "MIT"
+readme = "README.md"
+
+[lib]
+name = "fuzzy_search_rs"
+crate-type = ["cdylib", "rlib"]
+
+[dependencies]
+pyo3 = { version = "0.30", features = ["extension-module", "abi3-py310"] }
+rayon = "1"
+rustc-hash = "2"
+
+[profile.release]
+lto = "fat"
+codegen-units = 1

fuzzy_search_rs-0.1.0/PKG-INFO

@@ -0,0 +1,5 @@
+Metadata-Version: 2.4
+Name: fuzzy-search-rs
+Version: 0.1.0
+Summary: Rust backend for the skipgram scan and candidate building of fuzzy-search
+Requires-Python: >=3.10

fuzzy_search_rs-0.1.0/README.md

@@ -0,0 +1,102 @@
+# fuzzy-search-rs
+
+Optional Rust backend for `FuzzyPhraseSearcher` in [fuzzy-search](../README.md). It returns exactly the
+same matches as the Python implementation, about 35 times faster on one thread, and runs batches on
+several threads without the GIL.
+
+## Wheels
+
+The extension is built as an abi3 wheel: one wheel per platform works on every CPython from 3.10 on.
+The workflow `.github/workflows/rust-wheels.yml` builds and tests wheels for Linux (x86_64 and aarch64,
+manylinux2014; x86_64 musllinux), macOS (universal2: Intel and Apple silicon) and Windows (x64 and
+arm64), and uploads them as workflow artifacts. Publishing to PyPI is not automated.
+
+## Build and install
+
+Needs a Rust toolchain and [maturin](https://www.maturin.rs/). From this directory:
+
+```bash
+maturin build --release
+```
+
+Other platforms from macOS (with rustup's toolchain first on `PATH`; Homebrew's `rust` has only the
+host target):
+
+```bash
+rustup target add x86_64-apple-darwin x86_64-unknown-linux-gnu
+```
+
+```bash
+maturin build --release --target universal2-apple-darwin
+```
+
+```bash
+pip install ziglang && maturin build --release --target x86_64-unknown-linux-gnu --zig --compatibility manylinux2014
+```
+
+Cross-building aarch64 Linux with zig currently fails at link time (zig rejects rustc's
+`--fix-cortex-a53-843419` flag), and Windows wheels need a Windows machine; CI builds both natively.
+
+```bash
+uv pip install --python ../.venv/bin/python --reinstall target/wheels/fuzzy_search_rs-*.whl
+```
+
+or, with pip, `pip install target/wheels/fuzzy_search_rs-*.whl`.
+
+## Use
+
+Select the backend in the searcher config:
+
+```python
+from fuzzy_search import FuzzyPhraseSearcher
+
+searcher = FuzzyPhraseSearcher(phrase_model=phrases, config={'backend': 'auto'})
+matches = searcher.find_matches(text)                          # one text
+results = searcher.find_matches_batch(texts, workers=0)        # many texts, one thread per CPU
+```
+
+- `'auto'` (default): Rust when installed and possible, otherwise Python.
+- `'python'`: the Python implementation.
+- `'rust'`: the Rust backend; raises `RuntimeError` if it is not installed or cannot reproduce the
+  searcher's settings.
+
+With the Rust backend, `find_matches_batch(texts, workers=n)` uses `n` threads (0: one per CPU) instead
+of worker processes, so no `if __name__ == '__main__':` guard is needed.
+
+## What runs in Rust
+
+`find_matches` runs exact matching in Python, then Rust does, per text:
+
+1. the skipgram scan (`Searcher.find_skipgram_matches`),
+2. candidate building (`get_skipmatch_candidates`, `CandidatePartial`),
+3. word-boundary adjustment (`adjust_match_offsets`) and per-phrase overlap filtering
+   (`filter_overlapping_phrase_candidates`),
+4. scoring (`PhraseMatch.add_scores`: character overlap, ngram overlap, `Levenshtein.ratio`).
+
+Python then builds `PhraseMatch` objects only for candidates that pass the thresholds, and runs the
+remaining steps (distractors, combining with exact matches, offset thresholds, overlap filtering).
+
+The Rust code mirrors the Python code step by step, including quirks such as group ordering, the
+`next_start_offset and ...` truthiness test and `Counter` entries that drop to zero. Offsets are Unicode
+code points, as in Python slicing.
+
+## Limits and differences
+
+- `ngram_size` up to 6 (skipgrams are packed into 128-bit keys). Larger sizes fall back to Python
+  (`'auto'`) or raise (`'rust'`).
+- Phrases with `max_end_offset` fall back to Python, which raises an `AttributeError` for them.
+- `debug > 0` always runs the Python code, so its debug output is unchanged.
+- With `ignorecase`, texts whose lowercase has a different length (such as `İ`) or that contain a capital
+  sigma `Σ` are scored in Python, because there slicing the lowercased text differs from lowercasing a slice.
+- Where Python would raise `ZeroDivisionError` or `IndexError` on degenerate input (a phrase shorter
+  than `ngram_size`), the Rust backend raises `ValueError`.
+
+## Verification
+
+- `test/test_search_rust_backend.py` compares both backends over a grid of settings and texts, including
+  threaded batches, and checks backend selection and pickling.
+- The regression fixture runs with either backend:
+
+```bash
+.venv/bin/python -m pytest test/regression --regression-backend rust --regression-exact --regression-full
+```

fuzzy_search_rs-0.1.0/pyproject.toml

@@ -0,0 +1,12 @@
+[build-system]
+requires = ["maturin>=1.5,<2"]
+build-backend = "maturin"
+
+[project]
+name = "fuzzy-search-rs"
+version = "0.1.0"
+description = "Rust backend for the skipgram scan and candidate building of fuzzy-search"
+requires-python = ">=3.10"
+
+[tool.maturin]
+features = ["pyo3/extension-module"]