Age | Commit message (Collapse) | Author |
|
BUG=pdfium:488
Review-Url: https://codereview.chromium.org/1936383002
|
|
The Skia variant generates a couple of warnings on the linux build.
One of the warning silenced is an unused variable. Once PDFs are
identified that exercise this variable, the dummy code will be
replaced.
R=tsepez@chromium.org
Review-Url: https://codereview.chromium.org/1962683002
|
|
Review-Url: https://codereview.chromium.org/1954593004
|
|
Chromium defines WIN32_LEAN_AND_MEAN on Windows. Third party library
libtiff also defines it. So we undefine it before compiling libtiff
code for GYP build.
Also, remove _CRT_SECURE_NO_WARNINGS macro since it is redefined in
zlib library as well. Our code no longer needs it.
After fixing the above, re-enable warning flag 4005 which alerts about
"macro redefinition".
For GN build, we disable warning 4005 for compiling libtiff
code before we can figure out another way to do this.
Review-Url: https://codereview.chromium.org/1954773002
|
|
This fixes a build issue where MEMORY_TOOL_REPLACES_ALLOCATOR may already be
set.
Review-Url: https://codereview.chromium.org/1951383002
|
|
This Cl updates all the CLASSHASH defines to use an enum class. A few defines
where unabled and were either updated or removed if the number no longer exists
in the code base.
Review-Url: https://codereview.chromium.org/1946213003
|
|
TBR=tsepez@chromium.org,dsinclair@chromium.org
Review URL: https://codereview.chromium.org/1949383002 .
|
|
ASSERT(statement) will not execute statement in release builds. So
need to move statements outside ASSERT().
TBR=tsepez@chromium.org,dsinclair@chromium.org
Review URL: https://codereview.chromium.org/1949613004 .
|
|
The reason to disable 'treating warnings as errors' earilier is due to
c4267 warnings, which we already disabled uniformly in pdfium.gyp.
Re-enable this now so that we can have the same level of compilation
warnings/errors on 32 bit and 64 bit builds.
Also fix two c4390 warnings on Windows release builds. Their warning
messages are "empty controlled statement found; is this the intent?"
Review-Url: https://codereview.chromium.org/1949303002
|
|
This Cl updates FWL_ERR to be an FWL_Error enum class. It removes FX_ERR and
replaces it with FWL_Error as well as the values were the same.
There were many places where we either returned other values for FWL_ERR
results.
This Cl is the same as: https://codereview.chromium.org/1943413002/ but I seem
to have messed up the base URL in that one.
TBR=tsepez@chromium.org
Review-Url: https://codereview.chromium.org/1952693003
|
|
Review-Url: https://codereview.chromium.org/1951873003
|
|
Review-Url: https://codereview.chromium.org/1952853002
|
|
data_binding.pdf is much closer now that events are sent.
It still diffs due to some bg color issues, not that the
box is unchecked.
Review-Url: https://codereview.chromium.org/1952823002
|
|
This CL removes:
* IFWL_AdapterNative
* IFWL_Adapter{Widget|Thread}Mgr
* IFWL_WidgetMgrDelegate
* CFWL_SDAdapter{Widget|Thread}Mgr
Methods which just returned have also been removed.
Review-Url: https://codereview.chromium.org/1928963004
|
|
The m_dwData pointer was being used as a generic way to pass around data. There
were only two places we were reading that data. This Cl changes to have two
specific flags and removes the generic pointer.
Review-Url: https://codereview.chromium.org/1950973003
|
|
This Cl updates several of the theme defines to be constant values instead
of defines. Some of the items were re-ordered so the calculations would work
correctly.
Review-Url: https://codereview.chromium.org/1945183004
|
|
This CL converts defines into constants, enums, enum classes or removes them
as needed.
Review-Url: https://codereview.chromium.org/1938163002
|
|
Investigate results of:
git grep -ni 'return [(]*[a-z0-9_]* &[^&]'
git grep -ni 'BOOL.*= [(]*[a-z0-9_]* &[^&]'
Review-Url: https://codereview.chromium.org/1951653002
|
|
Passing bad args in pdfium_test -- all void* pointers look alike.
Make output less verbose while we're here. Turn on event sending
mode for corups tests. No effect until .evt files added to corpus.
Review-Url: https://codereview.chromium.org/1946873004
|
|
This Cl updates the README so GN is the recommended method to build PDFium in
a standalone fashion.
BUG=pdfium:485
Review-Url: https://codereview.chromium.org/1950103002
|
|
This Cl makes GN execute correctly for builds with pdf_use_skia=true enabled.
BUG=pdfium:487
Review-Url: https://codereview.chromium.org/1932683003
|
|
(patchset #1 id:1 of https://codereview.chromium.org/1946693002/ )
Reason for revert:
The culprit was found and confirmed, not this one.
Original issue's description:
> Revert of Relax a couple checks to allow certain non-standard PDF files. (patchset #1 id:1 of https://codereview.chromium.org/1926823002/ )
>
> Reason for revert:
> Speculatively revert due to high volume of crashes on Chromium.
>
> Original issue's description:
> > Relax a couple checks to allow certain non-standard PDF files.
> >
> > Some non-standard PDF files misuse the size of cross reference table,
> > and reuse some object number which the old one is still in use. PDFium
> > can relax the reusing of xref objects only since it is not referred in
> > the pdf document. When the size of cross reference table is larger
> > than defined, PDFium will try to continue other than abort.
> >
> > BUG=chromium:596947
> >
> > Committed: https://pdfium.googlesource.com/pdfium/+/cd1e9ff4f432cbc29ed279e6891fb7ddc2ea3734
>
> TBR=thestig@chromium.org,dsinclair@chromium.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
> BUG=chromium:596947
>
> Committed: https://pdfium.googlesource.com/pdfium/+/5fc4f31285c3a88fc157fd2d9b9cf2eb5c7cabed
TBR=thestig@chromium.org,dsinclair@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:596947
Review-Url: https://codereview.chromium.org/1947983002
|
|
https://codereview.chromium.org/1919283008/ )
Reason for revert:
Causing segv on certain PDFs which make the PDF load tests hang. crbug.com/608901
Original issue's description:
> Remove unneeded CPVT classes.
>
> - CPVT_Size is the same as CFX_PointF
> - CPVT_FloatRange is unused.
> - CPVT_ArrayTemplate is just a wrapper for CFX_ArrayTemplate.
>
> Committed: https://pdfium.googlesource.com/pdfium/+/a354eb517429e10d84abff65e455f0c183fe58e0
TBR=thestig@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
Review-Url: https://codereview.chromium.org/1947093002
|
|
All Release() did was invoke delete. Add some "overrides"
while we're at it.
Review-Url: https://codereview.chromium.org/1951573002
|
|
Review-Url: https://codereview.chromium.org/1944893003
|
|
Remove unused "dynamic" allocator (the scary one).
Use malloc/free wrapper allocator under #ifdef for CF/asan testing.
Rename IFX_MEMAllocator to IFX_MemoryAllocator (MEM in all caps
would imply that MEM was an acroynm, not an abbreviation).
Review-Url: https://codereview.chromium.org/1944093002
|
|
This Cl updates the FWL_WGTHITTEST_ defines to an enum class and fixes up any
code issues.
Review-Url: https://codereview.chromium.org/1948583002
|
|
(patchset #1 id:1 of https://codereview.chromium.org/1926823002/ )
Reason for revert:
Speculatively revert due to high volume of crashes on Chromium.
Original issue's description:
> Relax a couple checks to allow certain non-standard PDF files.
>
> Some non-standard PDF files misuse the size of cross reference table,
> and reuse some object number which the old one is still in use. PDFium
> can relax the reusing of xref objects only since it is not referred in
> the pdf document. When the size of cross reference table is larger
> than defined, PDFium will try to continue other than abort.
>
> BUG=chromium:596947
>
> Committed: https://pdfium.googlesource.com/pdfium/+/cd1e9ff4f432cbc29ed279e6891fb7ddc2ea3734
TBR=thestig@chromium.org,dsinclair@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=chromium:596947
Review-Url: https://codereview.chromium.org/1946693002
|
|
This reverts commit 038bf0b129e56212a6de0989dca1ae023bb17c5d.
In order to enable XFA from GN builds we need to be able to override
pdf_enable_xfa. To do this a build_overrides/pdfium.gni file needs to be created
in Chrome. To use that, //pdfium.gni reads the override values and sets them
into the defined arguments. We have to create a build_overrides/pdfium.gni file
in PDFium that sets the default overrides for PDFium builds.
BUG=chromium:62400
Review-Url: https://codereview.chromium.org/1936313002
|
|
Last part. Remove CFX_PtrArray typedef.
Review-Url: https://codereview.chromium.org/1941863002
|
|
Also fix typos and overrides.
Review-Url: https://codereview.chromium.org/1935793002
|
|
- CPVT_Size is the same as CFX_PointF
- CPVT_FloatRange is unused.
- CPVT_ArrayTemplate is just a wrapper for CFX_ArrayTemplate.
Review-Url: https://codereview.chromium.org/1919283008
|
|
Review-Url: https://codereview.chromium.org/1936733002
|
|
This CL adds in the tree status check so the CQ will not commit if the waterfall
closes the PDFium tree.
BUG=chromium:607625
Review-Url: https://codereview.chromium.org/1937943002
|
|
This CL removes the following and uses the only instance instead.
* IFDE_TxtEdtParag
* IFDE_TxtEdtEventSink
* IFDE_TxtEdtEngine
* IFDE_TxtEdtBuf
Review-Url: https://codereview.chromium.org/1927973003
|
|
https://codereview.chromium.org/1923333002/ )
Reason for revert:
Blocking the roll
Original issue's description:
> Allow overriding GN build flags
>
> In order to enable XFA from GN builds we need to be able to override
> pdf_enable_xfa. To do this a build_overrides/pdfium.gni file needs to be created
> in Chrome. To use that, we have to create a build_overrides/pdfium.gni file
> in PDFium that just uses the default values.
>
> BUG=chromium:62400
>
> Committed: https://pdfium.googlesource.com/pdfium/+/1ae572e7e35b52ce4897d1715c005a1b0f79526a
TBR=thakis@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=chromium:62400
Review-Url: https://codereview.chromium.org/1941563002
|
|
GYP and GN builds share some common utilities. GYP has three utilities
in its own directory which are also synced in build directory now.
We will use utilities from build directory instead of keeping our own
copies.
Compared to our own copies, utilities in build directory
are either same or newer.
find_depot_tools.py is identical;
vs_toolchain.py and find_sdk.py have some minor updates.
Review-Url: https://codereview.chromium.org/1933843002
|
|
Do more cleanup in modified files.
Review-Url: https://codereview.chromium.org/1938613003
|
|
Converted one place to unique_ptr to avoid redundant cleanup.
Review-Url: https://codereview.chromium.org/1937593002
|
|
Add a section about Windows Visual Studio toolchain setting.
Also announce we are using the Visual Studio toolchain from depot_tools
as default.
Review-Url: https://codereview.chromium.org/1891383002
|
|
This also removes another hand-written bubblesort in favor of
the std::sort() STL function.
Review-Url: https://codereview.chromium.org/1937513002
|
|
Some non-standard PDF files misuse the size of cross reference table,
and reuse some object number which the old one is still in use. PDFium
can relax the reusing of xref objects only since it is not referred in
the pdf document. When the size of cross reference table is larger
than defined, PDFium will try to continue other than abort.
BUG=chromium:596947
Review-Url: https://codereview.chromium.org/1926823002
|
|
Remaining uses are all in fxbarcode/.
Review-Url: https://codereview.chromium.org/1937453002
|
|
BUG=607739
Review-Url: https://codereview.chromium.org/1934483002
|
|
XFA edition.
Review-Url: https://codereview.chromium.org/1925363002
|
|
Review-Url: https://codereview.chromium.org/1926303002
|
|
compilation (patchset #1 id:1 of https://codereview.chromium.org/1922373003/ )
Reason for revert:
I fixed Clang build config, and updated drm binary. Should work this time.
Original issue's description:
> Revert of Use visual studio toolchain from depot_tools for PDFium compilation (patchset #3 id:40001 of https://codereview.chromium.org/1897523002/ )
>
> Reason for revert:
> Clang build is broken. drm build has some problem. Will revert this change for now.
>
> Original issue's description:
> > Use visual studio toolchain from depot_tools for PDFium compilation
> >
> > Change to use visual studio toolchain from depot_tools by default.
> > Setting DEPOT_TOOLS_WIN_TOOLCHAIN=0 allows compilation using system
> > toolchain as before.
> >
> > Using toolchain from depot_tools unifies the compilation
> > environment, and brings the benefits of automated update, bug fixes
> > etc.
> >
> > Committed: https://pdfium.googlesource.com/pdfium/+/590f2d9e057a0d5b17a9706affd3c6115265021b
>
> TBR=brucedawson@chromium.org,thestig@chromium.org,tsepez@chromium.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
>
> Committed: https://pdfium.googlesource.com/pdfium/+/df96690d4e3799536b981e3673d64018fa5fd037
TBR=brucedawson@chromium.org,thestig@chromium.org,tsepez@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/1927373002
|
|
Remove unused m_pdf417byteSegments, which was coped into from
an empty array and never again referenced.
Review-Url: https://codereview.chromium.org/1927253002
|
|
This version contains fixes that are required for running binaries
compiled with Visual Studio 2015.
R=zhaoqin@chromium.org
Review URL: https://codereview.chromium.org/1933493002 .
|
|
Review-Url: https://codereview.chromium.org/1932703003
|