An initial curvecp implementation. This implementation is not complete,

An initial curvecp implementation.  This implementation is not complete,
but is good enough to start collaboration.

BUG=none
TEST=curvecp_unittests
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=85753

[Peter Mayo, 2011-05-18 18:30:44]

http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc
File net/curvecp/messenger.cc (right):

http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc#newcod...
net/curvecp/messenger.cc:281: if (body_length < 0)
Checking to see if an unsigned int is less than zero produces a warning in my
compile, which is treated as an error and fails.

[willchan no longer on Chromium, 2011-05-18 19:23:22]

http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc
File net/curvecp/messenger.cc (right):

http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc#newcod...
net/curvecp/messenger.cc:281: if (body_length < 0)
On 2011/05/18 18:30:44, Peter Mayo wrote:
> Checking to see if an unsigned int is less than zero produces a warning in my
> compile, which is treated as an error and fails.

Hah, yes, this looks like a bug :) Mike should fix this at some point. I'm
surprised you are hitting this though, are you building all targets? We aren't
linking this into chrome and bots are green.

[Peter Mayo (wrong one), 2011-05-18 20:08:28]

I asked my team that question too, but have not dived into find the answer
myself yet.  Will soon.

(FYI Touch bots are not green, for this and other reasons)

Peter.

On Wed, May 18, 2011 at 3:23 PM, <willchan@chromium.org> wrote:

>
> http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc
> File net/curvecp/messenger.cc (right):
>
>
>
http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc#newcod...
> net/curvecp/messenger.cc:281: if (body_length < 0)
> On 2011/05/18 18:30:44, Peter Mayo wrote:
>
>> Checking to see if an unsigned int is less than zero produces a
>>
> warning in my
>
>> compile, which is treated as an error and fails.
>>
>
> Hah, yes, this looks like a bug :) Mike should fix this at some point.
> I'm surprised you are hitting this though, are you building all targets?
> We aren't linking this into chrome and bots are green.
>
>
> http://codereview.chromium.org/7037022/
>

[mbelshe, 2011-05-18 20:27:50]

fix pending.

not sure why you are seeing the error and i am not.  you must have stiffer
warnings enabled somehow

thanks for the report!

On Wed, May 18, 2011 at 1:08 PM, Peter Mayo <petermayo@google.com> wrote:

> I asked my team that question too, but have not dived into find the answer
> myself yet.  Will soon.
>
> (FYI Touch bots are not green, for this and other reasons)
>
> Peter.
>
>
> On Wed, May 18, 2011 at 3:23 PM, <willchan@chromium.org> wrote:
>
>>
>> http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc
>> File net/curvecp/messenger.cc (right):
>>
>>
>>
http://codereview.chromium.org/7037022/diff/1/net/curvecp/messenger.cc#newcod...
>> net/curvecp/messenger.cc:281: if (body_length < 0)
>> On 2011/05/18 18:30:44, Peter Mayo wrote:
>>
>>> Checking to see if an unsigned int is less than zero produces a
>>>
>> warning in my
>>
>>> compile, which is treated as an error and fails.
>>>
>>
>> Hah, yes, this looks like a bug :) Mike should fix this at some point.
>> I'm surprised you are hitting this though, are you building all targets?
>> We aren't linking this into chrome and bots are green.
>>
>>
>> http://codereview.chromium.org/7037022/
>>
>
>