[Ipe-discuss] Problem opening file (please help)

[Otfried Cheong]

Tim Hutt wrote:
>> Error parsing document 'ipe-theory-seminar.pdf' at character offset 117022
> 
> Good old helpful error messages! It never seems worth writing detailed
> ones at the time...

This is as accurate as it can get: Ipe tells you exactly where it could 
no longer parse the file.

This error should only happen when you prepare Ipe input manually or 
using some other software.  If Ipe cannot read it, it tells you where 
the problem happens, and then you should know how to fix that.

Unfortunately the problem also occurs when there is a bug in Ipe - as in 
this case.  It turns out that Ipe cannot handle \" in a page title.

> One thing you could try is running it through pdf2ipe or maybe
> ipe2ipe. It probably won't come out exactly the same but worth a try
> if you're desperate.

If Ipe cannot read a file, then ipetoipe cannot read it either - it's 
the same code.  pdftoipe would work, as it only uses the PDF markup, but 
you'd loose all Ipe markup information.

> Another desperate option: Get a hex editor and go to offset 117022.

This doesn't work either, because the offset that Ipe gives you is a 
character offset in the XML stream.  It has nothing to do with indices 
into the PDF file.

Ipe 7 contains a new utility "ipeextract" that can retrieve the XML 
stream from an Ipe-generated PDF file or EPS file.  It doesn't try to 
parse the XML stream, and so it also works even if Ipe can no longer 
read the file.   You can then debug the XML stream.

Ipeextract isn't yet available - like the rest of Ipe 7 - but I'm 
working on it...

Cheers,
   Otfried