[major] update pdfjs-dist dep from 3.9.179 to 4.2.67

Switch from CommonJS to ESM to use the new pdfjs-dist builds.
Switch from ts-node to tsx for running tests because ts-node doesn't work with ESM still.
Remove test coverage testing since it doesn't work with tsx.
Update .nvmrc to 22 because pdfjs-dist doesn't work below that.
Remove BinaryData export beacuse pdfjs-dist no longer exports it. Fix the data input to match pdfjs-dist's new type.
Delete globalThis.pdfjsWorker after reading a PDF as pdfjs-dist doesn't clean it up itself.

Update all other deps as well.

.nvmrc

@@ -1 +1 @@
-20
+22

README.md

@@ -1,6 +1,8 @@
 # PDF Text Reader
 
-Dead simple PDF text reader.
+Dead simple PDF text reader for Node.js. Uses Mozilla's [`pdfjs-dist`](https://www.npmjs.com/package/pdfjs-dist) package.
+
+Requires ESM and Node.js v22 or greater. (These are requirements from Mozilla's `pdf-dist` package itself.)
 
 # Install
 
@@ -43,8 +45,6 @@ See [the types](https://github.com/electrovir/pdf-text-reader/tree/master/src/re
 
 # Details
 
-This uses Mozilla's [`pdf.js`](https://github.com/mozilla/pdf.js/) package through its [`pdfjs-dist`](https://www.npmjs.com/package/pdfjs-dist) distribution on npm.
-
 This package simply reads the output of `pdfjs.getDocument` and sorts it into lines based on text position in the document. It also inserts spaces for text on the same line that is far apart horizontally and new lines in between lines that are far apart vertically.
 
 Example:

configs/mocha.config.js → configs/mocha.config.cjs

@@ -3,6 +3,7 @@ const baseOptions = require('virmator/base-configs/base-mocharc.js');
 /** @type {import('mocha').MochaOptions} */
 const mochaConfig = {
     ...baseOptions,
+    require: ['tsx'],
 };
 
 module.exports = mochaConfig;

configs/ncu.config.ts

@@ -6,11 +6,6 @@ export const ncuConfig: RunOptions = {
     // exclude these
     reject: [
         ...baseNcuConfig.reject,
-        /**
-         * Different versions of this have global pollution issues we're currently on a version that
-         * doesn't.
-         */
-        'pdfjs-dist',
     ],
     // include only these
     filter: [],

package.json

@@ -1,6 +1,6 @@
 {
     "name": "pdf-text-reader",
-    "version": "4.0.1",
+    "version": "5.0.0",
     "description": "Dead simple pdf text reader",
     "keywords": [
         "pdf",
@@ -22,54 +22,52 @@
         "name": "electrovir",
         "url": "https://github.com/electrovir"
     },
+    "type": "module",
     "main": "dist/index.js",
     "types": "dist/index.d.ts",
     "scripts": {
-        "compile": "virmator compile",
+        "compile": "rm -rf dist && tsc --pretty",
         "docs": "virmator docs",
-        "format": "virmator format",
+        "format": "prettier --color --cache --cache-strategy content \"./**/*.+(cjs|css|graphql|html|js|json|jsx|less|md|mjs|scss|toml|ts|tsx|yaml|yml)\"",
         "publish": "virmator publish \"npm run compile && npm run test:all\"",
-        "spellcheck": "virmator spellcheck",
-        "test": "virmator test",
-        "test:all": "concurrently --colors --kill-others-on-fail -c auto --names types,tests,spelling,format,docs,deps \"npm run test:types\" \"npm run test:coverage\" \"npm run test:spelling\" \"npm run test:format\" \"npm run test:docs\" \"npm run test:deps\"",
-        "test:coverage": "npm run test coverage",
+        "test": "mocha --colors --config 'configs/mocha.config.cjs'",
+        "test:all": "concurrently --colors --kill-others-on-fail -c auto --names types,tests,spelling,format,docs,deps \"npm run test:types\" \"npm run test\" \"npm run test:spelling\" \"npm run test:format\" \"npm run test:docs\" \"npm run test:deps\"",
         "test:deps": "virmator deps check",
         "test:docs": "virmator docs check",
-        "test:format": "virmator format check",
+        "test:format": "npm run format -- --check",
         "test:spelling": "virmator spellcheck",
-        "test:types": "tsc --noEmit",
-        "test:web": "virmator test-web"
+        "test:types": "tsc --noEmit"
     },
     "dependencies": {
-        "pdfjs-dist": "3.9.179"
+        "pdfjs-dist": "4.2.67"
     },
     "devDependencies": {
         "@electrovir/nyc": "^15.1.0-fix0",
         "@istanbuljs/nyc-config-typescript": "^1.0.2",
-        "@types/chai": "^4.3.10",
-        "@types/mocha": "^10.0.4",
-        "@types/node": "20.9.2",
-        "chai": "^4.3.10",
-        "cspell": "^8.0.0",
-        "dependency-cruiser": "^15.3.0",
-        "esbuild": "^0.19.6",
-        "istanbul-smart-text-reporter": "^1.1.2",
-        "markdown-code-example-inserter": "^0.3.2",
-        "mocha": "^10.2.0",
+        "@types/chai": "^4.3.16",
+        "@types/mocha": "^10.0.6",
+        "@types/node": "20.12.10",
+        "chai": "^5.1.0",
+        "cspell": "^8.8.0",
+        "dependency-cruiser": "^16.3.1",
+        "esbuild": "^0.21.0",
+        "istanbul-smart-text-reporter": "^1.1.4",
+        "markdown-code-example-inserter": "^1.0.0",
+        "mocha": "^10.4.0",
         "mocha-spec-reporter-with-file-names": "^0.0.3",
         "npm-check-updates": "~16.12.3",
-        "prettier": "^3.1.0",
-        "prettier-plugin-interpolated-html-tags": "^1.0.2",
-        "prettier-plugin-jsdoc": "^1.1.1",
-        "prettier-plugin-multiline-arrays": "^3.0.0",
+        "prettier": "^3.2.5",
+        "prettier-plugin-interpolated-html-tags": "^1.0.5",
+        "prettier-plugin-jsdoc": "^1.3.0",
+        "prettier-plugin-multiline-arrays": "^3.0.4",
         "prettier-plugin-organize-imports": "^3.2.4",
-        "prettier-plugin-packagejson": "^2.4.6",
-        "prettier-plugin-sort-json": "^3.1.0",
-        "prettier-plugin-toml": "^1.0.0",
-        "ts-node": "^10.9.1",
-        "type-fest": "^4.8.1",
-        "typedoc": "^0.25.3",
-        "typescript": "^5.2.2",
-        "virmator": "^11.1.1"
+        "prettier-plugin-packagejson": "^2.5.0",
+        "prettier-plugin-sort-json": "^4.0.0",
+        "prettier-plugin-toml": "^2.0.1",
+        "tsx": "^4.9.3",
+        "type-fest": "^4.18.2",
+        "typedoc": "^0.25.13",
+        "typescript": "^5.4.5",
+        "virmator": "^11.5.2"
     }
 }

src/read-pdf.test.ts

@@ -1,7 +1,7 @@
 import {assert} from 'chai';
-import {existsSync} from 'fs';
-import {join} from 'path';
-import {ReadonlyDeep} from 'type-fest';
+import {existsSync} from 'node:fs';
+import {join} from 'node:path';
+import type {ReadonlyDeep} from 'type-fest';
 import {PdfProgressData, readPdfPages, readPdfText} from './read-pdf';
 import {nodeModulesDir, sampleFilesDir} from './repo-paths.test-helper';
 

src/read-pdf.ts

@@ -1,14 +1,13 @@
-import {join} from 'path';
+import {join} from 'node:path';
 import {getDocument} from 'pdfjs-dist';
 import type {
-    BinaryData,
     DocumentInitParameters,
     PDFPageProxy,
     TextItem,
 } from 'pdfjs-dist/types/src/display/api';
-import {RequireExactlyOne} from 'type-fest';
+import type {RequireExactlyOne} from 'type-fest';
 
-export type {BinaryData, DocumentInitParameters} from 'pdfjs-dist/types/src/display/api';
+export type {DocumentInitParameters} from 'pdfjs-dist/types/src/display/api';
 
 /** A single page within a PDF file. */
 export type PdfPage = {
@@ -55,7 +54,7 @@ export type ReadPdfTextParams = PartialWithUndefined<{
         /** URL to the PDF. */
         url: string;
         /** PDF file data that has already been read from a PDF file. */
-        data: BinaryData;
+        data: DocumentInitParameters['data'];
         /** All other options that the Mozilla `pdfjs-dist` package supports. */
         allOptions: DocumentInitParameters;
     }>;
@@ -98,6 +97,9 @@ export async function readPdfPages({
         pages.push(await parsePage(page));
     }
 
+    /** This is populated by the pdfjs-dist package. We're deleting it here to prevent memory leaks. */
+    delete (globalThis as any).pdfjsWorker;
+
     return pages;
 }
 

src/repo-paths.test-helper.ts

@@ -1,12 +1,13 @@
-import {dirname, join, resolve} from 'path';
+import {dirname, join, resolve} from 'node:path';
+import {fileURLToPath} from 'node:url';
 
 /**
  * Path to the repo's root. Does not use the package name because the source code could
  * theoretically be cloned into any folder. "src" is used for the ts source code files (so they CAN
  * be run directly without transpiling it into JS) and "dist" is used for the transpiled JS output
  * directory.
  */
-const repoRootDir = dirname(__dirname);
+const repoRootDir = dirname(dirname(fileURLToPath(import.meta.url)));
 
 export const sampleFilesDir = join(repoRootDir, 'test-files');
-export const nodeModulesDir = resolve(__dirname, '..', 'node_modules', 'pdfjs-dist');
+export const nodeModulesDir = resolve(repoRootDir, 'node_modules', 'pdfjs-dist');

tsconfig.json

@@ -1,5 +1,6 @@
 {
     "compilerOptions": {
+        "module": "ES2022",
         "outDir": "./dist",
         "rootDir": "./src"
     },