Use the types from stdint.h in MD5 API.

Use the types from stdint.h in MD5 API.

BUG=138542
TEST=base_unittests --gtest_filter=MD5*
R=thakis@chromium.org

Committed: https://crrev.com/66e103df9bce01de86298644ef4572dbf9635276
Cr-Commit-Position: refs/heads/master@{#323277}

[tfarina, 2015-04-01 12:23:30]

PTAL

[Nico, 2015-04-01 14:49:49]

lgtm with comments addressed

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc
File base/md5.cc (right):

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc#newcode27
base/md5.cc:27: #include <stdint.h>  // For intptr_t.
I think we usually don't do "// For $FOO" comments on headers. Can you remove
these two comments?

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc#newcode44
base/md5.cc:44: ((unsigned)buf[1]<<8 | buf[0]);
nit: indent 2

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc#newcode68
base/md5.cc:68: register uint32_t a, b, c, d;
If you're touching this line anyways, can you remove the "register" too? That
keyword was deprecated in c++11. (optional)

[tfarina, 2015-04-01 15:51:28]

Thanks for the review! Pushing to CQ.

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc
File base/md5.cc (right):

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc#newcode27
base/md5.cc:27: #include <stdint.h>  // For intptr_t.
On 2015/04/01 14:49:49, Nico wrote:
> I think we usually don't do "// For $FOO" comments on headers. Can you remove
> these two comments?

I meant to remove those when I copied/pasted, but that is what you get for doing
things on mid-night.

Done.

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc#newcode44
base/md5.cc:44: ((unsigned)buf[1]<<8 | buf[0]);
On 2015/04/01 14:49:49, Nico wrote:
> nit: indent 2

clang-formatted.

Done.

https://codereview.chromium.org/1051703002/diff/1/base/md5.cc#newcode68
base/md5.cc:68: register uint32_t a, b, c, d;
On 2015/04/01 14:49:49, Nico wrote:
> If you're touching this line anyways, can you remove the "register" too? That
> keyword was deprecated in c++11. (optional)

Done.

[tfarina, 2015-04-01 15:51:39]

The CQ bit was checked by tfarina@chromium.org

[tfarina, 2015-04-01 15:51:39]

The patchset sent to the CQ was uploaded after l-g-t-m from thakis@chromium.org
Link to the patchset: https://codereview.chromium.org/1051703002/#ps20001
(title: "thakis review")

[commit-bot: I haz the power, 2015-04-01 15:52:03]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1051703002/20001

[commit-bot: I haz the power, 2015-04-01 17:36:38]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-04-01 17:37:12]

Patchset 2 (id:??) landed as
https://crrev.com/66e103df9bce01de86298644ef4572dbf9635276
Cr-Commit-Position: refs/heads/master@{#323277}

[Nico, 2015-04-01 17:38:40]

https://codereview.chromium.org/1051703002/diff/20001/base/md5.cc
File base/md5.cc (right):

https://codereview.chromium.org/1051703002/diff/20001/base/md5.cc#newcode43
base/md5.cc:43: t = (uint32_t)((unsigned)buf[3] << 8 | buf[2]) << 16 |
Hm, running clang-format on files that aren't clang-formatted is weird. Now the
inner block has an indent of 2 whikle the rest of the file uses an indent of 8.

[tfarina, 2015-04-01 17:57:48]

https://codereview.chromium.org/1051703002/diff/20001/base/md5.cc
File base/md5.cc (right):

https://codereview.chromium.org/1051703002/diff/20001/base/md5.cc#newcode43
base/md5.cc:43: t = (uint32_t)((unsigned)buf[3] << 8 | buf[2]) << 16 |
On 2015/04/01 17:38:40, Nico wrote:
> Hm, running clang-format on files that aren't clang-formatted is weird. Now
the
> inner block has an indent of 2 whikle the rest of the file uses an indent of
8.

I'm open to do anything you prefer. I think we have some options here. Just let
me know and I will upload another patch. Sorry for not realizing you were not
asking me to run clang-format :/

1- go back to 8 spaces with indentation fixed.
2- clang-format the whole file (I agree this is worse).
3- make this function 2 spaces indented.

WebRTC's MD5 implementation seems to have indented everything 2 spaces:
https://chromium.googlesource.com/external/webrtc/+/master/webrtc/base/md5.cc

[Nico, 2015-04-01 18:02:13]

On 2015/04/01 17:57:48, tfarina wrote:
> https://codereview.chromium.org/1051703002/diff/20001/base/md5.cc
> File base/md5.cc (right):
> 
> https://codereview.chromium.org/1051703002/diff/20001/base/md5.cc#newcode43
> base/md5.cc:43: t = (uint32_t)((unsigned)buf[3] << 8 | buf[2]) << 16 |
> On 2015/04/01 17:38:40, Nico wrote:
> > Hm, running clang-format on files that aren't clang-formatted is weird. Now
> the
> > inner block has an indent of 2 whikle the rest of the file uses an indent of
> 8.
> 
> I'm open to do anything you prefer. I think we have some options here. Just
let
> me know and I will upload another patch. Sorry for not realizing you were not
> asking me to run clang-format :/
> 
> 1- go back to 8 spaces with indentation fixed.
> 2- clang-format the whole file (I agree this is worse).
> 3- make this function 2 spaces indented.

Probably easiest to just clang-format the whole file.