Printing: Temporary fix on Linux using dummy print settings until we get the ...

Printing: Temporary fix on Linux using dummy print settings until we get the entire Linux printing pipeline to match Win/Mac.

BUG=none
TEST=printing works.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=61556

[Lei Zhang, 2010-10-05 03:51:17]

The last round of refactoring broke printing for me on Linux, i.e. on 7.0.536.2
dev. It looks like GTK is returning all zeros for the DPI and page sizes, so we
use DPI as a heuristic - if it's 0, then assume all the values are bad, and go
with the dummy values we were using previously.

Since we now fill in the dummy values early on, we no longer have to fill it
once per page in PrintWebViewHelper::PrintPage(), so I'm removing
FillDefaultPrintParams() which is effectively dead code.

[James Hawkins, 2010-10-05 20:16:06]

LGTM

http://codereview.chromium.org/3619002/diff/1/3
File printing/print_settings.cc (right):

http://codereview.chromium.org/3619002/diff/1/3#newcode164
printing/print_settings.cc:164: // Use dummy values if we cannot get valid
values.
Please put a TODO stating that this is a hack and needs to be fixed properly.
This path of code will be removed once I finish the refactoring and we actually
let the user pick a printer/settings.

[Lei Zhang, 2010-10-05 20:20:15]

http://codereview.chromium.org/3619002/diff/1/3
File printing/print_settings.cc (right):

http://codereview.chromium.org/3619002/diff/1/3#newcode164
printing/print_settings.cc:164: // Use dummy values if we cannot get valid
values.
On 2010/10/05 20:16:06, James Hawkins wrote:
> Please put a TODO stating that this is a hack and needs to be fixed properly.
> This path of code will be removed once I finish the refactoring and we
actually
> let the user pick a printer/settings.

Done.