summaryrefslogtreecommitdiff
path: root/core
AgeCommit message (Collapse)Author
2016-09-20Re-land "Make CPDF_IndirectObjectHolder use unique_ptr to objects""tsepez
This reverts commit 81e1e3fd2d33478733e47bd007b76fac1a663e74. Review-Url: https://codereview.chromium.org/2353013003
2016-09-20Fix memory leaking on ClosePage.art-snake
BUG=79367, 48791 The fonts was not cleared after unloading pages. Test pdf: http://www.nasa.gov/pdf/750614main_NASA_FY_2014_Budget_Estimates-508.pdf For this file, we have ~5 fonts per page, which equal ~1 Mb per page. In this PDF we have 670 pages, as result after slow scrolling(reading) full document we have ~600 Mb fonts data in memory. Memory usage of PDF Plugin: before this CL: ~660 Mb after this CL: ~100 Mb This is last part of reverted CL: Original CL: https://codereview.chromium.org/2158023002 Revert reason: BUG=647612 Fix bug CL: https://codereview.chromium.org/2350193003 Previous CL: https://codereview.chromium.org/2350243002 Review-Url: https://codereview.chromium.org/2351193002
2016-09-20Assert that dictionary can own the objects it is given.tsepez
Upon indirect object holder destruction, all indirect objects are destroyed -- currently by order of increasing object number -- but ideally without ordering constraints. So currently, we can get away with a dictionary pointing directly at an indirect object with a higher number. It gets destroyed first, invoking Release() on its subordinates, which skips destroying them if they are indirect objects. But we don't want to rely on this artifact of destruction order. Should it happen to be reversed, the dictionary would invoke Release() on freed memory. Interestingly, CPDF_Array skirts the issue by replacing any indirect objects it is given with references. Not clear whether we should do the same thing for dictionaries, or remove it from arrays. The technique certainly complicates understanding ownership. The one violation found is in the unittest that broke the previous CL which tried to use unique_ptrs in indirect object holder. Review-Url: https://codereview.chromium.org/2353093002
2016-09-20Refactor CFX_FontCache to have only one in GE Module.art-snake
After this CL: only one global CFX_FontCache used. Any cached items from it, are released, when they are not being used. This is restore part of reverted CL: Original CL: https://codereview.chromium.org/2158023002 Revert reason: BUG=647612 Fix bug CL: https://codereview.chromium.org/2350193003 Review-Url: https://codereview.chromium.org/2350243002
2016-09-20Fix nullptr dereference in FPDF_GenerateAP().thestig
BUG=645954 Review-Url: https://codereview.chromium.org/2355733002
2016-09-20Fix stack exhaustion in CPDF_PSProc::Parse()tsepez
BUG=648059 Review-Url: https://codereview.chromium.org/2350013003
2016-09-19Revert "Make CPDF_IndirectObjectHolder use unique_ptr to objects"Tom Sepez
This reverts commit c8544d634a1993e2592e41458be215fcd0956031. TBR=dsinclair@chromium.org Review URL: https://codereview.chromium.org/2355683002 .
2016-09-19Make CPDF_IndirectObjectHolder use unique_ptr to objectstsepez
The objects it is given are owned by it and are simply deleted without regard to Release() used by others. Review-Url: https://codereview.chromium.org/2350263002
2016-09-19Remove CPDF_Object::Destroy { delete this; }chromium/2866tsepez
We can delete this just fine on our own. Review-Url: https://codereview.chromium.org/2355593002
2016-09-19Fix "heap use after free" bug.art-snake
BUG=647612 Review-Url: https://codereview.chromium.org/2350193003
2016-09-19Clean up CPDF_Stream.tsepez
Replace the CPDF_Stream(nullptr, 0, nullptr) pattern with a default ctor. Remove unused parameters from CPDF_Stream::SetData(). Both are always passed as FALSE. CPDF_Stream declared its own m_GenNum, which shadowed the one in CPDF_Object. It was used only to distinguish file/memory streams, so add a bool explicitly for this purpose. Remove the union, it would be sad if we confused user data with a C++ object with virtual function calls. Use unique_ptrs with appropriate deleters to manage memory. Review-Url: https://codereview.chromium.org/2347993002
2016-09-19Remove dead code in CCodec_IccModulekcwu
TEST=build pdfium and chromium BUG=pdfium:599 Review-Url: https://codereview.chromium.org/2355523002
2016-09-19Revert of Pdfium: Fix fonts leaking on ClosePage. (patchset #10 id:180001 of ↵dsinclair
https://codereview.chromium.org/2158023002/ ) Reason for revert: Causes heap-use-after-free. See crbug.com/647612. Original issue's description: > Fix memory leaking on ClosePage. > CFX_FontCache refactoring: > after this CL: Only one global CFX_FontCache used. Any cached items from it, are released, when its are not used. > > BUG=79367,48791 > > The fonts was not cleared after unloading pages. > > Test pdf: > > http://www.nasa.gov/pdf/750614main_NASA_FY_2014_Budget_Estimates-508.pdf > > For this file, we have ~5 fonts per page, which equal ~1 Mb per page. > In this PDF we have 670 pages, as result after slow scrolling(reading) full document we have ~600 Mb fonts data in memory. > > memory usage of PDF Plugin: > before this CL: ~660 Mb > after this CL: ~100 Mb > > Committed: https://pdfium.googlesource.com/pdfium/+/cde5101eb15b24519e89fa500fe37038bc8e2201 TBR=tsepez@chromium.org,brucedawson@chromium.org,npm@chromium.org,art-snake@yandex-team.ru # Not skipping CQ checks because original CL landed more than 1 days ago. BUG=79367,48791 Review-Url: https://codereview.chromium.org/2350763002
2016-09-19Remove duplicated charset definitions, and move them to fx_font.hnpm
PWL_FontMap does not need its own charset definitions. fx_edit.h does not need to define DEFAULT_CHARSET. XFA have their own definitions. They look different in that most are MAC or MSWin charset definitions. So they are left untouched. public/fpdf_sysfontinfo.h duplicate ones were left untouched due to being in public folder. Review-Url: https://codereview.chromium.org/2347313002
2016-09-19Fix compare between signed and unsigned values in ↵stackexploit
CPDF_ImageRenderer::StartDIBSource. Correct the compare logic in CPDF_ImageRenderer::StartDIBSource() by using size_t instead of int. BUG=chromium:645036 R=ochang@chromium.org Review-Url: https://codereview.chromium.org/2323663002
2016-09-19Avoid nullptr access in sycc422_to_rgb and sycc420_to_rgbstackexploit
BUG=648127 Review-Url: https://codereview.chromium.org/2351623002
2016-09-16Reduce duplicated code in CPDF_Document by adding auxiliary methodschromium/2865chromium/2864chromium/2863npm
ProcessbCJK and CalculateFontDesc methods are used to reduce the code duplication between AddFont and AddWindowsFont methods. Review-Url: https://codereview.chromium.org/2341373003
2016-09-15Fix memory leaking on ClosePage.chromium/2862art-snake
CFX_FontCache refactoring: after this CL: Only one global CFX_FontCache used. Any cached items from it, are released, when its are not used. BUG=79367,48791 The fonts was not cleared after unloading pages. Test pdf: http://www.nasa.gov/pdf/750614main_NASA_FY_2014_Budget_Estimates-508.pdf For this file, we have ~5 fonts per page, which equal ~1 Mb per page. In this PDF we have 670 pages, as result after slow scrolling(reading) full document we have ~600 Mb fonts data in memory. memory usage of PDF Plugin: before this CL: ~660 Mb after this CL: ~100 Mb Review-Url: https://codereview.chromium.org/2158023002
2016-09-15Add string pools to save storage.tsepez
Adds string hashes so CFX strings will interoperate with STL unordered containers. These will be employed per-document in a subsequent cl. BUG=pdfium:597 Review-Url: https://codereview.chromium.org/2341683005
2016-09-15Use either /RECT or /QuadPoints for annotation coordinates, depending on /APtonikitoo
On Acrobat, if "/AP" is present on a text markup definition, the coordinates used to draw the annotation come from "/Rect values, whereas if "/AP" is not defined, the array defined in /QuadPoints is used to grab the annotation coordinates from. PDFium, on the other hand, uses "/Rect" regardless of presence or absence of "/AP". CL fixes PDFium to work similarly to Acrobat, in this case. TEST=testing/resources/pixel/bug_585_*.in BUG=pdfium:585 Review-Url: https://codereview.chromium.org/2289293005
2016-09-15Use ToUnicode mapping even when unicode is 0.npm
CPDF_Font::UnicodeFromCharcode returns 0 only if ToUnicode map maps the charcode to 0. CPDF_SimpleFont::UnicodeFromCharcode and CPDF_CID_Font:: UnicodeFromCharCode return 0 only if the call to CPDF_Font returns 0. In other cases, these methods return an empty string. So when processing text, a 0 return from the method should not be replaced with the charcode. BUG=pdfium:583 Review-Url: https://codereview.chromium.org/2342073002
2016-09-15Add short-cut in CFX_{Byte,Wide}String::Operator<()tsepez
Strings are never less than themselves, and this will occur given the shared CoW nature of these strings. Review-Url: https://codereview.chromium.org/2347433004
2016-09-15Rename dictionary set and get methodsdsinclair
This Cl makes the Get and Set methods consistenly use {G|S}et<Type>For. BUG=pdfium:596 Review-Url: https://codereview.chromium.org/2334323005
2016-09-14Delete unused parameter nWordStyle from GetCharWidth()chromium/2861npm
Deleted parameter from callers, who also do not use it. Review-Url: https://codereview.chromium.org/2337973004
2016-09-14Map glyph bitmap in LookUpGlyphBitmap even when unsuccessfulnpm
CFX_FaceCache::LookUpGlyphBitMap should remember its unsuccessful glyph bitmap calculations. This is especially important if there are a lot of repeated requests for the same glyph with large bitmaps (which means long calculations in freetype). With this CL, the pdf in the bug below will be rendered quickly. BUG=chromium:638856 Review-Url: https://codereview.chromium.org/2338883004
2016-09-14Cleanup CPDFXFA_Document.dsinclair
Remove unused params and methods. Cleanup formatting. Review-Url: https://codereview.chromium.org/2322003002
2016-09-13Sort include entries.dsinclair
This CL updates all of the includes to be correctly sorted. A PRESUBMIT warning is added (from chromium) that will warn if the includes are in the wrong order on upload. Review-Url: https://codereview.chromium.org/2337293002
2016-09-13Add check for trivial string equalitytsepez
BUG= Review-Url: https://codereview.chromium.org/2338623005
2016-09-13CFX_FloatPoint default constructor and equals operatorschromium/2860tsepez
Review-Url: https://codereview.chromium.org/2341453002
2016-09-13Remove more strcmp/memcmp usage.tsepez
Review-Url: https://codereview.chromium.org/2340513002
2016-09-13Stop converting widestring -> c_str -> widestring in several places.tsepez
Avoids a needless alloc and copy. Review-Url: https://codereview.chromium.org/2338553002
2016-09-13Get rid of memcmp in CFieldTree::Lookup()tsepez
Wide strings compare just fine on their own. Review-Url: https://codereview.chromium.org/2331233002
2016-09-13Make ipdf_formnotify a pure virtual interfacetsepez
Removes some unused code for the default implementation, Review-Url: https://codereview.chromium.org/2332983002
2016-09-12Remove GetDictBy("DR") from annot/field dictionariesnpm
We should not use DR field from annotation/field dictionaries, since these are not supposed to have one. In PDF spec 1.7, DR is defined only for the form dictionary, in 8.6.1. In addition, note 118 in Appendix H says "In PDF 1.2, an additional entry in the field dictionary, DR, was defined but was never implemented. Beginning with PDF 1.5, this entry is obsolete and should be ignored." The changes in CBA_FontMap fix the bug below. The changes in cpvt_generateap do not affect the bug but follow the reasoning from the above paragraph. BUG=pdfium:538 Review-Url: https://codereview.chromium.org/2337673002
2016-09-12Verify value of prec before usingdsinclair
The fx_codec_jpx_opj code will attempt to do a 1 << (prec - 1). If the prec value is >=32 then that shift will overflow the int value. This CL adds a check that prec is < 32 before attempting the shift. BUG=chromium:633208 Review-Url: https://codereview.chromium.org/2334823002
2016-09-12Refactor CPDF_Document by creating new methodsnpm
- Methods GetPagesDict, ProcessNonbCJK, CalculateFlags, and CalculateEncodingDict created to reduce duplicated code. - Code nits Review-Url: https://codereview.chromium.org/2323793003
2016-09-12change memcpy to memmove for potential nearby addressestracy_jiang
BUG=645186 Review-Url: https://codereview.chromium.org/2326103002
2016-09-12fix some uninitialized variableshong_zhang
BUG=627399 Review-Url: https://codereview.chromium.org/2328003002
2016-09-12Make CFX_Obeservable::Observer into a pointer-ish type;tsepez
This may be a better design because it avoids having a level of indirection that the Observer required. Review-Url: https://codereview.chromium.org/2326763002
2016-09-09Define behaviors of FPDF_RenderPageBitmap_Retail and FPDF_FFLDraw.chromium/2858chromium/2857chromium/2856jaepark
Previously, PDFium only supported widget annotations to draw forms. As we've implemented other annotations, the behavior of FPDF_RenderPageBitmap_Retail and FPDF_FFLDraw changed. So, this CL clearly defines what needs to be done in FPDF_RenderPageBitmap_Retail and FPDF_FFLDraw. This CL first assumes that PDFium users will always call FPDF_RenderPageBitmap_Retail and FPDF_FFLDraw to render PDF pages, because otherwise they are not able to support PDF forms. FPDF_RenderPageBitmap_Retail should only deal with non-widget annotations, such as highlight, underline, text, etc. If FPDF_ANNOT flag is passed, non-widget annotations are drawn. Otherwise, they are hidden. FPDF_FFLDraw should only deal with annotations that requires user-interaction, such as widget annotations and popup annotation. Since popup annotation is associated with non-widget annotation, they should not be drawn if the associated annotation is hidden. Thus, if FPDF_ANNOT flag is passed, popup annotations are drawn. Otherwise, they are hidden. Widget annotations should be always drawn regardless of FPDF_ANNOT flag since they need to be always displayed for PDF forms. Also, roll DEPS for testing/corpus to 8485b30. BUG=pdfium:594 Review-Url: https://codereview.chromium.org/2323203002
2016-09-08explicit operator bool for CFX_RetainPtr and CFX_CountRefchromium/2855tsepez
Review-Url: https://codereview.chromium.org/2324733003
2016-09-08Delete unused methods in CPDF_Documentnpm
Review-Url: https://codereview.chromium.org/2323933002
2016-09-08Check whether the annotation content is empty using CFX_WideString.jaepark
CFX_ByteString cannot properly check whether the contents string is empty because the first two bytes of text strings encoded in Unicode are always ASCII 254 followed by 255. So if we get contents in CFX_ByteString, the length will always be 2. Also, roll DEPS for testing/corpus to 608bf04. Review-Url: https://codereview.chromium.org/2293403003
2016-09-08Fix memory management errors for font loading and copyingweili
A few issues are fixed: --Change variable |m_bLogic| in CFX_Font to |m_bShallowCopy| to reflect its meaning better; --For a shallow copy of font, we must guarantee that the copied font will not be deleted until the shallow copy is deleted. So need to increase the src font's refcount when copying it; --The stream |m_pOwnedStream| needs to have matched new/delete These errors need to be fixed before we can properly delete all the fonts to address the leaks. BUG=pdfium:242 Review-Url: https://codereview.chromium.org/2322043002
2016-09-08Replace CJS_Runtime::Observer with CFX_Runtime<CJS_Runtime>::Observertsepez
Previously, the observer set a flag which conditionalized a null pointer return. Now, we just clear the pointer. Destruction order matters, so add a method to trigger notifications. Review-Url: https://codereview.chromium.org/2322743002
2016-09-08Switch to ValueOrDiedsinclair
We know the values are always valid at this point, so use ValueOrDie instead of ValueOrDefault. BUG=chromium:637984 Review-Url: https://codereview.chromium.org/2319343003
2016-09-07Make Observers into a templated classtsepez
Review-Url: https://codereview.chromium.org/2311343003
2016-09-07Verify image dimentions before usingdsinclair
Verify the provided image size is within bounds before loading. BUG=chromium:639160 Review-Url: https://codereview.chromium.org/2323473002
2016-09-07Verify pattern start values.dsinclair
When calculating the starting x and y for a pattern it is possible to overflow the int value. Use checked math to make sure we don't overflow. BUG=chromium:637984 Review-Url: https://codereview.chromium.org/2317283002
2016-09-07Maybe avoid a copy in CPDF_RenderStatus::SelectClipPathtsepez
Use the shared copy-on-write version instead of the underlying CFX_GraphStateData since it may not always get modified. Review-Url: https://codereview.chromium.org/2316823002
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-23 18:29:37 +0000