If poppler is disable and native pdf is enabled, we still need SkPDFRasterizer.cpp

If poppler is disable and native pdf is enabled, we still need SkPDFRasterizer.cpp

TBR=reed@google.com
NOTREECHECKS=true
NOTRY=true

Committed: http://code.google.com/p/skia/source/detail?r=13025

[commit-bot: I haz the power, 2014-01-10 18:42:19]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/vandebo@chromium.org/133583004/1

[reed1, 2014-01-10 18:44:09]

can we explicitly try this against the 10.8 builder, since that is the one that
broke earlier?

[vandebo (ex-Chrome), 2014-01-10 18:44:49]

On 2014/01/10 18:44:09, reed1 wrote:
> can we explicitly try this against the 10.8 builder, since that is the one
that
> broke earlier?

I tried it locally - repro'd the break and this fixed it.

[reed1, 2014-01-10 18:45:37]

you may have to commit this manually, since the CQ will wait while the tree is
green

[reed1, 2014-01-10 18:45:47]

On 2014/01/10 18:45:37, reed1 wrote:
> you may have to commit this manually, since the CQ will wait while the tree is
> green

I mean red

[epoger, 2014-01-10 18:52:32]

On 2014/01/10 18:45:47, reed1 wrote:
> On 2014/01/10 18:45:37, reed1 wrote:
> > you may have to commit this manually, since the CQ will wait while the tree
is
> > green
> 
> I mean red

If you add a "NOTREECHECKS=True" line in the CL description, the CQ will commit
it even if the tree is red.

I have filed https://code.google.com/p/skia/issues/detail?id=2013 ('document
NOTREECHECKS, NOTRY, etc. somewhere in skiadocs')

[reed1, 2014-01-10 18:53:59]

On 2014/01/10 18:52:32, epoger wrote:
> On 2014/01/10 18:45:47, reed1 wrote:
> > On 2014/01/10 18:45:37, reed1 wrote:
> > > you may have to commit this manually, since the CQ will wait while the
tree
> is
> > > green
> > 
> > I mean red
> 
> If you add a "NOTREECHECKS=True" line in the CL description, the CQ will
commit
> it even if the tree is red.
> 
> I have filed https://code.google.com/p/skia/issues/detail?id=2013 ('document
> NOTREECHECKS, NOTRY, etc. somewhere in skiadocs')

Sometimes (perhaps in this case) we need to commit a tree-breaker fix, as part
of several fixes, meaning even w/ this change, the trybot may still see red (but
we should commit anyway). In those cases I think we need a manual commit, yes?

[epoger, 2014-01-10 18:56:54]

On 2014/01/10 18:53:59, reed1 wrote:
> On 2014/01/10 18:52:32, epoger wrote:
> > On 2014/01/10 18:45:47, reed1 wrote:
> > > On 2014/01/10 18:45:37, reed1 wrote:
> > > > you may have to commit this manually, since the CQ will wait while the
> tree
> > is
> > > > green
> > > 
> > > I mean red
> > 
> > If you add a "NOTREECHECKS=True" line in the CL description, the CQ will
> commit
> > it even if the tree is red.
> > 
> > I have filed https://code.google.com/p/skia/issues/detail?id=2013 ('document
> > NOTREECHECKS, NOTRY, etc. somewhere in skiadocs')
> 
> Sometimes (perhaps in this case) we need to commit a tree-breaker fix, as part
> of several fixes, meaning even w/ this change, the trybot may still see red
(but
> we should commit anyway). In those cases I think we need a manual commit, yes?

I don't think so, if you specify both NOTREECHECKS and NOTRY.

NOTRY = don't bother with trybots before attempting commit
NOTREECHECKS = attempt commit even if the tree is red

[reed1, 2014-01-10 18:59:30]

On 2014/01/10 18:56:54, epoger wrote:
> On 2014/01/10 18:53:59, reed1 wrote:
> > On 2014/01/10 18:52:32, epoger wrote:
> > > On 2014/01/10 18:45:47, reed1 wrote:
> > > > On 2014/01/10 18:45:37, reed1 wrote:
> > > > > you may have to commit this manually, since the CQ will wait while the
> > tree
> > > is
> > > > > green
> > > > 
> > > > I mean red
> > > 
> > > If you add a "NOTREECHECKS=True" line in the CL description, the CQ will
> > commit
> > > it even if the tree is red.
> > > 
> > > I have filed https://code.google.com/p/skia/issues/detail?id=2013
('document
> > > NOTREECHECKS, NOTRY, etc. somewhere in skiadocs')
> > 
> > Sometimes (perhaps in this case) we need to commit a tree-breaker fix, as
part
> > of several fixes, meaning even w/ this change, the trybot may still see red
> (but
> > we should commit anyway). In those cases I think we need a manual commit,
yes?
> 
> I don't think so, if you specify both NOTREECHECKS and NOTRY.
> 
> NOTRY = don't bother with trybots before attempting commit
> NOTREECHECKS = attempt commit even if the tree is red

Ah, so then the CQ is really just C (no queue)?

[epoger, 2014-01-10 19:04:52]

> > I don't think so, if you specify both NOTREECHECKS and NOTRY.
> > 
> > NOTRY = don't bother with trybots before attempting commit
> > NOTREECHECKS = attempt commit even if the tree is red
> 
> Ah, so then the CQ is really just C (no queue)?

AFAIK, Y

[commit-bot: I haz the power, 2014-01-10 19:08:55]

CQ is trying da patch. Follow status at
https://skia-tree-status.appspot.com/cq/vandebo@chromium.org/133583004/1

[commit-bot: I haz the power, 2014-01-10 19:09:03]

Change committed as 13025

[vandebo (ex-Chrome), 2014-01-10 19:09:34]

On 2014/01/10 19:09:03, I haz the power (commit-bot) wrote:
> Change committed as 13025

Sorry for the delay, I thought TBR would have done several of those.

[epoger, 2014-01-10 19:12:33]

On 2014/01/10 19:09:34, vandebo wrote:
> On 2014/01/10 19:09:03, I haz the power (commit-bot) wrote:
> > Change committed as 13025
> 
> Sorry for the delay, I thought TBR would have done several of those.

I can see why you'd think that.

A restatement of the commands, as I understand them:

NOTREECHECKS = don't let a red tree stop you from committing
NOTRY = don't bother with trybots before attempting commit
TBR = don't let a lack-of-LGTM stop you from committing (if author has commit
rights)