@retorquere/bibtex-parser 6.4.17 → 6.4.18

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@retorquere/bibtex-parser",
-  "version": "6.4.17",
+  "version": "6.4.18",
   "description": "BibTeX parser",
   "main": "index.js",
   "typings": "./index.d.ts",
@@ -75,6 +75,8 @@
     "jabref.js",
     "README.md",
     "crossref.json",
+    "sentence-case.js",
+    "sentence-case.d.ts",
     "fields.json",
     "strings.bib",
     "unabbrev.json"

package/sentence-case.d.ts

@@ -0,0 +1,8 @@
+export type TextRange = {
+    start: number;
+    end: number;
+    description?: string;
+};
+export declare function restore(text: string, orig: string, preserve: TextRange[]): string;
+export declare function toSentenceCase(text: string, ignoreHTML?: boolean): string;
+//# sourceMappingURL=sentence-case.d.ts.map

package/sentence-case.js

@@ -0,0 +1,96 @@
+"use strict";
+var __importDefault = (this && this.__importDefault) || function (mod) {
+    return (mod && mod.__esModule) ? mod : { "default": mod };
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.toSentenceCase = exports.restore = void 0;
+/* eslint-disable @typescript-eslint/no-unsafe-argument */
+const xregexp_1 = __importDefault(require("xregexp"));
+function restore(text, orig, preserve) {
+    for (const { start, end } of preserve) {
+        text = text.substring(0, start) + orig.substring(start, end) + text.substring(end);
+    }
+    return text;
+}
+exports.restore = restore;
+class SentenceCaser {
+    constructor() {
+        this.acronym = (0, xregexp_1.default)('^(\\p{Lu}[.])+(?=$|[\\P{L}])');
+        this.quoted = (0, xregexp_1.default)('^"[^"]+"(?=$|[\\P{L}])');
+        this.innerCaps = (0, xregexp_1.default)('\\p{Ll}\\p{Lu}');
+        this.allCaps = (0, xregexp_1.default)('^\\p{Lu}+$');
+        this.aint = (0, xregexp_1.default)("^\\p{L}n't(?=$|[\\P{L}])"); // isn't
+        this.word = (0, xregexp_1.default)('^\\p{L}+(-\\p{L}+)*'); // also match gallium-nitride as one word
+        this.and = (0, xregexp_1.default)('^\\p{Lu}&\\p{Lu}(?=$|[\\P{L}])'); // Q&A
+    }
+    convert(text, ignoreHTML = false) {
+        this.input = text;
+        this.result = '';
+        this.sentenceStart = true;
+        const preserve = [];
+        if (ignoreHTML) {
+            this.input.replace(/[^<>]<[^>]+>/g, (match, i) => {
+                preserve.push({ start: i + 1, end: i + match.length });
+                // replace markup by the preceding char
+                return match[0].repeat(match.length);
+            });
+            this.input = this.input.replace(/<[^>]+>[^<>]/g, (match, i) => {
+                preserve.push({ start: i, end: i + match.length - 1 });
+                // replace markup by the following char
+                return match[match.length - 1].repeat(match.length);
+            });
+        }
+        this.input = this.input.replace(/[;:]\s+A\s/g, match => match.toLowerCase());
+        this.input = this.input.replace(/[–—]\s*A\s/g, match => match.toLowerCase());
+        let m;
+        while (this.input) {
+            if (m = xregexp_1.default.exec(this.input, this.quoted)) { // "Hello There"
+                this.add(m[0], undefined, true);
+            }
+            else if (m = xregexp_1.default.exec(this.input, this.acronym)) { // U.S.
+                this.add(m[0], undefined, true);
+            }
+            else if (m = xregexp_1.default.exec(this.input, this.aint)) { // isn't
+                this.add(m[0], undefined, false);
+            }
+            else if (m = xregexp_1.default.exec(this.input, this.word)) {
+                this.add(m[0], '-', false);
+            }
+            else if (m = xregexp_1.default.exec(this.input, this.and)) {
+                this.add(m[0], undefined, true);
+            }
+            else {
+                this.add(this.input[0], undefined, false);
+            }
+        }
+        return restore(this.result, text, preserve);
+    }
+    add(word, splitter, keep) {
+        if (splitter) {
+            word = word.split(splitter).map((part, i) => {
+                if ((keep || this.sentenceStart) && i === 0)
+                    return part;
+                if (xregexp_1.default.exec(part, this.innerCaps))
+                    return part;
+                if (xregexp_1.default.exec(part, this.allCaps))
+                    return part;
+                return part.toLowerCase();
+            }).join(splitter);
+        }
+        else {
+            if (!keep)
+                word = word.toLowerCase();
+        }
+        this.result += word;
+        this.input = this.input.substr(word.length);
+        if (!word.match(/^\s+$/)) {
+            this.sentenceStart = !!word.match(/^[.?!]$/) || (word.length === 2 && word[1] === '.'); // Vitamin A. Vitamin B.
+        }
+    }
+}
+const sentenceCaser = new SentenceCaser;
+function toSentenceCase(text, ignoreHTML = false) {
+    return sentenceCaser.convert(text, ignoreHTML);
+}
+exports.toSentenceCase = toSentenceCase;
+//# sourceMappingURL=sentence-case.js.map