Quantcast

fontforge's test printing facility

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

fontforge's test printing facility

Ryuji Suzuki
I fix several commercial fonts that are not quite usable because of
bad spacing. There are lots of them... Here's a functionality I wish
FF had, which makes this sort of work a lot easier.

Functions to do the following:

1. print sample text without kerning, without ligature
2. print sample text with kerning, without ligature
3. print sample text with kerning and with ligature

4. Functions 1, 2, and 3 wherein the sample text is as follows:

OAO OBO OCO ODO ... OZO
OaO ObO OcO OdO ... OzO
HAH HBH HCH HDH ... HZH
HaH HbH HcH HdH ... HzH
oAo oBo oCo oDo ... oZo
oao obo oco odo ... ozo
nAn nBn nCn nDn ... nZn
nan nbn ncn ndn ... nzn

xAx xBx xCx xDx ... xZx
xax xbx xcx xdx ... xzx
where x is any user specified character

101 111 121 ... 191
000 010 020 ... 090

The reason I want to make sure letters fit reasonably well without
kerning is because there are several broken applications like
OpenOffice. OpenOffice Writer seems to use kerning when available but
Draw doesn't seems to. I'm VERY annoyed by it so I'm ditching
OpenOffice. But I like to make sure my fonts look reasonable without
kerning, partly to minimize necessary number of kerning pairs. Then
ligature... I have to make sure the font looks good with or without
ligature because there are plenty of broken applications, including
OpenOffice. This one, even OO Writer doesn't seem to work with
ligature. (How can people use OpenOffice? It seems to die if I have
too many weights of a font family available to it as well...)

That "test pattern" seems to work very nicely when fitting types. I
use somewhat personalized version of Walter Tracy's method, and that
test pattern is found to be very useful for me. I prefer to work with
10-14pt on paper print. When I use metric view window with comfortable
size for screen, I often miss small problems. (Is this just me?)

In relation to my previous question, I really wish FF had a way to cut
and paste a-zA-Z as well as 0-9 with kern information because I like
to fit letters in regular style, and then work with Small Caps, and
then text figures... while cutting and pasting glyphs. (Unless there
is a more sophisticated way to do so... convert them all to OpenType,
but I already asked that question a long time ago and there wasnt an
easy solution for it.)

I feel bad making a lengthy wish list and have little to contribute
myself... But thanks for such a nice tool!

Ryuji Suzuki


-------------------------------------------------------
This SF.Net email is sponsored by Yahoo.
Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
Search APIs Find out how you can build Yahoo! directly into your own
Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

Ryuji Suzuki
Related to it, I made a text file containing my "test pattern"
in fonttest.text and made a script like this:

fontforge -c 'Open($argv[1]);PrintSetup(5); PrintFont(3, 14, "/.../fonttest.text");' fontfile.sfd

However, line gets screwed up when I have pairs with kerning info.
(At least with negative kernings.)

This doesn't happen when I type in the same text in print dialog
window.

Thanks

Ryuji


-------------------------------------------------------
This SF.Net email is sponsored by Yahoo.
Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
Search APIs Find out how you can build Yahoo! directly into your own
Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

Ralf Stubner
Ryuji Suzuki wrote:
> Related to it, I made a text file containing my "test pattern" in
> fonttest.text and made a script like this:
>
> fontforge -c 'Open($argv[1]);PrintSetup(5); PrintFont(3, 14,
> "/.../fonttest.text");' fontfile.sfd
>
> However, line gets screwed up when I have pairs with kerning info.
> (At least with negative kernings.)

In what sense are lines screwed up? Is this specific to some font? Can
you provide a small text file that shows this error when used together
with say Vera, one of the URW fonts, ... This might help debugging this
problem.

cheerio
ralf




-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

Ryuji Suzuki
From: Ralf Stubner <[hidden email]>
Subject: Re: [Fontforge-users] fontforge's test printing facility
Date: Thu, 02 Jun 2005 21:33:05 +0200

> In what sense are lines screwed up?

The line seems to come back to the beginning of the line without carriage.
The rest of the line prints over the previous contents.

> Is this specific to some font? Can
> you provide a small text file that shows this error when used together
> with say Vera, one of the URW fonts, ... This might help debugging this
> problem.

THis is what I think you can reproduce (based on my observation).
(The font I work with is very minor commercial font with lots of
personalization so it wouldn't help you.)

You could take any font, give a pair (say VA) a kerning (say -45).
The problem should occur when you feed a text file comprising VA in the
script of the form I described.

The cases I observed are highly correlated with negative kerning.

Ryuji Suzuki


-------------------------------------------------------
This SF.Net email is sponsored by Yahoo.
Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
Search APIs Find out how you can build Yahoo! directly into your own
Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

Ralf Stubner
Ryuji Suzuki wrote:

> You could take any font, give a pair (say VA) a kerning (say -45).
> The problem should occur when you feed a text file comprising VA in the
> script of the form I described.

Ok,

fontforge -c 'Open("uplr8a.pfb");PrintSetup(5);PrintFont(4,14,"th VA");'

does give a screwed up line, the way you described it. Since URW
Palladio L Roman (that's uplr8a.pfb) has positve kerning between th and
negative between VA, both signs seem to be affected. PS production via
PrintSetup(3) does work fine, though. So the problem seems to be with
the code producing PDF.

cheerio
ralf




-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

kerning pairs (was fontforge's test printing facility)

Ryuji Suzuki
By the way is there a way to print/list out ALL kerning pairs in a
font? (preferrably with fontforge or any other open source programs)

Thanks for comfirming my report

Ryuji

PS. I use LaserWriter 8500 and 12/640PS which take level 2 PS and
that's probably why PS output didn't work for me. So I saved the
output in PDF and printed them.

From: Ralf Stubner <[hidden email]>
Subject: Re: [Fontforge-users] fontforge's test printing facility
Date: Thu, 02 Jun 2005 23:30:57 +0200

> Since URW
> Palladio L Roman (that's uplr8a.pfb) has positve kerning between th and
> negative between VA, both signs seem to be affected. PS production via
> PrintSetup(3) does work fine, though. So the problem seems to be with
> the code producing PDF.


-------------------------------------------------------
This SF.Net email is sponsored by Yahoo.
Introducing Yahoo! Search Developer Network - Create apps using Yahoo!
Search APIs Find out how you can build Yahoo! directly into your own
Applications - visit http://developer.yahoo.net/?fr=offad-ysdn-ostg-q22005
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

George Williams
In reply to this post by Ryuji Suzuki
On Wed, 2005-06-01 at 19:11, Ryuji Suzuki wrote:
> I fix several commercial fonts that are not quite usable because of
> bad spacing. There are lots of them... Here's a functionality I wish
> FF had, which makes this sort of work a lot easier.

> 1. print sample text without kerning, without ligature
I'm not sure what you are asking for. FontForge has a print sample
command.

If you want to print with out kerning or ligatures, fontforge also has
commands to remove kerning and ligatures from a font. If you then print
it will print without kerning or ligatures.
> 2. print sample text with kerning, without ligature
> 3. print sample text with kerning and with ligature
>
> 4. Functions 1, 2, and 3 wherein the sample text is as follows:
You can set the sample text to whatever you wish.




-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

George Williams
In reply to this post by Ralf Stubner
On Thu, 2005-06-02 at 14:30, Ralf Stubner wrote:
> fontforge -c 'Open("uplr8a.pfb");PrintSetup(5);PrintFont(4,14,"th VA");'
>
> does give a screwed up line, the way you described it. Since URW
> Palladio L Roman (that's uplr8a.pfb) has positve kerning between th and
> negative between VA, both signs seem to be affected. PS production via
> PrintSetup(3) does work fine, though. So the problem seems to be with
> the code producing PDF.
Yes, I had misread the pdf documentation and the command I was using do
change the cursor position didn't.

Thanks for the simple test case Ralf.

pdfkern.patch (8K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: fontforge's test printing facility

Ryuji Suzuki
In reply to this post by George Williams
From: George Williams <[hidden email]>
Subject: Re: [Fontforge-users] fontforge's test printing facility
Date: 08 Jun 2005 10:34:45 -0700

> I'm not sure what you are asking for. FontForge has a print sample
> command.

> If you want to print with out kerning or ligatures, fontforge also
> has commands to remove kerning and ligatures from a font. If you
> then print it will print without kerning or ligatures.

One possible implementation is to provide a way to on/off kerning and
another to on/off ligature substitution, as a part of test printing
dialog. Is it difficult to add a ligature substitution? One on/off for
all available ligatures would be sufficient for testing purposes.

> You can set the sample text to whatever you wish.

Well, until I figured out that I could do this in a script, it wasn't
a very practical option... but I learned how to do this bit of work.

Thanks

Ryuji


-------------------------------------------------------
This SF.Net email is sponsored by: NEC IT Guy Games.  How far can you shotput
a projector? How fast can you ride your desk chair down the office luge track?
If you want to score the big prize, get to know the little guy.  
Play to win an NEC 61" plasma display: http://www.necitguy.com/?r=20
_______________________________________________
Fontforge-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/fontforge-users
Loading...