Read PDFs

Create a reader for a file or compatible input stream, inspect it, and release its resources when finished.

var muhammara = require("@muhammara/native");
var reader = muhammara.createReader("input.pdf");

console.log(reader.getPagesCount());

reader.end();

For an encrypted input, provide its user or owner password when creating the reader:

var reader = muhammara.createReader("encrypted.pdf", {
  password: "open-password",
});

See Change PDF Passwords to create, re-encrypt, or remove encryption.

Call end() when the reader is no longer needed, after every object and stream parsed from it has been consumed. It closes the underlying file handle; skipping it leaves input.pdf locked on Windows, where the file then cannot be deleted or renamed. Calls on the reader after end() throw PDF reader has ended. Recipe releases the readers it opens itself, in endPDF().

Readers provide page counts, PDF level, trailers, page dictionaries, and low-level PDF objects. parsePage(index) exposes page boxes and rotation; parseNewObject(id) returns a PDF object that can be converted with methods such as toPDFDictionary() or toPDFArray().

extractPageText(pageIndex, limits?) enumerates text-showing operations with their decoded Unicode text, raw character codes, text matrix, and active font state, and extractPageContentItems(pageIndex, limits?) reports page-marking operations without reading text. Neither provides general visual-text or image-extraction. See Find Text Positions for extraction examples and limits.