pdf-lib/es/core/pdf-parser/parseDocument.js

Library for creating and modifying PDF files in JavaScript

/* eslint-disable no-constant-condition */
import { error } from '../../utils';
import parseHeader from './parseHeader';
import parseIndirectObj from './parseIndirectObj';
import parseLinearization from './parseLinearization';
import { parseTrailer, parseTrailerWithoutDict } from './parseTrailer';
import parseXRefTable from './parseXRefTable';
/**
 * Accepts an array of bytes as input. Parses indirect objects from the input bytes
 * until an xref table or trailer is found. The "onParseIndirectObj" parse
 * handler is called with each indirect object that is parsed.
 *
 * Returns a subarray of the input bytes with the bytes making up the parsed
 * indirect objects removed.
 */
var parseBodySection = function (input, index, parseHandlers) {
    var remainder = input;
    while (true) {
        var result = parseIndirectObj(remainder, index, parseHandlers);
        if (!result)
            break;
        remainder = result[1];
    }
    return remainder;
};
/**
 * Accepts an array of bytes as input. Checks to see if the first characters in the
 * input make up an xref table followed by a trailer, or just a trailer. The
 * "onParseXRefTable" and "onParseTrailer" parseHandlers will be called with the
 * parsed objects.
 *
 * Returns a subarray of the input bytes with the bytes making up the parsed
 * objects removed.
 */
var parseFooterSection = function (input, index, parseHandlers) {
    var remainder = input;
    // Try to parse the XRef table (some PDFs omit the XRef table)
    var parsedXRef = parseXRefTable(input, parseHandlers);
    if (parsedXRef)
        remainder = parsedXRef[1];
    // Try to parse the trailer with and without dictionary, because some
    // malformatted documents are missing the dictionary.
    var parsedTrailer = parseTrailer(remainder, index, parseHandlers) ||
        parseTrailerWithoutDict(remainder, index, parseHandlers);
    if (!parsedTrailer)
        return undefined;
    remainder = parsedTrailer[1];
    return remainder;
};
/**
 * Accepts an array of bytes comprising a PDF document as input. Parses all the
 * objects in the file in a sequential fashion, beginning with the header and
 * ending with the last trailer.
 *
 * The XRef tables/streams in the input are not used to locate and parse objects
 * as needed. Rather, the whole document is parsed and stored in memory at once.
 */
var parseDocument = function (input, index, parseHandlers) {
    var cleaned = input;
    // Parse the document header
    var remainder;
    _a = parseHeader(cleaned, parseHandlers) || error('PDF is missing a header'), remainder = _a[1];
    // If document is linearized, we'll need to parse the linearization
    // dictionary and First-Page XRef table/stream next...
    var linearizationMatch = parseLinearization(remainder, index, parseHandlers);
    if (linearizationMatch)
        remainder = linearizationMatch[1];
    // Parse each body of the document and its corresponding footer.
    // (if document does not have update sections, loop will only occur once)
    while (remainder) {
        remainder = parseBodySection(remainder, index, parseHandlers);
        remainder = parseFooterSection(remainder, index, parseHandlers);
    }
    var _a;
};
export default parseDocument;