pdf.js/src/core/document.js
Jonas Jenwald b8ec518a1e Split the existing PDFFunction in two classes, a private PDFFunction and a public PDFFunctionFactory, and utilize the latter in PDFDocument to allow various code to access the methods of PDFFunction`
*Follow-up to PR 8909.*

This requires us to pass around `pdfFunctionFactory` to quite a lot of existing code, however I don't see another way of handling this while still guaranteeing that we can access `PDFFunction` as freely as in the old code.

Please note that the patch passes all tests locally (unit, font, reference), and I *very* much hope that we have sufficient test-coverage for the code in question to catch any typos/mistakes in the re-factoring.
2017-09-29 15:30:53 +02:00

633 lines
20 KiB
JavaScript

/* 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.
*/
import { Catalog, ObjectLoader, XRef } from './obj';
import { Dict, isDict, isName, isStream } from './primitives';
import {
info, isArrayBuffer, isNum, isSpace, isString, MissingDataException, OPS,
shadow, stringToBytes, stringToPDFString, Util, warn
} from '../shared/util';
import { NullStream, Stream, StreamsSequenceStream } from './stream';
import { OperatorList, PartialEvaluator } from './evaluator';
import { AnnotationFactory } from './annotation';
import { calculateMD5 } from './crypto';
import { Linearization } from './parser';
import { PDFFunctionFactory } from './function';
var Page = (function PageClosure() {
var DEFAULT_USER_UNIT = 1.0;
var LETTER_SIZE_MEDIABOX = [0, 0, 612, 792];
function isAnnotationRenderable(annotation, intent) {
return (intent === 'display' && annotation.viewable) ||
(intent === 'print' && annotation.printable);
}
function Page({ pdfManager, xref, pageIndex, pageDict, ref, fontCache,
builtInCMapCache, pdfFunctionFactory, }) {
this.pdfManager = pdfManager;
this.pageIndex = pageIndex;
this.pageDict = pageDict;
this.xref = xref;
this.ref = ref;
this.fontCache = fontCache;
this.builtInCMapCache = builtInCMapCache;
this.pdfFunctionFactory = pdfFunctionFactory;
this.evaluatorOptions = pdfManager.evaluatorOptions;
this.resourcesPromise = null;
var uniquePrefix = 'p' + this.pageIndex + '_';
var idCounters = {
obj: 0,
};
this.idFactory = {
createObjId() {
return uniquePrefix + (++idCounters.obj);
},
};
}
Page.prototype = {
getPageProp: function Page_getPageProp(key) {
return this.pageDict.get(key);
},
getInheritedPageProp: function Page_getInheritedPageProp(key, getArray) {
var dict = this.pageDict, valueArray = null, loopCount = 0;
var MAX_LOOP_COUNT = 100;
getArray = getArray || false;
// Always walk up the entire parent chain, to be able to find
// e.g. \Resources placed on multiple levels of the tree.
while (dict) {
var value = getArray ? dict.getArray(key) : dict.get(key);
if (value !== undefined) {
if (!valueArray) {
valueArray = [];
}
valueArray.push(value);
}
if (++loopCount > MAX_LOOP_COUNT) {
warn('getInheritedPageProp: maximum loop count exceeded for ' + key);
return valueArray ? valueArray[0] : undefined;
}
dict = dict.get('Parent');
}
if (!valueArray) {
return undefined;
}
if (valueArray.length === 1 || !isDict(valueArray[0])) {
return valueArray[0];
}
return Dict.merge(this.xref, valueArray);
},
get content() {
return this.getPageProp('Contents');
},
get 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
// we return an empty dictionary.
return shadow(this, 'resources',
this.getInheritedPageProp('Resources') || Dict.empty);
},
get mediaBox() {
var mediaBox = this.getInheritedPageProp('MediaBox', true);
// Reset invalid media box to letter size.
if (!Array.isArray(mediaBox) || mediaBox.length !== 4) {
return shadow(this, 'mediaBox', LETTER_SIZE_MEDIABOX);
}
return shadow(this, 'mediaBox', mediaBox);
},
get cropBox() {
var cropBox = this.getInheritedPageProp('CropBox', true);
// Reset invalid crop box to media box.
if (!Array.isArray(cropBox) || cropBox.length !== 4) {
return shadow(this, 'cropBox', this.mediaBox);
}
return shadow(this, 'cropBox', cropBox);
},
get userUnit() {
var obj = this.getPageProp('UserUnit');
if (!isNum(obj) || obj <= 0) {
obj = DEFAULT_USER_UNIT;
}
return shadow(this, 'userUnit', obj);
},
get view() {
// 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."
var mediaBox = this.mediaBox, cropBox = this.cropBox;
if (mediaBox === cropBox) {
return shadow(this, 'view', mediaBox);
}
var intersection = Util.intersect(cropBox, mediaBox);
return shadow(this, 'view', intersection || mediaBox);
},
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 (Array.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(() => {
let objectLoader = new ObjectLoader(this.resources, keys, this.xref);
return objectLoader.load();
});
},
getOperatorList({ handler, task, intent, renderInteractiveForms, }) {
var contentStreamPromise = this.pdfManager.ensure(this,
'getContentStream');
var resourcesPromise = this.loadResources([
'ExtGState',
'ColorSpace',
'Pattern',
'Shading',
'XObject',
'Font'
// ProcSet
// Properties
]);
var partialEvaluator = new PartialEvaluator({
pdfManager: this.pdfManager,
xref: this.xref,
handler,
pageIndex: this.pageIndex,
idFactory: this.idFactory,
fontCache: this.fontCache,
builtInCMapCache: this.builtInCMapCache,
options: this.evaluatorOptions,
pdfFunctionFactory: this.pdfFunctionFactory,
});
var dataPromises = Promise.all([contentStreamPromise, resourcesPromise]);
var pageListPromise = dataPromises.then(([contentStream]) => {
var opList = new OperatorList(intent, handler, this.pageIndex);
handler.send('StartRenderPage', {
transparency: partialEvaluator.hasBlendModes(this.resources),
pageIndex: this.pageIndex,
intent,
});
return partialEvaluator.getOperatorList({
stream: contentStream,
task,
resources: this.resources,
operatorList: opList,
}).then(function () {
return opList;
});
});
// Fetch the page's annotations and add their operator lists to the
// page's operator list to render them.
var annotationsPromise = this.pdfManager.ensure(this, 'annotations');
return Promise.all([pageListPromise, annotationsPromise]).then(
function ([pageOpList, annotations]) {
if (annotations.length === 0) {
pageOpList.flush(true);
return pageOpList;
}
// Collect the operator list promises for the annotations. Each promise
// is resolved with the complete operator list for a single annotation.
var i, ii, opListPromises = [];
for (i = 0, ii = annotations.length; i < ii; i++) {
if (isAnnotationRenderable(annotations[i], intent)) {
opListPromises.push(annotations[i].getOperatorList(
partialEvaluator, task, renderInteractiveForms));
}
}
return Promise.all(opListPromises).then(function(opLists) {
pageOpList.addOp(OPS.beginAnnotations, []);
for (i = 0, ii = opLists.length; i < ii; i++) {
pageOpList.addOpList(opLists[i]);
}
pageOpList.addOp(OPS.endAnnotations, []);
pageOpList.flush(true);
return pageOpList;
});
});
},
extractTextContent({ handler, task, normalizeWhitespace,
sink, combineTextItems, }) {
var contentStreamPromise = this.pdfManager.ensure(this,
'getContentStream');
var resourcesPromise = this.loadResources([
'ExtGState',
'XObject',
'Font'
]);
var dataPromises = Promise.all([contentStreamPromise, resourcesPromise]);
return dataPromises.then(([contentStream]) => {
var partialEvaluator = new PartialEvaluator({
pdfManager: this.pdfManager,
xref: this.xref,
handler,
pageIndex: this.pageIndex,
idFactory: this.idFactory,
fontCache: this.fontCache,
builtInCMapCache: this.builtInCMapCache,
options: this.evaluatorOptions,
pdfFunctionFactory: this.pdfFunctionFactory,
});
return partialEvaluator.getTextContent({
stream: contentStream,
task,
resources: this.resources,
normalizeWhitespace,
combineTextItems,
sink,
});
});
},
getAnnotationsData: function Page_getAnnotationsData(intent) {
var annotations = this.annotations;
var annotationsData = [];
for (var i = 0, n = annotations.length; i < n; ++i) {
if (!intent || isAnnotationRenderable(annotations[i], intent)) {
annotationsData.push(annotations[i].data);
}
}
return annotationsData;
},
get annotations() {
var annotations = [];
var annotationRefs = this.getInheritedPageProp('Annots') || [];
for (var i = 0, n = annotationRefs.length; i < n; ++i) {
var annotationRef = annotationRefs[i];
var annotation = AnnotationFactory.create(this.xref, annotationRef,
this.pdfManager,
this.idFactory);
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() {
var FINGERPRINT_FIRST_BYTES = 1024;
var EMPTY_FINGERPRINT = '\x00\x00\x00\x00\x00\x00\x00' +
'\x00\x00\x00\x00\x00\x00\x00\x00\x00';
function PDFDocument(pdfManager, arg) {
var stream;
if (isStream(arg)) {
stream = arg;
} else if (isArrayBuffer(arg)) {
stream = new Stream(arg);
} else {
throw new Error('PDFDocument: Unknown argument type');
}
if (stream.length <= 0) {
throw new Error('PDFDocument: stream must have data');
}
this.pdfManager = pdfManager;
this.stream = stream;
this.xref = new XRef(stream, pdfManager);
let evaluatorOptions = pdfManager.evaluatorOptions;
this.pdfFunctionFactory = new PDFFunctionFactory({
xref: this.xref,
isEvalSupported: evaluatorOptions.isEvalSupported,
});
}
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);
var version = this.catalog.catDict.get('Version');
if (isName(version)) {
this.pdfFormatVersion = version.name;
}
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 || !Array.isArray(fields) || fields.length === 0) &&
!this.xfa) {
// no fields and no XFA -- not a form (?)
this.acroForm = null;
}
}
} catch (ex) {
if (ex instanceof MissingDataException) {
throw 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 (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);
}
if (!this.pdfFormatVersion) {
// 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);
var pageFactory = {
createPage: (pageIndex, dict, ref, fontCache, builtInCMapCache) => {
return new Page({
pdfManager: this.pdfManager,
xref: this.xref,
pageIndex,
pageDict: dict,
ref,
fontCache,
builtInCMapCache,
pdfFunctionFactory: this.pdfFunctionFactory,
});
},
};
this.catalog = new Catalog(this.pdfManager, this.xref, pageFactory);
},
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) {
if (err instanceof MissingDataException) {
throw 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 = '';
var idArray = xref.trailer.get('ID');
if (Array.isArray(idArray) && idArray[0] && isString(idArray[0]) &&
idArray[0] !== EMPTY_FINGERPRINT) {
hash = stringToBytes(idArray[0]);
} else {
if (this.stream.ensureRange) {
this.stream.ensureRange(0,
Math.min(FINGERPRINT_FIRST_BYTES, this.stream.end));
}
hash = calculateMD5(this.stream.bytes.subarray(0,
FINGERPRINT_FIRST_BYTES), 0, FINGERPRINT_FIRST_BYTES);
}
for (var i = 0, n = hash.length; i < n; i++) {
var hex = hash[i].toString(16);
fileID += hex.length === 1 ? '0' + hex : hex;
}
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;
})();
export {
Page,
PDFDocument,
};