Add paper size reporting for CUPS printers (to let user to select one in the Print Preview).

Add paper size reporting for CUPS printers (to let user to select one in the Print Preview).

BUG=239879
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=277098

[Vitaly Buka (NO REVIEWS), 2014-06-12 19:32:52]

lgtm

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
File chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc (right):

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:50: bool fuzzy_match) {
fuzzy_match -> size_only_match?

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:78: if
(PaperSizeMatch(gtk_paper_size, media, fuzzy_match)) {
No need for arg bool fuzzy_match

GtkPaperSize* fuzzy_matched = NULL;
for (GList* p = gtk_paper_sizes; p && p->data; p = g_list_next(p)) {
       GtkPaperSize* gtk_paper_size = static_cast<GtkPaperSize*>(p->data);
       if (PaperSizeMatch(gtk_paper_size, media, false)) {
         return gtk_paper_size;
       }
       if (!fuzzy_matched && PaperSizeMatch(gtk_paper_size, media, true)) {
         return fuzzy_matched = gtk_paper_size;
       }
     }
     return fuzzy_matched;
   }

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper.cc
File printing/backend/cups_helper.cc (right):

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper...
printing/backend/cups_helper.cc:38: const double kMicronsInPoint = 352.777778;
kHundrethsMMPerInch / kPointsPerInch * 10

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper...
printing/backend/cups_helper.cc:408: static_cast<int>(ppd->sizes[i].width *
kMicronsInPoint + 0.5),
std::lrint ?

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper...
printing/backend/cups_helper.cc:410: if (paper_size_microns.width() > 0 &&
paper_size_microns.height() > 0) {
should we report continuous paper?

[Aleksey Shlyapnikov, 2014-06-12 21:21:12]

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
File chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc (right):

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:50: bool fuzzy_match) {
On 2014/06/12 19:32:51, Vitaly Buka wrote:
> fuzzy_match -> size_only_match?

I think fuzzy_match conveys the purpose better.

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:78: if
(PaperSizeMatch(gtk_paper_size, media, fuzzy_match)) {
On 2014/06/12 19:32:51, Vitaly Buka wrote:
> No need for arg bool fuzzy_match
> 
> GtkPaperSize* fuzzy_matched = NULL;
> for (GList* p = gtk_paper_sizes; p && p->data; p = g_list_next(p)) {
>        GtkPaperSize* gtk_paper_size = static_cast<GtkPaperSize*>(p->data);
>        if (PaperSizeMatch(gtk_paper_size, media, false)) {
>          return gtk_paper_size;
>        }
>        if (!fuzzy_matched && PaperSizeMatch(gtk_paper_size, media, true)) {
>          return fuzzy_matched = gtk_paper_size;
>        }
>      }
>      return fuzzy_matched;
>    }

It changes the semantics. My idea is to traverse the entire list first looking
for the exact match and only if it fails, look for the fuzzy match. In your
version we'll stop of first fuzzy match, even if there's a perfect match further
down the list.

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper.cc
File printing/backend/cups_helper.cc (right):

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper...
printing/backend/cups_helper.cc:38: const double kMicronsInPoint = 352.777778;
On 2014/06/12 19:32:51, Vitaly Buka wrote:
> kHundrethsMMPerInch / kPointsPerInch * 10

Done.

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper...
printing/backend/cups_helper.cc:408: static_cast<int>(ppd->sizes[i].width *
kMicronsInPoint + 0.5),
On 2014/06/12 19:32:51, Vitaly Buka wrote:
> std::lrint ?

I'd prefer predictable rounding behavior.

https://codereview.chromium.org/334763002/diff/1/printing/backend/cups_helper...
printing/backend/cups_helper.cc:410: if (paper_size_microns.width() > 0 &&
paper_size_microns.height() > 0) {
On 2014/06/12 19:32:51, Vitaly Buka wrote:
> should we report continuous paper?

And what print preview is supposed to do about it? I guess it's not capable of
handling it at the moment.

[Vitaly Buka (NO REVIEWS), 2014-06-12 21:23:01]

i see no new patch

[Aleksey Shlyapnikov, 2014-06-12 21:24:25]

Justin: for changes in chrome/common/chrome_utility_messages.h, separating paper
name into vendor_id (platform specific paper size id) and paper display name.

[Aleksey Shlyapnikov, 2014-06-12 21:27:12]

Elliot: for chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc changes

[Vitaly Buka (NO REVIEWS), 2014-06-12 21:34:08]

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
File chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc (right):

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:50: bool fuzzy_match) {
fuzzy is good when you can't explain logic.
Here is size only match.

On 2014/06/12 21:21:12, Aleksey Shlyapnikov wrote:
> On 2014/06/12 19:32:51, Vitaly Buka wrote:
> > fuzzy_match -> size_only_match?
> 
> I think fuzzy_match conveys the purpose better.

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:78: if
(PaperSizeMatch(gtk_paper_size, media, fuzzy_match)) {
Semantic exactly the same.
It just remember fuzzy match and returns it only if no exact match is possible.

On 2014/06/12 21:21:12, Aleksey Shlyapnikov wrote:
> On 2014/06/12 19:32:51, Vitaly Buka wrote:
> > No need for arg bool fuzzy_match
> > 
> > GtkPaperSize* fuzzy_matched = NULL;
> > for (GList* p = gtk_paper_sizes; p && p->data; p = g_list_next(p)) {
> >        GtkPaperSize* gtk_paper_size = static_cast<GtkPaperSize*>(p->data);
> >        if (PaperSizeMatch(gtk_paper_size, media, false)) {
> >          return gtk_paper_size;
> >        }
> >        if (!fuzzy_matched && PaperSizeMatch(gtk_paper_size, media, true)) {
> >          return fuzzy_matched = gtk_paper_size;
> >        }
> >      }
> >      return fuzzy_matched;
> >    }
> 
> It changes the semantics. My idea is to traverse the entire list first looking
> for the exact match and only if it fails, look for the fuzzy match. In your
> version we'll stop of first fuzzy match, even if there's a perfect match
further
> down the list.

[Aleksey Shlyapnikov, 2014-06-12 22:56:33]

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
File chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc (right):

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:50: bool fuzzy_match) {
On 2014/06/12 21:34:07, Vitaly Buka wrote:
> fuzzy is good when you can't explain logic.
> Here is size only match.
> 
> On 2014/06/12 21:21:12, Aleksey Shlyapnikov wrote:
> > On 2014/06/12 19:32:51, Vitaly Buka wrote:
> > > fuzzy_match -> size_only_match?
> > 
> > I think fuzzy_match conveys the purpose better.
> 

Here it also means "ignore PPD name".

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:78: if
(PaperSizeMatch(gtk_paper_size, media, fuzzy_match)) {
On 2014/06/12 21:34:08, Vitaly Buka wrote:
> Semantic exactly the same.
> It just remember fuzzy match and returns it only if no exact match is
possible.
> 
> On 2014/06/12 21:21:12, Aleksey Shlyapnikov wrote:
> > On 2014/06/12 19:32:51, Vitaly Buka wrote:
> > > No need for arg bool fuzzy_match
> > > 
> > > GtkPaperSize* fuzzy_matched = NULL;
> > > for (GList* p = gtk_paper_sizes; p && p->data; p = g_list_next(p)) {
> > >        GtkPaperSize* gtk_paper_size = static_cast<GtkPaperSize*>(p->data);
> > >        if (PaperSizeMatch(gtk_paper_size, media, false)) {
> > >          return gtk_paper_size;
> > >        }
> > >        if (!fuzzy_matched && PaperSizeMatch(gtk_paper_size, media, true))
{
> > >          return fuzzy_matched = gtk_paper_size;
> > >        }
> > >      }
> > >      return fuzzy_matched;
> > >    }
> > 
> > It changes the semantics. My idea is to traverse the entire list first
looking
> > for the exact match and only if it fails, look for the fuzzy match. In your
> > version we'll stop of first fuzzy match, even if there's a perfect match
> further
> > down the list.
> 

Addressed your concern.

[Vitaly Buka (NO REVIEWS), 2014-06-12 23:06:05]

lgtm

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
File chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc (right):

https://codereview.chromium.org/334763002/diff/1/chrome/browser/ui/libgtk2ui/...
chrome/browser/ui/libgtk2ui/print_dialog_gtk2.cc:50: bool fuzzy_match) {
Thanks.

On 2014/06/12 22:56:33, Aleksey Shlyapnikov wrote:
> On 2014/06/12 21:34:07, Vitaly Buka wrote:
> > fuzzy is good when you can't explain logic.
> > Here is size only match.
> > 
> > On 2014/06/12 21:21:12, Aleksey Shlyapnikov wrote:
> > > On 2014/06/12 19:32:51, Vitaly Buka wrote:
> > > > fuzzy_match -> size_only_match?
> > > 
> > > I think fuzzy_match conveys the purpose better.
> > 
> 
> Here it also means "ignore PPD name".

[jschuh, 2014-06-13 12:48:18]

ipc security lgtm

[Elliot Glaysher, 2014-06-13 17:10:00]

lgtm

[Aleksey Shlyapnikov, 2014-06-13 17:13:45]

The CQ bit was checked by alekseys@chromium.org

[commit-bot: I haz the power, 2014-06-13 17:14:41]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/alekseys@chromium.org/334763002/40001

[commit-bot: I haz the power, 2014-06-13 22:01:02]

Change committed as 277098