Check printing in windows?

[Thomas Troesch]

On Sat, Mar 17, 2012 at 5:41 PM, David T. <sunfish62 at yahoo.com> wrote:

> Thomas--
>
> Thanks. As neither a check printing person, nor a Windows person, I am
> flying double blind (which is fine for medical trials, but not so great for
> software debugging).
>
> Now, help me understand: you say it doesn't fix anything. How not? Is it
> that this modification works with Windows, but fails or breaks the form on
> other OSes? Is it that there is some underlying unexplained glitch that
> messes up all such printing in Windows?
>
>

There is some adage about a one-eyed man....  ;)

The windows bug is that all check printing is scaled down by a factor of
8.33 ( at least on my machine ).  This happens to be 600/72 which is my
printer DPI  divided by the DPI that checks are defined in.  I don't even
know if a different printer DPI would change the factor.  I can't test that.

The workaround file scales everything up by 8.33.  The font size in the
file is set to 83.3 - which should print as a 10pt font after the bug does
its thing.  The same is true for all the other parameters in the file.
 This works for most things.  It doesn't work, for instance, with the print
date feature, required by Canada.  I think that grid printing also doesn't
work.  And of course it doesn't fit the documentation.

If the file is used on Linux, everything will be 8.33 times too big.  There
are other problems with creating custom check formats with the user
interface.

As for underlying glitches I can only guess that the graphic context used
for check printing has the wrong scale only on windows.  I think there was
a similar bug for printing reports on windows a while back.


> Is there some way to incorporate the workaround into the general file with
> a test for Windows that would allow the kludge to be implemented only on
> that OS?
>
>

Doing that would still leave bugs and some very strange documentation!  But
your question is really for the developers mail list.  I have no idea what
that would take, but I would be disappointed with that solution.  I have no
doubt that it would be easier for a developer with a windows setup to fix
the problem correctly and completely.

Just my $.02.

Thomas


 David
>
>   *From:* Thomas Troesch <ttroesch at gmail.com>
> *To:* John Ralls <jralls at ceridwen.us>
> *Cc:* David T. <sunfish62 at yahoo.com>; "gnucash-user at gnucash.org" <
> gnucash-user at gnucash.org>
> *Sent:* Saturday, March 17, 2012 4:53 PM
>
> *Subject:* Re: Check printing in windows?
>
> A little clarity is in order.
>
> The file that dniesby posted is his implementation of a workaround for
> windows check printing described in
> https://bugzilla.gnome.org/show_bug.cgi?id=631611 .  It doesn't really
> fix anything.  It provides a check format for a 3-part non-voucher check
> that overcomes the check printing bug on windows.  This is only one of
> several formats that are distributed with GnuCash.  If you are not on
> windows, the format is not usable, and the numbers are NOT what is
> described in the documentation.  But it works(around) - almost completely.
>
> The bug lists confirmation of 'small check printing' for XP, Vista, Vista
> Home Edition, Windows 7 32 and 64 bit.  If any Windows users DON'T have the
> problem, that would be interesting.  It may lead to some insight on the
> problem.
>
> Its a very good thing that dniesby is sharing his format here.  But it is
> not a solution to the problem.  And I personally can't see a way to include
> the format in the distribution that would make any sense.
>
> John - are you saying there are no developers that have a windows
> environment to debug with? Or that there are, but they don't see the
> problem?
>
> Thomas
>
> On Sat, Mar 17, 2012 at 2:43 PM, John Ralls <jralls at ceridwen.us> wrote:
>
>
> On Mar 17, 2012, at 1:14 PM, David T. wrote:
>
> > The point is that Gnucash will continue to be bundled and distributed
> with the Windows printing flaw until someone submits a bugfix patch that
> can be tested. So Windows users will continue to know about your fixed file
> (buried in the user-list archives), and download and install it before they
> can print checks under Windows.
> >
> >
> > Wouldn't it just be great if the fixed file were a part of the
> application, and Windows users wouldn't have to do all that?
> >
>
> The fundamental problem for the developers is that none of us can
> replicate the problem. It doesn't happen for everyone, so we need to know
> if this patch in fact fixes the problem for more people than just dniesby.
>
> Regards,
> John Ralls
>
>
> _______________________________________________
> gnucash-user mailing list
> gnucash-user at gnucash.org
> https://lists.gnucash.org/mailman/listinfo/gnucash-user
> -----
> Please remember to CC this list on all your replies.
> You can do this by using Reply-To-List or Reply-All.
>
>
>
>
>