Forma Small PDF upload problem?

[james_]

I'm using Calibre with my Forma, when I upload a very small PDF it seems to not appear in the list of books on the Forma, even when I can see it listed in the directory on the device along with all the other books. Any idea what's happening?

(Typically I might convert an e-ticket to a PDF and use them on my e-reader if I'm travelling, or maybe convert a recipe to PDF so that I don't need to have scraps of paper floating around the kitchen.)

I must be doing something stupid.. any suggestions please!!

[DevonHess]

I've never run into this problem. Can you post one of your PDFs here (one that you own the rights to) so we can take a look?

I'm also curious, why do you use your Kobo for scraps like that? A smartphone seems infinitely more convenient.

[Quoth]

Originally Posted by james_

(Typically I might convert an e-ticket to a PDF and use them on my e-reader if I'm travelling, or maybe convert a recipe to PDF so that I don't need to have scraps of paper floating around the kitchen.)

I must be doing something stupid.. any suggestions please!!

Convert to ePub. PDFs are for 10" or larger tablets or proofing work for paper.

[shalym]

Originally Posted by Quoth

Convert to ePub. PDFs are for 10" or larger tablets or proofing work for paper.

How would converting to epub work for an e-ticket? How well do you think it would show the barcode (or QR code) in an epub, rather than a pdf? How long would it take to convert an e-ticket to an epub compared to printing it to a pdf?

Shari

[davidfor]

Originally Posted by Quoth

Convert to ePub. PDFs are for 10" or larger tablets or proofing work for paper.

Of course, you could set the page size in the conversion to closer to the screen size. These devices do handle PDF well if you don't need to zoom.

[Quoth]

Originally Posted by davidfor

Of course, you could set the page size in the conversion to closer to the screen size. These devices do handle PDF well if you don't need to zoom.

True if you are creating a PDF and the image fits, but if something isn't already a PDF, or isn't being sent to POD or a users printer, why on earth would you make a PDF rather than an epub?

[james_]

Originally Posted by DevonHess

I've never run into this problem. Can you post one of your PDFs here (one that you own the rights to) so we can take a look?

I'm also curious, why do you use your Kobo for scraps like that? A smartphone seems infinitely more convenient.

I'll try and create one and share later. The eTicket I have may still have some validity and I'd prefer not to share that.

An e-ink screen is considerably better/easier for scanners to read. For evidence of this, you just need to go visit my local train station and watch everyone who uses an e-ticket try to get through the barrier with a mobile phone. Usually there is at least one person or two swearing a lot and looking red faced.

I don't carry around my e-reader the whole time, but if I happen to have one on me for reading then why not stick the tickets on it... the Yotaphone is perfect for that situation but it's really too small to read on comfortably.

[james_]

Originally Posted by Quoth

Convert to ePub. PDFs are for 10" or larger tablets or proofing work for paper.

Ironically I just used my Boox Note instead as it displayed the PDF e-ticket just fine, and I had some books I was reading on that. I have no idea why, but the e-Ticket was actually A4 in size and a PDF, but I guess that's not unusual.

[davidfor]

Originally Posted by Quoth

True if you are creating a PDF and the image fits, but if something isn't already a PDF, or isn't being sent to POD or a users printer, why on earth would you make a PDF rather than an epub?

It depends on the source. I usually save the receipts for on-line payments as PDFs because that works and means I don't have to worry about formatting if I need to see it later. And some sites off a PDF download or emailed the receipt as a PDF.

[davidfor]

Originally Posted by james_

I'm using Calibre with my Forma, when I upload a very small PDF it seems to not appear in the list of books on the Forma, even when I can see it listed in the directory on the device along with all the other books. Any idea what's happening?

(Typically I might convert an e-ticket to a PDF and use them on my e-reader if I'm travelling, or maybe convert a recipe to PDF so that I don't need to have scraps of paper floating around the kitchen.)

I must be doing something stupid.. any suggestions please!!

It should just work. The things I can think of are:

Where are you putting it? Older firmware didn't scan directories whose name started with a dot. The recent firmware does, but, you can add a config file option to ignore some directories.

Some characters in the file name can prevent the book from being seen. If you are using calibre, it will normalise the name to make sure this doesn't happen. Of course, the extension has to be PDF.

Does the importing screen show when you disconnect the device? If it doesn't, it can mean the files aren't being seen. If it does, does the progress bar go across the screen? Are other books you add being imported?

You can turn logging on and that might show what happens. You need to turn on Developers mode. Open the search screen and enter "devmodeon". That should set the logging to cover the import. When you next connect, there will be a file in the .kobo directory with the name starting with "syslog" and the rest a timestamp. Put the PDF on the device, disconnect and let the import happen, or not happen. Then reconnect and look at the latest log file. There will be a lot of lines with "Pruning file" for each of the books already on the device. And hopefully one for the book you added with a hint of what went wrong.

[james_]

Originally Posted by davidfor

It should just work. The things I can think of are:

Where are you putting it? Older firmware didn't scan directories whose name started with a dot. The recent firmware does, but, you can add a config file option to ignore some directories.

Some characters in the file name can prevent the book from being seen. If you are using calibre, it will normalise the name to make sure this doesn't happen. Of course, the extension has to be PDF.

Does the importing screen show when you disconnect the device? If it doesn't, it can mean the files aren't being seen. If it does, does the progress bar go across the screen? Are other books you add being imported?

You can turn logging on and that might show what happens. You need to turn on Developers mode. Open the search screen and enter "devmodeon". That should set the logging to cover the import. When you next connect, there will be a file in the .kobo directory with the name starting with "syslog" and the rest a timestamp. Put the PDF on the device, disconnect and let the import happen, or not happen. Then reconnect and look at the latest log file. There will be a lot of lines with "Pruning file" for each of the books already on the device. And hopefully one for the book you added with a hint of what went wrong.

Thanks! I attempted to create a tiny PDF for sharing but it uploaded no problem. I'm not seeing the import 'flash' when the ticket was uploaded by itself. Other books transfer no problem The ticket appears on the Kobo with author of 'unknown' - (It's for a Warhol exhibition...)

Here's what is in the log:
Oct 30 21:51:49 adobehost: epub: static AdobeHost* AdobeHost::createInstance(const QString&) unknown magic "file:///mnt/onboard/Unknown/Warhol 06580249c2952a6aec6847cfe6eb3299 - Unknown.pdf" "485454502f312e30"
Oct 30 21:51:49 nickel: ( 18124.095 @ 0x1ac2298) void AdobeParser:arse() didn't get any metadata!
Oct 30 21:52:09 nickel: ( 18136.379 @ 0x1997b48) QObject::disconnect: Unexpected null parameter
Oct 30 21:52:09 nickel: ( 18136.380 @ 0x1997b48 / ui.warning) cancelled before full sync has started

Now, I have attached metadata to the file before when attempting a previous upload, question is, what is 'unknown magic'?

[davidfor]

Originally Posted by james_

Thanks! I attempted to create a tiny PDF for sharing but it uploaded no problem. I'm not seeing the import 'flash' when the ticket was uploaded by itself. Other books transfer no problem The ticket appears on the Kobo with author of 'unknown' - (It's for a Warhol exhibition...)

Here's what is in the log:
Oct 30 21:51:49 adobehost: epub: static AdobeHost* AdobeHost::createInstance(const QString&) unknown magic "file:///mnt/onboard/Unknown/Warhol 06580249c2952a6aec6847cfe6eb3299 - Unknown.pdf" "485454502f312e30"
Oct 30 21:51:49 nickel: ( 18124.095 @ 0x1ac2298) void AdobeParser:arse() didn't get any metadata!
Oct 30 21:52:09 nickel: ( 18136.379 @ 0x1997b48) QObject::disconnect: Unexpected null parameter
Oct 30 21:52:09 nickel: ( 18136.380 @ 0x1997b48 / ui.warning) cancelled before full sync has started

Now, I have attached metadata to the file before when attempting a previous upload, question is, what is 'unknown magic'?

Just tried adding a couple of small PDFs to see what happened. Including one with the above file name. They were imported correctly.

I'd say that the "unknown magic" is something missing in the book. What are you using to produce it? Are you printing it somehow from a browser, or something else? And would it be possible to see one? If you are doing this by printing from the browser, you could try something else such as this page. Or, upload a file somewhere and PM me the link. Of course, I probably can't do anything with it other than reproduce the error.

[james_]

PM sent .. see what you make of it :-)

[compurandom]

I've also had problems with some pdfs not being recognized by my kobo.
I suspect the problem was that the pdf was encrypted in some way the kobo doesn't support.

Since the book never showed up on the kobo after I copied it over, I kind of forgot about it and never investigated.

So, I just copied the original document to my kobo. I then converted it with calibre's epub-epub, and converted it with ghostscript pdfwrite with -dCompatibilityLevel=1.4

I noticed that calibre's conversion log had this in it:

 Document has copy-protection bit set.

Both of the conversions were recognized, the original was not.

I've used the compatibility level conversion before with some pdfs that were recognized on the kobo but some graphical elements were missing in diagrams.

For this particular document, the calibre pdf conversion scrambled the document a bit. I'm not sure if I like or dislike how it scrambled it. I might have to play with that more on other documents that have been problematic.

edit: Not surprisingly, the pdf->epub conversion looks about the same as the pdf->pdf conversion, scrambled in pretty much exactly the same way.

[davidfor]

Originally Posted by james_

PM sent .. see what you make of it :-)

The good news is that it failed with the same error here. Looking at it, there is no metadata in it, which is what the next line says. I don't know if the lack of metadata caused the problem, or if the first error is something else.

Other than working out how to set the metadata, I'm not sure were to go from here.