/* -*- Mode: Java; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ /* vim: set shiftwidth=2 tabstop=2 autoindent cindent expandtab: */ /* Copyright 2012 Mozilla Foundation * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. */ /* globals assert, calculateMD5, Catalog, Dict, error, info, isArray, isArrayBuffer, isName, isStream, isString, createPromiseCapability, Linearization, NullStream, PartialEvaluator, shadow, Stream, Lexer, StreamsSequenceStream, stringToPDFString, stringToBytes, Util, XRef, MissingDataException, Promise, Annotation, ObjectLoader, OperatorList */ 'use strict'; var Page = (function PageClosure() { var LETTER_SIZE_MEDIABOX = [0, 0, 612, 792]; function Page(pdfManager, xref, pageIndex, pageDict, ref, fontCache) { this.pdfManager = pdfManager; this.pageIndex = pageIndex; this.pageDict = pageDict; this.xref = xref; this.ref = ref; this.fontCache = fontCache; this.idCounters = { obj: 0 }; this.resourcesPromise = null; } Page.prototype = { getPageProp: function Page_getPageProp(key) { return this.pageDict.get(key); }, getInheritedPageProp: function Page_inheritPageProp(key) { var dict = this.pageDict; var value = dict.get(key); while (value === undefined) { dict = dict.get('Parent'); if (!dict) { break; } value = dict.get(key); } return value; }, get content() { return this.getPageProp('Contents'); }, get resources() { var value = this.getInheritedPageProp('Resources'); // For robustness: The spec states that a \Resources entry has to be // present, but can be empty. Some document omit it still. In this case // return an empty dictionary: if (value === undefined) { value = Dict.empty; } return shadow(this, 'resources', value); }, get mediaBox() { var obj = this.getInheritedPageProp('MediaBox'); // Reset invalid media box to letter size. if (!isArray(obj) || obj.length !== 4) { obj = LETTER_SIZE_MEDIABOX; } return shadow(this, 'mediaBox', obj); }, get view() { var mediaBox = this.mediaBox; var cropBox = this.getInheritedPageProp('CropBox'); if (!isArray(cropBox) || cropBox.length !== 4) { return shadow(this, 'view', mediaBox); } // From the spec, 6th ed., p.963: // "The crop, bleed, trim, and art boxes should not ordinarily // extend beyond the boundaries of the media box. If they do, they are // effectively reduced to their intersection with the media box." cropBox = Util.intersect(cropBox, mediaBox); if (!cropBox) { return shadow(this, 'view', mediaBox); } return shadow(this, 'view', cropBox); }, get annotationRefs() { return shadow(this, 'annotationRefs', this.getInheritedPageProp('Annots')); }, get rotate() { var rotate = this.getInheritedPageProp('Rotate') || 0; // Normalize rotation so it's a multiple of 90 and between 0 and 270 if (rotate % 90 !== 0) { rotate = 0; } else if (rotate >= 360) { rotate = rotate % 360; } else if (rotate < 0) { // The spec doesn't cover negatives, assume its counterclockwise // rotation. The following is the other implementation of modulo. rotate = ((rotate % 360) + 360) % 360; } return shadow(this, 'rotate', rotate); }, getContentStream: function Page_getContentStream() { var content = this.content; var stream; if (isArray(content)) { // fetching items var xref = this.xref; var i, n = content.length; var streams = []; for (i = 0; i < n; ++i) { streams.push(xref.fetchIfRef(content[i])); } stream = new StreamsSequenceStream(streams); } else if (isStream(content)) { stream = content; } else { // replacing non-existent page content with empty one stream = new NullStream(); } return stream; }, loadResources: function Page_loadResources(keys) { if (!this.resourcesPromise) { // TODO: add async getInheritedPageProp and remove this. this.resourcesPromise = this.pdfManager.ensure(this, 'resources'); } return this.resourcesPromise.then(function resourceSuccess() { var objectLoader = new ObjectLoader(this.resources.map, keys, this.xref); return objectLoader.load(); }.bind(this)); }, getOperatorList: function Page_getOperatorList(handler, intent) { var self = this; var pdfManager = this.pdfManager; var contentStreamPromise = pdfManager.ensure(this, 'getContentStream', []); var resourcesPromise = this.loadResources([ 'ExtGState', 'ColorSpace', 'Pattern', 'Shading', 'XObject', 'Font' // ProcSet // Properties ]); var partialEvaluator = new PartialEvaluator(pdfManager, this.xref, handler, this.pageIndex, 'p' + this.pageIndex + '_', this.idCounters, this.fontCache); var dataPromises = Promise.all([contentStreamPromise, resourcesPromise]); var pageListPromise = dataPromises.then(function(data) { var contentStream = data[0]; var opList = new OperatorList(intent, handler, self.pageIndex); handler.send('StartRenderPage', { transparency: partialEvaluator.hasBlendModes(self.resources), pageIndex: self.pageIndex, intent: intent }); return partialEvaluator.getOperatorList(contentStream, self.resources, opList).then(function () { return opList; }); }); var annotationsPromise = pdfManager.ensure(this, 'annotations'); return Promise.all([pageListPromise, annotationsPromise]).then( function(datas) { var pageOpList = datas[0]; var annotations = datas[1]; if (annotations.length === 0) { pageOpList.flush(true); return pageOpList; } var annotationsReadyPromise = Annotation.appendToOperatorList( annotations, pageOpList, pdfManager, partialEvaluator, intent); return annotationsReadyPromise.then(function () { pageOpList.flush(true); return pageOpList; }); }); }, extractTextContent: function Page_extractTextContent() { var handler = { on: function nullHandlerOn() {}, send: function nullHandlerSend() {} }; var self = this; var pdfManager = this.pdfManager; var contentStreamPromise = pdfManager.ensure(this, 'getContentStream', []); var resourcesPromise = this.loadResources([ 'ExtGState', 'XObject', 'Font' ]); var dataPromises = Promise.all([contentStreamPromise, resourcesPromise]); return dataPromises.then(function(data) { var contentStream = data[0]; var partialEvaluator = new PartialEvaluator(pdfManager, self.xref, handler, self.pageIndex, 'p' + self.pageIndex + '_', self.idCounters, self.fontCache); return partialEvaluator.getTextContent(contentStream, self.resources); }); }, getAnnotationsData: function Page_getAnnotationsData() { var annotations = this.annotations; var annotationsData = []; for (var i = 0, n = annotations.length; i < n; ++i) { annotationsData.push(annotations[i].getData()); } return annotationsData; }, get annotations() { var annotations = []; var annotationRefs = (this.annotationRefs || []); for (var i = 0, n = annotationRefs.length; i < n; ++i) { var annotationRef = annotationRefs[i]; var annotation = Annotation.fromRef(this.xref, annotationRef); if (annotation) { annotations.push(annotation); } } return shadow(this, 'annotations', annotations); } }; return Page; })(); /** * The `PDFDocument` holds all the data of the PDF file. Compared to the * `PDFDoc`, this one doesn't have any job management code. * Right now there exists one PDFDocument on the main thread + one object * for each worker. If there is no worker support enabled, there are two * `PDFDocument` objects on the main thread created. */ var PDFDocument = (function PDFDocumentClosure() { function PDFDocument(pdfManager, arg, password) { if (isStream(arg)) { init.call(this, pdfManager, arg, password); } else if (isArrayBuffer(arg)) { init.call(this, pdfManager, new Stream(arg), password); } else { error('PDFDocument: Unknown argument type'); } } function init(pdfManager, stream, password) { assert(stream.length > 0, 'stream must have data'); this.pdfManager = pdfManager; this.stream = stream; var xref = new XRef(this.stream, password, pdfManager); this.xref = xref; } function find(stream, needle, limit, backwards) { var pos = stream.pos; var end = stream.end; var strBuf = []; if (pos + limit > end) { limit = end - pos; } for (var n = 0; n < limit; ++n) { strBuf.push(String.fromCharCode(stream.getByte())); } var str = strBuf.join(''); stream.pos = pos; var index = backwards ? str.lastIndexOf(needle) : str.indexOf(needle); if (index === -1) { return false; /* not found */ } stream.pos += index; return true; /* found */ } var DocumentInfoValidators = { get entries() { // Lazily build this since all the validation functions below are not // defined until after this file loads. return shadow(this, 'entries', { Title: isString, Author: isString, Subject: isString, Keywords: isString, Creator: isString, Producer: isString, CreationDate: isString, ModDate: isString, Trapped: isName }); } }; PDFDocument.prototype = { parse: function PDFDocument_parse(recoveryMode) { this.setup(recoveryMode); try { // checking if AcroForm is present this.acroForm = this.catalog.catDict.get('AcroForm'); if (this.acroForm) { this.xfa = this.acroForm.get('XFA'); var fields = this.acroForm.get('Fields'); if ((!fields || !isArray(fields) || fields.length === 0) && !this.xfa) { // no fields and no XFA -- not a form (?) this.acroForm = null; } } } catch (ex) { info('Something wrong with AcroForm entry'); this.acroForm = null; } }, get linearization() { var linearization = null; if (this.stream.length) { try { linearization = Linearization.create(this.stream); } catch (err) { if (err instanceof MissingDataException) { throw err; } info(err); } } // shadow the prototype getter with a data property return shadow(this, 'linearization', linearization); }, get startXRef() { var stream = this.stream; var startXRef = 0; var linearization = this.linearization; if (linearization) { // Find end of first obj. stream.reset(); if (find(stream, 'endobj', 1024)) { startXRef = stream.pos + 6; } } else { // Find startxref by jumping backward from the end of the file. var step = 1024; var found = false, pos = stream.end; while (!found && pos > 0) { pos -= step - 'startxref'.length; if (pos < 0) { pos = 0; } stream.pos = pos; found = find(stream, 'startxref', step, true); } if (found) { stream.skip(9); var ch; do { ch = stream.getByte(); } while (Lexer.isSpace(ch)); var str = ''; while (ch >= 0x20 && ch <= 0x39) { // < '9' str += String.fromCharCode(ch); ch = stream.getByte(); } startXRef = parseInt(str, 10); if (isNaN(startXRef)) { startXRef = 0; } } } // shadow the prototype getter with a data property return shadow(this, 'startXRef', startXRef); }, get mainXRefEntriesOffset() { var mainXRefEntriesOffset = 0; var linearization = this.linearization; if (linearization) { mainXRefEntriesOffset = linearization.mainXRefEntriesOffset; } // shadow the prototype getter with a data property return shadow(this, 'mainXRefEntriesOffset', mainXRefEntriesOffset); }, // Find the header, remove leading garbage and setup the stream // starting from the header. checkHeader: function PDFDocument_checkHeader() { var stream = this.stream; stream.reset(); if (find(stream, '%PDF-', 1024)) { // Found the header, trim off any garbage before it. stream.moveStart(); // Reading file format version var MAX_VERSION_LENGTH = 12; var version = '', ch; while ((ch = stream.getByte()) > 0x20) { // SPACE if (version.length >= MAX_VERSION_LENGTH) { break; } version += String.fromCharCode(ch); } // removing "%PDF-"-prefix this.pdfFormatVersion = version.substring(5); return; } // May not be a PDF file, continue anyway. }, parseStartXRef: function PDFDocument_parseStartXRef() { var startXRef = this.startXRef; this.xref.setStartXRef(startXRef); }, setup: function PDFDocument_setup(recoveryMode) { this.xref.parse(recoveryMode); this.catalog = new Catalog(this.pdfManager, this.xref); }, get numPages() { var linearization = this.linearization; var num = linearization ? linearization.numPages : this.catalog.numPages; // shadow the prototype getter return shadow(this, 'numPages', num); }, get documentInfo() { var docInfo = { PDFFormatVersion: this.pdfFormatVersion, IsAcroFormPresent: !!this.acroForm, IsXFAPresent: !!this.xfa }; var infoDict; try { infoDict = this.xref.trailer.get('Info'); } catch (err) { info('The document information dictionary is invalid.'); } if (infoDict) { var validEntries = DocumentInfoValidators.entries; // Only fill the document info with valid entries from the spec. for (var key in validEntries) { if (infoDict.has(key)) { var value = infoDict.get(key); // Make sure the value conforms to the spec. if (validEntries[key](value)) { docInfo[key] = (typeof value !== 'string' ? value : stringToPDFString(value)); } else { info('Bad value in document info for "' + key + '"'); } } } } return shadow(this, 'documentInfo', docInfo); }, get fingerprint() { var xref = this.xref, hash, fileID = ''; if (xref.trailer.has('ID')) { hash = stringToBytes(xref.trailer.get('ID')[0]); } else { hash = calculateMD5(this.stream.bytes.subarray(0, 100), 0, 100); } for (var i = 0, n = hash.length; i < n; i++) { fileID += hash[i].toString(16); } return shadow(this, 'fingerprint', fileID); }, getPage: function PDFDocument_getPage(pageIndex) { return this.catalog.getPage(pageIndex); }, cleanup: function PDFDocument_cleanup() { return this.catalog.cleanup(); } }; return PDFDocument; })();