Merge pull request #91 from 2Toad/jp-issue-87

Fixes #87: Add support for multiple languages

Author: JasonPierce

.prettierignore

@@ -9,3 +9,4 @@ node_modules
 .nvmrc
 .prettierignore
 *.md
+Dockerfile

README.md

@@ -4,7 +4,7 @@
 [![Downloads](https://img.shields.io/npm/dm/@2toad/profanity.svg)](https://www.npmjs.com/package/@2toad/profanity)
 [![Build status](https://github.com/2toad/profanity/actions/workflows/ci.yml/badge.svg)](https://github.com/2Toad/Profanity/actions/workflows/nodejs.yml)
 
-A JavaScript profanity filter with full TypeScript support
+A multi-language profanity filter with full TypeScript support
 
 ## Getting Started
 
@@ -45,12 +45,35 @@ Create an instance of the Profanity class to change the default options:
 import { Profanity } from '@2toad/profanity';
 
 const profanity = new Profanity({
+    languages: ['de'],
     wholeWord: false,
     grawlix: '*****',
     grawlixChar: '$',
 });
 ```
 
+### languages
+
+By default, this is set to `['en']` (English). You can change the default to any [supported language](./supported-languages.md), including multiple languages:
+
+```JavaScript
+const profanity = new Profanity({
+    languages: ["en", "de"],
+});
+```
+
+You can override this option by specifying the languages in `exists` or `censor`:
+
+```JavaScript
+profanity.exists('Je suis un connard', ["fr"]);
+// true
+
+profanity.censor('I like big butts and je suis un connard', CensorType.Word, ["en", "de", "fr"]);
+// I like big @#$%&! and je suis un @#$%&!
+```
+
+If no languages are specified in the method call, it will use the languages specified in the options.
+
 ### wholeWord
 
 By default, this is set to `true` so profanity only matches on whole words:
@@ -112,7 +135,6 @@ profanity.censor('I like big butts and I cannot lie', CensorType.AllVowels);
 // I like big b$tts and I cannot lie
 ```
 
-
 ## Customize the word list
 
 Add words:

package.json

@@ -1,7 +1,7 @@
 {
   "name": "@2toad/profanity",
   "version": "2.5.0",
-  "description": "A JavaScript profanity filter with full TypeScript support",
+  "description": "A multi-language profanity filter with full TypeScript support",
   "homepage": "https://github.com/2Toad/Profanity",
   "author": "2Toad",
   "license": "MIT",
@@ -17,7 +17,7 @@
     "pretest": "npm run build",
     "test": "mocha -r ts-node/register tests/**/*.spec.ts",
     "test:watch": "npm run test -- --watch",
-    "benchmark": "ts-node src/benchmark/benchmark.ts",
+    "benchmark": "docker-compose -f ./src/benchmark/docker-compose.yml up --build",
     "lint": "eslint . --cache",
     "lint:fix": "eslint . --fix",
     "format": "prettier . --write",

src/benchmark/Dockerfile

@@ -0,0 +1,11 @@
+FROM node:20.17-alpine3.19
+
+WORKDIR /app
+
+COPY ../../package*.json ./
+
+RUN npm install
+
+COPY ../../ ./
+
+CMD ["npx", "ts-node", "src/benchmark/benchmark.ts"]

src/benchmark/benchmark.ts

@@ -64,6 +64,10 @@ const { smallCleanText, smallProfaneText, largeCleanText, largeProfaneText } = t
 const defaultProfanity = new Profanity();
 const partialMatchProfanity = new Profanity({ wholeWord: false });
 
+// Pre-cache regexes
+defaultProfanity.exists("foo");
+partialMatchProfanity.exists("bar");
+
 // Benchmark exists() function
 suite
   .add("exists() - small clean text", () => {

src/benchmark/docker-compose.yml

@@ -0,0 +1,13 @@
+services:
+  benchmark:
+    build:
+      context: ../..
+      dockerfile: ./src/benchmark/Dockerfile
+    deploy:
+      resources:
+        limits:
+          cpus: "1.0"
+          memory: 512M
+        reservations:
+          cpus: "1.0"
+          memory: 512M

src/benchmark/results.md

@@ -3,24 +3,41 @@
 ## Test Environment
 
 - **OS**: Windows 11 - WSL2 (Ubuntu 22.04.4 LTS)
-- **CPU**: AMD Ryzen 9 5900HX 3.30 GHz
-- **RAM**: 64 GB
+- **CPU**: AMD Ryzen 9 5900HX 3.30 GHz (Benchmark constrained to 1 CPU core)
+- **RAM**: 64 GB (Benchmark constrained to 512 MB)
 
 ### Benchmarks
 
+#### v3.0.0
+```
+Using test data: v1
+exists() - small clean text x 2,263,763 ops/sec ±3.96% (83 runs sampled)
+exists() - small profane text x 1,831,670 ops/sec ±3.09% (86 runs sampled)
+exists() - large clean text x 38,185 ops/sec ±2.82% (84 runs sampled)
+exists() - large profane text x 686,951 ops/sec ±2.11% (87 runs sampled)
+exists() - partial match, small profane text x 1,624,503 ops/sec ±8.02% (78 runs sampled)
+censor() - Word, small profane text x 915,620 ops/sec ±6.16% (83 runs sampled)
+censor() - FirstChar, small profane text x 1,275,945 ops/sec ±2.68% (77 runs sampled)
+censor() - FirstVowel, small profane text x 902,065 ops/sec ±3.43% (81 runs sampled)
+censor() - AllVowels, small profane text x 942,445 ops/sec ±2.94% (84 runs sampled)
+censor() - Word, large profane text x 5,578 ops/sec ±2.17% (86 runs sampled)
+censor() - partial match, Word, small profane text x 869,941 ops/sec ±7.91% (82 runs sampled)
+Fastest: exists() - small clean text
+```
+
 #### v2.4.0
 ```
 Using test data: v1
-exists() - small clean text x 7,384,356 ops/sec ±1.24% (95 runs sampled)
-exists() - small profane text x 6,347,800 ops/sec ±1.25% (90 runs sampled)
-exists() - large clean text x 49,978 ops/sec ±0.56% (93 runs sampled)
-exists() - large profane text x 1,216,505 ops/sec ±2.03% (81 runs sampled)
-exists() - partial match, small profane text x 5,319,125 ops/sec ±1.04% (93 runs sampled)
-censor() - Word, small profane text x 1,899,374 ops/sec ±0.54% (95 runs sampled)
-censor() - FirstChar, small profane text x 3,233,749 ops/sec ±1.40% (87 runs sampled)
-censor() - FirstVowel, small profane text x 1,894,666 ops/sec ±0.92% (92 runs sampled)
-censor() - AllVowels, small profane text x 1,697,305 ops/sec ±2.07% (92 runs sampled)
-censor() - Word, large profane text x 9,563 ops/sec ±0.91% (87 runs sampled)
-censor() - partial match, Word, small profane text x 1,597,856 ops/sec ±1.19% (92 runs sampled)
+exists() - small clean text x 3,838,466 ops/sec ±3.34% (81 runs sampled)
+exists() - small profane text x 2,557,317 ops/sec ±7.47% (74 runs sampled)
+exists() - large clean text x 41,031 ops/sec ±2.82% (83 runs sampled)
+exists() - large profane text x 799,283 ops/sec ±2.16% (83 runs sampled)
+exists() - partial match, small profane text x 3,013,455 ops/sec ±5.68% (88 runs sampled)
+censor() - Word, small profane text x 1,328,481 ops/sec ±2.17% (86 runs sampled)
+censor() - FirstChar, small profane text x 2,197,796 ops/sec ±5.86% (84 runs sampled)
+censor() - FirstVowel, small profane text x 1,184,065 ops/sec ±4.31% (75 runs sampled)
+censor() - AllVowels, small profane text x 1,105,599 ops/sec ±7.69% (77 runs sampled)
+censor() - Word, large profane text x 5,594 ops/sec ±6.02% (85 runs sampled)
+censor() - partial match, Word, small profane text x 1,031,901 ops/sec ±2.86% (81 runs sampled)
 Fastest: exists() - small clean text
 ```