Fix compiling of sync on linux.

Fix compiling of sync on linux.

[chron_chromium.org, 2009-09-21 23:31:11]

http://codereview.chromium.org/211019/diff/1/5
File chrome/browser/sync/syncable/directory_backing_store.cc (right):

http://codereview.chromium.org/211019/diff/1/5#newcode29
Line 29: const int kDirectoryBackingStoreBusyTimeoutMs =
std::numeric_limits<int>::max();
How did this come up during this CL?

http://codereview.chromium.org/211019/diff/1/7
File chrome/browser/sync/syncable/syncable.cc (right):

http://codereview.chromium.org/211019/diff/1/7#newcode104
Line 104: return base::strcasecmp(reinterpret_cast<const char *>(a),
Are a and b guaranteed to be null terminated? Does strcasecmp handle that
correctly?

http://codereview.chromium.org/211019/diff/1/7#newcode801
Line 801: DCHECK(1 == num_erased);
Why this change?

[zork, 2009-09-22 00:02:11]

On Mon, Sep 21, 2009 at 4:31 PM, <chron@chromium.org> wrote:

>
> http://codereview.chromium.org/211019/diff/1/5
> File chrome/browser/sync/syncable/directory_backing_store.cc (right):
>
> http://codereview.chromium.org/211019/diff/1/5#newcode29
> Line 29: const int kDirectoryBackingStoreBusyTimeoutMs =
> std::numeric_limits<int>::max();
> How did this come up during this CL?
>

If this is declared in the header file, all other files that include it
error with an unused variable warning.


>
> http://codereview.chromium.org/211019/diff/1/7
> File chrome/browser/sync/syncable/syncable.cc (right):
>
> http://codereview.chromium.org/211019/diff/1/7#newcode104
> Line 104: return base::strcasecmp(reinterpret_cast<const char *>(a),
> Are a and b guaranteed to be null terminated? Does strcasecmp handle
> that correctly?
>

Changed to use strncasecmp instead.


>
> http://codereview.chromium.org/211019/diff/1/7#newcode801
> Line 801: DCHECK(1 == num_erased);
> Why this change?


DCHECK_EQ causes signed/unsigned mismatch warnings, or something similar.


>
>
> http://codereview.chromium.org/211019
>

[chron_chromium.org, 2009-09-22 00:15:12]

LGTM