Correct iOS device build for latest Xcode beta.

Correct iOS device build for latest Xcode beta.

Stop referring to dylib directly, and instead use the linker shortcut.
With the latest Xcode beta, some dylibs are now named .tbd, so
referencing the actual path for simulator (dylib) and device (tbd) would
make for a very complicated build file.

Using the linker short name bypases all of that.

BUG=499098
TBRs for small iOS only gyp changes.
TBR=scottmg@chromium.org (libxml)
TBR=michaeln@chromium.org (sqlite)

Committed: https://crrev.com/23bb216c5f30782e97f6a51bcafb53f6607ab051
Cr-Commit-Position: refs/heads/master@{#333978}

[justincohen, 2015-06-10 20:53:23]

justincohen@chromium.org changed reviewers:
+ sdefresne@chromium.org

[justincohen, 2015-06-10 20:53:24]

This seems to be a cleaner attempt to work around the new library files in the
latest Xcode beta.

[justincohen, 2015-06-10 21:14:19]

justincohen@chromium.org changed reviewers:
+ stuartmorgan@chromium.org

[justincohen, 2015-06-10 21:14:20]

stuartmorgan@ We use similar syntax in base.gyp.  WDYT?

[rohitrao (ping after 24h), 2015-06-10 21:18:34]

rohitrao@chromium.org changed reviewers:
+ rohitrao@chromium.org

[rohitrao (ping after 24h), 2015-06-10 21:18:35]

Is there a bug for this?  Can you update the description (and the bug) to
mention that the extension varies between device and simulator (.tbd vs .dylib),
and that passing the -l flag Just Works?

[stuartmorgan, 2015-06-10 21:26:43]

Huh, so we do. I'm fine with the approach, but as Rohit said we really should
have a bug tracking this since this is at least the second patch.

[justincohen, 2015-06-11 00:40:44]

Updated the comment and added a bug, and changed to use the Xcode correct format
of OTHER_LDFLAGS.  Xcode and ninja should both compile now, but waiting for the
trybots to confirm.

PTAL!

[justincohen, 2015-06-11 11:28:29]

justincohen@chromium.org changed reviewers:
+ droger@chromium.org

[droger, 2015-06-11 11:35:35]

LGTM
Approach looks good, but please wait for sdefresne approval too, he's more
familiar than me with this.

[sdefresne, 2015-06-11 14:13:11]

lgtm with a nit (so that it is easier to add other values in the xcode_settings)

https://codereview.chromium.org/1179573004/diff/20001/ios/crnet/crnet_consume...
File ios/crnet/crnet_consumer/crnet_consumer.gyp (right):

https://codereview.chromium.org/1179573004/diff/20001/ios/crnet/crnet_consume...
ios/crnet/crnet_consumer/crnet_consumer.gyp:55: ]
nit: "]," and everywhere else

[justincohen, 2015-06-11 14:25:12]

https://codereview.chromium.org/1179573004/diff/20001/ios/crnet/crnet_consume...
File ios/crnet/crnet_consumer/crnet_consumer.gyp (right):

https://codereview.chromium.org/1179573004/diff/20001/ios/crnet/crnet_consume...
ios/crnet/crnet_consumer/crnet_consumer.gyp:55: ]
On 2015/06/11 14:13:11, sdefresne wrote:
> nit: "]," and everywhere else

Done.

[justincohen, 2015-06-11 15:04:56]

justincohen@chromium.org changed reviewers:
+ rsleevi@chromium.org

[justincohen, 2015-06-11 15:04:56]

rsleevi@ for /net OWNERS, small iOS only gyp change.

PTAL!

[Ryan Sleevi, 2015-06-11 16:13:47]

lgtm

[justincohen, 2015-06-11 16:13:56]

The CQ bit was checked by justincohen@chromium.org

[justincohen, 2015-06-11 16:13:57]

The patchset sent to the CQ was uploaded after l-g-t-m from droger@chromium.org,
sdefresne@chromium.org
Link to the patchset: https://codereview.chromium.org/1179573004/#ps40001
(title: "nit")

[commit-bot: I haz the power, 2015-06-11 16:15:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1179573004/40001

[commit-bot: I haz the power, 2015-06-11 17:39:17]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-06-11 17:40:11]

Patchset 3 (id:??) landed as
https://crrev.com/23bb216c5f30782e97f6a51bcafb53f6607ab051
Cr-Commit-Position: refs/heads/master@{#333978}