Am 22.10.2015 um 18:01 schrieb Tilman Hausherr:
Am 22.10.2015 um 08:21 schrieb Mario Alten:
Thanks, this online validator shows the same problem.
So you do not consider this as a (newly introduced) bug, but rather a
fix or improvement compared to the old version, right?
When I look at https://issues.apache.org/jira/browse/PDFBOX-2403and
the comment (written at 31/Oct/14 19:15) which says
"Please open a new issue if the current problem "Invalid Color space,
The operator "RG" can't be used with CMYK Profile",
I am wondering if this new jira entry exists!?
I have no idea why I wrote that, and don't know of an issue. Maybe
that was related to two messages by Ralf that he deleted?
Upload the document somewhere or send it to me, tilman at snafu dot de.
And yes, I consider the behaviour an improvement. Note that the two
messages are related:
2.1.2 : Invalid graphics object, /N entry of ICC profile is 4 but the
ICC profile has 3 components
2.4.1 : Invalid Color space, The operator "RG" can't be used with CMYK
Profile
The first message is self-explanatory. The second means that the
operator "RG" is in the content stream, but that currently a CMYK
stroking colorspace is selected (see in the OutputIntent dictionary at
the document level).
Thanks for the file. It is really like I wrote. Your content stream has
this at Root/Pages/Kids/[0]/Contents/[1] :
0 Tw
0 Tc
0 0 0 rg
0 0 0 RG
Your output profile at Root/OutputIntents/[0]/DestOutputProfile/N has
a value of 4. "rg" and "RG" are rgb commands. N=4 means 4 colors,
typically CMYK. (However the profile itself is RGB. Change the N to 3
and it works fine with us and with PDF Tools)
Re OutputIntents : "(Optional; PDF 1.4) An array of output intent
dictionaries that shall specify the colour characteristics of output
devices on which the document might be rendered (see 14.11.5, "Output
Intents")."
and
"Output intents (PDF 1.4) provide a means for matching the colour
characteristics of a PDF document with those of a target output device
or production environment in which the document will be printed. The
optional OutputIntents entry in the document catalogue (see 7.7.2,
“Document Catalog”) holds an array of output intent dictionaries, each
describing the colour reproduction characteristics of a possible output
device or production condition. The contents of these dictionaries may
vary for different devices and conditions"
It is of course surprising that Adobe would claim the file is OK. Their
PDF/A check is by Callas, and they're among the best in the PDF/A
business. Maybe they ignore the "N" value?
Tilman
Tilman
Regards
-----Ursprüngliche Nachricht-----
Von: Tilman Hausherr [mailto:[email protected]]
Gesendet: Mittwoch, 21. Oktober 2015 19:59
An: [email protected]
Betreff: Re: PDF/A marked as invalid (pdfbox 2.0.0-RC1)
We fixed many flaws in the 2.0 version, and did not port all
improvements to 1.8. Your attachment didn't go through. Please check
your file with http://www.pdf-tools.com/pdf/validate-pdfa-online.aspx
to get a second opinion.
Tilman
Am 21.10.2015 um 19:51 schrieb Mario Alten:
>
> Dear ladies and gentlemen.
>
> The validation of a quite simple PDF/A works fine with PDF-Box 1.8,
> but not anymore with pdfbox 2.0.0-RC1.
>
> I get the following error message(s):
>
> /2.1.2 : Invalid graphics object, /N entry of ICC profile is 4 but the
> ICC profile has 3 components/
>
> /2.4.1 : Invalid Color space, The operator "RG" can't be used with
> CMYK Profile/
>
> Is that a known issue within pdfbox, or is it a real problem of my PDF
> which was “just” not recognized with the old version of pdfbox? If
> needed, I can send the actual pdf (which actually uses RBG color
> profile, although the error message above indicates different).
> Attached is the preflight report which does not show any problems.
>
> Regards
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
<mailto:[email protected]>
> For additional commands, e-mail: [email protected]
<mailto:[email protected]>
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
