You are viewing a plain text version of this content. The canonical link for it is here.
Posted to fop-users@xmlgraphics.apache.org by MatthiasR <ma...@gmx.at> on 2009/07/29 16:47:15 UTC

Re: [Trunk] CMYK TIFF Problem in PDF output - Spot colors

Sorry for insisting... Has anybody a solution for this problem?

I'd try to fix this by myself in the FOP/Commons code if someone can point
me to the right direction.

Is it possible to implement support for 5 color TIFF or EPS files (CMYK +
Spot color)?

And how much work would it be to implement support for spot colors? (e.g.
rgb-icc(0,0,0,#Separation, PantoneXYZ))

Regards,
Matthias Reischenbacher


MatthiasR wrote:
> 
> Hi,
> 
> I've built the fop and commons trunk in order to check if there are some
> improvements regarding CMYK TIFFs. But unfortunately my tiff images are
> not displayed as expected. I see some kind of greyscale image without
> recognizing the original image in the PDF output.
> 
> JM wrote in the commons change list that an issue had been fixed
> ("Switched from linear RGB to sRGB for the fallback color model, in order
> to avoid trouble with some images (CMYK TIFF, for example)."). But using
> the xmlgraphics trunk doesn't seem to make any difference in my case.
> 
> I wanted to try the TIFF codec of the JAI Image IO Tools but since there
> is no 64-bit build for windows, that's not a solution for me.
> 
> Apparently no tiff compression is used (I checked with IrfanView).
> 
> I'm running on: Windows Server 2003 64-Bit and JDK 1.6 64-Bit.
> 
> Any help is greatly appreciated....
> 
> Thanks
> Matthias Reischenbacher
> 

-- 
View this message in context: http://www.nabble.com/-Trunk--CMYK-TIFF-Problem-in-PDF-output-tp24042100p24720610.html
Sent from the FOP - Users mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org


Re: [Trunk] CMYK TIFF Problem in PDF output - Spot colors

Posted by MatthiasR <ma...@gmx.at>.
Hi Jeremias,

thanks for you post. I'll start to implement support for spot colors in
order to get more familiar with fop code. Later on I'll start to research
the points you mentioned about using four of five color TIFFs.

I read your post about SVGP and the different color naming approaches. I'd
suggest to use Antenna House's color naming:
rgb-icc([num, num, num,] #Separation, <spot-color-name>[, tint-value]
[,C,M,Y,K]) 

If you prefer another name pattern, just let me know.

Inside the ICCColorFunction class I would remap the function name to
something else (rgb-icc() --> named-color()) in order to keep the code
separated from fop-rgb-icc().

I'm not quite sure how to implement the output part, so I guess I'll start
reading the PDF specification... :-)

Regards,
Matthias


Jeremias Maerki-2 wrote:
> 
> Hi Matthias
> 
> On 29.07.2009 16:47:15 MatthiasR wrote:
>> 
>> Sorry for insisting... Has anybody a solution for this problem?
> 
> Pretty advanced stuff, you know.
> 
>> I'd try to fix this by myself in the FOP/Commons code if someone can
>> point
>> me to the right direction.
>> 
>> Is it possible to implement support for 5 color TIFF or EPS files (CMYK +
>> Spot color)?
> 
> If you embed EPS in a PostScript file, no problem. But EPS in any other
> output format is currently not possible due to lack of a suitably
> licensed PostScript interpreter written in Java.
> 
> CMYK TIFFs have already caused a number of problems. I've never seen one
> with 5 components. A number of problems present themselves here:
> - You have to have a TIFF codec that can handle more than 3 or 4 color
> components and the interpretation of the spot color.
> - You have to find out how such a bitmap would be represented with
> Java's BufferedImage.
> - If you have the image in memory as a BufferedImage, you will then have
> to extend FOP to properly handle the encoding in the various target
> format FOP supports.
> 
> So I guess it is possible but without any further research I don't know
> how much work this will be.
> 
>> And how much work would it be to implement support for spot colors? (e.g.
>> rgb-icc(0,0,0,#Separation, PantoneXYZ))
> 
> Not sure. Probably not all too much. You'd need to:
> - specify how exactly to specify these colors (see [1])
> - Support the parsing of the color function
> - Possibly extend FOP's ColorExt (or create a new Color class) to hold
> the information.
> - Implement the spot color in every output format you need that feature
> for.
> - Make sure a decent fallback is used for all other output formats.
> 
> Since we already support DeviceCMYK to a certain degree, that could
> probably serve as an example.
> 
> [1] The XSL and SVG specifactions currently don't support spot/named
> colors. Everyone creates their own specification. XSL and SVG don't even
> use the same pattern to specify colors but they've started talking about
> unifying that. See my feedback on the SVG Print last call with a summary
> of how named colors are supported in the various languages and
> implementations:
> http://www.nabble.com/Feedback-on-SVG-Print-1.2-WD-2007-12-21-p15300614.html
> So, an eye should be kept on what's happening in the XSL and SVG working
> groups at W3C.
> 
> HTH
> 
>> Regards,
>> Matthias Reischenbacher
>> 
>> 
>> MatthiasR wrote:
>> > 
>> > Hi,
>> > 
>> > I've built the fop and commons trunk in order to check if there are
>> some
>> > improvements regarding CMYK TIFFs. But unfortunately my tiff images are
>> > not displayed as expected. I see some kind of greyscale image without
>> > recognizing the original image in the PDF output.
>> > 
>> > JM wrote in the commons change list that an issue had been fixed
>> > ("Switched from linear RGB to sRGB for the fallback color model, in
>> order
>> > to avoid trouble with some images (CMYK TIFF, for example)."). But
>> using
>> > the xmlgraphics trunk doesn't seem to make any difference in my case.
>> > 
>> > I wanted to try the TIFF codec of the JAI Image IO Tools but since
>> there
>> > is no 64-bit build for windows, that's not a solution for me.
>> > 
>> > Apparently no tiff compression is used (I checked with IrfanView).
>> > 
>> > I'm running on: Windows Server 2003 64-Bit and JDK 1.6 64-Bit.
>> > 
>> > Any help is greatly appreciated....
>> > 
>> > Thanks
>> > Matthias Reischenbacher
>> > 
>> 
>> -- 
>> View this message in context:
>> http://www.nabble.com/-Trunk--CMYK-TIFF-Problem-in-PDF-output-tp24042100p24720610.html
>> Sent from the FOP - Users mailing list archive at Nabble.com.
>> 
> 
> 
> Jeremias Maerki
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
> For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org
> 
> 
> 

-- 
View this message in context: http://www.nabble.com/-Trunk--CMYK-TIFF-Problem-in-PDF-output-tp24042100p24740686.html
Sent from the FOP - Users mailing list archive at Nabble.com.


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org


Re: [Trunk] CMYK TIFF Problem in PDF output - Spot colors

Posted by Jeremias Maerki <de...@jeremias-maerki.ch>.
Hi Matthias

On 29.07.2009 16:47:15 MatthiasR wrote:
> 
> Sorry for insisting... Has anybody a solution for this problem?

Pretty advanced stuff, you know.

> I'd try to fix this by myself in the FOP/Commons code if someone can point
> me to the right direction.
> 
> Is it possible to implement support for 5 color TIFF or EPS files (CMYK +
> Spot color)?

If you embed EPS in a PostScript file, no problem. But EPS in any other
output format is currently not possible due to lack of a suitably
licensed PostScript interpreter written in Java.

CMYK TIFFs have already caused a number of problems. I've never seen one
with 5 components. A number of problems present themselves here:
- You have to have a TIFF codec that can handle more than 3 or 4 color
components and the interpretation of the spot color.
- You have to find out how such a bitmap would be represented with
Java's BufferedImage.
- If you have the image in memory as a BufferedImage, you will then have
to extend FOP to properly handle the encoding in the various target
format FOP supports.

So I guess it is possible but without any further research I don't know
how much work this will be.

> And how much work would it be to implement support for spot colors? (e.g.
> rgb-icc(0,0,0,#Separation, PantoneXYZ))

Not sure. Probably not all too much. You'd need to:
- specify how exactly to specify these colors (see [1])
- Support the parsing of the color function
- Possibly extend FOP's ColorExt (or create a new Color class) to hold
the information.
- Implement the spot color in every output format you need that feature
for.
- Make sure a decent fallback is used for all other output formats.

Since we already support DeviceCMYK to a certain degree, that could
probably serve as an example.

[1] The XSL and SVG specifactions currently don't support spot/named
colors. Everyone creates their own specification. XSL and SVG don't even
use the same pattern to specify colors but they've started talking about
unifying that. See my feedback on the SVG Print last call with a summary
of how named colors are supported in the various languages and
implementations:
http://www.nabble.com/Feedback-on-SVG-Print-1.2-WD-2007-12-21-p15300614.html
So, an eye should be kept on what's happening in the XSL and SVG working
groups at W3C.

HTH

> Regards,
> Matthias Reischenbacher
> 
> 
> MatthiasR wrote:
> > 
> > Hi,
> > 
> > I've built the fop and commons trunk in order to check if there are some
> > improvements regarding CMYK TIFFs. But unfortunately my tiff images are
> > not displayed as expected. I see some kind of greyscale image without
> > recognizing the original image in the PDF output.
> > 
> > JM wrote in the commons change list that an issue had been fixed
> > ("Switched from linear RGB to sRGB for the fallback color model, in order
> > to avoid trouble with some images (CMYK TIFF, for example)."). But using
> > the xmlgraphics trunk doesn't seem to make any difference in my case.
> > 
> > I wanted to try the TIFF codec of the JAI Image IO Tools but since there
> > is no 64-bit build for windows, that's not a solution for me.
> > 
> > Apparently no tiff compression is used (I checked with IrfanView).
> > 
> > I'm running on: Windows Server 2003 64-Bit and JDK 1.6 64-Bit.
> > 
> > Any help is greatly appreciated....
> > 
> > Thanks
> > Matthias Reischenbacher
> > 
> 
> -- 
> View this message in context: http://www.nabble.com/-Trunk--CMYK-TIFF-Problem-in-PDF-output-tp24042100p24720610.html
> Sent from the FOP - Users mailing list archive at Nabble.com.
> 


Jeremias Maerki


---------------------------------------------------------------------
To unsubscribe, e-mail: fop-users-unsubscribe@xmlgraphics.apache.org
For additional commands, e-mail: fop-users-help@xmlgraphics.apache.org