Forma Small PDF upload problem?

[james_]

thanks... I guess with PDFs you add the metadata in a PDF specific way, as when I've added metadata in Calibre, it still didn't work. Probably because that metadata just sits in the Calibre library, rather than being embedded in the document.

Still, it is slightly irritating that you can't use the Kobo to stash these tickets if you happen to be travelling and reading at the same time...

[davidfor]

Originally Posted by james_

thanks... I guess with PDFs you add the metadata in a PDF specific way, as when I've added metadata in Calibre, it still didn't work. Probably because that metadata just sits in the Calibre library, rather than being embedded in the document.

I was expecting calibre to add the title and author when I sent the book to the device. I know it can't update all the metadata, but I thought it did the title and author at least. I don't have any tools for editing PDF, as I only use them for things like this. Or for manuals.

Still, it is slightly irritating that you can't use the Kobo to stash these tickets if you happen to be travelling and reading at the same time...

To me, the error says that these files are malformed in some way. It is likely that the RMSDK is stricter in how it handles these files than a PC based PDF reader.

Sorry, I'm not sure where to go from here. Other than recommending that you report the error to Kobo. They should look at it and see if it can be fixed.

[davidfor]

It's pretty obvious what is wrong when it is pointed out.

The start of the file is:

HTTP/1.0 200 OK
Cache-Control: no-cache
Content-Type:  application/pdf
Date:          Thu, 24 Sep 2020 09:29:24 GMT

%PDF-1.4

There are some HTTP headers at the start of the file. How they got there, I don't know. It could be the ticketing system, or it could be your browser. It seems other systems are ignoring those extra characters, but, the RMSDK used by the Kobo devices does not. A quick read of the PDF page in Wikipedia has:

A PDF file starts with a header containing the magic number and the version of the format such as %PDF-1.7.

That explains the error message. I don't think you should be expecting a fix for broken files.

[compurandom]

Originally Posted by davidfor

It's pretty obvious what is wrong when it is pointed out.

The start of the file is:

Interesting... the start of my file is a CR LF and then the %PDF, so same problem. Interesting that so many tools ignore this.

[james_]

Thanks David for all your investigation. I'm going to ask the people producing the tickets to fix it, as clearly this isn't a "Kobo" problem.

However, clearly Kobo have a particularly inept programming team. If pretty much every other device out there can read this PDF then you have to ask why Kobo have such a badly written PDF parser that it drops dead at the slightest sign of trouble.

[compurandom]

I wouldn't call the kobo programming team inept for a pdf reader that was probably written by Adobe rather than kobo, and strictly follows the published PDF standard rather than accepting any sloppily constructed pdf off the web.

[davidfor]

Originally Posted by james_

Thanks David for all your investigation. I'm going to ask the people producing the tickets to fix it, as clearly this isn't a "Kobo" problem.

However, clearly Kobo have a particularly inept programming team. If pretty much every other device out there can read this PDF then you have to ask why Kobo have such a badly written PDF parser that it drops dead at the slightest sign of trouble.

It is inept to follow well published standard? Sorry, but no. Producing files that don't follow the standards is inept. Handling files that don't follow the standards without even a warning, again, inept. But, making sure your system works even when crap is thrown at it, not inept. And deciding to throw away the crap rather than take a guess at how to handle it, again, not inept. Especially when doing this in a resource constrained environment.