I see. Fop 0.20.3 was a bit different. I didn't have that special JPEG handler for optimized JPEG import when I remeber correctly. It was decoded through JIMI and reencoded to PDF exactly like PNG files. Anyway, we have to take a look at your environment. What JVMs are in use when you call FOP either from the batch or from your application? If you remember the mail a few days ago about IBM's JVM that caused a corrupt PDF. The same could be happening here. Other than that, I'm currently out of ideas because the JPEG's contents are more or less copied to the PDF.
On 17.02.2003 15:04:19 Jiri_Nejedly wrote: > > I'm not really sure that the jpg file format is what is wrong. Several > arguments : > - Both in current (0.20.5) and recent version (0.20.3), I have one special > picture with question mark, which I use in case, when there's picture > missing in database. In FOP 0.20.3 was all ok, and this picture hasn't > changed since. > (See attached file: pictx.jpg) > - there's a few files that are GIFs , not JPGs - also invisible ! > - why then calling fop from batch is OK ? > > I set up a very simple fo file using above picture: > > <?xml version="1.0"?> > <fo:root xmlns:fo="http://www.w3.org/1999/XSL/Format";> > <fo:layout-master-set> > <fo:simple-page-master page-height="29.7cm" page-width="21cm" > margin-top="1cm" margin-bottom="2cm" margin-left="2.5cm" margin-right=" > 2.5cm" master-name="first"> > <fo:region-body margin-top="1cm" margin-bottom="1cm"/> > <fo:region-before extent="1cm"/> > <fo:region-after extent="1cm"/> > </fo:simple-page-master> > </fo:layout-master-set> > <fo:page-sequence master-reference="first"> > <fo:flow flow-name="xsl-region-body"> > <fo:block> > <fo:external-graphic width="2cm" height="2.5cm" src="c:\pictx.jpg"/> > </fo:block> > </fo:flow> > </fo:page-sequence> > </fo:root> > > Here's output from fop.bat (ok) > (See attached file: FromBatch.pdf) > Here's output from embedding (blind) > (See attached file: FromEmbed.pdf) > I compared these two files and found several differences. I think some > bytes in area where the picture is stored are changed - for example: > 023E 81 -> F3 > 0283 83 -> F3 > 0288 88 -> F3 > .... > On the first sight, all bytes that are changed, are changed (overwritten) > to 'F3'. > Something is wrong with PDF, but not with the source JPG ... > > Jiri Nejedly > > >Ok, I get it. Looking into the PDF using a text editor I can see that > >your JPEG image is right there in the PDF. The problem probably is that > >it's a special JPEG subformat (JFIF, Exif) that's currently not > >supported by FOP. > > > >By the way, I didn't get "no pictures", I got "Insufficient data for an > >image" as an error message from Adobe Acrobat Reader. That was the clue > >I needed to look at the PDF. > > > >You've got three alternatives: > >1. Convert the JPEG to a "normal" JPEG or another format such as PNG or > >TIFF. > >2. Disable the use of our JPEG implementation in the Java code and leave > >the handling of the JPEG to JIMI or JAI. That means some hacking and not > >a two minute job, I guess. > >3. Add support for your particular JPEG format to our JPEG > >implementation. Jeremias Maerki --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
