win: avoid warnings in string_number_conversion_test.cc

win: avoid warnings in string_number_conversion_test.cc

R=mark@chromium.org
BUG=crashpad:1
Committed: https://chromium.googlesource.com/crashpad/crashpad/+/4263334db8f8085697cbcecb67966bb19440ec3e

[Mark Mentovai, 2014-12-16 14:17:47]

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
File util/stdlib/string_number_conversion_test.cc (right):

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
util/stdlib/string_number_conversion_test.cc:56: #if defined(COMPILER_MSVC)
build/build_config.h for this if it’s still necessary—see the next comment.

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
util/stdlib/string_number_conversion_test.cc:61: {"-0x80000000", true,
-2147483648},
Would static_cast<>() also silence the warning? I think that’d be preferable.

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
util/stdlib/string_number_conversion_test.cc:217: // Ensure that embedded NUL
characters are treated as bad input.
The existing comment was duplicated from above, so your note about splitting the
string should be too.

[scottmg, 2014-12-16 17:53:56]

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
File util/stdlib/string_number_conversion_test.cc (right):

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
util/stdlib/string_number_conversion_test.cc:61: {"-0x80000000", true,
-2147483648},
On 2014/12/16 14:17:47, Mark Mentovai wrote:
> Would static_cast<>() also silence the warning? I think that’d be preferable.

Cast inside i.e. -static_cast<int>(2147483648) avoids the warning but looks a
little strange.

... It's going to be 0x80000000 regardless of -, perhaps just writing that?

https://codereview.chromium.org/807463004/diff/1/util/stdlib/string_number_co...
util/stdlib/string_number_conversion_test.cc:217: // Ensure that embedded NUL
characters are treated as bad input.
On 2014/12/16 14:17:47, Mark Mentovai wrote:
> The existing comment was duplicated from above, so your note about splitting
the
> string should be too.

Done.

[Mark Mentovai, 2014-12-16 18:42:56]

LGTM

https://codereview.chromium.org/807463004/diff/20001/util/stdlib/string_numbe...
File util/stdlib/string_number_conversion_test.cc (right):

https://codereview.chromium.org/807463004/diff/20001/util/stdlib/string_numbe...
util/stdlib/string_number_conversion_test.cc:56: {"-0x80000000", true,
-static_cast<int>(2147483648)},
This does look very strange. I guess I was misunderstanding what MSVS was
warning about initially.

I meant static_cast<int>(-2147483648), but I can see now that this wouldn’t work
either.

Let’s just use std::numeric_limits<int>::min() like we do for the decimal case
on line 40.

[scottmg, 2014-12-16 18:59:56]

https://codereview.chromium.org/807463004/diff/20001/util/stdlib/string_numbe...
File util/stdlib/string_number_conversion_test.cc (right):

https://codereview.chromium.org/807463004/diff/20001/util/stdlib/string_numbe...
util/stdlib/string_number_conversion_test.cc:56: {"-0x80000000", true,
-static_cast<int>(2147483648)},
On 2014/12/16 18:42:56, Mark Mentovai wrote:
> This does look very strange. I guess I was misunderstanding what MSVS was
> warning about initially.
> 
> I meant static_cast<int>(-2147483648), but I can see now that this wouldn’t
work
> either.
> 
> Let’s just use std::numeric_limits<int>::min() like we do for the decimal case
> on line 40.

Done.

[scottmg, 2014-12-16 19:00:26]

Committed patchset #3 (id:40001) manually as
4263334db8f8085697cbcecb67966bb19440ec3e (presubmit successful).