Refactor the URLResponseInfo to use new design

Refactor the URLResponseInfo to use new design

This puts all of the URLResponseInfo attributes in a struct so it can be sent over IPC in one message rather than requiring one sync IPC per attribute access.

This includes a new example of streaming to a file that I used to do some manual tests of this change.

I added and improved some documentation that I noticed when trying to write the example.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=168273

[bbudge, 2012-10-12 01:27:06]

LGTM w/ comments and nits

http://codereview.chromium.org/10993031/diff/12001/ppapi/cpp/file_io.h
File ppapi/cpp/file_io.h (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/cpp/file_io.h#newcod...
ppapi/cpp/file_io.h:105: /// class if it goes out of scope, BUT the callback
factory can NOT cancel
This sentance is hard to understand, as 'it' is ambiguous. I think you mean the
callback factory goes out of scope. If so, perhaps you could say:

If you use a completion callback factory and it goes out of scope, ...

http://codereview.chromium.org/10993031/diff/12001/ppapi/cpp/file_io.h#newcod...
ppapi/cpp/file_io.h:116: /// possibility for writing into invalid memory.
I may lack some understanding here. I get how the buffer should outlive the
FileIO resource, but I don't understand about 'one class' and how copying the
resource makes any difference.

http://codereview.chromium.org/10993031/diff/12001/ppapi/examples/url_loader/...
File ppapi/examples/url_loader/stream_to_file.cc (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/examples/url_loader/...
ppapi/examples/url_loader/stream_to_file.cc:123: // to keep things simple.
I don't know how messy it would be but I bet people who were copying this
example into their apps would appreciate the code to read the whole thing, or at
least a comment about where to find this example code.

http://codereview.chromium.org/10993031/diff/12001/ppapi/proxy/url_response_i...
File ppapi/proxy/url_response_info_resource.h (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/proxy/url_response_i...
ppapi/proxy/url_response_info_resource.h:23: // Will take an additional
reference to the file ref data if there is one.
This is repeated below.

http://codereview.chromium.org/10993031/diff/12001/ppapi/proxy/url_response_i...
ppapi/proxy/url_response_info_resource.h:27: // is passed from the caller and is
taken over by this object.
It might be helpful to state that file_ref_resource could be null, if the
request didn't have the stream-as-file property.

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/ppb_url_loader...
File ppapi/thunk/ppb_url_loader_api.h (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/ppb_url_loader...
ppapi/thunk/ppb_url_loader_api.h:57: // to to that object to the caller.
s/to to/to

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/ppb_url_respon...
File ppapi/thunk/ppb_url_response_info_api.h (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/ppb_url_respon...
ppapi/thunk/ppb_url_response_info_api.h:23: // if there is a file ref.
Maybe worth it to say "if body_as_file_ref is non-null".

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/resource_creat...
File ppapi/thunk/resource_creation_api.h (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/resource_creat...
ppapi/thunk/resource_creation_api.h:89: // Passed a reference to the
file_ref_response, which is a process-locak
s/Passed/Passes
s/file_ref_response/file_ref_resource
s/locak/local

http://codereview.chromium.org/10993031/diff/12001/ppapi/thunk/resource_creat...
ppapi/thunk/resource_creation_api.h:90: // resource corresponsing to the
body_as_file_ref host resource in |data|,
s/corresponsing/corresponding

[brettw, 2012-10-12 19:34:53]

http://codereview.chromium.org/10993031/diff/12001/ppapi/examples/url_loader/...
File ppapi/examples/url_loader/stream_to_file.cc (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/examples/url_loader/...
ppapi/examples/url_loader/stream_to_file.cc:123: // to keep things simple.
It's kind of annoying to read files like this, so I'd rather leave it out of
here since it's not what we're trying to show in this example. I've mentioned
that you should see a file I/O example while being vague, since it seems we
don't currently have a good one :( We'll have to work on this.

http://codereview.chromium.org/10993031/diff/12001/ppapi/proxy/url_response_i...
File ppapi/proxy/url_response_info_resource.h (right):

http://codereview.chromium.org/10993031/diff/12001/ppapi/proxy/url_response_i...
ppapi/proxy/url_response_info_resource.h:23: // Will take an additional
reference to the file ref data if there is one.
On 2012/10/12 01:27:06, bbudge1 wrote:
> This is repeated below.

Actually not. It's completely contradicted below! :(

[brettw, 2012-11-15 19:30:18]

+jschuh for _messages.h