Fix four annoying warnings:

Fix four annoying warnings:
Two "set but unused", one of which is surely an artifact from
copying code around, and the other which ought to be used for
the sake of clarity.

Two are unknown "optimize" pragmas, remove them since the code
has been shipped for years on other platforms under full optimization.

R=thestig@chromium.org

Committed: https://pdfium.googlesource.com/pdfium/+/e6406b39d09f765f5f1a1530a3a737f17a0666c1

[Tom Sepez, 2015-05-20 21:33:11]

tsepez@chromium.org changed reviewers:
+ thestig@chromium.org

[Tom Sepez, 2015-05-20 21:33:11]

Lei, simple review.

[Lei Zhang, 2015-05-20 21:43:16]

+hans for the Clang question.

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_e...
File core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp (right):

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_e...
core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp:217: _DCTEncodeBitmap(pMaskDict,
pMaskBitmap, pParam ? pParam->nQuality : 75, mask_buf, mask_size);
BTW, did you notice  _DCTEncodeBitmap() and _JBIG2EncodeBitmap() are empty
functions??

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_e...
core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp:227: if (pMaskDict) {
BTW, this is always true now, which means |bDeleteMask| will always be false.

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
File core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp (right):

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp:76: #if _FXM_PLATFORM_ ==
_FX_PLATFORM_WINDOWS_
Hans: do you know if Clang for Windows will support this #pragma? If not, maybe
this should check for something like MSC_VER?

[hans, 2015-05-20 21:53:06]

hans@chromium.org changed reviewers:
+ hans@chromium.org

[hans, 2015-05-20 21:53:07]

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
File core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp (right):

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp:76: #if _FXM_PLATFORM_ ==
_FX_PLATFORM_WINDOWS_
On 2015/05/20 21:43:16, Lei Zhang wrote:
> Hans: do you know if Clang for Windows will support this #pragma? If not,
maybe
> this should check for something like MSC_VER?

It doesn't support it currently, and I don't see it getting support anytime
soon.

Note that clang on Windows also sneakily defines _MSC_VER, so you'd probably
want to check that __clang__ is not defined too.

On a higher level, why is this pragma needed? If the code relies on
optimizations being disabled to function properly, that sounds extremely dodgy.

[Tom Sepez, 2015-05-21 00:07:24]

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_e...
File core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp (right):

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_e...
core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp:217: _DCTEncodeBitmap(pMaskDict,
pMaskBitmap, pParam ? pParam->nQuality : 75, mask_buf, mask_size);
On 2015/05/20 21:43:16, Lei Zhang wrote:
> BTW, did you notice  _DCTEncodeBitmap() and _JBIG2EncodeBitmap() are empty
> functions??
No, I didn't. Seems like we might write uninitialized memory if those paths get
taken. But I dont think it can happen because icompress is passed only as false
in one place, and recursively here.  I'll make a follow-up CL to kill that code.

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_e...
core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp:227: if (pMaskDict) {
On 2015/05/20 21:43:16, Lei Zhang wrote:
> BTW, this is always true now, which means |bDeleteMask| will always be false.

Right, will remove the indent, and will remove the assignment to false.  I want
this variable to track the path taken above, and nothing else.

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
File core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp (right):

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp:76: #if _FXM_PLATFORM_ ==
_FX_PLATFORM_WINDOWS_
On 2015/05/20 21:43:16, Lei Zhang wrote:
> Hans: do you know if Clang for Windows will support this #pragma? If not,
maybe
> this should check for something like MSC_VER?

Remove it and HFTB.

[Lei Zhang, 2015-05-21 00:14:09]

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
File core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp (right):

https://codereview.chromium.org/1148353002/diff/20001/core/src/fpdfapi/fpdf_p...
core/src/fpdfapi/fpdf_page/fpdf_page_func.cpp:76: #if _FXM_PLATFORM_ ==
_FX_PLATFORM_WINDOWS_
On 2015/05/21 00:07:23, Tom Sepez wrote:
> On 2015/05/20 21:43:16, Lei Zhang wrote:
> > Hans: do you know if Clang for Windows will support this #pragma? If not,
> maybe
> > this should check for something like MSC_VER?
> 
> Remove it and HFTB.

Do you want to ask Jun about this? He may know why this was added in the first
place. AFAICT, this has been in the old internal SVN repo since the beginning.

[Lei Zhang, 2015-05-21 00:29:36]

lgtm

https://codereview.chromium.org/1148353002/diff/40001/core/src/fpdfapi/fpdf_e...
File core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp (right):

https://codereview.chromium.org/1148353002/diff/40001/core/src/fpdfapi/fpdf_e...
core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp:228: CPDF_Stream* pMaskStream =
NULL;
declare below where first used??

[Lei Zhang, 2015-05-21 00:34:52]

And the CL description needs to be updated re: pragmas.

[Tom Sepez, 2015-05-21 21:53:21]

https://codereview.chromium.org/1148353002/diff/40001/core/src/fpdfapi/fpdf_e...
File core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp (right):

https://codereview.chromium.org/1148353002/diff/40001/core/src/fpdfapi/fpdf_e...
core/src/fpdfapi/fpdf_edit/fpdf_edit_image.cpp:228: CPDF_Stream* pMaskStream =
NULL;
On 2015/05/21 00:29:36, Lei Zhang wrote:
> declare below where first used??

Done.

[Tom Sepez, 2015-05-21 21:54:23]

Committed patchset #4 (id:60001) manually as
e6406b39d09f765f5f1a1530a3a737f17a0666c1 (presubmit successful).