Re: Postscript printer errors with rounded barlines?

[Graham Percival]

On Sat, Mar 17, 2012 at 07:38:29PM +, Keith OHara wrote:
> David Kastrup  gnu.org> writes:
> 
> > "Prior Emails"? 
> 
> Here :
> 
> The resulting patch solved the problem of thick barlines, but did not 
> make it into version 2.14.

True.  I've lost track of whether this is discussing current devel
or 2.14, but to save anybody else searching, it looks like that
discussion produced this commit:
  5291daf785cd215145473781612732de94890ba0
added between 2.15.4 and 2.15.6.

- Graham

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Keith OHara]

David Kastrup  gnu.org> writes:

> Todd Lewis  email.unc.edu> writes:
> 
> > The work-around [...] indicated in prior emails to avoid
> > rectstroke with a 0.0 linewidth, 
> 
> "Prior Emails"? 

Here :

The resulting patch solved the problem of thick barlines, but did not 
make it into version 2.14.

> Anyway, starting with version 2.15.31, the patch 
> [...]
> is in the repository.  No "previous mail" was consulted, the issue is a
> different one.  

Different, but similar, and affecting the same piece of code.



___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[David Kastrup]

Todd Lewis  writes:

> The work-around is to change /draw_round_box in
> ps/music-drawing-routines.ps as indicated in prior emails to avoid
> rectstroke with a 0.0 linewidth, the behavior of which is device
> dependent.

"Prior Emails"?  Anyway, starting with version 2.15.31, the patch

commit 4666869ad53760ead32eba6c5dc75e43c69f4c87
Author: David Kastrup 
Date:   Mon Feb 27 23:29:24 2012 +0100

Issue 2328: Ledger lines in PNG output have white interior

Don't stroke rectangles with height or width 0 in draw_round_box

diff --git a/ps/music-drawing-routines.ps b/ps/music-drawing-routines.ps
index d7450e45..3e62f03 100644
--- a/ps/music-drawing-routines.ps
+++ b/ps/music-drawing-routines.ps
@@ -98,15 +98,20 @@ bind def
 dup
0.0 gt {
setlinewidth % w h x y
-   0 setlinecap
-   1 setlinejoin
-
rmoveto % w h
-   currentpoint % w h x1 y1
-   4 2 roll % x1 y1 w h
-   4 copy
-   rectfill
-   rectstroke
+   2 copy 0 ne exch 0 ne and
+   {
+   0 setlinecap
+   1 setlinejoin
+   currentpoint % w h x1 y1
+   4 2 roll % x1 y1 w h
+   4 copy
+   rectfill
+   rectstroke
+   } {
+   1 setlinecap
+   rlineto stroke
+   } ifelse
} {
pop % w h x y
rmoveto % w h

is in the repository.  No "previous mail" was consulted, the issue is a
different one.  So it would very much appear like a failure in
communication.  When and in what context did you write your "previous
mail"?

-- 
David Kastrup


___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Todd Lewis]

> On Sat, Jul 23, 2011 at 7:45 PM, Jan Warchoł wrote:
>
>> I'm not experienced with PostScript, but i have an impression that
>> this may be related to an issue discussed earlier - see here
>> http://lists.gnu.org/archive/html/lilypond-user/2007-07/msg00485.html
>> and here
>> http://lists.gnu.org/archive/html/lilypond-devel/2007-07/msg00072.html
>> I see that something about draw_round_box was said there...
>
> I think this is an other issue.  Due to rounding, PDF viewers can err
> the placement of the barline by a pixel. This is insolvable, as there
> is no way to hint that the barline should coincide with the staffline,
> yet we need to go the end of the barline to get the correct effect in
> the corners in staves.
>
> The one-pixel error is very noticeable on-screen due to
> low-resolution, but on a 1200dpi printer,  a single dot error is about
> 0.02 mm, ie. not noticeable.

I can assure you this is the same issue and not a rounding error. I've also
found a way to reliably reproduce the problem, and a work-around. I've always
been frustrated that my printed output has these huge bar lines, while the
on-screen preview (evince on Ubuntu) looks fine. I've found that if I preview in
evince (where it looks good), select File->Print and print to a .ps file, then
open that .ps file in evince, the bar lines have become obnoxiously fat again.

The work-around is to change /draw_round_box in ps/music-drawing-routines.ps as
indicated in prior emails to avoid rectstroke with a 0.0 linewidth, the behavior
of which is device dependent.

After applying that fix tonight, I have for the first time produce printed (on
paper) bar lines thinner than 1pt, and I am very happy.


Now, if I can just find a way to left-align the first lyrics in each system I'll
have nothing left to complain about. :)
--
Todd


___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Jan Warchoł]

2011/7/27 Han-Wen Nienhuys :
> On Wed, Jul 27, 2011 at 3:23 PM, Jan Warchoł
>  wrote:
>
>>> I think this is an other issue.  Due to rounding, PDF viewers can err
>>> the placement of the barline by a pixel. This is insolvable, as there
>>> is no way to hint that the barline should coincide with the staffline,
>>> yet we need to go the end of the barline to get the correct effect in
>>> the corners in staves.
>>
>> I'd accept this if it were not for the fact that Finale seems to
>> handle this perfectly.  I've never experienced any sticking out
>> barlines in Finale pdfs, neither on screen nor in print (i can send
>> you some if you want).  Also, the thickness of the stems and barlines
>> is consistent in Finale pdfs (i.e. it looks like second image from
>> here http://lists.gnu.org/archive/html/lilypond-devel/2007-07/msg00076.html
>> while Lily output looks like first one).
>> I'd say that the fact that Finale does something this basic better
>> than LilyPond is a serious insult to us, given how much Finale sucks
>> in these matters (for example in flag attachment, see attached).
>
> You can make the problem disappear by
>
> - hard coding the output to a certain resolution

Umm, i don't understand what you mean...

> - making the bar go to the middle of the staffline, rather than the outside.

I have a patch doing this:
http://codereview.appspot.com/4809057

> both have disadvantages. (what does finale's barline at the end of a
> staff look like?)

Sharp and pointy, example attached.

cheers,
Janek


Canticorum.pdf
Description: Adobe PDF document
___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Wed, Jul 27, 2011 at 3:23 PM, Jan Warchoł
 wrote:

>> I think this is an other issue.  Due to rounding, PDF viewers can err
>> the placement of the barline by a pixel. This is insolvable, as there
>> is no way to hint that the barline should coincide with the staffline,
>> yet we need to go the end of the barline to get the correct effect in
>> the corners in staves.
>
> I'd accept this if it were not for the fact that Finale seems to
> handle this perfectly.  I've never experienced any sticking out
> barlines in Finale pdfs, neither on screen nor in print (i can send
> you some if you want).  Also, the thickness of the stems and barlines
> is consistent in Finale pdfs (i.e. it looks like second image from
> here http://lists.gnu.org/archive/html/lilypond-devel/2007-07/msg00076.html
> while Lily output looks like first one).
> I'd say that the fact that Finale does something this basic better
> than LilyPond is a serious insult to us, given how much Finale sucks
> in these matters (for example in flag attachment, see attached).

You can make the problem disappear by

- hard coding the output to a certain resolution
- making the bar go to the middle of the staffline, rather than the outside.

both have disadvantages. (what does finale's barline at the end of a
staff look like?)

-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Jan Warchoł]

2011/7/26 Han-Wen Nienhuys :
> On Sat, Jul 23, 2011 at 7:45 PM, Jan Warchoł
>  wrote:
>
>> I'm not experienced with PostScript, but i have an impression that
>> this may be related to an issue discussed earlier - see here
>> http://lists.gnu.org/archive/html/lilypond-user/2007-07/msg00485.html
>> and here
>> http://lists.gnu.org/archive/html/lilypond-devel/2007-07/msg00072.html
>> I see that something about draw_round_box was said there...
>
> I think this is an other issue.  Due to rounding, PDF viewers can err
> the placement of the barline by a pixel. This is insolvable, as there
> is no way to hint that the barline should coincide with the staffline,
> yet we need to go the end of the barline to get the correct effect in
> the corners in staves.

I'd accept this if it were not for the fact that Finale seems to
handle this perfectly.  I've never experienced any sticking out
barlines in Finale pdfs, neither on screen nor in print (i can send
you some if you want).  Also, the thickness of the stems and barlines
is consistent in Finale pdfs (i.e. it looks like second image from
here http://lists.gnu.org/archive/html/lilypond-devel/2007-07/msg00076.html
while Lily output looks like first one).
I'd say that the fact that Finale does something this basic better
than LilyPond is a serious insult to us, given how much Finale sucks
in these matters (for example in flag attachment, see attached).

> The one-pixel error is very noticeable on-screen due to
> low-resolution, but on a 1200dpi printer,  a single dot error is about
> 0.02 mm, ie. not noticeable.

I'm not sure whether you are referring to how it looked before your
patch or how it looks after it.
Before your patch it was certainly noticeable enough - even my friend
Luke, who doesn't care that much about the details, said it irked him.
As for how it looks after your patch - I hoped to be able to do some
prints on my choir's printer to see how things looks like now, but it
ran out of toner - new cartridge should be available in next week.
Besides, many printers have resolutions lower than 1200dpi.

cheers,
Janek
<>___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Sat, Jul 23, 2011 at 7:45 PM, Jan Warchoł
 wrote:

> I'm not experienced with PostScript, but i have an impression that
> this may be related to an issue discussed earlier - see here
> http://lists.gnu.org/archive/html/lilypond-user/2007-07/msg00485.html
> and here
> http://lists.gnu.org/archive/html/lilypond-devel/2007-07/msg00072.html
> I see that something about draw_round_box was said there...

I think this is an other issue.  Due to rounding, PDF viewers can err
the placement of the barline by a pixel. This is insolvable, as there
is no way to hint that the barline should coincide with the staffline,
yet we need to go the end of the barline to get the correct effect in
the corners in staves.

The one-pixel error is very noticeable on-screen due to
low-resolution, but on a 1200dpi printer,  a single dot error is about
0.02 mm, ie. not noticeable.

-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Francisco Vila]

2011/7/22 Graham Percival :
> On Thu, Jul 21, 2011 at 11:45:51PM -0300, Han-Wen Nienhuys wrote:
>>
>> This should be backported.  Do we have a special procedure for that nowadays?
>
> Yes, add "backport" to a google tracker issue about it.  But I
> think you just missed the deadline for 2.14.3

You mean 2.14.2

-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Karl Hammar]

Han-Wen Nienhuys:
> On Thu, Jul 21, 2011 at 12:31 PM, Karl Hammar  wrote:
...
> > After "dup" there is two "blot"s, "gt" consumes one and "setlinewidth"
> > the other.
> 
> I'm sorry - I misunderstood; I thought you saw a problem with the code
> rather than the comments.

No problem. I see no problem with the code, it was just the comment
that didn't match the code.

Regards,
/Karl Hammar

---
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57



___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Graham Percival]

On Thu, Jul 21, 2011 at 11:45:51PM -0300, Han-Wen Nienhuys wrote:
> 
> This should be backported.  Do we have a special procedure for that nowadays?

Yes, add "backport" to a google tracker issue about it.  But I
think you just missed the deadline for 2.14.3, and I'm not certain
if we'll have any more 2.14 releases.  First 2.16 release
candidate is in 10 days.

Cheers,
- Graham

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Thu, Jul 21, 2011 at 11:27 PM, Han-Wen Nienhuys  wrote:
>> After "dup" there is two "blot"s, "gt" consumes one and "setlinewidth"
>> the other.
>
> I'm sorry - I misunderstood; I thought you saw a problem with the code
> rather than the comments.
>
> pushed as 5291daf785cd215145473781612732de94890ba0

This should be backported.  Do we have a special procedure for that nowadays?

-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Thu, Jul 21, 2011 at 12:31 PM, Karl Hammar  wrote:
> Han-Wen Nienhuys:
>> On Thu, Jul 21, 2011 at 8:11 AM, Karl Hammar  wrote:
>> > Han-Wen Nienhuys:
>> >> Werner, can you have a look at http://codereview.appspot.com/4819041 ?
>> > There is no blot on the stack below (as indicated by the comment),
>> there is; the dup puts it on the stack.  The comment indicate the
>> state after the call.
>
>   /draw_round_box % width height x y blot
>   {
> w h x y b
>         dup
> w h x y b b
>         0.0 gt
> w h x y b a_bool  % the a_bool it consumed by the ifelse below
>         {
> w h x y b
>                 setlinewidth % w h x y
> w h x y
>                 0 setlinecap
>                 1 setlinejoin
> w h x y
>                 rmoveto % b w h
> w h
>                 currentpoint % b w h x1 y1
>                 4 2 roll % b x1 y1 w h
>                 4 copy
>                 rectfill
>                 rectstroke
>         } {
>                 pop % w h x y
>                 rmoveto % w h
>                 currentpoint % w h x1 y1
>                 4 2 roll % x1 y1 w h
>                 rectfill
>         } ifelse
>   } bind def
>
> After "dup" there is two "blot"s, "gt" consumes one and "setlinewidth"
> the other.

I'm sorry - I misunderstood; I thought you saw a problem with the code
rather than the comments.

pushed as 5291daf785cd215145473781612732de94890ba0
-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Karl Hammar]

Han-Wen Nienhuys:
> On Thu, Jul 21, 2011 at 8:11 AM, Karl Hammar  wrote:
> > Han-Wen Nienhuys:
> >> Werner, can you have a look at http://codereview.appspot.com/4819041 ?
> > There is no blot on the stack below (as indicated by the comment),
> there is; the dup puts it on the stack.  The comment indicate the
> state after the call.

   /draw_round_box % width height x y blot
   {
w h x y b
 dup
w h x y b b
 0.0 gt
w h x y b a_bool  % the a_bool it consumed by the ifelse below
 {
w h x y b
 setlinewidth % w h x y
w h x y
 0 setlinecap
 1 setlinejoin
w h x y
 rmoveto % b w h
w h
 currentpoint % b w h x1 y1
 4 2 roll % b x1 y1 w h
 4 copy
 rectfill
 rectstroke
 } {
 pop % w h x y
 rmoveto % w h
 currentpoint % w h x1 y1
 4 2 roll % x1 y1 w h
 rectfill
 } ifelse
   } bind def

After "dup" there is two "blot"s, "gt" consumes one and "setlinewidth"
the other.

///

Also if there were a "blot" at "rmoveto" time due to a "dup", it would
be at top of stack, not below "w h" as indicated by the comment.

///

If it would behave as the comment indicate, then

 w h x y 0draw_round_box => -
 w h x y blot draw_round_box => blot

e.i. it would leave a lone "blot" at the top of the stack if it's > 0.

Regards,
/Karl Hammar

---
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57



___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Thu, Jul 21, 2011 at 8:11 AM, Karl Hammar  wrote:
> Han-Wen Nienhuys:
> ...
>> Werner, can you have a look at http://codereview.appspot.com/4819041 ?
>
>   /draw_round_box % width height x y blot
>   {
>  -       setlinewidth % w h x y
>  -       0 setlinecap
>  -       1 setlinejoin
>  +        dup
>  +       0.0 gt {
>  +               setlinewidth % w h x y
>  +               0 setlinecap
>  +               1 setlinejoin
>  +
>
> There is no blot on the stack below (as indicated by the comment),

there is; the dup puts it on the stack.  The comment indicate the
state after the call.


> You don't seem to use this, why defining it?
>
>  +/draw_box % width height x y
>  +{

debugging fart. Will remove.

-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Karl Hammar]

Han-Wen Nienhuys:
...
> Werner, can you have a look at http://codereview.appspot.com/4819041 ?

   /draw_round_box % width height x y blot
   {
  -   setlinewidth % w h x y
  -   0 setlinecap
  -   1 setlinejoin
  +dup
  +   0.0 gt {
  +   setlinewidth % w h x y
  +   0 setlinecap
  +   1 setlinejoin
  +

There is no blot on the stack below (as indicated by the comment),
it was swallowed by setlinewidth.

  +   rmoveto % b w h
  +   currentpoint % b w h x1 y1
  +   4 2 roll % b x1 y1 w h
  +   4 copy
  +   rectfill
  +   rectstroke
  +   } {
  +   pop % w h x y
  +   rmoveto % w h
  +   currentpoint % w h x1 y1
  +   4 2 roll % x1 y1 w h
  +   rectfill
  +   } ifelse
  +} bind def

You don't seem to use this, why defining it?

  +/draw_box % width height x y
  +{
  rmoveto % w h
  currentpoint % w h x1 y1
  4 2 roll % x1 y1 w h
  4 copy
  rectfill
  -   rectstroke
   } bind def

Regards,
/Karl Hammar

---
Aspö Data
Lilla Aspö 148
S-742 94 Östhammar
Sweden
+46 173 140 57



___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Werner LEMBERG]

> Werner, can you have a look at http://codereview.appspot.com/4819041 ?

It looks OK, thanks!  Currently, I get a server error on Rietveld, so
I have to comment here.  Please check spaces vs. tabs.


Werner

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Wed, Jul 20, 2011 at 1:07 AM, Werner LEMBERG  wrote:
>
>> The patch below fixes it, and seems safe for inclusion, but it would
>> be nice if we could understand something more.
>>
>> Werner, any insights?
>>
>> +     0.0 gt { 0 } { 2 } ifelse
>> +     setlinecap
>
> Sorry, no.  Perhaps someone can ask on a GhostScript mailing list,
> showing the demo file which exhibits the problem.

Werner, can you have a look at http://codereview.appspot.com/4819041 ?

Thanks!

(I am assuming you are somewhat fluent in Postscript)

-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Han-Wen Nienhuys]

On Wed, Jul 20, 2011 at 1:07 AM, Werner LEMBERG  wrote:
>> The patch below fixes it, and seems safe for inclusion, but it would
>> be nice if we could understand something more.
>>
>> Werner, any insights?
>>
>> +     0.0 gt { 0 } { 2 } ifelse
>> +     setlinecap
>
> Sorry, no.  Perhaps someone can ask on a GhostScript mailing list,
> showing the demo file which exhibits the problem.

I asked around (the ex-maintainer of GS works at google); we concluded
that it was better to use fill for non-rounded boxes, rather than
doing filldraw.

-- 
Han-Wen Nienhuys - han...@xs4all.nl - http://www.xs4all.nl/~hanwen

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Werner LEMBERG]

> The patch below fixes it, and seems safe for inclusion, but it would
> be nice if we could understand something more.
> 
> Werner, any insights?
> 
> + 0.0 gt { 0 } { 2 } ifelse
> + setlinecap

Sorry, no.  Perhaps someone can ask on a GhostScript mailing list,
showing the demo file which exhibits the problem.


Werner

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Francisco Vila]

2011/7/19 Francisco Vila :

> Do you think it is related to the evince-lilypond-barline problem?

Aaand --- no, it is not related. here is my print scan from your Wagner example.

  http://paconet.org/lilypond/barlines.png

Sorry.
-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel

Re: Postscript printer errors with rounded barlines?

[Francisco Vila]

2011/7/19 Han-Wen Nienhuys :
> Hi guys,
>
> I've noticed that some printer postscript interpreters add
> linethickness to barlines when printed, even though they create with
> draw_round_box with line thickness 0.0.  The result is that barlines
> noticeably stick out of the staff; the error is in the printer, since
> GV at 64x mag shows that everything is OK, and show with both the PDF
> and the PS output.  Unfortunately, a small example does not trigger,
> but the attached file when run through lilypond at commit
> 4e55f40a1bf65633fedc62d09d91d8e3c0247efe does (PDF attached).   By the
> amount it sticks out, I suspect it is somehow picking up the
> blotdiameter of another routine, but I have no idea how that could
> happen.
>
> I have seen this on my home printer (HP LJ 2100, with HPs postscript
> extension card, configured as generic postscript in CUPS) and the
> printer at my office, (Gestetner mp c2550).  This is a regression, as
> I am sure it worked correctly at some point on my printer before, but
> it I can't see anything obvious in the change log.
>
> The patch below fixes it, and seems safe for inclusion, but it would
> be nice if we could understand something more.

Do you think it is related to the evince-lilypond-barline problem? If
poppler acts like a printer, that would explain thick barlines both on
screen and on printer when printed from evince. I have been suffering
that issue for AGES now, see

  http://lists.gnu.org/archive/html/lilypond-user/2009-01/msg00696.html

It would be absolutely great if the stock PDF viewer that comes with
gnome/ubuntu were able to display/print simple lilypond scores
correctly.  Sorry if it sounds stupid and/or it is unrelated.
-- 
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com

___
lilypond-devel mailing list
lilypond-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-devel