Merge pull request #527 from YuukiToriyama/release/v0.1.24

release/v0.1.24をmainブランチにマージ

.github/workflows/code-quality-check.yaml

@@ -25,10 +25,3 @@ jobs:
           reporter: 'github-pr-review'
           filter_mode: 'nofilter'
           github_token: ${{ secrets.GITHUB_TOKEN }}
-      - name: Run benchmark
-        uses: boa-dev/criterion-compare-action@v3
-        with:
-          token: ${{ secrets.GITHUB_TOKEN }}
-          branchName: ${{ github.base_ref }}
-          cwd: 'core'
-          benchName: 'core_benchmark'

.github/workflows/run-test.yaml

@@ -49,3 +49,16 @@ jobs:
       - name: Build check for wasm crate
         working-directory: wasm
         run: wasm-pack build --target web --scope toriyama
+
+  msrv:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Install minimum supported version
+        uses: dtolnay/rust-toolchain@master
+        with:
+          toolchain: 1.75.0
+      - name: Basic build
+        run: cargo build --verbose
+      - name: Build docs
+        run: cargo doc --verbose

.github/workflows/upload-npmjs.yaml

@@ -9,6 +9,8 @@ jobs:
   publish:
     runs-on: ubuntu-latest
     environment: npmjs
+    permissions:
+      id-token: write
     defaults:
       run:
         working-directory: wasm
@@ -48,6 +50,6 @@ jobs:
       - name: Upload wasm to npmjs.com
         run: |
           cd pkg
-          npm publish --access public
+          npm publish --provenance --access public
         env:
           NODE_AUTH_TOKEN: ${{ secrets.NPMJS_REGISTRY_TOKEN }}

Cargo.toml

@@ -8,7 +8,7 @@ members = [
 resolver = "2"
 
 [workspace.package]
-version = "0.1.23"
+version = "0.1.24"
 edition = "2021"
 description = "A Rust Library to parse japanese addresses."
 repository = "https://github.com/YuukiToriyama/japanese-address-parser"

README.md

@@ -2,7 +2,7 @@
 
 [![Docs](https://docs.rs/japanese-address-parser/badge.svg)](https://docs.rs/japanese-address-parser)
 [![Crates.io (latest)](https://img.shields.io/crates/v/japanese-address-parser)](https://crates.io/crates/japanese-address-parser)
-![Rust Version](https://img.shields.io/badge/rust%20version-%3E%3D1.73.0-orange)
+![Rust Version](https://img.shields.io/badge/rust%20version-%3E%3D1.75.0-orange)
 [![Unit test & Integration test](https://github.com/YuukiToriyama/japanese-address-parser/actions/workflows/run-test.yaml/badge.svg?branch=main)](https://github.com/YuukiToriyama/japanese-address-parser/actions/workflows/run-test.yaml)
 
 A Rust library for parsing Japanese addresses.
@@ -48,20 +48,7 @@ fn main() {
 This crate is designed to be buildable for `wasm32-unknown-unknown` with `wasm-pack`.
 Pre-compiled wasm module is available on npmjs.com
 
-```bash
-npm install @toriyama/japanese-address-parser
-```
-
-```javascript
-import init, {Parser} from "@toriyama/japanese-address-parser"
-
-init().then(() => {
-    const parser = new Parser()
-    parser.parse("東京都千代田区丸の内1-1-1").then(parseResult => {
-        console.log(JSON.stringify(parseResult, null, "\t"))
-    })
-})
-```
+You can run this crate on your browser. For more details, see [wasm module's README](wasm/README.md).
 
 ## Python support(experimental)
 

core/Cargo.toml

@@ -9,11 +9,10 @@ license.workspace = true
 readme = "../README.md"
 keywords.workspace = true
 categories.workspace = true
-rust-version = "1.73.0"
+rust-version = "1.75.0"
 
 [lib]
 crate-type = ["rlib", "cdylib"]
-bench = false
 
 [features]
 default = ["city-name-correction"]
@@ -23,12 +22,8 @@ format-house-number = []
 eliminate-whitespaces = []
 experimental = []
 
-[[bench]]
-name = "core_benchmark"
-harness = false
-
 [dependencies]
-itertools = "0.13.0"
+itertools = "0.13.0" # 互換性のために残してあるが、`core::parser::adapter`を削除する際に忘れずに削除する
 log.workspace = true
 rapidfuzz = "0.5.0"
 regex = { version = "1.10.6", default-features = false, features = ["std", "unicode-perl"] }

core/src/adapter.rs

@@ -0,0 +1 @@
+pub mod orthographical_variant_adapter;

core/src/adapter/orthographical_variant_adapter.rs

@@ -0,0 +1,160 @@
+#[derive(Clone)]
+pub enum OrthographicalVariant {
+    の,
+    ツ,
+    ケ,
+    薮,
+    崎,
+    檜,
+    龍,
+    竈,
+    嶋,
+    舘,
+    脊,
+    渕,
+    己,
+    槇,
+    治,
+    佛,
+    澤,
+    塚,
+    恵,
+    穂,
+    梼,
+    蛍,
+    與,
+    瀧,
+    籠,
+    濱,
+    祗,
+    曾,
+}
+
+impl OrthographicalVariant {
+    fn value(&self) -> &[char] {
+        match self {
+            OrthographicalVariant::の => &['の', 'ノ', '之'],
+            OrthographicalVariant::ツ => &['ツ', 'ッ'],
+            OrthographicalVariant::ケ => &['ケ', 'ヶ', 'が', 'ガ'],
+            OrthographicalVariant::薮 => &['薮', '藪', '籔'],
+            OrthographicalVariant::崎 => &['崎', '﨑'],
+            OrthographicalVariant::檜 => &['桧', '檜'],
+            OrthographicalVariant::龍 => &['龍', '竜'],
+            OrthographicalVariant::竈 => &['竈', '竃', '釜'],
+            OrthographicalVariant::嶋 => &['嶋', '島'],
+            OrthographicalVariant::舘 => &['舘', '館'],
+            OrthographicalVariant::脊 => &['脊', '背'],
+            OrthographicalVariant::渕 => &['渕', '淵'],
+            OrthographicalVariant::己 => &['己', '巳'],
+            OrthographicalVariant::槇 => &['槇', '槙'],
+            OrthographicalVariant::治 => &['治', '冶'],
+            OrthographicalVariant::佛 => &['佛', '仏'],
+            OrthographicalVariant::澤 => &['澤', '沢'],
+            OrthographicalVariant::塚 => &['塚', '塚'],
+            OrthographicalVariant::恵 => &['恵', '惠'],
+            OrthographicalVariant::穂 => &['穂', '穗'],
+            OrthographicalVariant::梼 => &['梼', '檮'],
+            OrthographicalVariant::蛍 => &['蛍', '螢'],
+            OrthographicalVariant::與 => &['與', '与'],
+            OrthographicalVariant::瀧 => &['瀧', '滝'],
+            OrthographicalVariant::籠 => &['籠', '篭'],
+            OrthographicalVariant::濱 => &['濱', '浜'],
+            OrthographicalVariant::祗 => &['祗', '祇'],
+            OrthographicalVariant::曾 => &['曾', '曽'],
+        }
+    }
+
+    fn permutations(&self) -> Vec<(char, char)> {
+        let characters = self.value();
+        let mut permutations: Vec<(char, char)> = vec![];
+        for n in 0..characters.len() {
+            for m in 0..characters.len() {
+                if n != m {
+                    permutations.push((characters[n], characters[m]));
+                }
+            }
+        }
+        permutations
+    }
+}
+
+pub struct OrthographicalVariantAdapter {
+    pub variant_list: Vec<OrthographicalVariant>,
+}
+
+impl OrthographicalVariantAdapter {
+    pub fn apply(self, input: &str, region_name: &str) -> Option<(String, String)> {
+        // 必要なパターンのみを選別する
+        let variant_list: Vec<&OrthographicalVariant> = self
+            .variant_list
+            .iter()
+            .filter(|v| v.value().iter().any(|&c| input.contains(c)))
+            .collect();
+        if variant_list.is_empty() {
+            return None;
+        }
+
+        // マッチ候補を容れておくためのVector
+        let mut candidates: Vec<String> = vec![region_name.to_string()];
+        // パターンを一つづつ検証していく
+        for variant in variant_list {
+            let mut semi_candidates: Vec<String> = vec![];
+            // variantから順列を作成
+            // ["ケ", "ヶ", "が"] -> (ケ, ヶ), (ケ, が), (ヶ, ケ), (ヶ, が), (が, ケ), (が, ヶ)
+            for (a, b) in variant.permutations() {
+                for candidate in candidates.iter().filter(|x| x.contains(a)) {
+                    let modified_candidate = modify_specific_character(candidate, a, b);
+                    if input.starts_with(&modified_candidate) {
+                        // マッチすれば早期リターン
+                        return Some((
+                            region_name.to_string(),
+                            input
+                                .chars()
+                                .skip(modified_candidate.chars().count())
+                                .collect(),
+                        ));
+                    } else {
+                        // マッチしなければsemi_candidatesに置き換え後の文字列をpush
+                        semi_candidates.push(modified_candidate);
+                    }
+                }
+            }
+            candidates = semi_candidates;
+            candidates.push(region_name.to_string());
+        }
+        None
+    }
+}
+
+fn modify_specific_character(text: &str, from: char, to: char) -> String {
+    text.chars()
+        .map(|x| if x == from { to } else { x })
+        .collect()
+}
+
+#[cfg(test)]
+mod tests {
+    use crate::adapter::orthographical_variant_adapter::OrthographicalVariant;
+
+    #[test]
+    fn permutations() {
+        let variant = OrthographicalVariant::ケ;
+        assert_eq!(
+            variant.permutations(),
+            vec![
+                ('ケ', 'ヶ'),
+                ('ケ', 'が'),
+                ('ケ', 'ガ'),
+                ('ヶ', 'ケ'),
+                ('ヶ', 'が'),
+                ('ヶ', 'ガ'),
+                ('が', 'ケ'),
+                ('が', 'ヶ'),
+                ('が', 'ガ'),
+                ('ガ', 'ケ'),
+                ('ガ', 'ヶ'),
+                ('ガ', 'が'),
+            ]
+        );
+    }
+}

core/src/lib.rs

@@ -13,6 +13,7 @@ compile_error! {
     "The `blocking` feature is not supported with wasm target."
 }
 
+mod adapter;
 #[deprecated(since = "0.1.23", note = "This module will be deleted in v0.2")]
 pub mod api;
 pub(crate) mod domain;

core/src/parser.rs

@@ -11,6 +11,7 @@ use crate::interactor::geolonia::{GeoloniaInteractor, GeoloniaInteractorImpl};
 use crate::tokenizer::{End, Tokenizer};
 use serde::Serialize;
 
+#[deprecated(since = "0.1.24", note = "This module will be deleted in v0.2")]
 pub mod adapter;
 
 impl From<Tokenizer<End>> for Address {

core/src/tokenizer/read_city.rs

@@ -1,7 +1,7 @@
-use crate::domain::common::token::{append_token, Token};
-use crate::parser::adapter::orthographical_variant_adapter::{
-    OrthographicalVariantAdapter, OrthographicalVariants, Variant,
+use crate::adapter::orthographical_variant_adapter::{
+    OrthographicalVariant, OrthographicalVariantAdapter,
 };
+use crate::domain::common::token::{append_token, Token};
 use crate::tokenizer::{CityNameFound, CityNameNotFound, PrefectureNameFound, Tokenizer};
 use std::marker::PhantomData;
 
@@ -29,29 +29,29 @@ impl Tokenizer<PrefectureNameFound> {
         }
 
         // ここまでで市区町村名が読み取れない場合は、表記ゆれを含む可能性を検討する
-        let mut variant_list = vec![Variant::ケ];
+        let mut variant_list = vec![OrthographicalVariant::ケ];
         match self.get_prefecture_name() {
             Some("青森県") => {
-                variant_list.push(Variant::舘);
+                variant_list.push(OrthographicalVariant::舘);
             }
             Some("宮城県") => {
-                variant_list.push(Variant::竈);
+                variant_list.push(OrthographicalVariant::竈);
             }
             Some("茨城県") => {
-                variant_list.push(Variant::龍);
-                variant_list.push(Variant::嶋);
+                variant_list.push(OrthographicalVariant::龍);
+                variant_list.push(OrthographicalVariant::嶋);
             }
             Some("東京都") => {
-                variant_list.push(Variant::檜);
+                variant_list.push(OrthographicalVariant::檜);
             }
             Some("兵庫県") => {
-                variant_list.push(Variant::塚);
+                variant_list.push(OrthographicalVariant::塚);
             }
             Some("高知県") => {
-                variant_list.push(Variant::梼);
+                variant_list.push(OrthographicalVariant::梼);
             }
             Some("福岡県") => {
-                variant_list.push(Variant::恵);
+                variant_list.push(OrthographicalVariant::恵);
             }
             _ => {}
         }