summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2016-05-18Remove Release() from CFDE_TxtEdit* classestsepez
Review-Url: https://codereview.chromium.org/1987223003
2016-05-18Remove Release() from CXFA_FWLThemetsepez
Cleanup virtual overrides along the way. Add final to other class deriving from pure virtual interface. Review-Url: https://codereview.chromium.org/1988373002
2016-05-18Remove Release() from CXFA_ScriptContexttsepez
Review-Url: https://codereview.chromium.org/1988263002
2016-05-18Cleanup XFA_HASHCODE usage.dsinclair
This CL cleans up several uses of XFA_HASHCODE. The defines have been converted into an enum. For the SavePackage call the type is now used as the param instead of a string. The callers pass in the correct type instead of doing an internal conversion. The GetXFAObject accepting a string was removed as it was unused. The other variant was changed to accept the XFA_HashCode type instead of uint32_t. GetPackageData was removed as it is unused. Review-Url: https://codereview.chromium.org/1989313002
2016-05-18Remove Release() from CFX_SAXReadertsepez
Review-Url: https://codereview.chromium.org/1991993002
2016-05-18Remove Release() from CXFA_RenderContexttsepez
Review-Url: https://codereview.chromium.org/1994733002
2016-05-18Remove Release() from CXFA_WidgetAccIteratortsepez
Review-Url: https://codereview.chromium.org/1990793004
2016-05-18Remove Release() from CXFA_ChecksumContexttsepez
Review-Url: https://codereview.chromium.org/1991913002
2016-05-18Split xfa/fee files into individual class files.dsinclair
This CL splits the xfa/fee files into individual class files and moves them into the xfa/fde directory where they belong. Review-Url: https://codereview.chromium.org/1994693002
2016-05-18Use std::vector in CPDF_StructTreeImpltsepez
Allows use of CFX_RetainPtrs in place of explicit refcounting. Review-Url: https://codereview.chromium.org/1985253002
2016-05-18Remove c_str() from widestring checksum variant.tsepez
Brings it back in line with bytestring version. Review-Url: https://codereview.chromium.org/1985223002
2016-05-18Pass objects instead of strings for undo/redo records.dsinclair
Currently the Undo/Redo records are serialized as byte strings and stored into a CFX_ByteStringArray. They are deserialized when used. This CL removes the serialization and stores the objects in a deque of unique pointers. Review-Url: https://codereview.chromium.org/1980293004
2016-05-17Remove some CFX_ArrayTemplate<> usage from fxjsetsepez
Allows use of otherwise incompatible unique_ptrs. Review-Url: https://codereview.chromium.org/1992513002
2016-05-17Replace XFA_N with valuedsinclair
The XFA_N define doesn't add anything over just using the value, replaced. The static arrays and method have been moved to an anonymous namespace. Review-Url: https://codereview.chromium.org/1992453002
2016-05-17Remove Release() from IFX_CharIter, use unique_ptrs.chromium/2742chromium/2741tsepez
Review-Url: https://codereview.chromium.org/1986373002
2016-05-17Don't use LCD antialiasing if Fontconfig doesn't support hintingthomasanderson
Some Freetype implementations (like the one packaged with Fedora) do not support hinting due to patents 6219025, 6239783, 6307566, 6225973, 6243070, 6393145, 6421054, 6282327, and 6624828; the latest one expires 10/7/19. This makes LCD antialiasing very ugly, so we instead fall back on NORMAL antialiasing. A before/after on Fedora: https://bugs.chromium.org/p/chromium/issues/detail?id=479400#c31 BUG=479400 Review-Url: https://codereview.chromium.org/1982263004
2016-05-17Rename CPDF_Struct{Element,Tree} to IPDF_tsepez
After all, what good is a naming convention unless you're going to blindly adhere to it? Review-Url: https://codereview.chromium.org/1981403002
2016-05-17XFA_WIDGETSTATUS cleanup.dsinclair
This CL moves all of the XFA_WIDGETSTATUS values into a single enum instead of multiple defines. The values are also normalized. The name was updated to be XFA_WidgetStatus. The XFA_WIDGETFILTER and XFA_LAYOUTSTATUS defines had to match up to XFA_WIDGETSTATUS. This Cl replaces those so we just have a single enum to work with. Review-Url: https://codereview.chromium.org/1986503002
2016-05-17Fix DrMemory bot error by properly initializing the variableweili
The variable needs to be initialized before using. Review-Url: https://codereview.chromium.org/1984323002
2016-05-17Allow multiprocess Dr Memory testsdsinclair
The Dr Memory tests currently run -j1 which makes corpus tests take a long time. Previously, multip-process corpus would override stdout which caused conflict when using Dr. Memory. That has been resolved so enable more parallelism. Review-Url: https://codereview.chromium.org/1981153003
2016-05-17Revert of Remove an unnecessary variable from CPDF_Creator (patchset #1 ↵weili
id:20001 of https://codereview.chromium.org/1988603002/ ) Reason for revert: The variable is still needed, will fix the other way. Original issue's description: > Remove an unnecessary variable from CPDF_Creator > > The variable may not be initialized before using, which caused > errors on DrMemory bot. Since it is not necessary, remove it. > > TBR=thestig@chromium.org > > Committed: https://pdfium.googlesource.com/pdfium/+/47b998b53edbb071bb55c909169be1070126fb22 TBR=dsinclair@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true Review-Url: https://codereview.chromium.org/1987783002
2016-05-17Combine corpus runner into test_runner.pydsinclair
This CL combines the corpus runner into the test_runner.py. This also enables the javascript and pixel runners to do a multi test runner. Review-Url: https://codereview.chromium.org/1952923002
2016-05-17Remove an unnecessary variable from CPDF_Creatorweili
The variable may not be initialized before using, which caused errors on DrMemory bot. Since it is not necessary, remove it. TBR=thestig@chromium.org Review-Url: https://codereview.chromium.org/1988603002
2016-05-16Use chromium_code standard for PDFium GYP compilationweili
Define and use chromium_code to be used in standalone PDFium GYP build so that PDFium code can have more stringent warning level. This is also enabled on GN build by default so that GYP and GN builds can have consistent compilation results. Also enable chromium_code for PDFium compilation in Chromium since most of the warnings are cleared. The left ones are clearly marked and will be addressed soon. A few more clean-ups for the build: -- Remove the suppression of sign-compare warnings for Clang since the code is clean and the warning can be re-enabled. -- Re-enable "treat warning as errors" on Mac -- Add a flag to make GCC build works as well. BUG=pdfium:29, pdfium:475 Review-Url: https://codereview.chromium.org/1985843002
2016-05-16Remove Release() { delete this; } in CFXA_Data{Importer,Exporter}tsepez
Review-Url: https://codereview.chromium.org/1988453002
2016-05-16Fix a nullptr deref in CPDF_BookmarkTree::GetFirstChild().thestig
BUG=590927 Review-Url: https://codereview.chromium.org/1988443002
2016-05-16Revert "Replace Release() { delete this; } in fde_xml_imp.h"Tom Sepez
This reverts commit fa34e805fd03ba81bcfe1148cf96b24fe63b39a0. Reason for revert: broke asan tests. TBR=thestig@chromium.org Review URL: https://codereview.chromium.org/1982843002 .
2016-05-16Replace Release() { delete this; } in fde_xml_imp.htsepez
Review-Url: https://codereview.chromium.org/1981003002
2016-05-16Use CFX_RetainPtr<> to refcount CPDFXFA_Pagetsepez
Review-Url: https://codereview.chromium.org/1984693002
2016-05-16Fix the code that causes warningsweili
These are the left or newly added code which causes compilation warnings of "signed and unsigned comparison". Need to fix them before I re-enable the warning flag. BUG=pdfium:29 Review-Url: https://codereview.chromium.org/1986533002
2016-05-16Remove XFA_WIDGETEVENT definesdsinclair
The XFA_WIDGETEVENT defines were used to allow one method to handle the PostAdd and PreRemove calls. The calls only shared setup code. This CL splits the WidgetEvent method into two, one for PostAdd and one for PreRemove and updates the only two callers to call the correct variant. Review-Url: https://codereview.chromium.org/1982033002
2016-05-16Remove { delete this; } anti-pattern from IXFA_WidgetIteratortsepez
Review-Url: https://codereview.chromium.org/1976123003
2016-05-16Clean up CPDF_Creator.thestig
- Used unique_ptr and initializer list. - Remove variables that never change in value. Review-Url: https://codereview.chromium.org/1976003003
2016-05-16Fix binding of null pointer dereference to reference typechromium/2740chromium/2739hans
A new Clang warning complains about this, and especially for the function returning a reference, it seems like a bad idea. BUG=none Review-Url: https://codereview.chromium.org/1986453002
2016-05-16Remove some c_str() calls from StringCs in xfa docs.tsepez
Avoids the risk of possibly getting a non-null terminated buffer if the StringC happened to be created by Mid(), etc. Doesn't seem to happen in practice. Review-Url: https://codereview.chromium.org/1983683003
2016-05-16Replace FX_WSTRC(L"xyz").c_str() with just L"xyz"tsepez
No need to make a WideStringC here at all. Review-Url: https://codereview.chromium.org/1984743003
2016-05-16Convert border style defines to an enum class.dsinclair
There were two defines, BBS_ and PBS_ for the various border styles in the system. They were the same, except PBS_ had an extra SHADOW define which was never used. This CL combines both of those into a single BorderStyle enum class and updates the code as needed. Also, removes ADDBIT, GETBIT unused defines. Updates barcode code to use the util.h defines instead of redefinition. fsdk_baseannot names starting with _ were cleaned up and some #defines moved to constants. Review-Url: https://codereview.chromium.org/1980973002
2016-05-16Readme update to make need for pdfium_is_standalone clear.dsinclair
Add a note about the need for pdfium_is_standalone and fix some spelling mistakes along the way. Review-Url: https://codereview.chromium.org/1979243002
2016-05-13Make CFX_ByteString(const CFX_ByteStringC&) explicit.tsepez
Add missing helper function to CFX_ByteTextBuf to avoid the anti-pattern CFX_ByteString(sBuf.AsStringC()), using the name "Make" to indicate there's an allocation going on in this case. Change some method arguments to take pre-existing ByteStrings where possible. Review-Url: https://codereview.chromium.org/1977093002
2016-05-13Make CFX_WideString(const CFX_WideString&) explicit.tsepez
BUG= Review-Url: https://codereview.chromium.org/1979723003
2016-05-13Fix GCC warnings to be ready for GCC standalone buildweili
ChromeOS still compile pdfium with GCC. Fix the code to be ready to have workable GCC standalone build. The build file change will be in a follow-up CL. One warning is about maybe using uninitialized variable, the other is about enum and non-enum in a conditional statement. Review-Url: https://codereview.chromium.org/1981593002
2016-05-13ReplaceSelections() methods never called.tsepez
Review-Url: https://codereview.chromium.org/1973883005
2016-05-13Templatize CFX_{Byte,Wide}StringCtsepez
Review-Url: https://codereview.chromium.org/1874773002
2016-05-13Make the Skia GN build work.thestig
On Linux, at least. BUG=pdfium:11 Review-Url: https://codereview.chromium.org/1971023007
2016-05-13Move some classes out of doc_vt.cpp into their own files.thestig
Review-Url: https://codereview.chromium.org/1979463002
2016-05-13allow Skia to be the default graphics enginecaryclark
PDFium uses fxge to denote the graphics engine used for general rendering. When pdf_use_skia=1 is set, use Skia as the default engine instead of antigrain. R=tsepez@chromium.org,dsinclair@chromium.org Review-Url: https://codereview.chromium.org/1975143002
2016-05-12Fix some misc nits.thestig
These were left over from after I broke a big "things that never return NULL" CL up into smaller ones. Review-Url: https://codereview.chromium.org/1960043003
2016-05-12Add GN bots into commit queuechromium/2738chromium/2737chromium/2736weili
Review-Url: https://codereview.chromium.org/1978453004
2016-05-12Add CFX_ByteStringC::CharAt() to avoid c_str() and casts.tsepez
Most of the time, we want to operate on chars as if they were unsigned, but there are a few places where we need the default (questionably signed) values. Consolidate the casting in a single place rather than forcing callers to get a char* ptr. BUG=pdfium:493 Review-Url: https://codereview.chromium.org/1972053003
2016-05-12Fix comment in fx_string.htsepez
Previous CLs have removed or marked explicit the constructors which allowed bad constructs to compile. I now get an error: candidate constructor not viable: expects an l-value for 1st argument when called as indicated in the comment. Review-Url: https://codereview.chromium.org/1975983002