mirrors/pdf.js
1
0
Fork 0
mirror of https://github.com/mozilla/pdf.js.git synced 2025-04-22 16:18:08 +02:00
Code Activity

Make XRef_indexObjects more robust against bad PDF files (issue 5752)

Browse source 
This patch improves the detection of `xref` in files where it is followed by an arbitrary whitespace character (not just a line-breaking char).
It also adds a check for missing whitespace, e.g. `1 0 obj<<`, to speed up `readToken` for the PDF file in the referenced issue.
Finally, the patch also replaces a bunch of magic numbers with suitably named constants.

Fixes 5752.

Also improves 6243, but there are still issues.
This commit is contained in:
Jonas Jenwald 2015-08-21 16:57:08 +02:00
parent 23cb01c8af
commit 56a43a3181
3 changed files with 22 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

17
src/core/obj.js
View file

@ -1013,9 +1013,12 @@ var XRef = (function XRefClosure() {
indexObjects: function XRef_indexObjects() {
// Simple scan through the PDF content to find objects,
// trailers and XRef streams.
var TAB = 0x9, LF = 0xA, CR = 0xD, SPACE = 0x20;
var PERCENT = 0x25, LT = 0x3C;
function readToken(data, offset) {
var token = '', ch = data[offset];
while (ch !== 13 && ch !== 10) {
while (ch !== LF && ch !== CR && ch !== LT) {
if (++offset >= data.length) {
break;
}
@ -1047,6 +1050,9 @@ var XRef = (function XRefClosure() {
var endobjBytes = new Uint8Array([101, 110, 100, 111, 98, 106]);
var xrefBytes = new Uint8Array([47, 88, 82, 101, 102]);
// Clear out any existing entries, since they may be bogus.
this.entries.length = 0;
var stream = this.stream;
stream.pos = 0;
var buffer = stream.getBytes();
@ -1054,23 +1060,24 @@ var XRef = (function XRefClosure() {
var trailers = [], xrefStms = [];
while (position < length) {
var ch = buffer[position];
if (ch === 32 || ch === 9 || ch === 13 || ch === 10) {
if (ch === TAB || ch === LF || ch === CR || ch === SPACE) {
++position;
continue;
}
if (ch === 37) { // %-comment
if (ch === PERCENT) { // %-comment
do {
++position;
if (position >= length) {
break;
}
ch = buffer[position];
} while (ch !== 13 && ch !== 10);
} while (ch !== LF && ch !== CR);
continue;
}
var token = readToken(buffer, position);
var m;
if (token === 'xref') {
if (token.indexOf('xref') === 0 &&
(token.length === 4 || /\s/.test(token[4]))) {
position += skipUntil(buffer, position, trailerBytes);
trailers.push(position);
position += skipUntil(buffer, position, startxrefBytes);