Use proper pair type in range-based for map iterations.

Use proper pair type in range-based for map iterations.

Since map<Key, Value>::value_type is std::pair<const Key, Value>, we
need to ensure that we use this type to avoid an extra copy/temporary
creation. This patch does this by explicitly using value_type where
it is clear what the type is, and adding const to the key where it's not.

Committed: https://crrev.com/49418e62a7ea52dfa99b3316b0e7b5cf79f98844
Cr-Commit-Position: refs/heads/master@{#311524}

[vmpstr, 2015-01-13 19:48:07]

vmpstr@chromium.org changed reviewers:
+ bauerb@chromium.org, rdsmith@chromium.org, sdefresne@chromium.org,
yoz@chromium.org

[vmpstr, 2015-01-13 19:48:08]

Please take a look.

+rdsmith for net/quic
+sdefresne for chrome/browser/history
+bauerb for chrome/browser/supervised_user
+yoz for extensions/browser

My personal preference would be to use const auto& here, so feel free to tell me
if you want me to do that.

[Randy Smith (Not in Mondays), 2015-01-13 19:54:53]

rdsmith@chromium.org changed reviewers:
+ rch@chromium.org

[Randy Smith (Not in Mondays), 2015-01-13 19:54:54]

net/quic -> rch.

[Ryan Hamilton, 2015-01-13 19:57:42]

net/quic: LGTM

[Yoyo Zhou, 2015-01-13 20:48:33]

yoz@chromium.org changed reviewers:
+ reillyg@chromium.org

[Yoyo Zhou, 2015-01-13 20:48:34]

api/hid -> reillyg
I'd be ok with auto, but Reilly is a more specific owner.

[Reilly Grant (use Gerrit), 2015-01-13 21:11:12]

hid_device_manager lgtm

I'm fine with whatever as long as it fits on one line. On the next lines I name
value_type.first and value_type.second. I suppose it depends on the compiler
optimizations whether it matters that I use int vs. const int&.

[Bernhard Bauer, 2015-01-13 23:16:40]

https://codereview.chromium.org/853533002/diff/1/chrome/browser/supervised_us...
File chrome/browser/supervised_user/supervised_user_url_filter.cc (right):

https://codereview.chromium.org/853533002/diff/1/chrome/browser/supervised_us...
chrome/browser/supervised_user/supervised_user_url_filter.cc:318: for (const
std::map<std::string, bool>::value_type& host_entry : host_map_) {
I would be in favor of const auto& here. In which case, c/b/supervised_user/
LGTM.

[vmpstr, 2015-01-13 23:37:07]

https://codereview.chromium.org/853533002/diff/1/chrome/browser/supervised_us...
File chrome/browser/supervised_user/supervised_user_url_filter.cc (right):

https://codereview.chromium.org/853533002/diff/1/chrome/browser/supervised_us...
chrome/browser/supervised_user/supervised_user_url_filter.cc:318: for (const
std::map<std::string, bool>::value_type& host_entry : host_map_) {
On 2015/01/13 23:16:40, Bernhard Bauer wrote:
> I would be in favor of const auto& here. In which case, c/b/supervised_user/
> LGTM.

Done.

[vmpstr, 2015-01-14 02:26:31]

vmpstr@chromium.org changed reviewers:
+ sky@chromium.org

[vmpstr, 2015-01-14 02:26:31]

+sky for chrome/browser/history/expire_history_backend_unittest.cc. Please take
a look.

[sdefresne, 2015-01-14 10:47:08]

lgtm

https://codereview.chromium.org/853533002/diff/20001/chrome/browser/history/e...
File chrome/browser/history/expire_history_backend_unittest.cc (right):

https://codereview.chromium.org/853533002/diff/20001/chrome/browser/history/e...
chrome/browser/history/expire_history_backend_unittest.cc:336: for (const
URLsDeletedNotificationList::value_type& pair :
nit: I slightly prefer const auto& here.

[vmpstr, 2015-01-14 17:31:58]

The CQ bit was checked by vmpstr@chromium.org

[vmpstr, 2015-01-14 17:32:16]

https://codereview.chromium.org/853533002/diff/20001/chrome/browser/history/e...
File chrome/browser/history/expire_history_backend_unittest.cc (right):

https://codereview.chromium.org/853533002/diff/20001/chrome/browser/history/e...
chrome/browser/history/expire_history_backend_unittest.cc:336: for (const
URLsDeletedNotificationList::value_type& pair :
On 2015/01/14 10:47:08, sdefresne wrote:
> nit: I slightly prefer const auto& here.

Done.

[commit-bot: I haz the power, 2015-01-14 17:32:33]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/853533002/40001

[commit-bot: I haz the power, 2015-01-14 19:08:04]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-01-14 19:09:29]

Patchset 3 (id:??) landed as
https://crrev.com/49418e62a7ea52dfa99b3316b0e7b5cf79f98844
Cr-Commit-Position: refs/heads/master@{#311524}