Age | Commit message (Collapse) | Author |
|
Edge closer to the goal of building PDFium with the chromium_code
configuration.
BUG=https://code.google.com/p/pdfium/issues/detail?id=29
R=bo_xu@foxitsoftware.com, thakis@chromium.org
Review URL: https://codereview.chromium.org/441763002
|
|
When newPos == file size, the current block will not be read or Get. If this block is a crucial part of the document (like m_pTrailer), the program will exit with parse error and
the document will not be rendered.
BUG=None
R=jun_fang@foxitsoftware.com
Review URL: https://codereview.chromium.org/440563003
|
|
No intended behavior change.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/436483002
|
|
It's unused, and it caused a warning about CPDFSDK_Widget::ResetAppearance()
failing to override it (since these two unrelated methods had the same name).
No intended behavior change.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/429483004
|
|
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.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/426763004
|
|
There are many warnings that look like:
error: 'CPWL_RadioButton::OnChar' hides overloaded virtual function [-Werror,-Woverloaded-virtual]
virtual FX_BOOL OnChar(FX_WORD nChar);
^
note: hidden overloaded virtual function 'CPWL_Wnd::OnChar' declared here: different number of parameters (2 vs 1)
virtual FX_BOOL OnChar(FX_WORD nChar, FX_DWORD nFlag);
^
It looks like someone added the nFlag parameter to the methods in CPWL_Wnd
at some point and missed to update all overloads This patch attempts to fix this:
It adds the parameter to all methods that look like they're trying to overload the base
class method, and renames the method in one case where it fairly clearly looks like
that it's not supposed to be an overload.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/416383004
|
|
fsdk_baseform.h:63:19: error: 'CPDFSDK_Widget::GetLayoutOrder' hides overloaded virtual function [-Werror,-Woverloaded-virtual]
virtual int GetLayoutOrder() {return 2;}
^
fsdk_baseannot.h:70:18: note: hidden overloaded virtual function 'CPDFSDK_Annot::GetLayoutOrder' declared here: different qualifiers (const vs none)
virtual int GetLayoutOrder() const { return 5; }
^
On Windows, I believe MSVS treats these as override since it's such a common and
easy mistake, but clang and gcc do what the standard specifies. Add a "const" to
the function in the subclass so that this is actually an override, as intended.
BUG=pdfium:29
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/427173002
|
|
fpdfview.cpp
BUG=397258
R=tsepez@chromium.org
Review URL: https://codereview.chromium.org/419063002
|
|
If somehow different length values could be obtained by two successive calls
to Doc_getFilePath() (and FieldBrowse() for that matter), and the method is
true to the API documentation that says "The return value always indicated
number of bytes required for the buffer, even when there is no buffer
specified, or the buffer size is less then required", then it is possible
to get a returned length describing memory beyond the current buffer.
We can make the corresponding JS_docGetFilePath() method more robust against
this case by applying better checks to the returned value.
This probably is unrelated since ASAN seems to be flagging the corresponding bug
as UAF, but doesn't hurt to make things more robust.
BUG=392956
R=jun_fang@foxitsoftware.com
Review URL: https://codereview.chromium.org/423233002
|
|
Follow-up from https://codereview.chromium.org/424883002/
- Remove some stray whitespace.
- Fix "else after return".
- Remove unused swResponse local.
- Treat unexpectedly large responses as errors.
BUG=
R=jun_fang@foxitsoftware.com
Review URL: https://codereview.chromium.org/423953002
|
|
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
|
|
Found by clang's -Wunused-variable, -Wunused-function, -Wunused-const-variable.
BUG=none
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/404653005
|
|
BUG=395266
R=jun_fang@foxitsoftware.com
Review URL: https://codereview.chromium.org/411713003
|
|
BUG=396255
R=jun_fang@foxitsoftware.com
Review URL: https://codereview.chromium.org/407243003
|
|
BUG=179413
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/410073002
|
|
The function is looking ahead N characters at both its "format" and "value"
strings without validating that accesses are in bounds. Add those validations.
There are also duplicate checks in the else-branches which re-test the inverse
of the if-branch. These are removed for simplicity.
I also tidied some stray whitespace in the function while I was at it.
BUG=393831
R=jun_fang@foxitsoftware.com
Review URL: https://codereview.chromium.org/395303004
|
|
Calling `delete` on an object of a type that has virtual functions but
not a virtual destructor is questionable: Since the object has virtual functions,
it likely has subclasses, so if it's deleted through the base pointer and the
destructor isn't virtual, the subclass destructor won't be called.
In most cases, the classes getting deleted can just be marked final to tell
the compiler that it can't possibly have subclasses (this also enables the
compiler to generate better code).
Two classes didn't have any sub- or superclasses but virtual functions -
this doesn't make sense, so make all methods of these classes non-virtual.
(Also delete an unused function on one of the two classes.)
In one case, a class actually did have a subclass that needs to be deleted
virtually, so mark one destructor as virtual.
BUG=none
R=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/370853002
|
|
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
|
|
Fixes a warning.
BUG=
TBR=bo_xu@foxitsoftware.com
Review URL: https://codereview.chromium.org/395293002
|
|
This change adds the support to extract "NumCopies", "PrintPageRange", "Duplex" viewer preferences for printing.
BUG=169120
R=bo_xu@foxitsoftware.com, vitalybuka@chromium.org
Review URL: https://codereview.chromium.org/345123002
|
|
BUG=381521
R=palmer@chromium.org
Review URL: https://codereview.chromium.org/383563002
|
|
This contains fixes for the following sorts of issues:
* Nonstandard extension: unnamed struct/union
This also changes the signature of FPDF_FillRect() to enable the cleanest
possible changes in https://codereview.chromium.org/372273005 .
BUG=81439
TEST=none
Original patch by Peter Kasting <pkasting@chromium.org>
Original code review: https://codereview.chromium.org/376003003/
R=pkasting@chromium.org
Review URL: https://codereview.chromium.org/377353002
|
|
Fixes for re-enabling more MSVC level 4 warnings: pdfium/ edition
BUG=81439
TBR=pkasting@chromium.org
Review URL: https://codereview.chromium.org/376043003
|
|
This contains fixes for the following sorts of issues:
* Nonstandard extension: unnamed struct/union
This also changes the signature of FPDF_FillRect() to enable the cleanest
possible changes in https://codereview.chromium.org/372273005 .
BUG=81439
TEST=none
Original patch by Peter Kasting <pkasting@chromium.org>
Original code review: https://codereview.chromium.org/376003003/
R=pkasting@chromium.org
Review URL: https://codereview.chromium.org/372383004
|
|
BUG=
R=palmer@chromium.org
Review URL: https://codereview.chromium.org/372473003
|
|
BUG=
R=thakis@chromium.org
Review URL: https://codereview.chromium.org/361553002
|
|
We'll use this for integer overflows going forward.
BUG=382606
R=jam@chromium.org
Review URL: https://codereview.chromium.org/349363005
|
|
BUG=382639
R=mdempsky@chromium.org
Review URL: https://codereview.chromium.org/354673002
|
|
This reverts commit d9713f05fdcecab8428d39034c6b84cd0bbd2920.
This is breaking compile.
|
|
We'll use this for integer overflows going forward.
BUG=382606
R=bo_xu@foxitsoftware.com, jschuh@chromium.org
Review URL: https://codereview.chromium.org/341533007
|
|
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
|
|
R=jam@chromium.org
Review URL: https://codereview.chromium.org/336563004
|
|
R=jam@chromium.org
Review URL: https://codereview.chromium.org/333753002
|
|
characters
BUG=
Review URL: https://codereview.chromium.org/301263002
|
|
BUG=security
Review URL: https://codereview.chromium.org/307653002
|
|
R=jam@chromium.org
Review URL: https://codereview.chromium.org/297993002
|
|
|
|
of file" warning.
|
|
into errors. It also makes it clearer to find usage of v8 in the library.
|
|
|