Pdf, calibre and metadata.

[aritza]

Originally Posted by itimpi

Many of those fields cannot be saved to the PDF. This is a limitation of the PDF format.

When you do a "Save to "Disk" you are also given the option of having the .opf file as well. This file will have ALL the metadata in it, and if the recipient has Calibre as well can be used to transfer it to their Calibre library

I see the following fields in the PDF file ("file properties"):

Title
Author
Subject
Keywords

And there there is a sectio for "Additional metadata":

Document title
Author
Author title
Description
Description writer
Keywords

plus many more, including "custom properties", of which you can create seemingly as many as you want.

So I do not understand why metadata couldn't be saved to the file, even if it requires some creative mapping.

My fear is that this lack of portability of metadata has me tied to Calibre, which for all I know could stop being developed at any time (or could start charging money, maybe even a lot of money), or a product that better serves my needs could come around.

I don't see why I seem to be the only one concerned about this.

[Starson17]

Originally Posted by aritza

I see the following fields in the PDF file ("file properties")...
plus many more, including "custom properties", of which you can create seemingly as many as you want.

So I do not understand why metadata couldn't be saved to the file, even if it requires some creative mapping.

Creative mapping may interfere with other programs usage of those fields. It's all in the .opf file anyway.

My fear is that this lack of portability of metadata has me tied to Calibre, which for all I know could stop being developed at any time (or could start charging money, maybe even a lot of money), or a product that better serves my needs could come around.

I don't see why I seem to be the only one concerned about this.

There's no secret plan to trap you into a proprietary format. Calibre is open source and licensed via GPL. You can always run any version for free. Even if you couldn't, all the data in Calibre is easily accessible in multiple ways. You can get it in opf files, in spreadsheets (several formats) in the SQL database, etc.

You have nothing to worry about.

[DoctorOhh]

Originally Posted by aritza

My fear is that this lack of portability of metadata has me tied to Calibre, which for all I know could stop being developed at any time (or could start charging money, maybe even a lot of money), or a product that better serves my needs could come around.

I don't see why I seem to be the only one concerned about this.

Ignorance is a powerful source of fear.

As Starson17 said about metadata

Originally Posted by Starson17

It's all in the .opf file anyway.

You can get it in opf files, in spreadsheets (several formats) in the SQL database, etc.

I've never seen a product that makes it so easy to grab any info. OPF stands for Open Packaging Format, you can read more about OPF here or here.

[itimpi]

Originally Posted by aritza;

My fear is that this lack of portability of metadata has me tied to Calibre, which for all I know could stop being developed at any time (or could start charging money, maybe even a lot of money), or a product that better serves my needs could come around.

I don't see why I seem to be the only one concerned about this.

The .opf file is in a XML format defined as part of the EPUB standard - probably the only one that is truly portable and open. Using this file to storeand transport metadata is the one that is most likely to protect you if you decide to switch away from Calibre.

[aritza]

Thank you for the reassurance. I think you will agree that emphasizing all of this would be a good way to promote Calibre.

Nonetheless, I would like to urge Kovid to please provide a simple and direct way to save information beyond Author and Title to a PDF in some way to the extent that it is possible to map fields.

Thank you!

[tonyc46]

I'm having the same problem. I'm trying to read a pdf in Android (Mantano Reader). I use Calibre Save to Disk (PDF format only) to my external SD card. Then I import with Mantano and don't get any of the metadata. Even many of the titles and authors are gibberish.

Thanks.

[Benji99]

I also have a problem with calibre and metadata in pdfs.

Actually, for me, in the few pdfs that I tried, the save to disk works, the metadata seems to be correctly added. The problem comes when I download books via the content server to my android Galaxy Nexus via either Mantano or Aldiko, the metadata doesn't come with it. I checked the downloaded pdf file and sure enough, the title and author fields are blank.

Is it possible for the content server to add the proper metadata before sending the file?
Also, as a side question, is it possible for applications such Mantano to download the .opf file as well in order to get all the metadata? (This would be awesome). If it's possible, I'll go start begging them to add that feature.

[kovidgoyal]

No, it isn't adding metadata to PDF is a very slow process, the entire PDF has to be read into memory and written out again, as such it is not suitable for a server.

[Benji99]

For some users that have reasonable specced servers, it would be a really nice option is having incorrect metadata in the reading applications is a pretty significant problem. Anyway I hope one day you can add an option for those that wish that add that feature.

That said, how about an option to write all tags to the pdf files in my library? Perhaps it would slow down Calibre too much if it do it live with all updates, but perhaps some option somewhere to update all tags in all files and let it run over night. This way the server wouldn't need to update the files when serving them.

Oh you forgot the answer my last question, so there's no way for the applications to have access to the data in the opf data along with the pdf file when retrieving it from the Content Server?

[kovidgoyal]

The content server makes all metadata avaibale in half a dozen different ways, whether the content server using app makes use of it or not, is up to it.

[dtrounson]

Hi guys, sorry for dragging this thread back from the dead again, but would it be possible for the save to disk command to somehow run the same process over the file actually IN the library? At the moment if I want to preserve any metadata I would need to save it outside, and then reimport it which seems a little time consuming?

Or I have the whole thing wrong...

[DoctorOhh]

Originally Posted by dtrounson

Hi guys, sorry for dragging this thread back from the dead again

Then why do it? At this point I'm unsure if you are talking about only PDF files or all files.

Originally Posted by dtrounson

would it be possible for the save to disk command to somehow run the same process over the file actually IN the library? At the moment if I want to preserve any metadata I would need to save it outside, and then reimport it which seems a little time consuming?

The Send to device feature embeds the same metadata in a book that the Save to disk feature does.

[dtrounson]

Well the thread is with regards to PDF so yes, I am just looking for the fastest way to ensure that the metadata is correctly embedded, with ePubs I tend to use the modifyEpub plugin to ensure consistency of my library, PDF's are a little trickier.

But I will figure the workflow for this some other way I think, it seems like the command line tool would work as a FAST method anyway

[aboettget]

You can use this script:
https://github.com/aboettger/ShellSc...ata_by_calibre

[Benji99]

Nice! will give it a shot shortly!