Merge pull request #10 from LucaCappelletti94/master

Adding limit parameter + serde + updated readme + const arity

.github/workflows/clippy.yml

@@ -0,0 +1,32 @@
+name: Clippy
+
+
+on:
+  push:
+    branches: ["master"]
+  pull_request:
+    branches: ["master"]
+
+env:
+  CARGO_TERM_COLOR: always
+
+jobs:
+  clippy:
+    runs-on: ubuntu-latest
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: Install Clippy
+        run:
+          rustup toolchain install nightly --component clippy
+      - name: Set up Rust
+        uses: actions-rs/toolchain@v1
+        with:
+          toolchain: nightly
+          override: true
+      - name: Run clippy
+        run: cargo clippy --all-features
+      - name: Run tests
+        run: cargo test --all-features
+      - name: Run tests release
+        run: cargo test --release --all-features

.gitignore

@@ -1,3 +1,8 @@
 /target/
 **/*.rs.bk
 Cargo.lock
+.DS_Store
+.vscode/
+test.txt
+rustc-ice-*
+*.graph

CONTRIBUTORS

@@ -1 +1,2 @@
 Hugo CHASTEL <hugo.chastel@orange.com>
+Luca Cappelletti <luca@ironprism.com>

Cargo.toml

@@ -1,20 +1,53 @@
 [package]
 name = "ngrammatic"
-version = "0.4.0"
-authors = ["Will Page <compenguy@gmail.com>"]
-description = "Character-oriented ngram generator and fuzzy matching library."
+version = "0.5.0"
+edition = "2021"
+authors = ["Will Page <compenguy@gmail.com>", "Luca Cappelletti <luca@ironprism.com>"]
+description = "Scalable n-gram generator for fuzzy similarity search with TF-IDF & efficient data structures."
 homepage = "https://github.com/compenguy/ngrammatic"
 repository = "https://github.com/compenguy/ngrammatic"
 documentation = "https://docs.rs/ngrammatic"
 readme = "README.md"
-categories = ["text-processing"]
-keywords = ["fuzzy", "ngrams", "shingles"]
+categories = ["algorithms", "text-processing", "data-structures"]
+keywords = ["fuzzy", "ngrams", "shingles", "tf-idf", "search"]
 license = "MIT"
 
 [lib]
 name = "ngrammatic"
 
 [dependencies]
+serde = { version = "1.0", features = ["derive"], optional = true }
+mem_dbg = { version = "0.1.8"}
+sux = {git = "https://github.com/LucaCappelletti94/sux-rs.git", no-default-features = true}
+half = {version="2.4.0", optional = true, features = ["zerocopy"]}
+rayon = {version="1.10.0", optional=true}
+trie-rs = {git = "https://github.com/LucaCappelletti94/trie-rs.git", optional = true, features = ["mem_dbg"]}
+webgraph = {git="https://github.com/vigna/webgraph-rs.git" }
 
-[package.metadata]
-msrv = "1.27.2"
+fxhash = "0.2.1"
+tempfile = "3.10.1"
+dsi-bitstream = "0.4.2"
+epserde = "0.4"
+log = "0.4.21"
+lender = "0.2.9"
+rand = "0.8.5"
+
+[dev-dependencies]
+flate2 = "1.0.28"
+indicatif = "0.17.8"
+# We also use the version of the ngrammatic library before the refactoring
+# to evaluate the changes in performance. The version we refer to is the 0.4.0
+# version of the library.
+ngrammatic_old = { version = "0.4.0", package="ngrammatic" }
+paste = "1.0.14"
+
+[features]
+default = ["rayon"]
+serde = ["dep:serde", "half/serde", "trie-rs/serde"]
+rayon = ["dep:rayon", "sux/rayon", "trie-rs/rayon"]
+
+[profile.release]
+overflow-checks = false   # Disable integer overflow checks.
+debug = false            # Include debug info.
+debug-assertions = false  # Enables debug assertions.
+opt-level = 3