Update waterfall templates to include Milo link.

Update waterfall templates to include Milo link.

Milo is currently ingesting all of the build data for all of the
waterfalls. Add a link to each waterfall with a parallel Milo view that
clicks through to Milo.

In order to do this, we add core "milo.js" and "milo.css" files to
"chromium"'s "public_html" folders. We also augment "layout.html" to
execute the "milo.js" injection function.

Every waterfall with its own "layout.html" template has been updated to
include the "milo.js" import and injection code.

Every waterfall with its own "public_html" directory has had the Milo
JavaScript, CSS, and image files symlinked to "chromium/" so they can
access the resources referenced by their "layout.html".

The injection function examines the current URL and decides if it maps
to a view Milo supports, namely a builder or build view. If it does, a
[LogDog] tab is injected into the header that links to the Milo
equivalent view.

The BuildBot base template is augmented to pass the master name to
"milo.js". This is the actual name used by Milo, as opposed to the
chromegw URL name which can occasionally vary. The modified template
parameter is protected with the "|default" filter function so masters
whose templates update but aren't restarted won't fail to render.

BUG=None
TEST=local
  - Started masters with affected "layout.html" pages locally, seem to
    be working!

R=estaab@chromium.org, iannucci@chromium.org

Committed: https://chromium.googlesource.com/chromium/tools/build/+/48c11b45c0c209234d68833b33c5b81dd282e633

[dnj, 2016-11-09 23:03:01]

PTAL.

Screenshots:
- Static: https://screenshot.googleplex.com/Agqf5hJ26fr.png
- On Hover: https://screenshot.googleplex.com/bSVU80Tm8Po.png

[estaab, 2016-11-09 23:24:56]

Looks great, just a couple of comments.

Could we split this into 2 CLs with 1) the new files + one master and 2) the
rest of the masters?

There have been a few rollouts across multiple masters that were a little rough
recently. Maybe infra masters first?

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
File masters/master.chromium/public_html/milo.js (right):

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
masters/master.chromium/public_html/milo.js:12: // Do we have a Milo
substitution avialable?
available

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
masters/master.chromium/public_html/milo.js:52: miloElement.innerHTML += `
Not sure if we're supposed to be using ES6 templates but I guess we can find out
if anyone is using an old browser and complains. :)

[dnj, 2016-11-10 00:22:55]

The problem with targeting masters has to do with the way templates are used.
Basically, waterfalls either:
1) Declare their own "templates/layout.html", in which case they are
independent.
2) Declare that they use another waterfall's "templates/layout.html" (usually
"chromium" waterfall's), in which case the inherit this.

Additionally, they can:
A) Use "chromium"'s "public_html" directory.
B) Use their own "public_html" directory.

[1] needs to land for at least one waterfall. Waterfalls in the [2] category
will inherit this update, so we need to find [2B] and update their "public_html"
to include the new JS/CSS. If we miss one, it'll show page errors in the console
but the page should continue to load thanks to the way I built the injection.

I can try and pull this back to the set of waterfalls that use the "chromium"
templates for the first round. WDYT?

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
File masters/master.chromium/public_html/milo.js (right):

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
masters/master.chromium/public_html/milo.js:12: // Do we have a Milo
substitution avialable?
On 2016/11/09 23:24:56, estaab wrote:
> available

Done.

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
masters/master.chromium/public_html/milo.js:52: miloElement.innerHTML += `
On 2016/11/09 23:24:56, estaab wrote:
> Not sure if we're supposed to be using ES6 templates but I guess we can find
out
> if anyone is using an old browser and complains. :)

SGTM. It's been in Chromium since M41, and we're on M53 right now, so I'm
completely okay with this :)

[estaab, 2016-11-10 00:29:27]

Just ones using the chromium template in the first pass sounds good. Just be
safe and use your discretion (the last time I said that on a CL there was a
short outage so I really mean it :).

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
File masters/master.chromium/public_html/milo.js (right):

https://codereview.chromium.org/2491643003/diff/1/masters/master.chromium/pub...
masters/master.chromium/public_html/milo.js:52: miloElement.innerHTML += `
On 2016/11/10 at 00:22:55, dnj wrote:
> On 2016/11/09 23:24:56, estaab wrote:
> > Not sure if we're supposed to be using ES6 templates but I guess we can find
out
> > if anyone is using an old browser and complains. :)
> 
> SGTM. It's been in Chromium since M41, and we're on M53 right now, so I'm
completely okay with this :)

Someone on the web platform team pushed back when I did it so I thought I'd
follow their lead and do a nudge. No real objections of my own. :)

[estaab, 2016-11-10 00:29:34]

lgtm

[dnj, 2016-11-10 05:34:53]

Constrained to just Chromium waterfalls. Moved the common resources into a
"common-resources" folder so we only need 1 symlink per waterfall.

I added this link to internal waterfalls preemptively, just in case some are
using the "chromium" templates directly. Worst case they sit latent:
https://chromereviews.googleplex.com/537537013

[iannucci, 2016-11-10 20:56:41]

lgtm

https://codereview.chromium.org/2491643003/diff/20001/masters/master.chromium...
File masters/master.chromium/templates/layout.html (right):

https://codereview.chromium.org/2491643003/diff/20001/masters/master.chromium...
masters/master.chromium/templates/layout.html:74: <!-- Identify opportunities to
redirect to Milo and populate. -->
this looks like a random extra comment?

[dnj, 2016-11-10 22:42:03]

https://codereview.chromium.org/2491643003/diff/20001/masters/master.chromium...
File masters/master.chromium/templates/layout.html (right):

https://codereview.chromium.org/2491643003/diff/20001/masters/master.chromium...
masters/master.chromium/templates/layout.html:74: <!-- Identify opportunities to
redirect to Milo and populate. -->
On 2016/11/10 20:56:41, iannucci wrote:
> this looks like a random extra comment?

Done.

[dnj, 2016-11-10 22:42:16]

The CQ bit was checked by dnj@chromium.org

[dnj, 2016-11-10 22:42:17]

The patchset sent to the CQ was uploaded after l-g-t-m from estaab@chromium.org,
iannucci@chromium.org
Link to the patchset: https://codereview.chromium.org/2491643003/#ps40001
(title: "Remove superfluous comment.")

[commit-bot: I haz the power, 2016-11-10 22:42:19]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-11-10 22:52:19]

Description was changed from

==========
Update waterfall templates to include Milo link.

Milo is currently ingesting all of the build data for all of the
waterfalls. Add a link to each waterfall with a parallel Milo view that
clicks through to Milo.

In order to do this, we add core "milo.js" and "milo.css" files to
"chromium"'s "public_html" folders. We also augment "layout.html" to
execute the "milo.js" injection function.

Every waterfall with its own "layout.html" template has been updated to
include the "milo.js" import and injection code.

Every waterfall with its own "public_html" directory has had the Milo
JavaScript, CSS, and image files symlinked to "chromium/" so they can
access the resources referenced by their "layout.html".

The injection function examines the current URL and decides if it maps
to a view Milo supports, namely a builder or build view. If it does, a
[LogDog] tab is injected into the header that links to the Milo
equivalent view.

The BuildBot base template is augmented to pass the master name to
"milo.js". This is the actual name used by Milo, as opposed to the
chromegw URL name which can occasionally vary. The modified template
parameter is protected with the "|default" filter function so masters
whose templates update but aren't restarted won't fail to render.

BUG=None
TEST=local
  - Started masters with affected "layout.html" pages locally, seem to
    be working!

R=estaab@chromium.org, iannucci@chromium.org
==========

to

==========
Update waterfall templates to include Milo link.

Milo is currently ingesting all of the build data for all of the
waterfalls. Add a link to each waterfall with a parallel Milo view that
clicks through to Milo.

In order to do this, we add core "milo.js" and "milo.css" files to
"chromium"'s "public_html" folders. We also augment "layout.html" to
execute the "milo.js" injection function.

Every waterfall with its own "layout.html" template has been updated to
include the "milo.js" import and injection code.

Every waterfall with its own "public_html" directory has had the Milo
JavaScript, CSS, and image files symlinked to "chromium/" so they can
access the resources referenced by their "layout.html".

The injection function examines the current URL and decides if it maps
to a view Milo supports, namely a builder or build view. If it does, a
[LogDog] tab is injected into the header that links to the Milo
equivalent view.

The BuildBot base template is augmented to pass the master name to
"milo.js". This is the actual name used by Milo, as opposed to the
chromegw URL name which can occasionally vary. The modified template
parameter is protected with the "|default" filter function so masters
whose templates update but aren't restarted won't fail to render.

BUG=None
TEST=local
  - Started masters with affected "layout.html" pages locally, seem to
    be working!

R=estaab@chromium.org, iannucci@chromium.org

Committed:
https://chromium.googlesource.com/chromium/tools/build/+/48c11b45c0c209234d68...
==========

[commit-bot: I haz the power, 2016-11-10 22:52:20]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/chromium/tools/build/+/48c11b45c0c209234d68...