| Age | Commit message (Collapse) | Author |
|---|
|
In most cases, we just CHECK() that no exception was thrown. Previously,
we'd just crash.
Ideally, this should all be fixed and the system should cope with those
exceptions, but that's beyond this CL.
R=tsepez@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1126203010
|
|
This involves adding some explicit c_str() calls. Doing so flagged
PDF_EncodeText() and FindOptionValue() as having suboptimal signatures, in
that we are often throwing away a perfectly fine length and recomputing it.
There are still some platform-specific code that needs the operator.
R=brucedawson@chromium.org
Review URL: https://codereview.chromium.org/1101933003
|
|
This reverts commit 15a62973b9b89c3e229cc0ab501c45967f91b325.
Reason for revert: broke build on windows, mac. I must have missed
some platform-specific conversions.
TBR=brucedawson@chromium.org
Review URL: https://codereview.chromium.org/1108883002
|
|
This involves adding some explicit c_str() calls. Doing so flagged
PDF_EncodeText() and FindOptionValue() as having suboptimal signatures, in
that we are often throwing away a perfectly fine length and recomputing it.
R=brucedawson@chromium.org
Review URL: https://codereview.chromium.org/1101933003
|
|
The code to validate the number of parameters happens inside each particular
method, rather than prior to method dispatch. As such, there's no point in
having this number take up space in the table.
Add some test to cover at least some of the per-method validations, and
update error messages to be more useful.
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/1084183008
|
|
The red-flag here is the explicit invocation of things like
params[1].operator CFX_WideString()
rather than
static_cast<CFX_WideString>(params[1])
to invoke the conversion. Turns out the above won't compile due to
ambiguity given the number of implicit constructors for widestrings.
CJS_Value has both constructors and assignment operators for the
primitive types, which means that conversions can take place
unexpectedly in both directions, a second red flag.
We don't want the compiler invoking these at will since it may hide
bugs. In fact, when they are removed, three such places were
discovered.
Also rename ToJSValue to ToV8Value to match the other ToV8xxxxx
functions added.
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/1096813008
|
|
This provides no benefit, and reduces transparency.
Along the way:
Kill off some unused/commented-out code.
Return void where a bool return doesn't make sense.
Remove a pointless template type.
Remove now unused constants and types.
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/971033002
|
|
This implements the previously unimplemented JS_Error() function.
Along the way:
- fix some IWYU when the include order in global.cpp was perturbed.
- remove some uses of JS_ErrorString, to increase transparency.
- use vp.IsSetting() in place of !vp.IsGetting() for clarity.
- specify an error string on several error return paths.
- add an error string for writing readonly properties.
- rename an error string constant to reflect the actual message.
- replace calls to variadic Format() with a function doing string appends.
- remove unused JS_GetClassName()
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/963193003
|
|
This is a purely mechanical change, no new functionality.
- Expand some macros which were merely a short-cut to save
typing but reduced transparency.
- Put GET_VALUE_TYPE() implementation into a .cpp file.
This is a portion of the patch from issue 908033002 at
patchset 40001 (http://crrev.com/908033002#ps40001)
R=brucedawson@chromium.org
Review URL: https://codereview.chromium.org/927263003
|
|
R=brucedawson@chromium.org
Review URL: https://codereview.chromium.org/837533003
|
|
PDFium static initializers must go. Static initializers are prohibited
by the style guide. They have negative consequences including increased
startup time (from pulling in additional code pages) and reduced sharing
of data pages (since the variables can't go in the read-only data
segment).
This change uses a template struct and typed enums to reproduce
JS_CalcHash at run-time. An unsigned long long constant and masking with
0xFFFFFFFF are used to avoid compile errors due to integer overflow of
compile-time constants.
The HashVerify class is used to check the results, necessary since none
of the functions in global.cpp are called when pdfium_test.exe runs.
const_expr would be a much cleaner way to implement this change but it
is not yet widely supported.
On the Windows release build this reduces the code size (.text
virtual size) by 0x240 (576) bytes, the .data section by 0x20 bytes
(for eight unsigned globals), and the .rdata section by 0x20 bytes
(the unneeded string savings, minus the eight unsigned globals now
being there).
BUG=441899
R=tsepez@chromium.org
Review URL: https://codereview.chromium.org/792043005
|
|
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
|
|
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
|
|
|
|
|
