Add the ability to select a source image to use in the code.

Add the ability to select a source image to use in the code.

A much farther ranging change than I suspected.

Basically add a 'source' integer to every Try, store that in the database with every Try, add the source to the computation of the hash, and load and use the 'source' value when navigating history.

BUG=skia:

Committed: http://code.google.com/p/skia/source/detail?r=14960

[mtklein, 2014-05-28 17:36:49]

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
File experimental/webtry/DESIGN.md (right):

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:147: source    INTEGER   DEFAULT 0                
 NOT NULL,
Having both "code" and "source" is a touch confusing.  Is it to late to go to
something like source_image_id and webtry.source_images?

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:169: CREATE TABLE sources (
Funky indent?

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:171: image     MEDIUMBLOB  DEFAULT ''             
   NOT NULL,
Can we note somehow how image is encoded?  Is it, the original source file (i.e.
must be sniffed, probably JPG, PNG, etc.) or decoded into some bitmap format
(ARGB8888?)

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/webt...
File experimental/webtry/webtry.go (right):

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/webt...
experimental/webtry/webtry.go:230: id        INTEGER     PRIMARY KEY            
   NOT NULL,
Is it just Rietveld, or is this weirdly indented?

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/webt...
experimental/webtry/webtry.go:480: m, _, err := image.Decode(f)
Might consider leaving everything in its native format, relying on Skia and the
browser to handle things?  We won't have to think about what image formats are
in the intersection of this server and Skia... it's just whatever Skia supports.

[jcgregorio, 2014-05-28 18:42:58]

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
File experimental/webtry/DESIGN.md (right):

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:147: source    INTEGER   DEFAULT 0                
 NOT NULL,
On 2014/05/28 17:36:50, mtklein wrote:
> Having both "code" and "source" is a touch confusing.  Is it to late to go to
> something like source_image_id and webtry.source_images?

Done.

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:169: CREATE TABLE sources (
On 2014/05/28 17:36:50, mtklein wrote:
> Funky indent?

Done.

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:171: image     MEDIUMBLOB  DEFAULT ''             
   NOT NULL,
On 2014/05/28 17:36:50, mtklein wrote:
> Can we note somehow how image is encoded?  Is it, the original source file
(i.e.
> must be sniffed, probably JPG, PNG, etc.) or decoded into some bitmap format
> (ARGB8888?)

Done.

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/webt...
File experimental/webtry/webtry.go (right):

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/webt...
experimental/webtry/webtry.go:230: id        INTEGER     PRIMARY KEY            
   NOT NULL,
On 2014/05/28 17:36:50, mtklein wrote:
> Is it just Rietveld, or is this weirdly indented?

Done.

https://codereview.chromium.org/294903017/diff/50001/experimental/webtry/webt...
experimental/webtry/webtry.go:480: m, _, err := image.Decode(f)
On 2014/05/28 17:36:50, mtklein wrote:
> Might consider leaving everything in its native format, relying on Skia and
the
> browser to handle things?  We won't have to think about what image formats are
> in the intersection of this server and Skia... it's just whatever Skia
supports.

I started down that path originally, but it opens a can of worms, we need to
sniff the uploaded images, guess their mimetype, then write out the images with
the right extension for that mimetype, and if it turns out to be a format that
Skia doesn't support then the user might have a very hard to debug issue later
on when they try to use that image.

[mtklein, 2014-05-29 15:26:19]

Just doc nits.  LGTM

https://codereview.chromium.org/294903017/diff/70001/experimental/webtry/DESI...
File experimental/webtry/DESIGN.md (right):

https://codereview.chromium.org/294903017/diff/70001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:139: GRANT SELECT, INSERT, UPDATE ON
webtry.sources      TO 'webtry'@'%';
Needs update?

https://codereview.chromium.org/294903017/diff/70001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:239: as a binary blog in the database.
blob?

[jcgregorio, 2014-05-29 15:37:32]

https://codereview.chromium.org/294903017/diff/70001/experimental/webtry/DESI...
File experimental/webtry/DESIGN.md (right):

https://codereview.chromium.org/294903017/diff/70001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:139: GRANT SELECT, INSERT, UPDATE ON
webtry.sources      TO 'webtry'@'%';
Good catch, done.

On 2014/05/29 15:26:19, mtklein wrote:
> Needs update?

https://codereview.chromium.org/294903017/diff/70001/experimental/webtry/DESI...
experimental/webtry/DESIGN.md:239: as a binary blog in the database.
On 2014/05/29 15:26:19, mtklein wrote:
> blob?

Done.

[jcgregorio, 2014-05-29 15:37:40]

The CQ bit was checked by jcgregorio@google.com

[commit-bot: I haz the power, 2014-05-29 15:38:43]

CQ is trying da patch. Follow status at
 https://skia-tree-status.appspot.com/cq/jcgregorio@google.com/294903017/80001

[commit-bot: I haz the power, 2014-05-29 15:58:12]

Change committed as 14960