139566 - "Number of copies" doesn't work in print dialog

Status: VERIFIED FIXED

(Core :: Printing: Output, defect)

Description

[Joshua Goldberg]

buildID: 2002031103

If I type a 2 in the Number of copies field of the print dialog, I still only
get one copy printed.

I don't attempt to use this feature often, but I tried this twice and it
happened both times

(apologies if this is a duplicate.  bug searching isn't working)

Comment 1

[Boris Zbarsky [:bzbarsky]]

This happens on Linux too.

Comment 2

[sujay]

works fine on Windows and Mac...only a problem on Linux and apparently SunOS.

Comment 3

[rods (gone)]

Roland? Is this suppose to work?

Comment 4

[Jim Crumley]

This is specific to the postscript printer driver. There is currently nowhere in
the code (mozilla/gfx/src/ps/nsPostScriptObj.cpp) that deals with number of
copies. So this affects all platforms that use that driver. 

The xprint driver allows for picking multiple copies, but has another different
bug for me at least choosing 2 copies gets me 4 copies printed out.

Comment 5

[Roland Mainz]

rods wrote:
> Roland? Is this suppose to work?
[CC:'ing myself...]

I remember I added support for number-of-copies in Xprint module long ago, but
AFAIK the current postscript module code doesn't care about the number of
copies... ;-(

Comment 6

[Roland Mainz]

Jim Crumley wrote:
> The xprint driver allows for picking multiple copies, but has another 
> different bug for me at least choosing 2 copies gets me 4 copies printed out.

Ouch. I hope noone tries to print 100 copies... =:-)

Can you file a bug for that issue, please ?
Thanks!

Comment 7

[Jim Crumley]

Ok the xprint bug is bug 140030.

> Ouch. I hope noone tries to print 100 copies... =:-)

Yeah, it seems to be geometric, so I think most printer trays would run out of
paper before printing 10000 copies .

On this bug, I'm working on a patch to fix this. I hope to post it by tomorrow -
I'm trying to generalize what I have a little bit.

Comment 8

[Jim Crumley]

Attachment: Allows for printed more than one copy for the postscript print module

At first I tried to make this too hard, but the final patch is pretty simple. 

This works for me on Solaris, but it should work on any platform that uses the
postscript print module.

Comment 9

[Roland Mainz]

Comment on attachment 81000 [details] [diff] [review]
Allows for printed more than one copy for the postscript print module

Patch looks good.

Please attach a PostScript job of http://www.mozilla.org/ with numcopies set to
three for testing (I do not trust GhostScript much, Solaris DPS is much
pickier... :)

r=[e-mail]

Comment 10

[Jim Crumley]

Attachment: PS print job with 1 copy selected

Attaching postscript file version of
http://www.cs.indiana.edu/dept/grad/handbook/Y790Form.txt with one copy
selected, for comparison with 3 copies version to be attached next.

Comment 11

[Jim Crumley]

Attachment: PS file showing selection of 3 copies

Postscript version of http://www.cs.indiana.edu/dept/grad/handbook/Y790Form.txt

showing the results of patch 81000 and choosing to print 3 copies.

Comment 12

[Marc Attinasi]

Comment on attachment 81000 [details] [diff] [review]
Allows for printed more than one copy for the postscript print module

sr=attinasi

Comment 13

[rods (gone)]

This would be good to get into 1.0, It would be odd for a user to request
multiple copies and only get one.

Comment 14

[Roland Mainz]

Just one comment based on the discussion how the PostScript driver in Xprint
implements (or better: should implement; the X.org sample implementation has a
bug, see bug 140030) the stuff:
AFAIK we may want to adjust the current patch to pass the number of copies not
within the document - instead we should use the number-of-copies-argument of the
spooler application (like "-#" if you use lpr) to make n copies of the job...

Comment 15

[Bradley Baetz (:bbaetz)]

using the lpr -# option won't work if you are printing to a file, though...

Comment 16

[Roland Mainz]

Bradley Baetz wrote:
> using the lpr -# option won't work if you are printing to a file, though...

That was part of the discussion (see comment #14). Adding number-of-copies is
useless when you print to a file... and adding the copy-count to the PostScript
document while printing to file gives a lot of headdaces (and may result in a
bad surprise (like that you can never get rid if this hardcoded copy count [1]
(without editing the postscript source))).

The current code operates at document level but it should operate at job level
(e.g. ask the spooer for the copies) ...

----

[1]=<joke>We may want to sell this as copy-protection that none can print the
docs. Attempt to print it will result in 500000 copies... =:-)</joke>

Comment 17

[Jim Crumley]

I had a halfway working patch to fix this bug by using the switches to lp/lpr,
but its much uglier. Since we allow the user to change the print command at
print time and since there are at least 2 common print commands on Unix (lp/lpr
- aren't ther e others as well?) with different command line options, its a
hassle to append the correct option to the print command. Also, I think we might
have to deal with VMS separately if we go to the command level.

Anyway, I think the PS driver is a different situation from the Xprint driver,
where the print command is not user editable from Mozilla.  For PS I think it
makes sense to let the user change the number of copies at the job level
themselves by editing the print command, or at the document level using the num
copies dialog. 

Comment 18

[Jaime Rodriguez, Jr.]

Let's get this one in, after RC2/beta is released. adt1.0.0- [adt2 RTM]

Comment 19

[sujay]

verified in 5/7 trunk build om linux.

I get two copies printed out when I enter "2" in that field.

Comment 20

[Jaime Rodriguez, Jr.]

from email from rods, "very small fix, no risk, simple writes the proper
postscript codes to indicate to print more than one copy ... tested by more than
one person, tested on more than one platform."

rods - can you provide a case on why this should be taken on the branch for RC3
(i.e. Is it a high visibility issue, etc.)? 

IMHO, I think this would be a very visible issue, as users will not be able to
print more than one copy of a document, and think it should be taken for RC3.

Comment 21

[scottputterman]

adding adt1.0.0+.  Please get drivers approval and then checkin to the 1.0 branch.

Comment 22

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Is there any chance this might do something weird on some printers?

Comment 23

[chris hofmann]

Comment on attachment 81000 [details] [diff] [review]
Allows for printed more than one copy for the postscript print module

a=chofmann,dbaron,valeski for the 1.0 branch.  we had a question about how good
the implimentation of standards in this area is for postscript printers.  any
chance this could work on some printers and not others?  we could use RC3 to
find out.  can we release not a request for every tester to do some multi-copy
postscript printing and update this bug if they see any problems..

Comment 24

[chris hofmann]

dropped release note request into http://bugzilla.mozilla.org/show_bug.cgi?id=133795

Comment 25

[sujay]

verified in 5/17 branch build.