From cjl at spamcop.net  Wed Sep  1 19:56:21 2010
From: cjl at spamcop.net (cjl at spamcop.net)
Date: Wed, 01 Sep 2010 20:56:21 +0100
Subject: [gs-devel] Ghostscript 9.00 release candidate 2
Message-ID: <20100901205621.bvml4c3um84so4so-pwy@webmail.spamcop.net>


We found a couple of issues we felt were too important to ignore for a  
release, so....

The second Ghostscript 9.00 release candidate archives are available at:
http://ghostscript.com/~chrisl/ghostscript-9.00rc2.tar.gz
http://ghostscript.com/~chrisl/ghostscript-9.00rc2.tar.xz

(depending on your preferred compression type)

please give it a try on your system if you're interested in helping test the
release-in-progress.


Chris

From john at wexfordpress.com  Wed Sep  1 20:20:48 2010
From: john at wexfordpress.com (John Culleton)
Date: Wed, 1 Sep 2010 16:20:48 -0400
Subject: [gs-devel] Need PDF X/1-a output.
Message-ID: <201009011620.49082.john@wexfordpress.com>

Many of the copies of Acrobat Distiller are sold for just one feature: 
conversion of a cover pdf to PDF X/1-a specifications. Scribus has this 
capability built in but for other methods of producing pdfs, including 
ghostscript pdftex, inkscape  etc. there is no easy way to get there.

The man page for ps2pdf refers to an html document but I have not 
found it yet.

If Ghostscript has this capability that would mean a lot of cash saved 
for self and small publishers. 
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html

From cjl at spamcop.net  Wed Sep  1 21:21:29 2010
From: cjl at spamcop.net (cjl at spamcop.net)
Date: Wed, 01 Sep 2010 22:21:29 +0100
Subject: [gs-devel] Need PDF X/1-a output.
In-Reply-To: <201009011620.49082.john@wexfordpress.com>
References: <201009011620.49082.john@wexfordpress.com>
Message-ID: <20100901222129.93j9nr6b4o848kgs-pwy@webmail.spamcop.net>


Apologies, I thought someone else had answered you previously on this.

The pdfwrite device isn't really my area, but the html document the  
manpage refers to is part of the Ghostscript documentation, and  
exactly where it is will depend on exactly how to get Ghostscript, but  
you can find the very latest version here:
http://svn.ghostscript.com/ghostscript/trunk/gs/doc/Ps2pdf.htm


We do currently support PDF/X-3 output, so that might be of help (see  
the above link).

Alternatively, if you really need PDF/X-1a, you could raise a feature  
request in our Bugzilla database:
http://bugs.ghostscript.com/

I hope that's of some help,

Chris


Quoting John Culleton <john at wexfordpress.com>:

> Many of the copies of Acrobat Distiller are sold for just one feature:
> conversion of a cover pdf to PDF X/1-a specifications. Scribus has this
> capability built in but for other methods of producing pdfs, including
> ghostscript pdftex, inkscape  etc. there is no easy way to get there.
>
> The man page for ps2pdf refers to an html document but I have not
> found it yet.
>
> If Ghostscript has this capability that would mean a lot of cash saved
> for self and small publishers.
> --
> John Culleton
> Wexford Press
> "Create Book Covers with Scribus"
> Printable E-book 38 pages $5.95
> http://www.booklocker.com/books/4055.html
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>


From ztiago at yahoo.com.br  Mon Sep  6 19:51:08 2010
From: ztiago at yahoo.com.br (Tiago Zaniquelli)
Date: Mon, 6 Sep 2010 12:51:08 -0700 (PDT)
Subject: [gs-devel] pdf/a searchable
Message-ID: <964716.65622.qm@web39506.mail.mud.yahoo.com>

Good Afternoon,

I am new to the list, and I have a problem to generate PDF / A Searchable.

I have a searchable PDF file, but when I try to turn it into PDF / A
I miss the part searchable, stay within the pdf / a image and not just the
text.

Has anyone had this problem?

Thanks

Tiago Zaniquelli


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ghostscript.com/pipermail/gs-devel/attachments/20100906/0d4c88f3/attachment.html>

From ken.sharp at artifex.com  Tue Sep  7 14:09:58 2010
From: ken.sharp at artifex.com (Ken Sharp)
Date: Tue, 07 Sep 2010 15:09:58 +0100
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <396352.32505.qm@web39503.mail.mud.yahoo.com>
References: <5.1.0.14.2.20100907130341.042f5468@pop.gmail.com>
Message-ID: <5.1.0.14.2.20100907150842.0420ae90@pop.gmail.com>

At 06:48 07/09/2010 -0700, you wrote:
>Hello Ken,
>
>This picture is before I make the conversion and it was made by OCR, but I 
>wanted to convert to PDF / A keeping the OCR, you know?
>
>as I do that? could you help me? how to get my command line?


As I said, you can't. The non-marking text is dropped, I think by the PDF 
interpreter, because it makes no marks on the page. Since pdfwrite never 
sees the invisible text, it can't put it in the output PDF.


             Ken


From ken.sharp at artifex.com  Tue Sep  7 16:03:50 2010
From: ken.sharp at artifex.com (Ken Sharp)
Date: Tue, 07 Sep 2010 17:03:50 +0100
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <401029.21829.qm@web39503.mail.mud.yahoo.com>
References: <5.1.0.14.2.20100907150842.0420ae90@pop.gmail.com>
Message-ID: <5.1.0.14.2.20100907170323.06302488@pop.gmail.com>

At 09:00 07/09/2010 -0700, you wrote:


>Do you know any way or tool I convert a PDF to Searchable PDF / A 
>searchable, can be paid to any component.

Adobe Acrobat would seem like your most likely candidate.



         Ken


From ken.sharp at artifex.com  Tue Sep  7 16:27:27 2010
From: ken.sharp at artifex.com (Ken Sharp)
Date: Tue, 07 Sep 2010 17:27:27 +0100
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <475814BB8706D34186ECDF6F2C82DAEA014B4563@BKEXCH.brancekrac
	hy.com>
References: <5.1.0.14.2.20100907150842.0420ae90@pop.gmail.com>
	<5.1.0.14.2.20100907170323.06302488@pop.gmail.com>
Message-ID: <5.1.0.14.2.20100907171948.04229850@pop.gmail.com>

At 11:14 07/09/2010 -0500, you wrote:


>Full Acrobat can be pricey.  First, I would try PDFill.  It is
>crippleware (puts a watermark on everything it saves until you activate)

It also needs a copy of Ghostscript (at least for the free PDF tools), and 
they don't pay Artifex any licence fees for that.

Forgive us for not recommending such software ;-)

It may be that Primo PDF (who we do like, because they licence Ghostscript) 
can do the PDF/A conversion also. As usual, Google is your friend...


>We purchased a 20-seat license three years ago and have had excellent
>results with it.

Its not me that's asking for a solution.....


             Ken


From soroush124 at yahoo.com  Tue Sep  7 23:25:10 2010
From: soroush124 at yahoo.com (Soroush Hassani)
Date: Tue, 7 Sep 2010 16:25:10 -0700 (PDT)
Subject: [gs-devel] gs command error
Message-ID: <600988.1166.qm@web112401.mail.gq1.yahoo.com>

Hi,
I am using below command to compress a pdf document 
http://simplewebcode.com/97.pdf :
 gs -dPDFSETTINGS=/screen -dCompatibilityLevel=1.5 -sDEVICE=pdfwrite -dNOPROMPT 
-dNOPAUSE -dBATCH -sOutputFile=output_path input_path

But I get below error:
GPL Ghostscript 8.71 (2010-02-10)
Copyright (C) 2010 Artifex Software, Inc.  All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
Processing pages 1 through 12.
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Error: /rangecheck in --.execform1--
Operand stack:
   --dict:11/20(L)--   1   11   --dict:2/2(L)--   --nostringval--   --dict:2/2(L 
                                                                                
       )--   128.389   776.785   136.689   528.777
Execution stack:
   %interp_exit   .runexec2   --nostringval--   --nostringval--   --nostringval- 
                                                                                
       -   2   %stopped_push   --nostringval--   --nostringval--   
--nostringval--   fa                                                             
                           lse   1   %stopped_push   1878   1   3   %oparray_pop 
  1877   1   3   %oparray_                                                       
                                 pop   1861   1   3   %oparray_pop   
--nostringval--   --nostringval--   7   1                                       
                                                    12   --nostringval--   
%for_pos_int_continue   --nostringval--   --nostringval--                       
                                                                    false   1   
%stopped_push   --nostringval--   --nostringval--   --nostringval               
                                                                         --   
%array_continue   --nostringval--   false   1   %stopped_push   --nostringv     
                                                                                
   al--   %loop_continue   --nostringval--   1233344   --nostringval--   1826   
4                                                                               
            12   %oparray_pop   --nostringval--   false   1   %stopped_push   
1825   4   12                                                                   
                        %oparray_pop   --nostringval--
Dictionary stack:
   --dict:1153/1684(ro)(G)--   --dict:1/20(G)--   --dict:76/200(L)--   --dict:76 
                                                                                
       /200(L)--   --dict:108/127(ro)(G)--   --dict:288/300(ro)(G)--   
--dict:22/25(L)-                                                                 
                       -   --dict:6/8(L)--   --dict:22/40(L)--   
--dict:1/1(ro)(G)--   --dict:7/15(L)--                                           
                                                --dict:1/1(ro)(G)--
Current allocation mode is local
Last OS error: 2
GPL Ghostscript 8.71: Unrecoverable error, exit code 1

Same error on Linux CentOS and WinXP and even on two release candidate of 
version 9 compiled on CentOS.

Any idea where is the problem?

Thanks
Soroush



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ghostscript.com/pipermail/gs-devel/attachments/20100907/0b079945/attachment.html>

From john at wexfordpress.com  Wed Sep  8 03:46:42 2010
From: john at wexfordpress.com (John Culleton)
Date: Tue, 7 Sep 2010 23:46:42 -0400
Subject: [gs-devel] gs command error
In-Reply-To: <600988.1166.qm@web112401.mail.gq1.yahoo.com>
References: <600988.1166.qm@web112401.mail.gq1.yahoo.com>
Message-ID: <201009072346.43001.john@wexfordpress.com>

On Tuesday 07 September 2010 19:25:10 Soroush Hassani wrote:
> Hi,
> I am using below command to compress a pdf document
> http://simplewebcode.com/97.pdf :
>  gs -dPDFSETTINGS=/screen -dCompatibilityLevel=1.5
> -sDEVICE=pdfwrite -dNOPROMPT -dNOPAUSE -dBATCH
> -sOutputFile=output_path input_path
>
> But I get below error:
> GPL Ghostscript 8.71 (2010-02-10)
> Copyright (C) 2010 Artifex Software, Inc.  All rights reserved.
> This software comes with NO WARRANTY: see the file PUBLIC for
> details. Processing pages 1 through 12.
> Page 1
> Page 2
> Page 3
> Page 4
> Page 5
> Page 6
> Error: /rangecheck in --.execform1--
> Operand stack:
>    --dict:11/20(L)--   1   11   --dict:2/2(L)--   --nostringval--  
> --dict:2/2(L
>
>        )--   128.389   776.785   136.689   528.777
> Execution stack:
>    %interp_exit   .runexec2   --nostringval--   --nostringval--  
> --nostringval-
>
>        -   2   %stopped_push   --nostringval--   --nostringval--
> --nostringval--   fa
>                            lse   1   %stopped_push   1878   1   3  
> %oparray_pop 1877   1   3   %oparray_
>                                  pop   1861   1   3   %oparray_pop
> --nostringval--   --nostringval--   7   1
>                                                     12  
> --nostringval-- %for_pos_int_continue   --nostringval--  
> --nostringval--
>                                                                    
> false   1 %stopped_push   --nostringval--   --nostringval--  
> --nostringval -- %array_continue   --nostringval--   false   1  
> %stopped_push   --nostringv
>
>    al--   %loop_continue   --nostringval--   1233344  
> --nostringval--   1826 4
>             12   %oparray_pop   --nostringval--   false   1  
> %stopped_push 1825   4   12
>                         %oparray_pop   --nostringval--
> Dictionary stack:
>    --dict:1153/1684(ro)(G)--   --dict:1/20(G)--  
> --dict:76/200(L)--   --dict:76
>
>        /200(L)--   --dict:108/127(ro)(G)--  
> --dict:288/300(ro)(G)-- --dict:22/25(L)-
>                        -   --dict:6/8(L)--   --dict:22/40(L)--
> --dict:1/1(ro)(G)--   --dict:7/15(L)--
>                                                 --dict:1/1(ro)(G)--
> Current allocation mode is local
> Last OS error: 2
> GPL Ghostscript 8.71: Unrecoverable error, exit code 1
>
> Same error on Linux CentOS and WinXP and even on two release
> candidate of version 9 compiled on CentOS.
>
> Any idea where is the problem?
>
> Thanks
> Soroush

How was the pdf file created? It looks like a defect in the file itself. 
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html

From soroush124 at yahoo.com  Wed Sep  8 04:34:29 2010
From: soroush124 at yahoo.com (Soroush Hassani)
Date: Tue, 7 Sep 2010 21:34:29 -0700 (PDT)
Subject: [gs-devel] gs command error
In-Reply-To: <201009072346.43001.john@wexfordpress.com>
References: <600988.1166.qm@web112401.mail.gq1.yahoo.com>
	<201009072346.43001.john@wexfordpress.com>
Message-ID: <503643.86292.qm@web112401.mail.gq1.yahoo.com>

Thanks John for checking and replying.
It was created and exported  to PDF with InDesign CS4,  Acrobat Reader and 
GSview 4.9 open and show it fine.
Which sort of defect?

Soroush



________________________________
From: John Culleton <john at wexfordpress.com>
To: gs-devel at ghostscript.com
Sent: Wed, September 8, 2010 8:16:42 AM
Subject: Re: [gs-devel] gs command error

On Tuesday 07 September 2010 19:25:10 Soroush Hassani wrote:
> Hi,
> I am using below command to compress a pdf document
> http://simplewebcode.com/97.pdf :
>  gs -dPDFSETTINGS=/screen -dCompatibilityLevel=1.5
> -sDEVICE=pdfwrite -dNOPROMPT -dNOPAUSE -dBATCH
> -sOutputFile=output_path input_path
>
> But I get below error:
> GPL Ghostscript 8.71 (2010-02-10)
> Copyright (C) 2010 Artifex Software, Inc.  All rights reserved.
> This software comes with NO WARRANTY: see the file PUBLIC for
> details. Processing pages 1 through 12.
> Page 1
> Page 2
> Page 3
> Page 4
> Page 5
> Page 6
> Error: /rangecheck in --.execform1--
> Operand stack:
>    --dict:11/20(L)--   1   11   --dict:2/2(L)--   --nostringval--  
> --dict:2/2(L
>
>        )--   128.389   776.785   136.689   528.777
> Execution stack:
>    %interp_exit   .runexec2   --nostringval--   --nostringval--  
> --nostringval-
>
>        -   2   %stopped_push   --nostringval--   --nostringval--
> --nostringval--   fa
>                            lse   1   %stopped_push   1878   1   3  
> %oparray_pop 1877   1   3   %oparray_
>                                  pop   1861   1   3   %oparray_pop
> --nostringval--   --nostringval--   7   1
>                                                     12  
> --nostringval-- %for_pos_int_continue   --nostringval--  
> --nostringval--
>                                                                    
> false   1 %stopped_push   --nostringval--   --nostringval--  
> --nostringval -- %array_continue   --nostringval--   false   1  
> %stopped_push   --nostringv
>
>    al--   %loop_continue   --nostringval--   1233344  
> --nostringval--   1826 4
>             12   %oparray_pop   --nostringval--   false   1  
> %stopped_push 1825   4   12
>                         %oparray_pop   --nostringval--
> Dictionary stack:
>    --dict:1153/1684(ro)(G)--   --dict:1/20(G)--  
> --dict:76/200(L)--   --dict:76
>
>        /200(L)--   --dict:108/127(ro)(G)--  
> --dict:288/300(ro)(G)-- --dict:22/25(L)-
>                        -   --dict:6/8(L)--   --dict:22/40(L)--
> --dict:1/1(ro)(G)--   --dict:7/15(L)--
>                                                 --dict:1/1(ro)(G)--
> Current allocation mode is local
> Last OS error: 2
> GPL Ghostscript 8.71: Unrecoverable error, exit code 1
>
> Same error on Linux CentOS and WinXP and even on two release
> candidate of version 9 compiled on CentOS.
>
> Any idea where is the problem?
>
> Thanks
> Soroush

How was the pdf file created? It looks like a defect in the file itself. 
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html
_______________________________________________
gs-devel mailing list
gs-devel at ghostscript.com
http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ghostscript.com/pipermail/gs-devel/attachments/20100907/a8e5e3e5/attachment-0001.html>

From ken.sharp at artifex.com  Wed Sep  8 07:03:33 2010
From: ken.sharp at artifex.com (Ken Sharp)
Date: Wed, 08 Sep 2010 08:03:33 +0100
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <20100907.OY6.45112400@192.6.1.181>
Message-ID: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>

At 18:47 07/09/2010 +0000, you wrote:
>Hi Ken,
>Have you been able to embed a barcode into a PDF?

I haven't tried. There are two ways to do this:

1) Use a barcode font

2) Create a graphic representation of a barcode using vector elements.

Of the two a barcode font is considerably the easier.


             Ken


From mail_ben_schmidt at yahoo.com.au  Wed Sep  8 07:49:04 2010
From: mail_ben_schmidt at yahoo.com.au (Ben Schmidt)
Date: Wed, 08 Sep 2010 17:49:04 +1000
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
Message-ID: <4C873FF0.70201@yahoo.com.au>

Of course, a third way is to embed a bitmap of the barcode.

I could barely believe it when this was how a barcode was delivered to me by some 
government organisation once. I'd chosen PS as my preferred format (rather than 
TIFF) as I assumed I would get a vector graphic, but no....

Ben.



On 8/09/10 5:03 PM, Ken Sharp wrote:
> At 18:47 07/09/2010 +0000, you wrote:
>> Hi Ken,
>> Have you been able to embed a barcode into a PDF?
>
> I haven't tried. There are two ways to do this:
>
> 1) Use a barcode font
>
> 2) Create a graphic representation of a barcode using vector elements.
>
> Of the two a barcode font is considerably the easier.
>
>
> Ken
>
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>


From ken.sharp at artifex.com  Wed Sep  8 08:02:03 2010
From: ken.sharp at artifex.com (Ken Sharp)
Date: Wed, 08 Sep 2010 09:02:03 +0100
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <4C873FF0.70201@yahoo.com.au>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
Message-ID: <5.1.0.14.2.20100908085903.06449e60@pop.gmail.com>

At 17:49 08/09/2010 +1000, Ben Schmidt wrote:

>Of course, a third way is to embed a bitmap of the barcode.

The quality of that solution is, of course, poor. It doesn't scale well 
(especially if there is any rotation), lossy image compression could 
corrupt the integrity of the coding. And its usually larger (data size) 
than using a vector form.

All of which is why I didn't really consider it as an option, though I too 
have seen it done ;-)

Are you trying to solve a specific problem ?


         Ken


From brian at aljex.com  Wed Sep  8 15:03:19 2010
From: brian at aljex.com (Brian K. White)
Date: Wed, 08 Sep 2010 11:03:19 -0400
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
Message-ID: <4C87A5B7.40001@aljex.com>

On 9/8/2010 3:03 AM, Ken Sharp wrote:
> At 18:47 07/09/2010 +0000, you wrote:
>> Hi Ken,
>> Have you been able to embed a barcode into a PDF?
>
> I haven't tried. There are two ways to do this:
>
> 1) Use a barcode font
>
> 2) Create a graphic representation of a barcode using vector elements.
>
> Of the two a barcode font is considerably the easier.

There is also a postscript _program_ that generates barcodes on the fly 
within a postcript interpreter.

The good:
It's free and open-source (mit/x)
It's dead easy to use
It's flexible as all heck
It works within gs
I don't remember for sure but I think it also did all the necessary 
encoding and checksumming that most barcode formats require so you don't 
have to do it in your app or in some intermediate 
post-application/pre-print processor script.

The bad:
It's about the least efficient way possible to generate a barcode in 
terms of cpu at run-time. An hp lj 1320n takes several solid seconds to 
render a little 12 byte barcode where it prints the exact same barcode 
as a pcl soft font instantly (same speed as plain text or just ejecting 
an empty page, and that's with including the font as part of every print 
job, not pre-loading the font in the printer and merely using it after 
that) GS on any typical pc or server renders it essentially instantly 
but that may or may not scale up very far. The host simply has more cpu 
than a printer and printing a single test form and having it come out 
fast does not mean much.

So, the pcl font also has the advantage that not all pcl printers have 
ps engines so it would work on more and cheaper printers. The ps barcode 
program has the advantage that it's way way way more flexible, free 
(unlike the barcode font), and if you don't mind hitting the host with 
some cpu work it means you can output to any printer gs or cups or linux 
in general can print to, which is even more and even cheaper printers.

http://www.terryburton.co.uk/barcodewriter/
http://www.mecsw.com/software/PCLFontKits.html

I wonder if there is a web graphics generator like gd that has a barcode 
function? It would seem to be a natural.

-- 
bkw

From martin at oneiros.de  Wed Sep  8 20:42:09 2010
From: martin at oneiros.de (=?ISO-8859-1?Q?Martin_Schr=F6der?=)
Date: Wed, 8 Sep 2010 22:42:09 +0200
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <4C87A5B7.40001@aljex.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<4C87A5B7.40001@aljex.com>
Message-ID: <AANLkTi=q3ZJTZMG23qc1dTSuG+Zp2K80nkX-AcosW=YW@mail.gmail.com>

2010/9/8 Brian K. White <brian at aljex.com>:
> http://www.terryburton.co.uk/barcodewriter/
> http://www.mecsw.com/software/PCLFontKits.html

And there's gnu barcode: http://www.gnu.org/software/barcode/

Best
   Martin

From john at wexfordpress.com  Wed Sep  8 20:45:57 2010
From: john at wexfordpress.com (John Culleton)
Date: Wed, 8 Sep 2010 16:45:57 -0400
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <4C87A5B7.40001@aljex.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<4C87A5B7.40001@aljex.com>
Message-ID: <201009081645.57351.john@wexfordpress.com>

On Wednesday 08 September 2010 11:03:19 Brian K. White wrote:
> On 9/8/2010 3:03 AM, Ken Sharp wrote:
> > At 18:47 07/09/2010 +0000, you wrote:
> >> Hi Ken,
> >> Have you been able to embed a barcode into a PDF?
> >
> > I haven't tried. There are two ways to do this:
> >
> > 1) Use a barcode font
> >
> > 2) Create a graphic representation of a barcode using vector
> > elements.
> >
> > Of the two a barcode font is considerably the easier.
>
> There is also a postscript _program_ that generates barcodes on the
> fly within a postcript interpreter.
>
> The good:
> It's free and open-source (mit/x)
> It's dead easy to use
> It's flexible as all heck
> It works within gs
> I don't remember for sure but I think it also did all the necessary
> encoding and checksumming that most barcode formats require so you
> don't have to do it in your app or in some intermediate
> post-application/pre-print processor script.
>
> The bad:
> It's about the least efficient way possible to generate a barcode
> in terms of cpu at run-time. An hp lj 1320n takes several solid
> seconds to render a little 12 byte barcode where it prints the
> exact same barcode as a pcl soft font instantly (same speed as
> plain text or just ejecting an empty page, and that's with
> including the font as part of every print job, not pre-loading the
> font in the printer and merely using it after that) GS on any
> typical pc or server renders it essentially instantly but that may
> or may not scale up very far. The host simply has more cpu than a
> printer and printing a single test form and having it come out fast
> does not mean much.
>
> So, the pcl font also has the advantage that not all pcl printers
> have ps engines so it would work on more and cheaper printers. The
> ps barcode program has the advantage that it's way way way more
> flexible, free (unlike the barcode font), and if you don't mind
> hitting the host with some cpu work it means you can output to any
> printer gs or cups or linux in general can print to, which is even
> more and even cheaper printers.
>
> http://www.terryburton.co.uk/barcodewriter/
> http://www.mecsw.com/software/PCLFontKits.html
>
> I wonder if there is a web graphics generator like gd that has a
> barcode function? It would seem to be a natural.

Wow, where to you people live? First there is an online barcode 
generator that I frequently use:
http://www.tux.org/~milgram/bookland
The output is an eps file, not a bitmap!
Next several DTP programs have built in barcode generators, such as 
Scribus and Inkscape. Again we are talking vector graphics. Finally if 
you like to fiddle with a barcode font in a document the TeXLive distro 
of TeX has an example somewhere. But that is doing it the hard way. 
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html

From martin at oneiros.de  Wed Sep  8 20:49:34 2010
From: martin at oneiros.de (=?ISO-8859-1?Q?Martin_Schr=F6der?=)
Date: Wed, 8 Sep 2010 22:49:34 +0200
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <201009081645.57351.john@wexfordpress.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<4C87A5B7.40001@aljex.com>
	<201009081645.57351.john@wexfordpress.com>
Message-ID: <AANLkTimb8ixLkNuun8WOGpabcoy25miBXKZj2HF4rjkr@mail.gmail.com>

2010/9/8 John Culleton <john at wexfordpress.com>:
> Wow, where to you people live? First there is an online barcode
> generator that I frequently use:

No HTML5 (JavaScript) generator with SVG output?

Best
   Martin

From john at wexfordpress.com  Wed Sep  8 21:05:49 2010
From: john at wexfordpress.com (John Culleton)
Date: Wed, 8 Sep 2010 17:05:49 -0400
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <AANLkTimb8ixLkNuun8WOGpabcoy25miBXKZj2HF4rjkr@mail.gmail.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<201009081645.57351.john@wexfordpress.com>
	<AANLkTimb8ixLkNuun8WOGpabcoy25miBXKZj2HF4rjkr@mail.gmail.com>
Message-ID: <201009081705.49768.john@wexfordpress.com>

On Wednesday 08 September 2010 16:49:34 Martin Schr?der wrote:
> 2010/9/8 John Culleton <john at wexfordpress.com>:
> > Wow, where to you people live? First there is an online barcode
> > generator that I frequently use:
>
> No HTML5 (JavaScript) generator with SVG output?
>
> Best
>    Martin
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel

Well which do you want? Inkscape certainly produces svg. You could 
build a barcode-containing svg file in Inkscape  and include it in your 
html5 document as an image.  The code for the bookland generator I 
mentioned above is written in python and can be downloaded. It is 
already integrated into a web page (see above) probably as a cgi-bin 
program.  In any case it gets the job done.

JPG is a bitmap format and is discouraged for barcodes. A vector 
format allows adjustment in size without degrading the quality. 

I am an applications guy. If I have a convenient way to generate 
barcode and it suits my purposes (book covers mostly) then I don't 
need to delve further.  I have already mentioned five paths, TeX, the 
Bookland online generator, the python code for that facility, Inkscape 
and Scribus.  Is there a need that they cannot fulfill?
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html

From martin at oneiros.de  Wed Sep  8 21:08:37 2010
From: martin at oneiros.de (=?ISO-8859-1?Q?Martin_Schr=F6der?=)
Date: Wed, 8 Sep 2010 23:08:37 +0200
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <201009081705.49768.john@wexfordpress.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<201009081645.57351.john@wexfordpress.com>
	<AANLkTimb8ixLkNuun8WOGpabcoy25miBXKZj2HF4rjkr@mail.gmail.com>
	<201009081705.49768.john@wexfordpress.com>
Message-ID: <AANLkTi=9gCwqf4hy+x8Y3W4bXCR7+1i7v+qXhVOsFspr@mail.gmail.com>

2010/9/8 John Culleton <john at wexfordpress.com>:
> On Wednesday 08 September 2010 16:49:34 Martin Schr?der wrote:
>> 2010/9/8 John Culleton <john at wexfordpress.com>:
>> > Wow, where to you people live? First there is an online barcode
>> > generator that I frequently use:
>>
>> No HTML5 (JavaScript) generator with SVG output?
>
> Well which do you want? Inkscape certainly produces svg. You could

JavaScript that generates canvas or SVG, i.e. a pure HTML5 solution.
It's certainly doable and probably already exists. :-)

Best
   Martin

PS: I know that this discussion has become completely OT. :-)

From brian at aljex.com  Wed Sep  8 21:32:14 2010
From: brian at aljex.com (Brian K. White)
Date: Wed, 08 Sep 2010 17:32:14 -0400
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <201009081705.49768.john@wexfordpress.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>	<201009081645.57351.john@wexfordpress.com>	<AANLkTimb8ixLkNuun8WOGpabcoy25miBXKZj2HF4rjkr@mail.gmail.com>
	<201009081705.49768.john@wexfordpress.com>
Message-ID: <4C8800DE.1040305@aljex.com>

On 9/8/2010 5:05 PM, John Culleton wrote:
> On Wednesday 08 September 2010 16:49:34 Martin Schr?der wrote:
>> 2010/9/8 John Culleton<john at wexfordpress.com>:
>>> Wow, where to you people live? First there is an online barcode
>>> generator that I frequently use:
>>
>> No HTML5 (JavaScript) generator with SVG output?
>>
>> Best
>>     Martin
>> _______________________________________________
>> gs-devel mailing list
>> gs-devel at ghostscript.com
>> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>
> Well which do you want? Inkscape certainly produces svg. You could
> build a barcode-containing svg file in Inkscape  and include it in your
> html5 document as an image.  The code for the bookland generator I
> mentioned above is written in python and can be downloaded. It is
> already integrated into a web page (see above) probably as a cgi-bin
> program.  In any case it gets the job done.
>
> JPG is a bitmap format and is discouraged for barcodes. A vector
> format allows adjustment in size without degrading the quality.
>
> I am an applications guy. If I have a convenient way to generate
> barcode and it suits my purposes (book covers mostly) then I don't
> need to delve further.  I have already mentioned five paths, TeX, the
> Bookland online generator, the python code for that facility, Inkscape
> and Scribus.  Is there a need that they cannot fulfill?


Did the OP say he wanted to generate a single barcode one time?, or to 
generate them manually?

On a list about a tool like gs, I assume all questions are about doing 
things non-interactively unless explicitly stated otherwise, since that 
is what GS itself does. Where do YOU live that you think some web page 
or gui desktop app is even the slightest bit useful for integrating 
barcodes (or searcheable text embedded in a pdf document which was the 
real question remember) in an aoutomated procedure or server print 
post-processor or even an embedded appliance? Some "applications guy". 
If I made my users go to a web page, or install some app like scribus 
which may or may not even exist for their OS, and type stuff in and 
download files to add a barcode to something my app generated, I 
wouldn't have very many users. ...as long as we're feeling free to toss 
insults around.

--
bkw

From john at wexfordpress.com  Thu Sep  9 17:53:56 2010
From: john at wexfordpress.com (John Culleton)
Date: Thu, 9 Sep 2010 13:53:56 -0400
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <4C8800DE.1040305@aljex.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<201009081705.49768.john@wexfordpress.com>
	<4C8800DE.1040305@aljex.com>
Message-ID: <201009091353.57021.john@wexfordpress.com>

On Wednesday 08 September 2010 17:32:14 Brian K. White wrote:
> On 9/8/2010 5:05 PM, John Culleton wrote:
> > On Wednesday 08 September 2010 16:49:34 Martin Schr?der wrote:
> >> 2010/9/8 John Culleton<john at wexfordpress.com>:
> >>> Wow, where to you people live? First there is an online barcode
> >>> generator that I frequently use:
> >>
> >> No HTML5 (JavaScript) generator with SVG output?
> >>
> >> Best
> >>     Martin
> >> _______________________________________________
> >> gs-devel mailing list
> >> gs-devel at ghostscript.com
> >> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
> >
> > Well which do you want? Inkscape certainly produces svg. You
> > could build a barcode-containing svg file in Inkscape  and
> > include it in your html5 document as an image.  The code for the
> > bookland generator I mentioned above is written in python and can
> > be downloaded. It is already integrated into a web page (see
> > above) probably as a cgi-bin program.  In any case it gets the
> > job done.
> >
> > JPG is a bitmap format and is discouraged for barcodes. A vector
> > format allows adjustment in size without degrading the quality.
> >
> > I am an applications guy. If I have a convenient way to generate
> > barcode and it suits my purposes (book covers mostly) then I
> > don't need to delve further.  I have already mentioned five
> > paths, TeX, the Bookland online generator, the python code for
> > that facility, Inkscape and Scribus.  Is there a need that they
> > cannot fulfill?
>
> Did the OP say he wanted to generate a single barcode one time?, or
> to generate them manually?
>
> On a list about a tool like gs, I assume all questions are about
> doing things non-interactively unless explicitly stated otherwise,
> since that is what GS itself does. Where do YOU live that you think
> some web page or gui desktop app is even the slightest bit useful
> for integrating barcodes (or searcheable text embedded in a pdf
> document which was the real question remember) in an aoutomated
> procedure or server print post-processor or even an embedded
> appliance? Some "applications guy". If I made my users go to a web
> page, or install some app like scribus which may or may not even
> exist for their OS, and type stuff in and download files to add a
> barcode to something my app generated, I wouldn't have very many
> users. ...as long as we're feeling free to toss insults around.
>
> --
> bkw
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel


GS is specific to GhostScript which is specific to PostScript and PDF. I 
don't think svg falls in that orbit.  For eps output of EAN 13 and EAN 
5 I again suggest the Bookland python program, which I have called 
from another program to produce 1,000 ISBN files in eps format.  It has 
already been adapted as a cgi-bin program by its author and you could 
adapt it further. 
http://www.tux.org/milgram-bin/isbn.py">

It would be possible to call pdftex from another program and feed it a 
file containing the requisite TeX code for creating a barcode. Again 
the output would be ps or pdf, not svg. 

If you must have svg output then I suggest Inkscape as the best place 
to look for that flavor of output.  Inkscape is an interactive program 
which runs on all three major platforms. Barcode formats are 
Datamatrix, EAN 5, EAN 8, EAN 13, UPC A, UPC E, Code 39, Code 39 ext, 
Code 93, Code 128 and so on. What code(s) were you interested in?

The source code for barcodes in Inkscape is probably borrowed from 
another Open Source program. I will inquire on inkscape-dev list. It 
won't be Javascript but many languages can be used in cgi-bin 
applications.






-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ghostscript.com/pipermail/gs-devel/attachments/20100909/58588ad1/attachment.html>

From john at wexfordpress.com  Thu Sep  9 19:26:09 2010
From: john at wexfordpress.com (John Culleton)
Date: Thu, 9 Sep 2010 15:26:09 -0400
Subject: [gs-devel] pdf/a searchable
In-Reply-To: <201009091353.57021.john@wexfordpress.com>
References: <5.1.0.14.2.20100908080204.042ac518@pop.gmail.com>
	<4C8800DE.1040305@aljex.com>
	<201009091353.57021.john@wexfordpress.com>
Message-ID: <201009091526.09967.john@wexfordpress.com>

On Thursday 09 September 2010 13:53:56 John Culleton wrote:

> The source code for barcodes in Inkscape is probably borrowed from
> another Open Source program. I will inquire on inkscape-dev list.
> It won't be Javascript but many languages can be used in cgi-bin
> applications.

On further exploration the python driver program is found here:
http://bazaar.launchpad.net/~inkscape.dev/inkscape/trunk/annotate/head%3A/share/extensions/render_barcode.py

bazaar is a replacment for svn.

Barcode.py calls some other modules, namely inkex and sys. Inkex is 
also included in the Inkscape source distribution and sys.py is a 
standard Python module. 
--
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html

From jmaki at multitech.com  Thu Sep  9 20:55:38 2010
From: jmaki at multitech.com (James Maki)
Date: Thu, 9 Sep 2010 15:55:38 -0500
Subject: [gs-devel] pdf/a searchable
Message-ID: <1CD05B980C94AC408B37C82BBABAA3E61407C2CC@mtsexchange.dc.multitech.prv>

Just do uboot-conf write device_type until it is gone

-----Original Message-----
From: gs-devel-bounces at ghostscript.com
[mailto:gs-devel-bounces at ghostscript.com] On Behalf Of John Culleton
Sent: Thursday, September 09, 2010 2:26 PM
To: gs-devel at ghostscript.com
Subject: Re: [gs-devel] pdf/a searchable


On Thursday 09 September 2010 13:53:56 John Culleton wrote:

> The source code for barcodes in Inkscape is probably borrowed from 
> another Open Source program. I will inquire on inkscape-dev list. It 
> won't be Javascript but many languages can be used in cgi-bin 
> applications.

On further exploration the python driver program is found here:
http://bazaar.launchpad.net/~inkscape.dev/inkscape/trunk/annotate/head%3
A/share/extensions/render_barcode.py

bazaar is a replacment for svn.

Barcode.py calls some other modules, namely inkex and sys. Inkex is 
also included in the Inkscape source distribution and sys.py is a 
standard Python module. 
--
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html
_______________________________________________
gs-devel mailing list
gs-devel at ghostscript.com
http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel

From jmaki at multitech.com  Fri Sep 10 13:45:41 2010
From: jmaki at multitech.com (James Maki)
Date: Fri, 10 Sep 2010 08:45:41 -0500
Subject: [gs-devel] pdf/a searchable
Message-ID: <1CD05B980C94AC408B37C82BBABAA3E61407C3CA@mtsexchange.dc.multitech.prv>

Ooops! Sorry I was in a hurry to reply to a bunch of emails and replied
to the wrong one. Ignore my last email.

-----Original Message-----
From: gs-devel-bounces at ghostscript.com
[mailto:gs-devel-bounces at ghostscript.com] On Behalf Of James Maki
Sent: Thursday, September 09, 2010 3:56 PM
To: John Culleton; gs-devel at ghostscript.com
Subject: Re: [gs-devel] pdf/a searchable


Just do uboot-conf write device_type until it is gone

-----Original Message-----
From: gs-devel-bounces at ghostscript.com
[mailto:gs-devel-bounces at ghostscript.com] On Behalf Of John Culleton
Sent: Thursday, September 09, 2010 2:26 PM
To: gs-devel at ghostscript.com
Subject: Re: [gs-devel] pdf/a searchable


On Thursday 09 September 2010 13:53:56 John Culleton wrote:

> The source code for barcodes in Inkscape is probably borrowed from
> another Open Source program. I will inquire on inkscape-dev list. It 
> won't be Javascript but many languages can be used in cgi-bin 
> applications.

On further exploration the python driver program is found here:
http://bazaar.launchpad.net/~inkscape.dev/inkscape/trunk/annotate/head%3
A/share/extensions/render_barcode.py

bazaar is a replacment for svn.

Barcode.py calls some other modules, namely inkex and sys. Inkex is 
also included in the Inkscape source distribution and sys.py is a 
standard Python module. 
--
-- 
John Culleton
Wexford Press
"Create Book Covers with Scribus"
Printable E-book 38 pages $5.95
http://www.booklocker.com/books/4055.html
_______________________________________________
gs-devel mailing list
gs-devel at ghostscript.com
http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
_______________________________________________
gs-devel mailing list
gs-devel at ghostscript.com
http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel

From h.goebel at goebel-consult.de  Sat Sep 11 22:12:46 2010
From: h.goebel at goebel-consult.de (Hartmut Goebel)
Date: Sun, 12 Sep 2010 00:12:46 +0200
Subject: [gs-devel] Can't make device bmp16 to use stdout set by set_stdio()
Message-ID: <4C8BFEDE.4020001@goebel-consult.de>

Hello,

not sure whether I've got the correct mailinglist.

I'm trying to use the Ghostscript API for generating images from
Postscript files. Since I'm going to post-process the images, I would
prefer using the API for a smaller memory footprint instead of spawning
a process and reading from a pipe.

The problem I'm facing is: I can not make device bmp16 to use stdout set
by set_stdio(). Below please find a mimimalisitc test program, based on
<http://pages.cs.wisc.edu/~ghost/doc/cvs/API.htm#set_stdio>. The program
collects Ghostscripts stdout into a buffer and print this buffer after
processing finished.

The effect I'm observing is: The Ghostcript banner gets collected, but
the image gets written to the original stdout.

What's going wrong here?

import StringIO
import sys
import ghostscript._gsprint as gs

postscript_doc = "0 0 moveto (Hello World) show showpage"

args = """test.py -dNOPAUSE -dBATCH -dSAFER
       -sDEVICE=bmp16 -g10x10 -sOutputFile=-
       """.split()

instance = gs.new_instance()

stdout = StringIO.StringIO() # create buffer for collecting output

# wrappers like in http://pages.cs.wisc.edu/~ghost/doc/cvs/API.htm#set_stdio
gsdll_stdin  = gs._wrap_stdin(sys.stdin)
gsdll_stdout = gs._wrap_stdout(stdout)
gsdll_stderr = gs._wrap_stderr(sys.stderr)

gs.set_stdio(instance, gsdll_stdin, gsdll_stdout, gsdll_stderr)

gs.init_with_args(instance, args)
gs.run_string(instance, postscript_doc)
gs.delete_instance(instance)

# print data collected from ghostscripts stdout
print 'Read data', repr(stdout.getvalue())


-- 
Sch?nen Gru? - Regards
Hartmut Goebel
Dipl.-Informatiker (univ.), CISSP, CSSLP

Goebel Consult
Spezialist f?r IT-Sicherheit in komplexen Umgebungen
http://www.goebel-consult.de

Monatliche Kolumne: http://www.cissp-gefluester.de/
Goebel Consult mit Mitglied bei http://www.7-it.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5220 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://ghostscript.com/pipermail/gs-devel/attachments/20100912/b2890ab6/attachment.bin>

From cjl at spamcop.net  Wed Sep 15 07:30:50 2010
From: cjl at spamcop.net (cjl at spamcop.net)
Date: Wed, 15 Sep 2010 08:30:50 +0100
Subject: [gs-devel] Ghostscript 9.00 relese candidate 3
Message-ID: <20100915083050.v5xa5vui1w40c8ks-pwy@webmail.spamcop.net>

Hi Folks,

A serious problem was reported against release candidate 2, so we've  
got another one - again, depending on your favoured compression:

http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.gz
http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.xz

please give it a try on your system if you're interested in helping test the
release-in-progress.

Regards,

Chris

From beebe at math.utah.edu  Wed Sep 15 15:29:42 2010
From: beebe at math.utah.edu (Nelson H. F. Beebe)
Date: Wed, 15 Sep 2010 09:29:42 -0600 (MDT)
Subject: [gs-devel]  remarks on ghostscript source code distributions
Message-ID: <CMM.0.95.0.1284564582.beebe@psi.math.utah.edu>

Thanks for this news today:

	http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.gz

However, can you folks please also supply companion digital signature
files?

Here is why package signatures are important:

    Several years ago, the Free Software Foundation's software archive
    site at ftp://ftp.gnu.org was broken into, and many packages were
    replaced by trojaned versions.  This was not discovered for
    several weeks, and when it was, the site was shut down for a
    couple of months while they recovered original presumed-safe
    versions, and then attached digital signatures to all of them.
    Many packages disappeared entirely, because it could not be
    determined whether they were trojaned or not.

I routinely digitally sign files that I post on our Web site, and in
addition, I supply MD5, SHA-1, and RipeMD-160 checksums, and digitally
sign the file that lists those checksums.  Examples can be found here:

	http://ftp.math.utah.edu/pub/bibsql/

Several years ago, I wrote up notes on the subject of digital
signatures and posted them at

	http://www.math.utah.edu/~beebe/PGP-notes.html

There is also a discussion of the topic in our book, Classic Shell
Scripting, section 10.7.4, pp 304ff.

-------------------------------------------------------------------------------
- Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
- University of Utah                    FAX: +1 801 581 4148                  -
- Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
- 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
- Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
-------------------------------------------------------------------------------

From cjl at spamcop.net  Wed Sep 15 15:33:50 2010
From: cjl at spamcop.net (Chris Liddell)
Date: Wed, 15 Sep 2010 16:33:50 +0100
Subject: [gs-devel] remarks on ghostscript source code distributions
In-Reply-To: <CMM.0.95.0.1284564582.beebe@psi.math.utah.edu>
References: <CMM.0.95.0.1284564582.beebe@psi.math.utah.edu>
Message-ID: <4C90E75E.4030408@spamcop.net>


We supply, IIRC, MD5 and SHA-1 checksums for releases, but haven't done
so previously for release candidates, as release candidates tend to be
very transient.

If you feel it's that important for release candidates, I'll consider
doing so in the future.

Chris


On 15/09/10 16:29, Nelson H. F. Beebe wrote:
> Thanks for this news today:
> 
> 	http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.gz
> 
> However, can you folks please also supply companion digital signature
> files?
> 
> Here is why package signatures are important:
> 
>     Several years ago, the Free Software Foundation's software archive
>     site at ftp://ftp.gnu.org was broken into, and many packages were
>     replaced by trojaned versions.  This was not discovered for
>     several weeks, and when it was, the site was shut down for a
>     couple of months while they recovered original presumed-safe
>     versions, and then attached digital signatures to all of them.
>     Many packages disappeared entirely, because it could not be
>     determined whether they were trojaned or not.
> 
> I routinely digitally sign files that I post on our Web site, and in
> addition, I supply MD5, SHA-1, and RipeMD-160 checksums, and digitally
> sign the file that lists those checksums.  Examples can be found here:
> 
> 	http://ftp.math.utah.edu/pub/bibsql/
> 
> Several years ago, I wrote up notes on the subject of digital
> signatures and posted them at
> 
> 	http://www.math.utah.edu/~beebe/PGP-notes.html
> 
> There is also a discussion of the topic in our book, Classic Shell
> Scripting, section 10.7.4, pp 304ff.
> 
> -------------------------------------------------------------------------------
> - Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
> - University of Utah                    FAX: +1 801 581 4148                  -
> - Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
> - 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
> - Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
> -------------------------------------------------------------------------------
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel

From mail_ben_schmidt at yahoo.com.au  Wed Sep 15 15:42:05 2010
From: mail_ben_schmidt at yahoo.com.au (Ben Schmidt)
Date: Thu, 16 Sep 2010 01:42:05 +1000
Subject: [gs-devel] remarks on ghostscript source code distributions
In-Reply-To: <4C90E75E.4030408@spamcop.net>
References: <CMM.0.95.0.1284564582.beebe@psi.math.utah.edu>
	<4C90E75E.4030408@spamcop.net>
Message-ID: <4C90E94D.1050803@yahoo.com.au>

The MD5/SHA1 hashes, of course, don't solve the problem Nelson referred to. They 
only ensure a successful download/data integrity, not authenticity. A digital 
signature is required to give assurance of authenticity.

Ben.



On 16/09/10 1:33 AM, Chris Liddell wrote:
>
> We supply, IIRC, MD5 and SHA-1 checksums for releases, but haven't done
> so previously for release candidates, as release candidates tend to be
> very transient.
>
> If you feel it's that important for release candidates, I'll consider
> doing so in the future.
>
> Chris
>
>
> On 15/09/10 16:29, Nelson H. F. Beebe wrote:
>> Thanks for this news today:
>>
>> 	http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.gz
>>
>> However, can you folks please also supply companion digital signature
>> files?
>>
>> Here is why package signatures are important:
>>
>>      Several years ago, the Free Software Foundation's software archive
>>      site at ftp://ftp.gnu.org was broken into, and many packages were
>>      replaced by trojaned versions.  This was not discovered for
>>      several weeks, and when it was, the site was shut down for a
>>      couple of months while they recovered original presumed-safe
>>      versions, and then attached digital signatures to all of them.
>>      Many packages disappeared entirely, because it could not be
>>      determined whether they were trojaned or not.
>>
>> I routinely digitally sign files that I post on our Web site, and in
>> addition, I supply MD5, SHA-1, and RipeMD-160 checksums, and digitally
>> sign the file that lists those checksums.  Examples can be found here:
>>
>> 	http://ftp.math.utah.edu/pub/bibsql/
>>
>> Several years ago, I wrote up notes on the subject of digital
>> signatures and posted them at
>>
>> 	http://www.math.utah.edu/~beebe/PGP-notes.html
>>
>> There is also a discussion of the topic in our book, Classic Shell
>> Scripting, section 10.7.4, pp 304ff.
>>
>> -------------------------------------------------------------------------------
>> - Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
>> - University of Utah                    FAX: +1 801 581 4148                  -
>> - Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
>> - 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
>> - Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
>> -------------------------------------------------------------------------------
>> _______________________________________________
>> gs-devel mailing list
>> gs-devel at ghostscript.com
>> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>


From cjl at spamcop.net  Wed Sep 15 15:49:08 2010
From: cjl at spamcop.net (Chris Liddell)
Date: Wed, 15 Sep 2010 16:49:08 +0100
Subject: [gs-devel] remarks on ghostscript source code distributions
In-Reply-To: <4C90E94D.1050803@yahoo.com.au>
References: <CMM.0.95.0.1284564582.beebe@psi.math.utah.edu>	<4C90E75E.4030408@spamcop.net>
	<4C90E94D.1050803@yahoo.com.au>
Message-ID: <4C90EAF4.3070700@spamcop.net>


I'll raise the question of digitally signing the archives at the next
Artifex staff meeting.

Chris


On 15/09/10 16:42, Ben Schmidt wrote:
> The MD5/SHA1 hashes, of course, don't solve the problem Nelson referred
> to. They only ensure a successful download/data integrity, not
> authenticity. A digital signature is required to give assurance of
> authenticity.
> 
> Ben.
> 
> 
> 
> On 16/09/10 1:33 AM, Chris Liddell wrote:
>>
>> We supply, IIRC, MD5 and SHA-1 checksums for releases, but haven't done
>> so previously for release candidates, as release candidates tend to be
>> very transient.
>>
>> If you feel it's that important for release candidates, I'll consider
>> doing so in the future.
>>
>> Chris
>>
>>
>> On 15/09/10 16:29, Nelson H. F. Beebe wrote:
>>> Thanks for this news today:
>>>
>>>     http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.gz
>>>
>>> However, can you folks please also supply companion digital signature
>>> files?
>>>
>>> Here is why package signatures are important:
>>>
>>>      Several years ago, the Free Software Foundation's software archive
>>>      site at ftp://ftp.gnu.org was broken into, and many packages were
>>>      replaced by trojaned versions.  This was not discovered for
>>>      several weeks, and when it was, the site was shut down for a
>>>      couple of months while they recovered original presumed-safe
>>>      versions, and then attached digital signatures to all of them.
>>>      Many packages disappeared entirely, because it could not be
>>>      determined whether they were trojaned or not.
>>>
>>> I routinely digitally sign files that I post on our Web site, and in
>>> addition, I supply MD5, SHA-1, and RipeMD-160 checksums, and digitally
>>> sign the file that lists those checksums.  Examples can be found here:
>>>
>>>     http://ftp.math.utah.edu/pub/bibsql/
>>>
>>> Several years ago, I wrote up notes on the subject of digital
>>> signatures and posted them at
>>>
>>>     http://www.math.utah.edu/~beebe/PGP-notes.html
>>>
>>> There is also a discussion of the topic in our book, Classic Shell
>>> Scripting, section 10.7.4, pp 304ff.
>>>
>>> -------------------------------------------------------------------------------
>>>
>>> - Nelson H. F. Beebe                    Tel: +1 801 581
>>> 5254                  -
>>> - University of Utah                    FAX: +1 801 581
>>> 4148                  -
>>> - Department of Mathematics, 110 LCB    Internet e-mail:
>>> beebe at math.utah.edu  -
>>> - 155 S 1400 E RM 233                       beebe at acm.org 
>>> beebe at computer.org -
>>> - Salt Lake City, UT 84112-0090, USA    URL:
>>> http://www.math.utah.edu/~beebe/ -
>>> -------------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> gs-devel mailing list
>>> gs-devel at ghostscript.com
>>> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>> _______________________________________________
>> gs-devel mailing list
>> gs-devel at ghostscript.com
>> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>>
> 
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel

From beebe at math.utah.edu  Wed Sep 15 16:13:20 2010
From: beebe at math.utah.edu (Nelson H. F. Beebe)
Date: Wed, 15 Sep 2010 10:13:20 -0600 (MDT)
Subject: [gs-devel] remarks on ghostscript source code distributions
In-Reply-To: <4C90E94D.1050803@yahoo.com.au>
Message-ID: <CMM.0.95.0.1284567200.beebe@psi.math.utah.edu>

>> The MD5/SHA1 hashes, of course, don't solve the problem..

That is why I digitally sign the file that contains them.  When there
are large numbers of files in a directory, it is easy to make one
signed file of checksums.  For single package files, a signature
(either .sig or .asc) for that package generally makes sense, and
is the procedure adopted by the Free Software Foundation, Apache,
and many other software sites.

-------------------------------------------------------------------------------
- Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
- University of Utah                    FAX: +1 801 581 4148                  -
- Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
- 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
- Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
-------------------------------------------------------------------------------

From abdulrazaqali at hotmail.com  Wed Sep 15 16:27:25 2010
From: abdulrazaqali at hotmail.com (Ali)
Date: Thu, 16 Sep 2010 00:27:25 +0800
Subject: [gs-devel] remarks on ghostscript source code distributions
In-Reply-To: <4C90E94D.1050803@yahoo.com.au>
References: <CMM.0.95.0.1284564582.beebe@psi.math.utah.edu>
	<4C90E75E.4030408@spamcop.net> <4C90E94D.1050803@yahoo.com.au>
Message-ID: <AANLkTi=RCtSk4+SOyBKryCHdqFDBK6gHDOE3ab70-Gy_@mail.gmail.com>

> A digital signature is required to give assurance of authenticity.

Which means 3-rd party goto be there.

ali

On Wed, Sep 15, 2010 at 11:42 PM, Ben Schmidt
<mail_ben_schmidt at yahoo.com.au> wrote:
> The MD5/SHA1 hashes, of course, don't solve the problem Nelson referred to.
> They only ensure a successful download/data integrity, not authenticity. A
> digital signature is required to give assurance of authenticity.
>
> Ben.
>
>
>
> On 16/09/10 1:33 AM, Chris Liddell wrote:
>>
>> We supply, IIRC, MD5 and SHA-1 checksums for releases, but haven't done
>> so previously for release candidates, as release candidates tend to be
>> very transient.
>>
>> If you feel it's that important for release candidates, I'll consider
>> doing so in the future.
>>
>> Chris
>>
>>
>> On 15/09/10 16:29, Nelson H. F. Beebe wrote:
>>>
>>> Thanks for this news today:
>>>
>>> ? ? ? ?http://www.ghostscript.com/~chrisl/ghostscript-9.00rc3.tar.gz
>>>
>>> However, can you folks please also supply companion digital signature
>>> files?
>>>
>>> Here is why package signatures are important:
>>>
>>> ? ? Several years ago, the Free Software Foundation's software archive
>>> ? ? site at ftp://ftp.gnu.org was broken into, and many packages were
>>> ? ? replaced by trojaned versions. ?This was not discovered for
>>> ? ? several weeks, and when it was, the site was shut down for a
>>> ? ? couple of months while they recovered original presumed-safe
>>> ? ? versions, and then attached digital signatures to all of them.
>>> ? ? Many packages disappeared entirely, because it could not be
>>> ? ? determined whether they were trojaned or not.
>>>
>>> I routinely digitally sign files that I post on our Web site, and in
>>> addition, I supply MD5, SHA-1, and RipeMD-160 checksums, and digitally
>>> sign the file that lists those checksums. ?Examples can be found here:
>>>
>>> ? ? ? ?http://ftp.math.utah.edu/pub/bibsql/
>>>
>>> Several years ago, I wrote up notes on the subject of digital
>>> signatures and posted them at
>>>
>>> ? ? ? ?http://www.math.utah.edu/~beebe/PGP-notes.html
>>>
>>> There is also a discussion of the topic in our book, Classic Shell
>>> Scripting, section 10.7.4, pp 304ff.
>>>
>>>
>>> -------------------------------------------------------------------------------
>>> - Nelson H. F. Beebe ? ? ? ? ? ? ? ? ? ?Tel: +1 801 581 5254
>>> ? ? ?-
>>> - University of Utah ? ? ? ? ? ? ? ? ? ?FAX: +1 801 581 4148
>>> ? ? ?-
>>> - Department of Mathematics, 110 LCB ? ?Internet e-mail:
>>> beebe at math.utah.edu ?-
>>> - 155 S 1400 E RM 233 ? ? ? ? ? ? ? ? ? ? ? beebe at acm.org
>>> ?beebe at computer.org -
>>> - Salt Lake City, UT 84112-0090, USA ? ?URL:
>>> http://www.math.utah.edu/~beebe/ -
>>>
>>> -------------------------------------------------------------------------------
>>> _______________________________________________
>>> gs-devel mailing list
>>> gs-devel at ghostscript.com
>>> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>>
>> _______________________________________________
>> gs-devel mailing list
>> gs-devel at ghostscript.com
>> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>>
>
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel
>
>

From beebe at math.utah.edu  Thu Sep 16 15:07:47 2010
From: beebe at math.utah.edu (Nelson H. F. Beebe)
Date: Thu, 16 Sep 2010 09:07:47 -0600 (MDT)
Subject: [gs-devel]  remarks on ghostscript-9.00rc3
Message-ID: <CMM.0.95.0.1284649667.beebe@psi.math.utah.edu>

I now have ghostscript-9.00rc3 built, validated, and installed on at
least 19 different Unix flavors in my test lab.  However, one only 3
or 4 does the usual recipe "./configure && make all check" work
directly.  Here is a summary of the problems that I found, and for
most, their solutions.

------------------------------------------------------------------------

I was finally able to get a successful build on MirBSD 10 IA-32, but
not without code modifications.  There are two problems with
autoconfiguration on this system.  The first is that the config.guess
script from even the latest autoconf (version 2.67) does not recognize
this O/S at all.  MirBSD is an offshoot of OpenBSD, so for most
packages, it is possible to just do

	env CC=mgcc ./configure --build i386-unknown-openbsd

Unfortunately, that fails with ghostscript, because of a second
problem: ghostscript has additional configure scripts below the top
level directory, and its top-level configure script fails to pass the
--build and --host settings down to child configure scripts.  That
can, and should, be fixed.

The solution proved to be to edit a copy of the most recent
config.guess file and add a code block for this O/S:

    % diff ~/bin/config.guess~ ~/bin/config.guess
    207a208,211
    >     *:MirBSD:*:*)
    >       UNAME_MACHINE_ARCH=`arch | sed 's/MirBSD.//'`
    >       echo ${UNAME_MACHINE_ARCH}-unknown-openbsd${UNAME_RELEASE}
    >       exit ;;

I then copied the revised config.guess file, and its companion
config.sub, to each of the directories in the ghostscript tree where
those files existed previously.  I then got a successful build,
validation, and installation of ghostscript-9.00 on MirBSD.

------------------------------------------------------------------------

On several systems, initial builds fail because the obj and bin
subdirectories are still not being created: that problem has existed
for far too many releases of ghostscript.  Once I did so manually, I
could restart the makes, and get the builds to complete.  Please, if
you don't want to change the top-level Makefile to create those two
essential subdirectories, then just include empty ones in the
ghostscript-x.y.z.tar.gz files, so that unpacking the distribution
creates them automatically.

------------------------------------------------------------------------

On Gentoo SPARC, ghostscript builds successfully, but so far, all of
the 9.00-rc* candidates get an immediate segment violation on startup,
and I have not been able to figure out why.  Rebuilds with
gcc-4-{1,2,3,5,6} produce executables that exhibit the same run-time
failure.

------------------------------------------------------------------------

On Solaris SPARC, with the solstudio-12.1 compilers installed here on
19-Aug-2010, a build with CFLAGS="-O -g" is successful, but execution
fails like this

    % dbx ./gs
    (dbx) run ../examples/tiger.eps
    Running: gs ../examples/tiger.eps
    (process id 3646)
    Reading libc_psr.so.1
    GPL Ghostscript RELEASE CANDIDATE 9.00 (2010-09-14)
    Copyright (C) 2010 Artifex Software, Inc.  All rights reserved.
    This software comes with NO WARRANTY: see the file PUBLIC for details.
    Unrecoverable error: limitcheck in token
    Operand stack:
	--nostringval--  --nostringval--  Fontmap.GS
    t at 1 (l at 1) signal SEGV (no mapping at the fault address) in interp at 0xebeb0
    0x000ebeb0: interp+0x0184:      lduh     [%i1], %o0
    (dbx) where
    current thread: t at 1
    =>[1] interp(0x60, 0x0, 0xff4a48, 0x3c00, 0x4fb, 0x1001a6c), at 0xebeb0
      [2] gs_call_interp(0xfd9bac, 0x0, 0x0, 0xffbfe5ac, 0xffbfe5a4, 0x3f80), at 0xeaff8
      [3] gs_interpret(0xfd9bac, 0x6a5400, 0x0, 0xffbfe5ac, 0xffbfe5a4, 0x1001948), at 0xeaef8
      [4] gs_main_run_string_begin(0xfd9b58, 0x0, 0xffbfe5ac, 0xffbfe5a4, 0x1000, 0x6a3c00), at 0xe2934
      [5] gs_main_run_string_with_length(0xfd9b58, 0x6a3ea8, 0x5e, 0x0, 0xffbfe5ac, 0xffbfe5a4), at 0xe28a0
      [6] gs_main_finit(0xfd9b58, 0xff, 0xfffffff3, 0x10067c0, 0x6d6a60, 0x0), at 0xe2fb4
      [7] main(0xff, 0xffbfe674, 0xffbfe680, 0xfd9b58, 0xfd99e0, 0x69d400), at 0x5804c

On a hunch, I did "make clean" and then rebuilt with CFLAGS=-g; that
version runs correctly, and has been installed.

Last weekend, I installed a test release of the newer solstudio-12.2
compilers on Solaris (SPARC, IA-32, and AMD64) and GNU/Linux (IA-32
and AMD64), and I plan to retry the Solaris SPARC build with higher
optimization levels.  I may report later on that experiment.

------------------------------------------------------------------------

On several systems, I still have to augment the EXTRALIBS list with
additional flags and libraries, e.g., on OpenBSD:

    % grep ^EXTRALIBS Makefile
    EXTRALIBS=-lm   -L/usr/X11R6/lib -lfontconfig -lexpat -lfreetype -lz

    % make EXTRALIBS='-lm -L/usr/X11R6/lib -lfontconfig -l expat -lfreetype \
          -lz -L/usr/local/lib -R/usr/local/lib -ljpeg

This happens despite building with PATH=/bin:/usr/bin; apparently, the
existence of libraries in the /usr/local tree causes ghostscript to
try to use those instead of its own jasper and jpeg code directories.
On many of our systems, it is impossible to temporarily rename the
/usr/local tree to hide it from the ghostscript build process, because
that action prevents user logins and may cause other running programs
to fail.

------------------------------------------------------------------------

On OpenBSD, I had to add obj/gsparamx.o to obj/ld.tr, and compile its
source file manually; I don't know why gsparamx.c is not included in
the list of files to be compiled.  I did not see that problem on any
other platform, or in builds of any earlier release of
ghostscript-{8,9}* on OpenBSD.

------------------------------------------------------------------------

Finally, it should be documented in doc/Install.htm that GNU make is
required.  On {Free,Mir,Net,Open}BSD, the default Berkeley
/usr/bin/make cannot correctly handle the ghostscript Makefiles, or
builds where object files are placed in a directory other than that of
the source files.

------------------------------------------------------------------------

P.S. As always, I can provide detailed build logs privately to any
in-house ghostscript developer to help debug the issues that I have
raised in this report.

-------------------------------------------------------------------------------
- Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
- University of Utah                    FAX: +1 801 581 4148                  -
- Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
- 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
- Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
-------------------------------------------------------------------------------

From cjl at spamcop.net  Thu Sep 16 15:52:44 2010
From: cjl at spamcop.net (Chris Liddell)
Date: Thu, 16 Sep 2010 16:52:44 +0100
Subject: [gs-devel] remarks on ghostscript-9.00rc3
In-Reply-To: <CMM.0.95.0.1284649667.beebe@psi.math.utah.edu>
References: <CMM.0.95.0.1284649667.beebe@psi.math.utah.edu>
Message-ID: <4C923D4C.3090901@spamcop.net>


Thanks. I'll add the MirBSD issue to the other ones to look at.

I believe I have all the relevant logs you made available to me
previously, but if there's new stuff from this round, then I would
appreciate seeing it.

Thanks again,

Chris


On 16/09/10 16:07, Nelson H. F. Beebe wrote:
> I now have ghostscript-9.00rc3 built, validated, and installed on at
> least 19 different Unix flavors in my test lab.  However, one only 3
> or 4 does the usual recipe "./configure && make all check" work
> directly.  Here is a summary of the problems that I found, and for
> most, their solutions.
> 
> ------------------------------------------------------------------------
> 
> I was finally able to get a successful build on MirBSD 10 IA-32, but
> not without code modifications.  There are two problems with
> autoconfiguration on this system.  The first is that the config.guess
> script from even the latest autoconf (version 2.67) does not recognize
> this O/S at all.  MirBSD is an offshoot of OpenBSD, so for most
> packages, it is possible to just do
> 
> 	env CC=mgcc ./configure --build i386-unknown-openbsd
> 
> Unfortunately, that fails with ghostscript, because of a second
> problem: ghostscript has additional configure scripts below the top
> level directory, and its top-level configure script fails to pass the
> --build and --host settings down to child configure scripts.  That
> can, and should, be fixed.
> 
> The solution proved to be to edit a copy of the most recent
> config.guess file and add a code block for this O/S:
> 
>     % diff ~/bin/config.guess~ ~/bin/config.guess
>     207a208,211
>     >     *:MirBSD:*:*)
>     >       UNAME_MACHINE_ARCH=`arch | sed 's/MirBSD.//'`
>     >       echo ${UNAME_MACHINE_ARCH}-unknown-openbsd${UNAME_RELEASE}
>     >       exit ;;
> 
> I then copied the revised config.guess file, and its companion
> config.sub, to each of the directories in the ghostscript tree where
> those files existed previously.  I then got a successful build,
> validation, and installation of ghostscript-9.00 on MirBSD.
> 
> ------------------------------------------------------------------------
> 
> On several systems, initial builds fail because the obj and bin
> subdirectories are still not being created: that problem has existed
> for far too many releases of ghostscript.  Once I did so manually, I
> could restart the makes, and get the builds to complete.  Please, if
> you don't want to change the top-level Makefile to create those two
> essential subdirectories, then just include empty ones in the
> ghostscript-x.y.z.tar.gz files, so that unpacking the distribution
> creates them automatically.
> 
> ------------------------------------------------------------------------
> 
> On Gentoo SPARC, ghostscript builds successfully, but so far, all of
> the 9.00-rc* candidates get an immediate segment violation on startup,
> and I have not been able to figure out why.  Rebuilds with
> gcc-4-{1,2,3,5,6} produce executables that exhibit the same run-time
> failure.
> 
> ------------------------------------------------------------------------
> 
> On Solaris SPARC, with the solstudio-12.1 compilers installed here on
> 19-Aug-2010, a build with CFLAGS="-O -g" is successful, but execution
> fails like this
> 
>     % dbx ./gs
>     (dbx) run ../examples/tiger.eps
>     Running: gs ../examples/tiger.eps
>     (process id 3646)
>     Reading libc_psr.so.1
>     GPL Ghostscript RELEASE CANDIDATE 9.00 (2010-09-14)
>     Copyright (C) 2010 Artifex Software, Inc.  All rights reserved.
>     This software comes with NO WARRANTY: see the file PUBLIC for details.
>     Unrecoverable error: limitcheck in token
>     Operand stack:
> 	--nostringval--  --nostringval--  Fontmap.GS
>     t at 1 (l at 1) signal SEGV (no mapping at the fault address) in interp at 0xebeb0
>     0x000ebeb0: interp+0x0184:      lduh     [%i1], %o0
>     (dbx) where
>     current thread: t at 1
>     =>[1] interp(0x60, 0x0, 0xff4a48, 0x3c00, 0x4fb, 0x1001a6c), at 0xebeb0
>       [2] gs_call_interp(0xfd9bac, 0x0, 0x0, 0xffbfe5ac, 0xffbfe5a4, 0x3f80), at 0xeaff8
>       [3] gs_interpret(0xfd9bac, 0x6a5400, 0x0, 0xffbfe5ac, 0xffbfe5a4, 0x1001948), at 0xeaef8
>       [4] gs_main_run_string_begin(0xfd9b58, 0x0, 0xffbfe5ac, 0xffbfe5a4, 0x1000, 0x6a3c00), at 0xe2934
>       [5] gs_main_run_string_with_length(0xfd9b58, 0x6a3ea8, 0x5e, 0x0, 0xffbfe5ac, 0xffbfe5a4), at 0xe28a0
>       [6] gs_main_finit(0xfd9b58, 0xff, 0xfffffff3, 0x10067c0, 0x6d6a60, 0x0), at 0xe2fb4
>       [7] main(0xff, 0xffbfe674, 0xffbfe680, 0xfd9b58, 0xfd99e0, 0x69d400), at 0x5804c
> 
> On a hunch, I did "make clean" and then rebuilt with CFLAGS=-g; that
> version runs correctly, and has been installed.
> 
> Last weekend, I installed a test release of the newer solstudio-12.2
> compilers on Solaris (SPARC, IA-32, and AMD64) and GNU/Linux (IA-32
> and AMD64), and I plan to retry the Solaris SPARC build with higher
> optimization levels.  I may report later on that experiment.
> 
> ------------------------------------------------------------------------
> 
> On several systems, I still have to augment the EXTRALIBS list with
> additional flags and libraries, e.g., on OpenBSD:
> 
>     % grep ^EXTRALIBS Makefile
>     EXTRALIBS=-lm   -L/usr/X11R6/lib -lfontconfig -lexpat -lfreetype -lz
> 
>     % make EXTRALIBS='-lm -L/usr/X11R6/lib -lfontconfig -l expat -lfreetype \
>           -lz -L/usr/local/lib -R/usr/local/lib -ljpeg
> 
> This happens despite building with PATH=/bin:/usr/bin; apparently, the
> existence of libraries in the /usr/local tree causes ghostscript to
> try to use those instead of its own jasper and jpeg code directories.
> On many of our systems, it is impossible to temporarily rename the
> /usr/local tree to hide it from the ghostscript build process, because
> that action prevents user logins and may cause other running programs
> to fail.
> 
> ------------------------------------------------------------------------
> 
> On OpenBSD, I had to add obj/gsparamx.o to obj/ld.tr, and compile its
> source file manually; I don't know why gsparamx.c is not included in
> the list of files to be compiled.  I did not see that problem on any
> other platform, or in builds of any earlier release of
> ghostscript-{8,9}* on OpenBSD.
> 
> ------------------------------------------------------------------------
> 
> Finally, it should be documented in doc/Install.htm that GNU make is
> required.  On {Free,Mir,Net,Open}BSD, the default Berkeley
> /usr/bin/make cannot correctly handle the ghostscript Makefiles, or
> builds where object files are placed in a directory other than that of
> the source files.
> 
> ------------------------------------------------------------------------
> 
> P.S. As always, I can provide detailed build logs privately to any
> in-house ghostscript developer to help debug the issues that I have
> raised in this report.
> 
> -------------------------------------------------------------------------------
> - Nelson H. F. Beebe                    Tel: +1 801 581 5254                  -
> - University of Utah                    FAX: +1 801 581 4148                  -
> - Department of Mathematics, 110 LCB    Internet e-mail: beebe at math.utah.edu  -
> - 155 S 1400 E RM 233                       beebe at acm.org  beebe at computer.org -
> - Salt Lake City, UT 84112-0090, USA    URL: http://www.math.utah.edu/~beebe/ -
> -------------------------------------------------------------------------------
> _______________________________________________
> gs-devel mailing list
> gs-devel at ghostscript.com
> http://ghostscript.com/cgi-bin/mailman/listinfo/gs-devel

From chris.liddell at artifex.com  Fri Sep 17 22:10:10 2010
From: chris.liddell at artifex.com (Chris Liddell)
Date: Fri, 17 Sep 2010 23:10:10 +0100
Subject: [gs-devel] Ghostscript and GhostPDL 9.00 released
Message-ID: <4C93E742.2010402@artifex.com>

Artifex Software, Inc. is happy to announce the release of GPL
Ghostscript 9.00 and GhostPDL 9.00.

This is the second in this series of stable Ghostscript and GhostPDL
releases.

Source packages are available from:

* http://ghostscript.com/releases/ghostscript-9.00.tar.xz
* http://ghostscript.com/releases/ghostscript-9.00.tar.gz
* http://ghostscript.com/releases/ghostpdl-9.00.tar.bz2

And from the following mirrors:

* https://sourceforge.net/projects/ghostscript/files/GPL%20Ghostscript/9.00


The following mirrors will be updated in due course:

* http://mirror.cs.wisc.edu/pub/mirrors/ghost/GPL/gs900/
* http://mirror.cs.wisc.edu/pub/mirrors/ghost/GPL/ghostpdl/
* http://code.google.com/p/ghostscript/downloads/list



SHA-1 checksums
* 0ef940b65fba776e7b00a23bd21b0483a45db31e  ghostscript-9.00.tar.gz

MD5 checksums

* 43d71a8e4d78ac99da10d3fb60f8fdbb  ghostscript-9.00.tar.gz

This is the first release in the stable 9.x series.

This release includes a move to an ICC-based color rendering workflow.
The design allows easy integration of 3rd party color management modules
(CMMs) and management of DeviceN and spot source colors with ICC
profiles as well as with non-ICC proprietary methods. The default CMM is
the well known littleCMS. Postscript color objects and non-ICC CIE-based
PDF color spaces are converted to equivalent ICC profiles enabling
complete color management for all color spaces by an ICC-based CMM. New
command line options enable the specification of gray, rgb and cmyk
default ICC profiles as well as output device ICC profiles. The new work
flow provides performance improvements in the rendering of images,
shadings and transparencies. In addition, the color conversions are
designed to work efficiently in multithreaded display list (c-list)
rendering through the use of a shared link cache. Finally, proper ICC
based rendering now occurs for ALL XPS objects including Named colors,
N-Channel colors and images with internally embedded ICC profiles.

The second major change in this release is that glyph rendering for all
font types except Type 3 is now handled by Freetype (via a much improved
and enhanced Font API). The most obvious benefit of this is a complete,
and proven implementation of the TrueType byte code interpreter (now
unencumbered by patents).

Another major change is that the Postscript interpreter's graphics state
now stores two separate colour space settings for stroking and filling,
to better support the stroke and fill colours required by the PDF
interpreter.

Finally, support for optional content has been added to the PDF interpreter.

This release also includes a number of improvements to the pdfwrite
device to improve reliability and quality with PCL and XPS input. Many
improvements to the support of JPX images in PDF files, PDF annotation
rendering and the usual number of general bug and robustness fixes.

For security reasons, Ghostscript no longer searches the current
directory by default. Use -P option to revert to old behavior.


From hswan at conocophillips.com  Thu Sep 30 18:53:41 2010
From: hswan at conocophillips.com (hswan)
Date: Thu, 30 Sep 2010 11:53:41 -0700 (PDT)
Subject: [gs-devel]  Trouble compiling Ghostscript 9.00
Message-ID: <29851272.post@talk.nabble.com>


I tried compiling Ghostscript 9.00 right out of the box with

./configure
make


Things worked OK until I got to the following:


gcc  -DHAVE_MKSTEMP -DHAVE_HYPOT -DHAVE_FILE64 -DHAVE_MKSTEMP64
-DHAVE_FONTCONFIG -DHAVE_LIBIDN -DHAVE_SETLOCALE -O2 -Wall -Wstrict-pro
totypes -Wundef -Wmissing-declarations -Wmissing-prototypes -Wwrite-strings
-Wno-strict-aliasing -Wdeclaration-after-statement -fno-bui ltin -fno-common
-DHAVE_STDINT_H -DGX_COLOR_INDEX_TYPE="unsigned long int"   -I./obj -I./base  
-o ./obj/gdevcups.o -c cups/gdevcups.c
cups/gdevcups.c: In function `cups_close':
cups/gdevcups.c:454: warning: implicit declaration of function `close'
cups/gdevcups.c: In function `cups_print_pages':
cups/gdevcups.c:2810: error: `CUPS_RASTER_WRITE_COMPRESSED' undeclared
(first use in this function)
cups/gdevcups.c:2810: error: (Each undeclared identifier is reported only
once
cups/gdevcups.c:2810: error: for each function it appears in.)
cups/gdevcups.c: In function `cups_put_params':
cups/gdevcups.c:2894: warning: unused variable `floatval'
cups/gdevcups.c:2910: warning: unused variable `found'
cups/gdevcups.c: In function `cups_get_color_comp_index':
cups/gdevcups.c:720: warning: case value `17' not in enumerated type
`cups_cspace_t'
gmake: *** [obj/gdevcups.o] Error 1


I'm running on Red Hat Enterprise Linux AS Release 4 (Nahant Update 8) on a
machine with 64-bit
addressing.


-- 
View this message in context: http://old.nabble.com/Trouble-compiling-Ghostscript-9.00-tp29851272p29851272.html
Sent from the Ghostscript - Dev mailing list archive at Nabble.com.