Delete some dead code.

Delete some dead code.

Found by a tweaked -Wunused-member-function Nuno Lopes is working on.

BUG=none
R=brettw@chromium.org, miket@chromium.org, rsleevi@chromium.org, sky@chromium.org
TBR=rogerta, sky

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=259767

[Nico, 2014-03-26 21:08:58]

brettw: base, gn
rogerta: google_apis
miket: devices
sky: ui
rsleevi: net

[miket_OOO, 2014-03-26 21:18:43]

> miket: devices

/device LGTM and thanks!

[Ryan Sleevi, 2014-03-26 21:22:44]

//net/cert LGTM
//net/socket not LGTM

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
File net/socket/ssl_client_socket_nss.cc (right):

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_nss.cc:360: for (size_t i = 0; i < other.size();
++i)
This = not lgtm

Not sure why Clang griped about this, but this change is inconsistent with how
line 361 works.

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_nss.cc:389: for (unsigned i = 0; i < size(); i++) {
Ditto this - inconsistent with using certs_[i] immediately below.

[Nico, 2014-03-26 21:35:12]

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
File net/socket/ssl_client_socket_nss.cc (right):

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_nss.cc:360: for (size_t i = 0; i < other.size();
++i)
On 2014/03/26 21:22:45, Ryan Sleevi wrote:
> This = not lgtm
> 
> Not sure why Clang griped about this, but this change is inconsistent with how
> line 361 works.

Why is this? It calls others.size(), not this->size().

(clang remarked that size() from line 333 was unused, so I figured I'd use it
instead of deleting it)

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_nss.cc:389: for (unsigned i = 0; i < size(); i++) {
On 2014/03/26 21:22:45, Ryan Sleevi wrote:
> Ditto this - inconsistent with using certs_[i] immediately below.

Ditto "I don't understand" :-)

[Ryan Sleevi, 2014-03-26 21:42:20]

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
File net/socket/ssl_client_socket_nss.cc (right):

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_nss.cc:360: for (size_t i = 0; i < other.size();
++i)
On 2014/03/26 21:35:12, Nico wrote:
> On 2014/03/26 21:22:45, Ryan Sleevi wrote:
> > This = not lgtm
> > 
> > Not sure why Clang griped about this, but this change is inconsistent with
how
> > line 361 works.
> 
> Why is this? It calls others.size(), not this->size().
> 
> (clang remarked that size() from line 333 was unused, so I figured I'd use it
> instead of deleting it)

Because it's inconsistently coupling implementations - size() [a public API]
with certs_ [an implementation detail]

If size() is used, it should be using operator[] (line 335)
If certs_[] is used, it should be coupled with certs_.size() (which it's doing
today)

I think we can just nuke size() altogether - ::Core now only uses empty() before
accessing [0], and the bits that accessed the other indices are rolled into this
class, via AsStringPieceVector().

[Nico, 2014-03-26 21:43:52]

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
File net/socket/ssl_client_socket_nss.cc (right):

https://codereview.chromium.org/213363003/diff/1/net/socket/ssl_client_socket...
net/socket/ssl_client_socket_nss.cc:360: for (size_t i = 0; i < other.size();
++i)
On 2014/03/26 21:42:20, Ryan Sleevi wrote:
> On 2014/03/26 21:35:12, Nico wrote:
> > On 2014/03/26 21:22:45, Ryan Sleevi wrote:
> > > This = not lgtm
> > > 
> > > Not sure why Clang griped about this, but this change is inconsistent with
> how
> > > line 361 works.
> > 
> > Why is this? It calls others.size(), not this->size().
> > 
> > (clang remarked that size() from line 333 was unused, so I figured I'd use
it
> > instead of deleting it)
> 
> Because it's inconsistently coupling implementations - size() [a public API]
> with certs_ [an implementation detail]
> 
> If size() is used, it should be using operator[] (line 335)
> If certs_[] is used, it should be coupled with certs_.size() (which it's doing
> today)
> 
> I think we can just nuke size() altogether - ::Core now only uses empty()
before
> accessing [0], and the bits that accessed the other indices are rolled into
this
> class, via AsStringPieceVector().

Ok, deleted size() instead.

[Ryan Sleevi, 2014-03-26 21:51:45]

LGTM

[brettw, 2014-03-26 22:12:03]

tools/gn LGTM

https://codereview.chromium.org/213363003/diff/50001/tools/gn/builder_unittes...
File tools/gn/builder_unittest.cc (right):

https://codereview.chromium.org/213363003/diff/50001/tools/gn/builder_unittes...
tools/gn/builder_unittest.cc:38: // Returns true if two loads have been
requested and they matchs the given
spelling matchs -> match

[Nico, 2014-03-26 22:17:13]

Thanks!

tbr'ing sky and rogerta

https://codereview.chromium.org/213363003/diff/50001/tools/gn/builder_unittes...
File tools/gn/builder_unittest.cc (right):

https://codereview.chromium.org/213363003/diff/50001/tools/gn/builder_unittes...
tools/gn/builder_unittest.cc:38: // Returns true if two loads have been
requested and they matchs the given
On 2014/03/26 22:12:04, brettw wrote:
> spelling matchs -> match

Done.

[Nico, 2014-03-26 22:17:18]

The CQ bit was checked by thakis@chromium.org

[commit-bot: I haz the power, 2014-03-26 22:18:28]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/thakis@chromium.org/213363003/70001

[sky, 2014-03-27 02:16:12]

LGTM

[Nico, 2014-03-27 02:46:34]

Committed patchset #5 manually as r259767 (presubmit successful).