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

Library for creating and modifying PDF files in JavaScript

import { PDFDictionary, PDFName } from '../pdf-objects';
import { PDFCatalog, PDFLinearizationParams, PDFPage, PDFPageTree, } from '../pdf-structures';
import { arrayToString, error, trimArrayAndRemoveComments } from '../../utils';
import { isIdentity, validate } from '../../utils/validate';
import parseArray from './parseArray';
import parseBool from './parseBool';
import parseHexString from './parseHexString';
import parseIndirectRef from './parseIndirectRef';
import parseName from './parseName';
import parseNull from './parseNull';
import parseNumber from './parseNumber';
import parseString from './parseString';
// prettier-ignore
var typeDict = function (dict) {
    if (dict.getMaybe('Linearized'))
        return PDFLinearizationParams.fromDict(dict);
    switch (dict.getMaybe('Type')) {
        case PDFName.from('Catalog'): return PDFCatalog.fromDict(dict);
        case PDFName.from('Pages'): return PDFPageTree.fromDict(dict);
        case PDFName.from('Page'): return PDFPage.fromDict(dict);
        default: return dict;
    }
};
/**
 * Accepts an array of bytes as input. Checks to see if the first characters in
 * the trimmed input make up a PDF Dictionary.
 *
 * If so, returns a tuple containing (1) an object representing the parsed
 * PDFDictionary and (2) a subarray of the input with the characters making up
 * the parsed dictionary removed. The "onParseDict" parse handler will also be
 * called with the PDFDictionary object.
 *
 * If not, null is returned.
 *
 * Note that the entries of the PDF Dictionary are recursively parsed, so the
 * appropriate parse handlers will be called when each entry of the dictionary
 * is parsed. The returned PDFDictionary's keys will be PDFName objects, and its
 * values will be PDFObjects.
 */
var parseDict = function (input, index, parseHandlers) {
    if (parseHandlers === void 0) { parseHandlers = {}; }
    var trimmed = trimArrayAndRemoveComments(input);
    if (arrayToString(trimmed, 0, 2) !== '<<')
        return undefined;
    var pdfDict = PDFDictionary.from(new Map(), index);
    // Recursively parse each entry in the dictionary
    var remainder = trimArrayAndRemoveComments(trimmed.subarray(2));
    while (arrayToString(trimArrayAndRemoveComments(remainder), 0, 2) !== '>>') {
        // Parse the key for this entry
        var _a = parseName(remainder) || error('Failed to parse dictionary key'), key = _a[0], r1 = _a[1];
        remainder = r1;
        // Parse the value for this entry
        var _b = parseName(remainder, parseHandlers) ||
            parseDict(remainder, index, parseHandlers) ||
            parseArray(remainder, index, parseHandlers) ||
            parseString(remainder, parseHandlers) ||
            parseIndirectRef(remainder, parseHandlers) ||
            parseNumber(remainder, parseHandlers) ||
            parseHexString(remainder, parseHandlers) ||
            parseBool(remainder, parseHandlers) ||
            parseNull(remainder, parseHandlers) ||
            error('Failed to parse dictionary value'), pdfObject = _b[0], r2 = _b[1];
        pdfDict.set(key, pdfObject);
        remainder = r2;
    }
    var remainderTrim = trimArrayAndRemoveComments(remainder);
    // Make sure the brackets are paired
    validate(arrayToString(remainderTrim, 0, 2), isIdentity('>>'), 'Mismatched brackets!');
    remainder = trimArrayAndRemoveComments(remainderTrim.subarray(2)); // Remove ending '>>' pair
    var typedDict = typeDict(pdfDict);
    if (parseHandlers.onParseDict)
        parseHandlers.onParseDict(typedDict);
    return [typedDict, remainder];
};
export default parseDict;