implements page-numbers-plan

tests/e2e/helpers.ts

@@ -123,6 +123,27 @@ export const getPageContentText = (document: PDFDocument, pageIndex: number) =>
 		.join('\n');
 };
 
+/**
+ * Decoded operands of all `Tj` operators. pdf-lib writes drawn text as hex
+ * strings (`<426C2E2031> Tj`), so the raw content stream is not readable.
+ */
+export const getPageTextValues = (document: PDFDocument, pageIndex: number) => {
+	const content = getPageContentText(document, pageIndex);
+	const values: string[] = [];
+
+	for (const match of content.matchAll(/<([0-9a-fA-F]+)>\s*Tj/g)) {
+		values.push(Buffer.from(match[1], 'hex').toString('latin1'));
+	}
+	for (const match of content.matchAll(/\(((?:\\.|[^()\\])*)\)\s*Tj/g)) {
+		values.push(match[1]);
+	}
+
+	return values;
+};
+
+export const getPageText = (document: PDFDocument, pageIndex: number) =>
+	getPageTextValues(document, pageIndex).join('\n');
+
 export const createEncryptedPdf = async (pageCount: number, label: string, password: string) => {
 	const document = await EncryptablePdfDocument.load(await createPdf(pageCount, label));
 	document.encrypt({