summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-04-12Rename CXFA_SimpleParser to CXFA_DocumentParserDan Sinclair
The CXFA_SimpleParser is handling the XFA document parsing. This CL renames the class to make it clearer that this is the XFA document parser. Change-Id: Ia290df2671fba9efa221ae997156ee4846749b02 Reviewed-on: https://pdfium-review.googlesource.com/30231 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-12Fold CXFA_DocumentParser into CXFA_FFDocDan Sinclair
The CXFA_DocumentParser is a thin wrapper around CXFA_SimpleParser. This CL folds CXFA_DocumentParser into a ParseDoc method of CXFA_FFDoc. Change-Id: I7e7073899436f0435c4ebfd548241ae842a2a276 Reviewed-on: https://pdfium-review.googlesource.com/30213 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-12Retrieve the XML tree in the DocumentParserDan Sinclair
This CL changes the document parser to store the parsed XML tree. That way we no longer need to store the pointer to the CXFA_SimpleParser in the CXFA_DocumentParser. Change-Id: I9272150e73613411cb1bb59c3cf77c7af6cc799d Reviewed-on: https://pdfium-review.googlesource.com/30211 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-12Allow retrieving the XML tree from the CFX_XMLDocDan Sinclair
This CL allows the CXFA_SimpleParser to retrieve the XML tree from the CFX_XMLDoc. This way, we don't have to keep the doc around and can store the pointer to the tree in the CXFA_SimpleParser. Change-Id: I5b478acbe61e6f1ca5fa04d03133a2b327a0cb1c Reviewed-on: https://pdfium-review.googlesource.com/30210 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-12Merge CFX_XMLParser and CFX_XMLSyntaxParserDan Sinclair
The CFX_XMLParser was a wrapper around the CFX_XMLSyntaxParser. This CL merges the SyntaxParser into protected/private methods if the XMLParser. Change-Id: If1519b5de55866ed14359dffd64dc12c36ee0244 Reviewed-on: https://pdfium-review.googlesource.com/30171 Reviewed-by: Ryan Harrison <rharrison@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-12Cleanup some CFX_XMLParser codeDan Sinclair
This CL cleans up minor nits in the CFX_XMLParser code. Change-Id: Ie19d12d3dcce16c9ce6088160ecdec3d9855c11f Reviewed-on: https://pdfium-review.googlesource.com/30170 Reviewed-by: Ryan Harrison <rharrison@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-12[xfa] Remove form checksum abilitiesDan Sinclair
This CL removes the code for verifying and creating checksums associated with form elements in XFA documents. This was the only code requiring the SAXReader which has also been removed. According to the XFA spec and application can decide which parts of the signatures are supported. This feature is being removed until we determine if/when it is needed. Bug: pdfium:1063 Change-Id: Iec2261282340f8fc72a1225d2e0d3e6ddf05edcb Reviewed-on: https://pdfium-review.googlesource.com/30150 Reviewed-by: Ryan Harrison <rharrison@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-11Remove CFX_DIBAttribute::m_strAuthorTom Sepez
It is set in a couple of places, but the value is never used for any purpose. Change-Id: I6fc0839bc14b21ee8217fcb3eadf6c252ad67aa7 Reviewed-on: https://pdfium-review.googlesource.com/30330 Reviewed-by: Lei Zhang <thestig@chromium.org> Commit-Queue: Tom Sepez <tsepez@chromium.org>
2018-04-11Fix crash when ColorSpace references itself directly.Henrique Nakashima
Also fixes any problems with cycles between colorspaces. Past fixes have solved problems with CPDF_DocPageData::GetColorSpace() calling itself and CPDF_DocPageData::GetColorSpace() calling CPDF_ColorSpace::Load() and vice versa. They have not solved CPDF_ColorSpace::Load() calling itself. This CL repurposes the |pVisited| set to ensure CPDF_ColorSpace::Load() does not try to load a colorspace as a dependency of itself and creates |pVisitedLocal| to ensure CPDF_DocPageData::GetColorSpace() does not create a similar circular dependency not involving CPDF_ColorSpace::Load(). Bug: chromium:828206 Change-Id: Ib2d0ec494be169135607f3651e0f70627b26ebd7 Reviewed-on: https://pdfium-review.googlesource.com/29810 Reviewed-by: Lei Zhang <thestig@chromium.org> Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
2018-04-11Fix issues with PDFium third_party/base/span.hTom Sepez
Remove stray const in operator[] that was introduced when downgrading from C++14 to C++11 syntax. Add missing Get() in first() that was introduced when converting to UnownedPtr. Prevent ASAN from flagging spans where the UnownedPtr points to byte N+1 of a N byte object, and the span is empty. This is legal in C for ordinary pointers so long as the pointer isn't de-referenced, but is not allowed per the rules for UnownedPtr. Change-Id: Ic143c5ef4e37c1cf86f0a3e5408be6e2076a85e2 Reviewed-on: https://pdfium-review.googlesource.com/30212 Commit-Queue: Tom Sepez <tsepez@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-11Remove use of GetBuffer()/ReleaseBuffer() when c_str() is sufficient.Tom Sepez
Greatly simplify the code which was neither changing allocations nor doing any actual work. Eventually GetBuffer() will return a span, and this makes fewer places to change. Comment AsFPDFWideString() as a place where we may want a redundant copy, and make available outside the .cpp file. Fix one "const" along the way. Change-Id: I187758b69a0ba3501b3622f6c06280793cd5d464 Reviewed-on: https://pdfium-review.googlesource.com/30011 Reviewed-by: dsinclair <dsinclair@chromium.org> Commit-Queue: Tom Sepez <tsepez@chromium.org>
2018-04-11Make cxfa_fmlexer.cpp resilient to null stringsTom Sepez
As currently written, the calculation of m_end will underflow when passed a {nullptr, 0} pair as input, and m_end becomes essentially unbounded. Change-Id: Id3249b201c446555d9aa4fa04e6a3c94a357cd99 Reviewed-on: https://pdfium-review.googlesource.com/30230 Reviewed-by: dsinclair <dsinclair@chromium.org> Commit-Queue: Tom Sepez <tsepez@chromium.org>
2018-04-11Fix trailing whitespace handling in CPDF_StreamParser::ReadInlineStream()Tom Sepez
After advancing past whitespace, the buffer may become empty. We were checking too early. Bug: 831498 Test: covered by fuzzers Change-Id: I6796b4723a05c2bcafe495dc7bc648213aa9cc22 Reviewed-on: https://pdfium-review.googlesource.com/30191 Commit-Queue: Tom Sepez <tsepez@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-11Reserve space to reduce memory operations while encoding barcodeRyan Harrison
In the test case from the bug, the majority of the time is being spent resizing Widestring internal buffers, since += is being called in a tight loop. Since the size of the input being mutated and stored is known, this CL reserves the space before hand to lower thrashing. This substantially improves runtime of this test case locally. BUG=chromium:802242 Change-Id: I5176dabc94634b4d6bc3e9425fe6469a5bf35a41 Reviewed-on: https://pdfium-review.googlesource.com/30190 Reviewed-by: Lei Zhang <thestig@chromium.org> Commit-Queue: Ryan Harrison <rharrison@chromium.org>
2018-04-11Add documentation about Valgrind LLD workaroundRyan Harrison
BUG=chromium:830706 Change-Id: I72ea6373d898953c10e523399655f1f630941b09 Reviewed-on: https://pdfium-review.googlesource.com/30110 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
2018-04-11Avoid stack overflow when loading CPDF_Function.chromium/3395Henrique Nakashima
CPDF_StitchFuncs that reference each other create a Load() loop. Maintaining a set of the visited CPDF_Objects during a Load() call tree prevents that. Bug: chromium:830221 Change-Id: I6f494da16c6d79f05870ff85cff38ff8fe69ecfe Reviewed-on: https://pdfium-review.googlesource.com/30050 Commit-Queue: dsinclair <dsinclair@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-10Add static_asserts for FX_RECT and FX_COLORREF.Lei Zhang
Make sure they match their win32 equivalent data types. Change-Id: I395054ce5c5ae110c78dbae30dd985fc498045fb Reviewed-on: https://pdfium-review.googlesource.com/30070 Reviewed-by: Tom Sepez <tsepez@chromium.org> Commit-Queue: Lei Zhang <thestig@chromium.org>
2018-04-10Load CIDToGIDMap stream for CID fonts if it exists.Lei Zhang
BUG=chromium:813705 Change-Id: I57265aad11d8f717f6ce33cc09241f4bef6381db Reviewed-on: https://pdfium-review.googlesource.com/29710 Reviewed-by: Nicolás Peña Moreno <npm@chromium.org> Commit-Queue: Lei Zhang <thestig@chromium.org>
2018-04-10Roll DEPS for Clang and buildRyan Harrison
Roll DEPS for Clang to d7c36b0a. https://chromium.googlesource.com/chromium/src/tools/clang/+log/7d56ff90..d7c36b0a Roll DEPS for build to 57e31076 https://chromium.googlesource.com/chromium/src/build/+log/74c3e063..57e31076 TBR=dsinclair@chromium.org,thestig@chromium.org Change-Id: I490ee17ad5d641917bb91349e25c46e2762481be Reviewed-on: https://pdfium-review.googlesource.com/30090 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Lei Zhang <thestig@chromium.org>
2018-04-10Add an assert to make sure all code is included in static libRyan Harrison
When building a complete static lib using pdf_is_complete_lib, assert that is_component_build=false, so that 'component' targets will be statically linked into the final product, instead of being dynamically linked in. We cannot have a rule like if pdf_is_complete_lib then is_component_build=false, because the is_component_build arg is defined in a file we import and overwriting the value later doesn't do what one expects, since the value has already been used in the imported file set things like env vars and build flags. Thus we assert the value and have a message about needing to set is_component_build=false. BUG=pdfium:1062 Change-Id: I83c430ca374e70c01f18e4b15a1eeb68aafa4444 Reviewed-on: https://pdfium-review.googlesource.com/30030 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Lei Zhang <thestig@chromium.org>
2018-04-10Remove CFX_Rect.Lei Zhang
It is rarely used and FX_RECT is the more common integer rect type. Change-Id: I7c5b875321c2d587becedcd058bb3a57fd1f0b61 Reviewed-on: https://pdfium-review.googlesource.com/30053 Reviewed-by: Tom Sepez <tsepez@chromium.org> Commit-Queue: Lei Zhang <thestig@chromium.org>
2018-04-10Change CFX_RenderDevice::FillRect() to take FX_RECT by const-ref.Lei Zhang
It currently takes const FX_RECT*, but the pointer is never nullptr. Change-Id: I571e9e8dd04756bc4daa25a61a5af8d1f902914b Reviewed-on: https://pdfium-review.googlesource.com/30052 Commit-Queue: Lei Zhang <thestig@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-10Change FillRectWithBlend methods to take FX_RECT by const-ref.Lei Zhang
They currently take const FX_RECT*, but the pointer is never nullptr. Also add a comment to explain why FX_RECT is the way it is. It has the same layout as a win32 RECT. Change-Id: Icf0e4c3eb25fe03317590a736578e053b9dccf7a Reviewed-on: https://pdfium-review.googlesource.com/30051 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-10Implement CPDFSDK_XFAWidgetHandler::OnKillFocus.Henrique Nakashima
This makes XFA widgets lose focus when a click is performed outside their area. Bug: chromium:820256 Change-Id: I7b70fafdc625089f3340a3494bfd51e5c8d0446d Reviewed-on: https://pdfium-review.googlesource.com/29860 Commit-Queue: Henrique Nakashima <hnakashima@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-10Break down CXFA_FFWidget::On{L|R}ButtonDown() into two steps.Henrique Nakashima
The first step, AcceptsFocusOnButtonDown(), returns whether the widget wants focus on a mouse click. The second step, On{L|R}ButtonDown(), handles the click after focus has been assigned. Bug: chromium:820256 Change-Id: I9da1f1a930f198fc17b24acb7f1e3e6ef7f12ae4 Reviewed-on: https://pdfium-review.googlesource.com/29951 Reviewed-by: Ryan Harrison <rharrison@chromium.org> Commit-Queue: Henrique Nakashima <hnakashima@chromium.org>
2018-04-10Merge CXFA_SimpleParser::{Start|Do}Parsechromium/3394Dan Sinclair
The StartParse and DoParse methods have been merged into a single Parse method. Change-Id: I1310bda90b853c11af8b8d1794de6b5c19e931cf Reviewed-on: https://pdfium-review.googlesource.com/30010 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-09Use ByteStringView / pdfium::span in CPDF font as appropriate.chromium/3393Tom Sepez
Change-Id: I92c7ba605bf95a9023ad046b8dddebe0a0592802 Reviewed-on: https://pdfium-review.googlesource.com/29992 Reviewed-by: dsinclair <dsinclair@chromium.org> Commit-Queue: Tom Sepez <tsepez@chromium.org>
2018-04-09Pass FX_RECT by const-ref in rendering code.Lei Zhang
Change-Id: I1232e6c61cbe696d90d87f461ebed0a7aea40893 Reviewed-on: https://pdfium-review.googlesource.com/29973 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-09Remove RenderDeviceDriverIface::GetCTM().Lei Zhang
It has no overrides and always returns the identity matrix. Many callers that use the returned value can be simplified, because they are scaling by 1 or concatenating an identity matrix. Change-Id: I7afb7214be210d02638644dfb9b58404420c2ef2 Reviewed-on: https://pdfium-review.googlesource.com/29972 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-09Cleanup CPDF_RenderStatus::GetObjectClippedRect().Lei Zhang
- Just return the rect directly and let the caller check for emptiness. - Get rid of a boolean parameter with always false argument. Change-Id: Ia33587db34c0a628047e59b7faa390c3ac57ec8a Reviewed-on: https://pdfium-review.googlesource.com/29971 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-09Merge CXFA_DocumentParser::{Start|Do}Parse methodsDan Sinclair
This CL combines the StartParse and DoParse methods of CXFA_DocumentParser into a single Parse method since they are always called together. Change-Id: I45f38b2730f47c4719ca4f54c32ee8db6f11470d Reviewed-on: https://pdfium-review.googlesource.com/29991 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-09Merge CXFA_FFDoc::{Start|Do|Stop}Load methodsDan Sinclair
The StartLoad, DoLoad and StopLoad methods are called in sequence from the CPDFXFA_Context. This CL merges the three methods into a single Load method. Change-Id: I7135502354fb4c64784c89fa39b74fc58c76b683 Reviewed-on: https://pdfium-review.googlesource.com/29990 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-09Use pdfium::span<> in CFX_BitStream, CPDF_SimpleParser.Tom Sepez
Get bounds checks in parsers automatically when using spans. Change-Id: I71fbe7b838435d455376db2f89817d807a9cdcfd Reviewed-on: https://pdfium-review.googlesource.com/29830 Commit-Queue: Tom Sepez <tsepez@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-09Make StringViewTemplate be based on pdfium::span<>.Tom Sepez
Change-Id: I13d3e462c86d035bbb992696aab523df5642533b Reviewed-on: https://pdfium-review.googlesource.com/29811 Commit-Queue: Tom Sepez <tsepez@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-09Revert "Remove CXFA nodes instead of CFX_XML nodes"dsinclair
This reverts commit 2b4ad9ff5bf1727d12ef2bcce7e3fa925bbe8978. Reason for revert: Causing crashes on windows: Rendering PDF file E:\b\build\slave\windows_xfa\build\pdfium\out\debug_xfa_v8\gen\pdfium\testing\pixel\static_list_box_caption.pdf. ==== C stack trace =============================== diff: 0.00% passed std::tuple<CXFA_Node * __ptr64 const & __ptr64>::tuple<CXFA_Node * __ptr64 const & __ptr64> [0x0000000141785700+48] ??@875fbed02c72fe689afc0ad24a247bf8@ [0x000000013F7BB443+131] ??$_Buynode@AEBUpiecewise_construct_t@std@@V?$tuple@AEBQEAVCXFA_Node@@@2@V?$tuple@$$V@2@@?$_Tree_comp_alloc@V?$_Tmap_traits@PEAVCXFA_Node@@V?$unique_ptr@V?$map@IV?$unique_ptr@U?$pair@V?$set@PEAVCXFA_Node@@U?$less@PEAVCXFA_Node@@@std@@V?$allocator@PEAVCXFA [0x0000000141785309+169] ??@2b14910fe4cc3866ca68e10bfed05253@ [0x000000013F7CCB09+185] CJX_Object::SetAttributeValue [0x000000013FAE0F7E+302] v8::internal::compiler::Int64Lowering::LowerNode [0x000000013FBB04C0+31920] CXFA_Document::DataMerge_CopyContainer [0x00000001417A43A3+9603] CXFA_Document::DataMerge_CopyContainer [0x00000001417A2048+552] CXFA_Document::DataMerge_CopyContainer [0x00000001417A3436+5654] CXFA_Document::DataMerge_CopyContainer [0x00000001417A1FEE+462] CXFA_Document::DoDataMerge [0x00000001417A5F96+3142] CXFA_FFDocView::StartLayout [0x00000001416BECCD+93] CPDFXFA_Context::LoadXFADoc [0x000000014169162D+669] FPDF_LoadXFA [0x0000000140AC5A2A+42] std::_Compressed_pair<std::default_delete<CFWL_WidgetProperties>,CFWL_WidgetProperties * __ptr64,1>::_Get_first [0x000000013F72A648+1432] main [0x000000013F727C05+1685] invoke_main [0x0000000141CDB924+52] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:79) __scrt_common_main_seh [0x0000000141CDB817+295] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:283) __scrt_common_main [0x0000000141CDB6DE+14] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:326) mainCRTStartup [0x0000000141CDB9B9+9] (f:\dd\vctools\crt\vcstartup\src\startup\exe_main.cpp:17) BaseThreadInitThunk [0x0000000077415A4D+13] RtlUserThreadStart [0x000000007764B831+33] Rendering PDF file E:\b\build\slave\windows_xfa\build\pdfium\out\debug_xfa_v8\gen\pdfium\testing\pixel\resolve_nodes.pdf. ==== C stack trace =============================== std::tuple<CXFA_Node * __ptr64 const & __ptr64>::tuple<CXFA_Node * __ptr64 const & __ptr64> [0x0000000141785700+48] ??@875fbed02c72fe689afc0ad24a247bf8@ [0x000000013F7BB443+131] ??$_Buynode@AEBUpiecewise_construct_t@std@@V?$tuple@AEBQEAVCXFA_Node@@@2@V?$tuple@$$V@2@@?$_Tree_comp_alloc@V?$_Tmap_traits@PEAVCXFA_Node@@V?$unique_ptr@V?$map@IV?$unique_ptr@U?$pair@V?$set@PEAVCXFA_Node@@U?$less@PEAVCXFA_Node@@@std@@V?$allocator@PEAVCXFA [0x0000000141785309+169] ??@2b14910fe4cc3866ca68e10bfed05253@ [0x000000013F7CCB09+185] CJX_Object::SetAttributeValue [0x000000013FAE0F7E+302] v8::internal::compiler::Int64Lowering::LowerNode [0x000000013FBB03F6+31718] CXFA_Document::DataMerge_UpdateBindingRelations [0x00000001417A48DA+1066] CXFA_Document::DataMerge_UpdateBindingRelations [0x00000001417A4ED8+2600] CXFA_Document::DataMerge_UpdateBindingRelations [0x00000001417A4ED8+2600] CXFA_Document::DataMerge_UpdateBindingRelations [0x00000001417A4ED8+2600] CXFA_Document::DataMerge_UpdateBindingRelations [0x00000001417A4ED8+2600] CXFA_Document::DataMerge_UpdateBindingRelations [0x00000001417A4512+98] CXFA_Document::DoDataMerge [0x00000001417A5FF8+3240] CXFA_FFDocView::StartLayout [0x00000001416BECCD+93] CPDFXFA_Context::LoadXFADoc [0x000000014169162D+669] FPDF_LoadXFA [0x0000000140AC5A2A+42] std::_Compressed_pair<std::default_delete<CFWL_WidgetProperties>,CFWL_WidgetProperties * __ptr64,1>::_Get_first [0x000000013F72A648+1432] main [0x000000013F727C05+1685] invoke_main [0x0000000141CDB924+52] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:79) __scrt_common_main_seh [0x0000000141CDB817+295] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:283) __scrt_common_main [0x0000000141CDB6DE+14] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:326) mainCRTStartup [0x0000000141CDB9B9+9] (f:\dd\vctools\crt\vcstartup\src\startup\exe_main.cpp:17) BaseThreadInitThunk [0x0000000077415A4D+13] RtlUserThreadStart [0x000000007764B831+33] FAILURE: static_list_box_caption.pdf; Command '['E:\\b\\build\\slave\\windows_xfa\\build\\pdfium\\out\\debug_xfa_v8\\pdfium_test.exe', '--send-events', '--png', '--md5', 'E:\\b\\build\\slave\\windows_xfa\\build\\pdfium\\out\\debug_xfa_v8\\gen\\pdfium\\testing\\pixel\\static_list_box_caption.pdf']' returned non-zero exit status -1073741819 Rendering PDF file E:\b\build\slave\windows_xfa\build\pdfium\out\debug_xfa_v8\gen\pdfium\testing\pixel\standard_symbols.pdf. ==== C stack trace =============================== std::tuple<CXFA_Node * __ptr64 const & __ptr64>::tuple<CXFA_Node * __ptr64 const & __ptr64> [0x0000000141785700+48] FAILURE: resolve_nodes.pdf; Command '['E:\\b\\build\\slave\\windows_xfa\\build\\pdfium\\out\\debug_xfa_v8\\pdfium_test.exe', '--send-events', '--png', '--md5', 'E:\\b\\build\\slave\\windows_xfa\\build\\pdfium\\out\\debug_xfa_v8\\gen\\pdfium\\testing\\pixel\\resolve_nodes.pdf']' returned non-zero exit status -1073741819 ??@875fbed02c72fe689afc0ad24a247bf8@ [0x000000013F7BB443+131] ??$_Buynode@AEBUpiecewise_construct_t@std@@V?$tuple@AEBQEAVCXFA_Node@@@2@V?$tuple@$$V@2@@?$_Tree_comp_alloc@V?$_Tmap_traits@PEAVCXFA_Node@@V?$unique_ptr@V?$map@IV?$unique_ptr@U?$pair@V?$set@PEAVCXFA_Node@@U?$less@PEAVCXFA_Node@@@std@@V?$allocator@PEAVCXFA [0x0000000141785309+169] ??@2b14910fe4cc3866ca68e10bfed05253@ [0x000000013F7CCB09+185] CJX_Object::SetAttributeValue [0x000000013FAE0F7E+302] v8::internal::compiler::Int64Lowering::LowerNode [0x000000013FBB04C0+31920] CXFA_Document::DataMerge_CopyContainer [0x00000001417A43A3+9603] CXFA_Document::DataMerge_CopyContainer [0x00000001417A2048+552] CXFA_Document::DataMerge_CopyContainer [0x00000001417A3436+5654] CXFA_Document::DataMerge_CopyContainer [0x00000001417A1FEE+462] CXFA_Document::DoDataMerge [0x00000001417A5F96+3142] CXFA_FFDocView::StartLayout [0x00000001416BECCD+93] CPDFXFA_Context::LoadXFADoc [0x000000014169162D+669] FPDF_LoadXFA [0x0000000140AC5A2A+42] std::_Compressed_pair<std::default_delete<CFWL_WidgetProperties>,CFWL_WidgetProperties * __ptr64,1>::_Get_first [0x000000013F72A648+1432] main [0x000000013F727C05+1685] invoke_main [0x0000000141CDB924+52] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:79) __scrt_common_main_seh [0x0000000141CDB817+295] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:283) __scrt_common_main [0x0000000141CDB6DE+14] (f:\dd\vctools\crt\vcstartup\src\startup\exe_common.inl:326) mainCRTStartup [0x0000000141CDB9B9+9] (f:\dd\vctools\crt\vcstartup\src\startup\exe_main.cpp:17) BaseThreadInitThunk [0x0000000077415A4D+13] RtlUserThreadStart [0x000000007764B831+33] FAILURE: standard_symbols.pdf; Command '['E:\\b\\build\\slave\\windows_xfa\\build\\pdfium\\out\\debug_xfa_v8\\pdfium_test.exe', '--send-events', '--png', '--md5', 'E:\\b\\build\\slave\\windows_xfa\\build\\pdfium\\out\\debug_xfa_v8\\gen\\pdfium\\testing\\pixel\\standard_symbols.pdf']' returned non-zero exit status -1073741819 Original change's description: > Remove CXFA nodes instead of CFX_XML nodes > > When we set data into the CJX nodes, we need to update the backing store > of the XFA nodes, which are the XML nodes. When we update the backing > store, there are cases where we need to remove the children (setting > CData on a node or setting the value of an Attribute which is backed by > a real node). > > Currently we call DeleteChildren on the XML node. This causes issues > when we cleanup the tree as XFA nodes may point to these deleted XML > nodes. > > Instead, this CL cleans up the XFA nodes which will in turn clean up the > XML nodes. This seems like the right thing to do as we no longer have > the backing nodes, so we should no longer have the XFA nodes on top of > them. > > Bug: chromium:813155 > Change-Id: I63d53fd56999ec9d3b3af2ab5952b374663f040c > Reviewed-on: https://pdfium-review.googlesource.com/29950 > Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> > Commit-Queue: dsinclair <dsinclair@chromium.org> TBR=thestig@chromium.org,dsinclair@chromium.org,hnakashima@chromium.org Change-Id: Ic6b2fbeb9f8217292da0a722e143d96f9972056a No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: chromium:813155 Reviewed-on: https://pdfium-review.googlesource.com/29974 Reviewed-by: dsinclair <dsinclair@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-09Make pdfium::span<> be based off of UnownedPtr<>.Tom Sepez
Because we can get the lifetime check for free if we do this. This requires adding some "constexpr/noexcept" to UnownedPtr to make the types line up with what span.h requires. Change-Id: I45918f8723122082036eed959f769644ab4c509f Reviewed-on: https://pdfium-review.googlesource.com/29672 Commit-Queue: Tom Sepez <tsepez@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-09Sort test cases in test_runner.py.Lei Zhang
Allow them to run a bit more consistently. Change-Id: I6b0f528a73f2c872813f4644aeae886ec04879f4 Reviewed-on: https://pdfium-review.googlesource.com/29930 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: Lei Zhang <thestig@chromium.org>
2018-04-09Move the CFX_XMLParser out of CXFA_SimpleParserDan Sinclair
This CL removes the CFX_XMLParser from CXFA_SimpleParser and constructs it in the CFX_XMLDoc. Change-Id: Id69b396f091892170a4c4654ac1d0fd65d1efbb6 Reviewed-on: https://pdfium-review.googlesource.com/29970 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-09Remove m_pOldFocusWidget from CXFA_FFDocView.Henrique Nakashima
It can be refactored to a local, then the local can be refactored out too. There should be no functional changes. Change-Id: I4b49cbf27736f856a567db5db952c03aba743818 Reviewed-on: https://pdfium-review.googlesource.com/29911 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Lei Zhang <thestig@chromium.org>
2018-04-09Remove CXFA nodes instead of CFX_XML nodesDan Sinclair
When we set data into the CJX nodes, we need to update the backing store of the XFA nodes, which are the XML nodes. When we update the backing store, there are cases where we need to remove the children (setting CData on a node or setting the value of an Attribute which is backed by a real node). Currently we call DeleteChildren on the XML node. This causes issues when we cleanup the tree as XFA nodes may point to these deleted XML nodes. Instead, this CL cleans up the XFA nodes which will in turn clean up the XML nodes. This seems like the right thing to do as we no longer have the backing nodes, so we should no longer have the XFA nodes on top of them. Bug: chromium:813155 Change-Id: I63d53fd56999ec9d3b3af2ab5952b374663f040c Reviewed-on: https://pdfium-review.googlesource.com/29950 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-09Move code to set XML value to CXFA_NodeDan Sinclair
This CL moves the code to set the XML node data from CJX_Node to CXFA_Node. The XML node is owned by the XFA node, not the CJX node so it makes more sense to have the modifications happen there. This combines the duplicate code in CJX_Node into a single SetToXML method. Bug: chromium:813155 Change-Id: I493725d1412688cb1a0d04bd9ae9fa5a36caebf3 Reviewed-on: https://pdfium-review.googlesource.com/29858 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: dsinclair <dsinclair@chromium.org>
2018-04-06Add span.h from chromium base.chromium/3392Tom Sepez
Allows indexing with better bounds-checking to occur. Some small modifications are required to deal with PDFium being intentionally held at C++11 compliance, not C++14. Use in one place as check on correctness. Change-Id: Id2875cf0a93980112bc536a93c4f9ec5306c0dac Reviewed-on: https://pdfium-review.googlesource.com/29671 Commit-Queue: Tom Sepez <tsepez@chromium.org> Reviewed-by: Chris Palmer <palmer@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-06Avoid uninitialized internal fields in V8 global objects.Tom Sepez
V8 won't do this for us when it creates a global object under the covers off of a template with a non-zero internal field count, instead just leaving it uninitialized. We were careful to set the iternal fields on the object we explicitly create, but there are these implicitly created ones as part of making a new context that need to be handled as well. BUG: pdfium:1005 Change-Id: Ic40bafc206ec5119cbceb58f0bb725693e7ddf80 Reviewed-on: https://pdfium-review.googlesource.com/29910 Reviewed-by: dsinclair <dsinclair@chromium.org> Commit-Queue: Tom Sepez <tsepez@chromium.org>
2018-04-06Fix typos in CXFA_FMParser.chromium/3391Lei Zhang
Change-Id: I0059e2ec25e90162a241a49e5d7b327c80330a33 Reviewed-on: https://pdfium-review.googlesource.com/29890 Commit-Queue: dsinclair <dsinclair@chromium.org> Reviewed-by: dsinclair <dsinclair@chromium.org>
2018-04-05Make Internal field usage in cfxjs_engine match README.docTom Sepez
As it turns out, this doesn't cause any bugs with the FXJS/FXJSE interaction since the magic values will never be present in the other slot, but the code looks wrong wrt. the document. Also fix an assert in FXJSE that our objects have two slots, and null appropriately (just a defensive measure). Also assert that one of our casts is valid. Change-Id: I3146fe58350da5e9b76e711d81480565dabd587f Reviewed-on: https://pdfium-review.googlesource.com/29859 Reviewed-by: dsinclair <dsinclair@chromium.org> Commit-Queue: Tom Sepez <tsepez@chromium.org>
2018-04-05Combine if/else blocks in CPDF_DIBSource::TranslateScanline24bpp().Lei Zhang
The two branches are very similar. Change-Id: I2a9502858e9ce997f28b17bc1051b71fa058c2c4 Reviewed-on: https://pdfium-review.googlesource.com/24350 Commit-Queue: Lei Zhang <thestig@chromium.org> Reviewed-by: Henrique Nakashima <hnakashima@chromium.org>
2018-04-05Split CPDF_DIBSource::TranslateScanline24bpp().Lei Zhang
Move some code into TranslateScanline24bppDefaultDecode(). Change-Id: I7e985c2abb9b2f9e41214531321f8f3846dd3a94 Reviewed-on: https://pdfium-review.googlesource.com/29857 Reviewed-by: Henrique Nakashima <hnakashima@chromium.org> Commit-Queue: Lei Zhang <thestig@chromium.org>
2018-04-05Fix const casts in CPDF_Encryptor.Lei Zhang
Change-Id: I4d7661716116969ae80fdac52f6b400a5ff68e50 Reviewed-on: https://pdfium-review.googlesource.com/29855 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-05Get rid of a const_cast in CPDF_FlateEncoder.Lei Zhang
Change-Id: I3ee8f8e9eb33d251dc00336bcd31e578e7ce5143 Reviewed-on: https://pdfium-review.googlesource.com/29854 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>
2018-04-05Remove some const_casts in CPDF_Creator.Lei Zhang
Change-Id: I7c639655c003527417255c134a4393d68cda3ef1 Reviewed-on: https://pdfium-review.googlesource.com/29852 Commit-Queue: Ryan Harrison <rharrison@chromium.org> Reviewed-by: Ryan Harrison <rharrison@chromium.org>