Jay,

take a look at my comments I provided in this question yesterday:

http://forums.adobe.com/thread/998312?tstart=0

You are running into the same problems. PDF is a binary file format, and

your byte offsets in the cross reference table are incorrect. Acrobat will

repair your file for you when you load it, and because at the time you want

to close the file, it's different than the one you have on disk (it

contains Acrobat's fixes), the application is prompting you to save it.

Once your byte offsets are correct, this will no longer happen.

Karl Heinz Kremer

PDF Acrobatics Without a Net

khk@khk.net

http://www.khkonsulting.com

Let's approach this from a different angle: Why are you trying to generate

a PDF file manually? Creating a PDF file this way is almost always a

pre-cursor to writing software to output PDF. It does not make any sense

for an end user to create a PDF based on the PDF spec. That's why we have

software that does that. Even if you need a PDF file with a specific

feature, there is usually no need to hand craft a file: You can just write

a PostScript file (which is a text format, and can be written based on the

PostScript spec) and then run it through a PSToPDF converter like

Distiller.

So again, why do you think you need to write a PDF file in a text editor?

Karl Heinz Kremer

PDF Acrobatics Without a Net

khk@khk.net

http://www.khkonsulting.com

Jay,

I don't quite understand what you want to accomplish, but it sounds like

some level of HTML to PDF conversion. Yes, there are existing solutions,

and they may seem to be expensive, but based on my own experience, they are

worth what you pay for (and more). It may seem that building your own

solutions could be cheaper, but again, based on my own experience, you will

waste a lot of time and money to come up with something that works as well

as anything that you can buy today.

Based on your problems coming up with a correct XRef table, it looks like

you do not have a deep enough understanding of the PDF spec to do this.

I've read the spec too often to count. It's not an easy to read document,

but you need to read and understand every part of it. There is no way

around it.

I work mostly on software that processes existing PDFs, and almost every

PDF file that is generated by somebody who thought that they needed to

re-invent the wheel and write their own PDF generator, they end up writing

corrupt PDF. Go with one of the established solutions, it will pay for

itself in the long run.

If you really want to write PDF in an editor, you need an editor that shows

you byte offsets. I use vim in it's binary mode.

Karl Heinz Kremer

PDF Acrobatics Without a Net

khk@khk.net

http://www.khkonsulting.com

Thanks for your answers!

@ Screen Name:

1) Ok, your right. I corrected it.

2) Ok, you probably are right. But in my cases it always was the same.

@Irosneth

Thanks for the documents. I will study the ISo-document. But mostly it dont describe the failures, so it isnt very helpful because then you have to check the whole document.... I think to buy a bok is a little oversized for me. I only want to show text in the simplest form it is possible.

Hm..ok, if you show me a good pdf-creator for winCE, i will use it. But i spent a day on searching for such a pdf creatorand there is no one.

I try to explain you my problem. In my company we produce machines. They produce statistics about that they do. Now i want to save these statistics in a pdf-file. But there is no pdf-creator, so we have to write one for our use. And i am a student and dont cost much, so i have this damn job to write such a pdf creator. And becaouse of we have many different machines, it should be as easy as it could be to adapt the file from the old machine to the statistics from the other machine. And i cant do this for the whole company, so the others must be able to adapt it too.

I hope you can understand it.

IT WORKS, its unbelievable, but it works

Im so happy! Many thanks to you!!! You had the right suggestion to make it work. THANKS!

My last problem was a uncorrect date in the info-Dictionary :-(