Use snprintf over sprintf for a few places

Use sstream over sprintf

gcc warns about sprintf being unsafe in some cases.
Also snprintf is not available under windows.

BUG=

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=249365

[boliu, 2014-02-04 04:23:32]

Huh, won't compile on windows :/

http://stackoverflow.com/questions/2915672/snprintf-and-visual-studio-2010

[no sievers, 2014-02-06 00:04:25]

On 2014/02/04 04:23:32, boliu wrote:
> Huh, won't compile on windows :/
> 
> http://stackoverflow.com/questions/2915672/snprintf-and-visual-studio-2010

Maybe this can be rewritten in C++, for example std::hex or std::stoi (with
base=16)?

[boliu, 2014-02-06 01:08:00]

Good idea...

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
gpu/command_buffer/client/gles2_implementation.cc:15: #include <string>
presubmit was warning <string.h> is in wrong order, I think it was a mistake and
should have been <string> in the first place

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
gpu/command_buffer/client/gles2_implementation.cc:125: this_in_hex_ = ss.str();
difference from previous code is that if this is null, then it used to print
"nil" whereas it now prints "0". Luckily that case should never happen.

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/commo...
File gpu/command_buffer/common/gles2_cmd_utils.cc (right):

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/commo...
gpu/command_buffer/common/gles2_cmd_utils.cc:695: std::stringstream ss;
way more verbose, but should match sprintf implementation

[no sievers, 2014-02-06 01:33:37]

lgtm

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
gpu/command_buffer/client/gles2_implementation.cc:125: this_in_hex_ = ss.str();
On 2014/02/06 01:08:00, boliu wrote:
> difference from previous code is that if this is null, then it used to print
> "nil" whereas it now prints "0". Luckily that case should never happen.

nit: can be simply 'this_in_hex_ = (std::hex << this).str()'

And probably don't need to include <sstream> then. (But maybe <ios>?)

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/commo...
File gpu/command_buffer/common/gles2_cmd_utils.cc (right):

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/commo...
gpu/command_buffer/common/gles2_cmd_utils.cc:695: std::stringstream ss;
On 2014/02/06 01:08:00, boliu wrote:
> way more verbose, but should match sprintf implementation

You can do: 
std::stringstream ss;
ss.fill('0');
ss.width(value < 0x10000 ? 4 : 8);
ss << std::hex << value;
return "0x" + ss.str();

[boliu, 2014-02-06 01:54:45]

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
File gpu/command_buffer/client/gles2_implementation.cc (right):

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/clien...
gpu/command_buffer/client/gles2_implementation.cc:125: this_in_hex_ = ss.str();
On 2014/02/06 01:33:37, sievers wrote:
> On 2014/02/06 01:08:00, boliu wrote:
> > difference from previous code is that if this is null, then it used to print
> > "nil" whereas it now prints "0". Luckily that case should never happen.
> 
> nit: can be simply 'this_in_hex_ = (std::hex << this).str()'
> 
> And probably don't need to include <sstream> then. (But maybe <ios>?)

Ehh, std::hex returns ios_base, which doesn't support operator<< (that's on
ostream) or str

If I'm reading the signatures correctly, in "std::cout << std::hex", std::hex is
a function which is passed to operator<< as a function pointer...oh c++....

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/commo...
File gpu/command_buffer/common/gles2_cmd_utils.cc (right):

https://codereview.chromium.org/152843003/diff/90001/gpu/command_buffer/commo...
gpu/command_buffer/common/gles2_cmd_utils.cc:695: std::stringstream ss;
On 2014/02/06 01:33:37, sievers wrote:
> On 2014/02/06 01:08:00, boliu wrote:
> > way more verbose, but should match sprintf implementation
> 
> You can do: 
> std::stringstream ss;
> ss.fill('0');
> ss.width(value < 0x10000 ? 4 : 8);
> ss << std::hex << value;
> return "0x" + ss.str();

Done, kind of amused that char* + string compiles...

[boliu, 2014-02-06 04:21:17]

The CQ bit was checked by [email protected]

[commit-bot: I haz the power, 2014-02-06 04:23:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/[email protected]/152843003/10003

[commit-bot: I haz the power, 2014-02-06 11:38:52]

Change committed as 249365