summaryrefslogtreecommitdiff
path: root/core/include/fpdfapi
diff options
context:
space:
mode:
authorJUN FANG <jun_fang@foxitsoftware.com>2015-08-05 17:07:07 -0700
committerJUN FANG <jun_fang@foxitsoftware.com>2015-08-05 17:22:12 -0700
commitb14283e6b7f2521070106b070eb74219dfd97e1b (patch)
tree3cebf550ee709715fb3fda32f16679e5a0cd1e48 /core/include/fpdfapi
parentce17b97a1240525b88d7f94c842eac0f7f4051fa (diff)
downloadpdfium-b14283e6b7f2521070106b070eb74219dfd97e1b.tar.xz
Merge to XFA: Early object closing in PDF via premature 'endobj' keyword
Adobe acrobat defines keywords "\nendstream\n" and "\nendobj\n" in object's stream. However, Pdfium searches "endstream" and "endobj" as the keywords in the streams of objects. Some words like "(endstream)" are wrongly treated as keywords in pdfium. This fix follows Adobe's solution. BUG=493126 R=tsepez@chromium.org Review URL: https://codereview.chromium.org/1219133005 .
Diffstat (limited to 'core/include/fpdfapi')
-rw-r--r--core/include/fpdfapi/fpdf_parser.h2
1 files changed, 2 insertions, 0 deletions
diff --git a/core/include/fpdfapi/fpdf_parser.h b/core/include/fpdfapi/fpdf_parser.h
index 7640c857fa..cb4e15e25d 100644
--- a/core/include/fpdfapi/fpdf_parser.h
+++ b/core/include/fpdfapi/fpdf_parser.h
@@ -297,6 +297,8 @@ class CPDF_SyntaxParser {
CFX_ByteString ReadHexString();
+ unsigned int ReadEOLMarkers(FX_FILESIZE pos);
+
CPDF_Stream* ReadStream(CPDF_Dictionary* pDict,
PARSE_CONTEXT* pContext,
FX_DWORD objnum,