Erster Commit

extract-pdf.js

@@ -0,0 +1,29 @@
+const fs = require('fs');
+const PDFParser = require("pdf2json");
+
+function extract(fileIn, fileOut) {
+    return new Promise((resolve, reject) => {
+        const pdfParser = new PDFParser(this, 1);
+
+        pdfParser.on("pdfParser_dataError", errData => reject(errData.parserError));
+        pdfParser.on("pdfParser_dataReady", pdfData => {
+            fs.writeFileSync(fileOut, pdfParser.getRawTextContent());
+            console.log(`Extracted: ${fileOut}`);
+            resolve();
+        });
+
+        pdfParser.loadPDF(fileIn);
+    });
+}
+
+async function main() {
+    const args = process.argv.slice(2);
+    if (args.length >= 2) {
+        await extract(args[0], args[1]);
+    } else {
+        await extract('./docs/ZF Global DESADV 1.3.9 EN Examples.pdf', './docs/zf_examples.txt');
+        await extract('./docs/ZF Global DESADV 1.3.9 EN.pdf', './docs/zf_guideline.txt');
+    }
+}
+
+main();