improve-boundry

.vscode/settings.json

@@ -1,5 +1,19 @@
 {
-    "deno.enable": true,
-    "deno.lint": true,
-    "deno.unstable": true
-}
+  "deno.lint": true,
+  "deno.enable": true,
+  "deno.unstable": true,
+  "search.exclude": {
+    "deno.lock": true,
+    "dist/**": true
+  },
+  "editor.defaultFormatter": "denoland.vscode-deno",
+  "[typescript]": {
+    "editor.defaultFormatter": "denoland.vscode-deno"
+  },
+  "[markdown]": {
+    "editor.defaultFormatter": "denoland.vscode-deno"
+  },
+  "[github-actions-workflow]": {
+    "editor.defaultFormatter": "esbenp.prettier-vscode"
+  }
+}

build_npm.ts

@@ -24,9 +24,7 @@ await build({
       type: "git",
       url: "git+https://github.com/uriva/silly-nlp.git",
     },
-    bugs: {
-      url: "https://github.com/uriva/silly-nlp/issues",
-    },
+    bugs: { url: "https://github.com/uriva/silly-nlp/issues" },
   },
   postBuild() {
     Deno.copyFileSync("./LICENSE", outDir + "/LICENSE");

src/index.test.ts

@@ -1,14 +1,13 @@
-import { sideLog } from "https://deno.land/x/gamla@43.0.0/src/debug.ts";
-import { regexpTimes } from "./index.ts";
-import { matchesRegexp } from "./index.ts";
 import {
   approximateSemanticEquality,
   capitalizedPrefix,
   capitalizedSuffix,
   cleanSpeakers,
+  matchesRegexp,
   ngramsOfAtLeastNWords,
   prefixesWithSuffix,
   quotedTexts,
+  regexpTimes,
   simplify,
   someKewyordMatches,
   suffixesWithPrefix,
@@ -23,7 +22,7 @@ const testFn =
   <F extends Func>(name: string, f: F) =>
   (cases: [Parameters<F>, ReturnType<F>][]) =>
     cases.forEach(([args, result]) =>
-      Deno.test(name, () => assertEquals(f(...args), result)),
+      Deno.test(name, () => assertEquals(f(...args), result))
     );
 
 const testUnaryFn =
@@ -66,8 +65,7 @@ testFn(
 Deno.test("suffixesWithPrefix", () =>
   assertEquals(suffixesWithPrefix(/from\s+/gi, "from the matrix"), [
     "the matrix",
-  ]),
-);
+  ]));
 
 Deno.test("capitalizedSuffix", () => {
   assertEquals(

src/index.ts

@@ -101,8 +101,7 @@ export const suffixesWithPrefix = (regex: RegExp, input: string) =>
   regExpLocations(regex, input).map(({ end }) => input.slice(end));
 
 export const majority =
-  (equivalence: (str: string) => string) =>
-  (elements: string[]): string => {
+  (equivalence: (str: string) => string) => (elements: string[]): string => {
     const counts: Record<string, number> = {};
     const original: Record<string, string> = {};
     for (const element of elements) {
@@ -194,17 +193,15 @@ export const simplify: (x: string) => string = pipe(
 const allEnglishWordsAsSet = new Set(englishWords);
 
 const fixMissingSpaceInOneWord = (x: string) =>
-  allEnglishWordsAsSet.has(x)
-    ? x
-    : letIn(
-        range(1, x.length - 1).find(
-          (index) =>
-            allEnglishWordsAsSet.has(x.slice(0, index)) &&
-            allEnglishWordsAsSet.has(x.slice(index)),
-        ),
-        (location) =>
-          location ? [x.slice(0, location), x.slice(location)].join(" ") : x,
-      );
+  allEnglishWordsAsSet.has(x) ? x : letIn(
+    range(1, x.length - 1).find(
+      (index) =>
+        allEnglishWordsAsSet.has(x.slice(0, index)) &&
+        allEnglishWordsAsSet.has(x.slice(index)),
+    ),
+    (location) =>
+      location ? [x.slice(0, location), x.slice(location)].join(" ") : x,
+  );
 
 const missingSpace = (x: string) =>
   x.split(/\s/).map(fixMissingSpaceInOneWord).join(" ");
@@ -276,7 +273,7 @@ export const selectionGroup = (x: RegExp) =>
 
 const bracketIfNeeded = (s: string) =>
   (s.startsWith("(") && s.endsWith(")")) ||
-  (s.startsWith("[") && s.endsWith("]"))
+    (s.startsWith("[") && s.endsWith("]"))
     ? s
     : `(?:${s})`;
 
@@ -312,7 +309,7 @@ const personName = [
 
 const hyphen = /[―-]/;
 
-const boundry = [/[@.-\s:/בלה[\]?&%$#=*,!()]/, /^/, /$/].reduce(regExpOr); // \b doesn't work for non ascii
+const boundry = [/[_@.-\s:/בלה[\]?&%$#=*,!()]/, /^/, /$/].reduce(regExpOr); // \b doesn't work for non ascii
 
 const speaker = [optional(hyphen), personName, /\s?:/, boundry].reduce(
   concatRegexp,