| Age | Commit message (Collapse) | Author |
|---|
|
m_sTimeMap is a global variable with a constructor and destructor, which
is not allowed. This change moves it to a function with a static pointer
so that it is constructed on demand and then leaked, thus avoiding
having startup and shutdown code.
This also fixes a worrisome bug caused by having m_sTimeMap defined in
a header file. Because m_sTimeMap was defined (and marked as static) in
a header file there were fifteen separate copies of it, one for each
source file which included the header file. This could easily lead to
bugs because a timer that was added from one source file would be
invisible to other source files.
Each instance of m_sTimeMap added four entries to the
dump-static-initializers.py report, for a total of sixty, so this fix
significantly cleans up that report.
BUG=441899
R=tsepez@chromium.org
Review URL: https://codereview.chromium.org/831903002
|
|
QeTable is a 752 byte array that was defined in a header file. This
caused it to be instantiated by the VC++ compiler 12 times, wasting
8,272 bytes of space in the data segment. Because 'const' implies
'static' this did not cause any duplicate symbol errors.
JSCONST_n*HASH are a set of eight variables that are defined in a header
file. This causes them to be replicated 15 times. The variables
themselves are tiny but they are dynamically initialized and this
dynamic initialization code is replicated 15 times.
When tested on pdfium_test.exe the effect of this change is to:
Reduce the .text (code) segment by 3,616 bytes.
Reduce the .rdata section by 8,656 bytes.
Reduce the total binary file size by 13312 bytes.
These are the worst offenders for pdf.dll as shown in:
https://drive.google.com/open?id=1BvubxoA2SU_2e4T5cq7jHTjc1TlT0qOndpIfX3DMeA8&authuser=0
This will also drastically simplify the list of work to be done
for bug 441899 (getting rid of initializers).
BUG=441988
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/802013002
|
|
FX_ArraySize macro.
pdfium has numerous places where the number of elements in an array is
calculated with expressions like:
sizeof(cFormats)/sizeof(FX_LPCWSTR)
This is suboptimal because it is verbose, it is easy to get wrong, and
it cannot be determined through casual inspection whether the code is
correct. It will give incorrect results if cFormats is a pointer instead
of an array and it will give incorrect results if FX_LPCWSTR is not the
type of the array elements.
The FX_WSTRC macro in fx_string.h which I fixed was particularly scary
because it would silently misbehave if passed a pointer.
The FX_ArraySize macro which I have added and started using (taken from
arraysize in v8's macros.h) is easier to use and will always give correct
results. If passed a pointer it will fail to compile.
For this change I only fixed instances of sizeof(FX_LPCWSTR). There
appear to be about 150 other places in the pdfium code that could
benefit from using FX_ArraySize.
R=bo_xu@foxitsoftware.com, tsepez@chromium.org
Review URL: https://codereview.chromium.org/729293003
|
|
Since casts to FX_LPCWSTR have been shown to hide bugs I tried removing
more of them, targeting those places where a cast was used to force a
conversion from CFX_WideString to FX_LPCWSTR, replacing these casts with
calls to the newly added .c_str() function. This revealed two places
where the cast was hiding a bug -- where ->c_str() was required instead!
This removes ~33 FX_LPCWSTR casts and there are ~31 left, many of which
will go away in some future change.
Also includes this change:
Removing unnecessary casts from wchar_t* to wchar_t*, by various names.
Original patch from Bruce Dawson(brucedawson@chromium.org)
R=bo_xu@foxitsoftware.com, tsepez@chromium.org
Review URL: https://codereview.chromium.org/733693003
|
|
Remove casts that merely cast from wchar_t* to wchar_t*. Sometimes the
types or casts are FX_LPCWSTR but the idea is the same. Excess casts
can (and have) hidden bugs so removing these may prevent future problems.
Original patch from Bruce Dawson(brucedawson@chromium.org)
R=bo_xu@foxitsoftware.com, tsepez@chromium.org
Review URL: https://codereview.chromium.org/730993002
|
|
BUG=425129
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/688303003
|
|
BUG=pdfium:28
R=thakis@chromium.org
Review URL: https://codereview.chromium.org/472563002
|
|
No intended behavior change.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/436483002
|
|
No intended behavior change.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/426763003
|
|
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/427353003
|
|
No intended behavior change.
- Remove more unused variables, functions, member variables.
- Put a few constructor initializers in the order they execute in.
- Add braces for subobject initializers.
- Fix a handful of signed / unsigned comparisons.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/429593005
|
|
It remains to call the PumpMessageLoop() method at a regular interval,
however, since nothing posts to the loop yet, that shouldn't be a
problem.
BUG=25
R=jam@chromium.org
Review URL: https://codereview.chromium.org/374123002
|
|
MSVC lexes L#macro_arg as a single wide string literal token, but
Clang and other C/C++ compliant lexers do not. There was already
a workaround to use implicit string concatenation for GCC, but there's
a simpler solution of token pasting the L onto the stringized macro
argument with 'L###macro_arg'. This works with Clang, GCC, and MSVC.
R=jun_fang@foxitsoftware.com, jam@chromium.org
BUG=82385
Original patch by Reid Kleckner <rnk@chromium.org>
Review URL: https://codereview.chromium.org/345643002
|
|
|
|
into errors. It also makes it clearer to find usage of v8 in the library.
|
|
|
