@chr33s/pdf-codepoints 5.0.1 → 5.0.2

package/scripts/update-data.ts

@@ -1,64 +0,0 @@
-import { Buffer } from "node:buffer";
-import { mkdir, readdir, stat, writeFile } from "node:fs/promises";
-import { dirname, join, relative, resolve } from "node:path";
-import { fileURLToPath } from "node:url";
-
-const unicodeVersion = process.env.UNICODE_VERSION ?? "12.0.0";
-const packageRoot = resolve(dirname(fileURLToPath(import.meta.url)), "..");
-const dataDir = join(packageRoot, "data");
-
-async function* walk(dir: string): AsyncGenerator<string> {
-  const entries = await readdir(dir, { withFileTypes: true });
-  for (const entry of entries) {
-    const entryPath = join(dir, entry.name);
-    if (entry.isDirectory()) {
-      for await (const child of walk(entryPath)) {
-        yield child;
-      }
-    } else {
-      yield entryPath;
-    }
-  }
-}
-
-async function ensureDir(path: string): Promise<void> {
-  await mkdir(path, { recursive: true });
-}
-
-async function downloadFile(source: string, destination: string): Promise<void> {
-  const response = await fetch(source);
-  if (!response.ok) {
-    throw new Error(`Failed to download ${source}: ${response.status} ${response.statusText}`);
-  }
-
-  const arrayBuffer = await response.arrayBuffer();
-  await ensureDir(dirname(destination));
-  await writeFile(destination, Buffer.from(arrayBuffer));
-}
-
-async function main(): Promise<void> {
-  console.log(`Fetching Unicode data files (version ${unicodeVersion})...`);
-
-  for await (const absolutePath of walk(dataDir)) {
-    if (!absolutePath.endsWith(".txt")) continue;
-
-    const relativePath = relative(dataDir, absolutePath).split("\\").join("/");
-    const url = `https://www.unicode.org/Public/${unicodeVersion}/ucd/${relativePath}`;
-
-    const stats = await stat(absolutePath).catch(() => null);
-    const previousSize = stats?.size ?? 0;
-
-    await downloadFile(url, absolutePath);
-
-    const newStats = await stat(absolutePath);
-    const sizeChange = newStats.size - previousSize;
-    const changeLabel =
-      sizeChange === 0 ? "unchanged" : sizeChange > 0 ? `+${sizeChange}` : `${sizeChange}`;
-    console.log(`Downloaded ${relativePath} (${newStats.size} bytes, ${changeLabel})`);
-  }
-}
-
-main().catch((error) => {
-  console.error(error);
-  process.exitCode = 1;
-});

package/test/parser.test.ts

@@ -1,77 +0,0 @@
-import { beforeAll, describe, expect, test } from "vitest";
-
-import parser, { type CodePointTable } from "../src/parser.js";
-
-let codePoints: CodePointTable;
-
-beforeAll(() => {
-  codePoints = parser();
-});
-
-describe("parser", () => {
-  test("loads metadata for basic Latin letters", () => {
-    const capitalA = codePoints[0x0041];
-    const smallA = codePoints[0x0061];
-
-    expect(capitalA).toBeDefined();
-    expect(capitalA?.name).toBe("LATIN CAPITAL LETTER A");
-    expect(capitalA?.category).toBe("Lu");
-    expect(capitalA?.block).toBe("Basic Latin");
-    expect(capitalA?.script).toBe("Latin");
-    expect(capitalA?.eastAsianWidth).toBe("Na");
-    expect(capitalA?.lowercase).toEqual([0x0061]);
-
-    expect(smallA).toBeDefined();
-    expect(smallA?.name).toBe("LATIN SMALL LETTER A");
-    expect(smallA?.category).toBe("Ll");
-    expect(smallA?.block).toBe("Basic Latin");
-    expect(smallA?.uppercase).toEqual([0x0041]);
-  });
-
-  test("marks compatibility decompositions and canonical compositions", () => {
-    const noBreakSpace = codePoints[0x00a0];
-    const combiningDiaeresis = codePoints[0x0308];
-
-    expect(noBreakSpace).toBeDefined();
-    expect(noBreakSpace?.isCompat).toBe(true);
-    expect(noBreakSpace?.decomposition).toEqual([0x0020]);
-
-    expect(combiningDiaeresis).toBeDefined();
-    expect(combiningDiaeresis?.isCompat).toBe(false);
-    expect(combiningDiaeresis?.compositions[0x0041]).toBe(0x00c4);
-  });
-
-  test("applies derived metadata from supplemental files", () => {
-    const combiningAcute = codePoints[0x0301];
-    const aegeanSixtyThousand = codePoints[0x10130];
-
-    expect(combiningAcute).toBeDefined();
-    expect(combiningAcute?.combiningClass).toBe(230);
-    expect(combiningAcute?.combiningClassName).toBe("Above");
-
-    expect(aegeanSixtyThousand).toBeDefined();
-    expect(aegeanSixtyThousand?.numeric).toBe("60000");
-  });
-
-  test("records normalization quick-check and conditional casing data", () => {
-    const dottedCapitalI = codePoints[0x0130];
-
-    expect(dottedCapitalI).toBeDefined();
-    expect(dottedCapitalI?.NFD_QC).toBe(1);
-    expect(dottedCapitalI?.NFKD_QC).toBe(1);
-    expect(dottedCapitalI?.caseConditions).not.toBeNull();
-    expect(dottedCapitalI?.caseConditions).toEqual(expect.arrayContaining(["az"]));
-  });
-
-  test("fills in range entries and Arabic joining metadata", () => {
-    const cjkExtensionA = codePoints[0x3401];
-    const arabicBeh = codePoints[0x0628];
-
-    expect(cjkExtensionA).toBeDefined();
-    expect(cjkExtensionA?.code).toBe(0x3401);
-
-    expect(arabicBeh).toBeDefined();
-    expect(arabicBeh?.joiningType).toBe("Dual_Joining");
-    expect(arabicBeh?.joiningGroup).toBe("BEH");
-  });
-});

package/tsconfig.json

@@ -1,10 +0,0 @@
-{
-  "extends": "../../tsconfig.json",
-  "compilerOptions": {
-    "rootDir": "src",
-    "outDir": "dist",
-    "declarationDir": "dist"
-  },
-  "include": ["src/**/*"],
-  "exclude": ["dist", "node_modules"]
-}

package/tsconfig.typecheck.json

@@ -1,14 +0,0 @@
-{
-  "extends": "./tsconfig.json",
-  "compilerOptions": {
-    "noEmit": true,
-    "rootDir": "."
-  },
-  "include": [
-    "**/*.ts",
-  ],
-  "exclude": [
-    "dist",
-    "node_modules"
-  ]
-}

package/vitest.config.ts

@@ -1,8 +0,0 @@
-import { defineConfig } from "vitest/config";
-
-export default defineConfig({
-  test: {
-    environment: "node",
-    include: ["test/**/*.test.ts"],
-  },
-});