Make the parser an implementation detail of the Document

Make the parser an implementation detail of the Document

This CL removes the parser accessors from the document to prevent folks from
grabbing at it.

R=ojan@chromium.org, eseidel@chromium.org

Committed: https://chromium.googlesource.com/external/mojo/+/da7467cfd227235cf64b103a2dc8c2803c931c69

[ojan, 2014-11-12 18:57:57]

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

[ojan, 2014-11-12 18:57:57]

lgtm

[eseidel, 2014-11-12 19:02:14]

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/dom/Document.h
File sky/engine/core/dom/Document.h (right):

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/dom/Document...
sky/engine/core/dom/Document.h:338: TextPosition textPosition() const;
What does this mean?

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/html/imports...
File sky/engine/core/html/imports/HTMLImportLoader.cpp (right):

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/html/imports...
sky/engine/core/html/imports/HTMLImportLoader.cpp:106:
m_document->startParsing()->parse(response->body.Pass());
Should this just take the mojo pipe in the startParsing() call?

[abarth-chromium, 2014-11-12 21:13:57]

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/dom/Document.h
File sky/engine/core/dom/Document.h (right):

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/dom/Document...
sky/engine/core/dom/Document.h:338: TextPosition textPosition() const;
On 2014/11/12 at 19:02:14, eseidel wrote:
> What does this mean?

It's the position (row and col) that's currently being parsed.  We should
probably use a better name...

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/html/imports...
File sky/engine/core/html/imports/HTMLImportLoader.cpp (right):

https://codereview.chromium.org/723583002/diff/1/sky/engine/core/html/imports...
sky/engine/core/html/imports/HTMLImportLoader.cpp:106:
m_document->startParsing()->parse(response->body.Pass());
On 2014/11/12 at 19:02:14, eseidel wrote:
> Should this just take the mojo pipe in the startParsing() call?

Yes, that would be sensible, but in a later CL in this series, I'm adding
another argument.  We can make startParsing pass these arguments through to the
parser, but it seemed easier to just return the parser...

[abarth-chromium, 2014-11-12 21:22:52]

Committed patchset #2 (id:20001) manually as
da7467cfd227235cf64b103a2dc8c2803c931c69.