Added support to the Windows cloud print proxy to print XPS documents directl...

Added support to the Windows cloud print proxy to print XPS documents directly without any conversion (only works on Windows 7). The proxy uses the Accept: header to specify what data types it can accept.

BUG=None.
TEST=When the cloud print service supports this, submitting an XPS document should go through to the Win 7 proxy without any conversions.


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=75286

[Scott Byer, 2011-02-17 00:31:01]

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
File chrome/service/cloud_print/print_system_cups.cc (right):

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
chrome/service/cloud_print/print_system_cups.cc:651: // TODO(sanjeevr):
Determine this dynamically.
A bug + link for the TODO?

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
File chrome/service/cloud_print/print_system_win.cc (right):

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
chrome/service/cloud_print/print_system_win.cc:584:
file_util::ReadFileToString(print_data_file_path, &document_data);
Does this function need to check that it's being called on the right thread
since it's got file I/O going on?

http://codereview.chromium.org/6523040/diff/7003/printing/backend/win_helper.cc
File printing/backend/win_helper.cc (right):

http://codereview.chromium.org/6523040/diff/7003/printing/backend/win_helper....
printing/backend/win_helper.cc:223: static bool initialized = InitImpl();
This isn't thread safe, really. Use LazyInstance if it really needs to be.

[sanjeevr, 2011-02-17 00:53:52]

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
File chrome/service/cloud_print/print_system_cups.cc (right):

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
chrome/service/cloud_print/print_system_cups.cc:651: // TODO(sanjeevr):
Determine this dynamically.
On 2011/02/17 00:31:02, Scott Byer wrote:
> A bug + link for the TODO?

Done.

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
File chrome/service/cloud_print/print_system_win.cc (right):

http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
chrome/service/cloud_print/print_system_win.cc:584:
file_util::ReadFileToString(print_data_file_path, &document_data);
On 2011/02/17 00:31:02, Scott Byer wrote:
> Does this function need to check that it's being called on the right thread
> since it's got file I/O going on?

This is the service process and not the browser process. We don't have the same
kind of restrictions about which thread to do file I/O on since there is no UI.

http://codereview.chromium.org/6523040/diff/7003/printing/backend/win_helper.cc
File printing/backend/win_helper.cc (right):

http://codereview.chromium.org/6523040/diff/7003/printing/backend/win_helper....
printing/backend/win_helper.cc:223: static bool initialized = InitImpl();
On 2011/02/17 00:31:02, Scott Byer wrote:
> This isn't thread safe, really. Use LazyInstance if it really needs to be.

Yes, I know this isn't thread-safe. But it does not need to be since the worst
thing that will happen is that the initialization will happen twice (the static
bool check is just an optimization anyway). This is the same pattern we follow
in XPSModule::Init.

[Scott Byer, 2011-02-17 17:17:10]

LGTM

On 2011/02/17 00:53:52, sanjeevr wrote:
>
http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
> File chrome/service/cloud_print/print_system_cups.cc (right):
> 
>
http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
> chrome/service/cloud_print/print_system_cups.cc:651: // TODO(sanjeevr):
> Determine this dynamically.
> On 2011/02/17 00:31:02, Scott Byer wrote:
> > A bug + link for the TODO?
> 
> Done.
> 
>
http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
> File chrome/service/cloud_print/print_system_win.cc (right):
> 
>
http://codereview.chromium.org/6523040/diff/7003/chrome/service/cloud_print/p...
> chrome/service/cloud_print/print_system_win.cc:584:
> file_util::ReadFileToString(print_data_file_path, &document_data);
> On 2011/02/17 00:31:02, Scott Byer wrote:
> > Does this function need to check that it's being called on the right thread
> > since it's got file I/O going on?
> 
> This is the service process and not the browser process. We don't have the
same
> kind of restrictions about which thread to do file I/O on since there is no
UI.
> 
>
http://codereview.chromium.org/6523040/diff/7003/printing/backend/win_helper.cc
> File printing/backend/win_helper.cc (right):
> 
>
http://codereview.chromium.org/6523040/diff/7003/printing/backend/win_helper....
> printing/backend/win_helper.cc:223: static bool initialized = InitImpl();
> On 2011/02/17 00:31:02, Scott Byer wrote:
> > This isn't thread safe, really. Use LazyInstance if it really needs to be.
> 
> Yes, I know this isn't thread-safe. But it does not need to be since the worst
> thing that will happen is that the initialization will happen twice (the
static
> bool check is just an optimization anyway). This is the same pattern we follow
> in XPSModule::Init.