Sakurai/pdf.js
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[api-minor] Let LinkAnnotation/PDFLinkService_getDestinationHash return a stringified version of the destination array for explicit destinations

Browse Source 
Currently for explicit destinations, compared to named destinations, we manually try to build a hash that often times is a quite poor representation of the *actual* destination. (Currently this only, kind of, works for `\XYZ` destinations.)
For PDF files using explicit destinations, this can make it difficult/impossible to obtain a link to a specific section of the document through the URL.

Note that in practice most PDF files, especially newer ones, use named destinations and these are thus unnaffected by this patch.
This patch also fixes an existing issue in `PDFLinkService_getDestinationHash`, where a named destination consisting of only a number would not be handled correctly.

With the added, and already existing, type checks in place for destinations, I really don't think that this patch exposes any "sensitive" internal destination code not already accessible through normal hash parameters.

*Please note:* Just trying to improve the algorithm that generates the hash is unfortunately not possible in general, since there are a number of cases where it will simply never work well.

 - First of all, note that `getDestinationHash` currently relies on the `_pagesRefCache`, hence it's possible that the hash returned is empty during e.g. ranged/streamed loading of a PDF file.

 - Second of all, the currently computed hash is actually dependent on the document rotation. With named destinations, the fetched internal destination array is rotational invariant (as it should be), but this will not hold in general for the hash. We can easily avoid this issue by using a stringified destination array.

 - Third of all, note that according to the PDF specification[1], `GoToR` destinations may actually contain explicit destination arrays. Since we cannot really construct a hash in `annotation.js`, we currently have no good way to support those. Even though this case seems *very* rare in practice (I've not actually seen such a PDF file), it's in the specification, and this patch allows us to support that for "free".

---
[1] http://www.adobe.com/content/dam/Adobe/en/devnet/acrobat/pdfs/PDF32000_2008.pdf#G11.1951685
This commit is contained in:
Jonas Jenwald 2016-05-15 12:12:47 +02:00
parent 01ab15a6f1
commit b354682dd6
5 changed files with 155 additions and 39 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/core/annotation.js
View File

@ -743,9 +743,17 @@ var LinkAnnotation = (function LinkAnnotationClosure() {
if (isName(remoteDest)) {
remoteDest = remoteDest.name;
}
if (isString(remoteDest) && isString(url)) {
if (isString(url)) {
var baseUrl = url.split('#')[0];
url = baseUrl + '#' + remoteDest;
if (isString(remoteDest)) {
// In practice, a named destination may contain only a number.
// If that happens, use the '#nameddest=' form to avoid the link
// redirecting to a page, instead of the correct destination.
url = baseUrl + '#' +
(/^\d+$/.test(remoteDest) ? 'nameddest=' : '') + remoteDest;
} else if (isArray(remoteDest)) {
url = baseUrl + '#' + JSON.stringify(remoteDest);
}
}
}
// The 'NewWindow' property, equal to `LinkTarget.BLANK`.

58
test/unit/annotation_layer_spec.js
View File

@ -268,8 +268,9 @@ describe('Annotation layer', function() {
var actionDict = new Dict();
actionDict.set('Type', Name.get('Action'));
actionDict.set('S', Name.get('GoToR'));
actionDict.set('F', '../../0021/002156/215675E.pdf');
actionDict.set('D', '15');
actionDict.set('F', '../../0013/001346/134685E.pdf');
actionDict.set('D', '4.3');
actionDict.set('NewWindow', true);
var annotationDict = new Dict();
annotationDict.set('Type', Name.get('Annot'));
@ -283,7 +284,58 @@ describe('Annotation layer', function() {
var data = annotation.data;
expect(data.annotationType).toEqual(AnnotationType.LINK);
expect(data.url).toBeUndefined();
expect(data.url).toBeUndefined(); // ../../0013/001346/134685E.pdf#4.3
expect(data.dest).toBeUndefined();
expect(data.newWindow).toEqual(true);
});
it('should correctly parse a GoToR action, with named destination',
function() {
var actionDict = new Dict();
actionDict.set('Type', Name.get('Action'));
actionDict.set('S', Name.get('GoToR'));
actionDict.set('F', 'http://www.example.com/test.pdf');
actionDict.set('D', '15');
var annotationDict = new Dict();
annotationDict.set('Type', Name.get('Annot'));
annotationDict.set('Subtype', Name.get('Link'));
annotationDict.set('A', actionDict);
var xrefMock = new XrefMock([annotationDict]);
var annotationRef = new Ref(495, 0);
var annotation = annotationFactory.create(xrefMock, annotationRef);
var data = annotation.data;
expect(data.annotationType).toEqual(AnnotationType.LINK);
expect(data.url).toEqual('http://www.example.com/test.pdf#nameddest=15');
expect(data.dest).toBeUndefined();
expect(data.newWindow).toBeFalsy();
});
it('should correctly parse a GoToR action, with explicit destination array',
function() {
var actionDict = new Dict();
actionDict.set('Type', Name.get('Action'));
actionDict.set('S', Name.get('GoToR'));
actionDict.set('F', 'http://www.example.com/test.pdf');
actionDict.set('D', [14, Name.get('XYZ'), null, 298.043, null]);
var annotationDict = new Dict();
annotationDict.set('Type', Name.get('Annot'));
annotationDict.set('Subtype', Name.get('Link'));
annotationDict.set('A', actionDict);
var xrefMock = new XrefMock([annotationDict]);
var annotationRef = new Ref(489, 0);
var annotation = annotationFactory.create(xrefMock, annotationRef);
var data = annotation.data;
expect(data.annotationType).toEqual(AnnotationType.LINK);
expect(data.url).toEqual('http://www.example.com/test.pdf#' +
'[14,{"name":"XYZ"},null,298.043,null]');
expect(data.dest).toBeUndefined();
expect(data.newWindow).toBeFalsy();
});

6
web/app.js
View File

@ -857,6 +857,7 @@ var PDFViewerApplication = {
pdfViewer.setDocument(pdfDocument);
var firstPagePromise = pdfViewer.firstPagePromise;
var pagesPromise = pdfViewer.pagesPromise;
var onePageRendered = pdfViewer.onePageRendered;
this.pageRotation = 0;
@ -962,9 +963,8 @@ var PDFViewerApplication = {
}
});
// outline depends on pagesRefMap
var promises = [pagesPromise, this.animationStartedPromise];
Promise.all(promises).then(function() {
Promise.all([onePageRendered, this.animationStartedPromise]).then(
function() {
pdfDocument.getOutline().then(function(outline) {
self.pdfOutlineViewer.render({ outline: outline });
});

116
web/pdf_link_service.js
View File

@ -29,6 +29,11 @@
var parseQueryString = uiUtils.parseQueryString;
var PageNumberRegExp = /^\d+$/;
function isPageNumber(str) {
return PageNumberRegExp.test(str);
}
/**
* @typedef {Object} PDFLinkServiceOptions
* @property {EventBus} eventBus - The application event bus.
@ -40,7 +45,7 @@ var parseQueryString = uiUtils.parseQueryString;
* @class
* @implements {IPDFLinkService}
*/
var PDFLinkService = (function () {
var PDFLinkService = (function PDFLinkServiceClosure() {
/**
* @constructs PDFLinkService
* @param {PDFLinkServiceOptions} options
@ -100,7 +105,7 @@ var PDFLinkService = (function () {
var self = this;
var goToDestination = function(destRef) {
// dest array looks like that: <page-ref> </XYZ|FitXXX> <args..>
// dest array looks like that: <page-ref> </XYZ|/FitXXX> <args..>
var pageNumber = destRef instanceof Object ?
self._pagesRefCache[destRef.num + ' ' + destRef.gen + ' R'] :
(destRef + 1);
@ -150,30 +155,15 @@ var PDFLinkService = (function () {
*/
getDestinationHash: function PDFLinkService_getDestinationHash(dest) {
if (typeof dest === 'string') {
return this.getAnchorUrl('#' + escape(dest));
// In practice, a named destination may contain only a number.
// If that happens, use the '#nameddest=' form to avoid the link
// redirecting to a page, instead of the correct destination.
return this.getAnchorUrl(
'#' + (isPageNumber(dest) ? 'nameddest=' : '') + escape(dest));
}
if (dest instanceof Array) {
var destRef = dest[0]; // see navigateTo method for dest format
var pageNumber = destRef instanceof Object ?
this._pagesRefCache[destRef.num + ' ' + destRef.gen + ' R'] :
(destRef + 1);
if (pageNumber) {
var pdfOpenParams = this.getAnchorUrl('#page=' + pageNumber);
var destKind = dest[1];
if (typeof destKind === 'object' && 'name' in destKind &&
destKind.name === 'XYZ') {
var scale = (dest[4] || this.pdfViewer.currentScaleValue);
var scaleNumber = parseFloat(scale);
if (scaleNumber) {
scale = scaleNumber * 100;
}
pdfOpenParams += '&zoom=' + scale;
if (dest[2] || dest[3]) {
pdfOpenParams += ',' + (dest[2] || 0) + ',' + (dest[3] || 0);
}
}
return pdfOpenParams;
}
var str = JSON.stringify(dest);
return this.getAnchorUrl('#' + escape(str));
}
return this.getAnchorUrl('');
},
@ -192,6 +182,7 @@ var PDFLinkService = (function () {
* @param {string} hash
*/
setHash: function PDFLinkService_setHash(hash) {
var pageNumber, dest;
if (hash.indexOf('=') >= 0) {
var params = parseQueryString(hash);
// borrowing syntax from "Parameters for Opening PDF Files"
@ -202,7 +193,6 @@ var PDFLinkService = (function () {
this.navigateTo(params.nameddest);
return;
}
var pageNumber, dest;
if ('page' in params) {
pageNumber = (params.page | 0) || 1;
}
@ -252,13 +242,23 @@ var PDFLinkService = (function () {
mode: params.pagemode
});
}
} else if (/^\d+$/.test(hash)) { // page number
this.page = hash;
} else { // named destination
if (this.pdfHistory) {
this.pdfHistory.updateNextHashParam(unescape(hash));
} else if (isPageNumber(hash)) { // Page number.
this.page = hash | 0;
} else { // Named (or explicit) destination.
dest = unescape(hash);
try {
dest = JSON.parse(dest);
} catch (ex) {}
if (typeof dest === 'string' || isValidExplicitDestination(dest)) {
if (this.pdfHistory) {
this.pdfHistory.updateNextHashParam(dest);
}
this.navigateTo(dest);
return;
}
this.navigateTo(unescape(hash));
console.error('PDFLinkService_setHash: \'' + unescape(hash) +
'\' is not a valid destination.');
}
},
@ -316,6 +316,60 @@ var PDFLinkService = (function () {
}
};
function isValidExplicitDestination(dest) {
if (!(dest instanceof Array)) {
return false;
}
var destLength = dest.length, allowNull = true;
if (destLength < 2) {
return false;
}
var page = dest[0];
if (!(typeof page === 'object' &&
typeof page.num === 'number' && (page.num | 0) === page.num &&
typeof page.gen === 'number' && (page.gen | 0) === page.gen) &&
!(typeof page === 'number' && (page | 0) === page && page >= 0)) {
return false;
}
var zoom = dest[1];
if (!(typeof zoom === 'object' && typeof zoom.name === 'string')) {
return false;
}
switch (zoom.name) {
case 'XYZ':
if (destLength !== 5) {
return false;
}
break;
case 'Fit':
case 'FitB':
return destLength === 2;
case 'FitH':
case 'FitBH':
case 'FitV':
case 'FitBV':
if (destLength !== 3) {
return false;
}
break;
case 'FitR':
if (destLength !== 6) {
return false;
}
allowNull = false;
break;
default:
return false;
}
for (var i = 2; i < destLength; i++) {
var param = dest[i];
if (!(typeof param === 'number' || (allowNull && param === null))) {
return false;
}
}
return true;
}
return PDFLinkService;
})();

2
web/pdf_viewer.js
View File

@ -587,6 +587,8 @@ var PDFViewer = (function pdfViewer() {
scale = Math.min(Math.abs(widthScale), Math.abs(heightScale));
break;
default:
console.error('PDFViewer_scrollPageIntoView: \'' + dest[1].name +
'\' is not a valid destination type.');
return;
}