question about epub opf file

[galaxyhub]

Is there a way to not include the
<meta name="calibre:user_metadata ....
entries in the opf. I would like to retain the dc: entries and the <meta content entries.

Also is there a batch way to strip the <meta name="calibre:user_metadata .... out of my epubs. I know the excellent Modify ePub plugin has an option to remove non dc: elements but this will strip out the series information that I need as well.

[carmenchu]

Comment + suggestion:
1. The .opf file of a book should not/could not be changed: it contains the data to restore the calibre database in case of accident (or so I understand)
2. How those data are copied to an .epub by convert/polish->update metada is another question.
There, I would like as well to delete the user information, AFTER having employed some of my <meta name="calibre:user_metadata:#customized"... . For instance:
<meta name="calibre:user_metadata:#ordate"...content... &quot;__value__&quot;: &quot{date}&quot; -> <dc:date opf:event="creation">{date}</dc:date>
I suspect this would require a plug-in, but I don't know where to begin--a template to get started?

[BetterRed]

Originally Posted by carmenchu

Comment + suggestion:
1. The .opf file of a book should not/could not be changed: it contains the data to restore the calibre database in case of accident (or so I understand)
2. How those data are copied to an .epub by convert/polish->update metada is another question.
There, I would like as well to delete the user information, AFTER having employed some of my <meta name="calibre:user_metadata:#customized"... . For instance:
<meta name="calibre:user_metadata:#ordate"...content... &quot;__value__&quot;: &quot{date}&quot; -> <dc:date opf:event="creation">{date}</dc:date>
I suspect this would require a plug-in, but I don't know where to begin--a template to get started?

1. no, you can do what you like to the content.opf file in the book within the bounds of the ePub specs; the calibre library manager won't know or care. It is the metadata.opf files in the book folders that are used to restore the database (if/when it gets broken - which has never happened to me in about 9 years).

2. The .opf files in the book are built from the content of the database, as are the metadata.opf files. But, don't equate the .opf file in the book and the metadata.opf file in the book folders. Whilst they have similarities they serve different purposes. And what's in the book will not reflect what's in the database if you don't Polish, Modify or Convert (ePub to ePub), but the metadata.opf file in the book folder will** always reflect what's in the database.

**if you do a bulk edit to many books, the updating of the metadata.opf files in the book folders happens in a low priority background task that may take a while to catch up.

BR

[carmenchu]

Originally Posted by carmenchu

Comment + suggestion:
1. The .opf file of a book should not/could not be changed: it contains the data to restore the calibre database in case of accident (or so I understand)
2. How those data are copied to an .epub by convert/polish->update metada is another question.
There, I would like as well to delete the user information, AFTER having employed some of my <meta name="calibre:user_metadata:#customized"... . For instance:
<meta name="calibre:user_metadata:#ordate"...content... &quot;__value__&quot;: &quot{date}&quot; -> <dc:date opf:event="creation">{date}</dc:date>
I suspect this would require a plug-in, but I don't know where to begin--a template to get started?

My fault: (1) was about metadata.opf. (2) Is what I would wish done (automatically) to the .opf file in the epub: doing it (and more) by hand every time.

[BetterRed]

You can use the Modify ePub plugin to strip the metadata back to Dublin Core elements - i.e. those prefixed by dc:.

The Sigil ePub editor has a GUI tool to add, change, delete metadata items within the embedded .opf, being a minimalist I use it to remove dc:description and dc:subjects.

Beyond that you can edit the embedded .opf with a text editor or one of the ePub editors. Or you can write, or hack an existing, plugin to meet your specific wants.

Take care

BR

[PeterT]

I wonder if this tool might help. https://www.mobileread.com/forums/sh...d.php?t=132607

[galaxyhub]

my main libraries are all epub only on onedrive, convert as needed then delete the other formats. So batch editing of opf would be ideal to remove all the opf mess excepting dc elements and series name and index.

[DNSB]

Originally Posted by galaxyhub

my main libraries are all epub only on onedrive, convert as needed then delete the other formats. So batch editing of opf would be ideal to remove all the opf mess excepting dc elements and series name and index.

One thing you might want to consider is moving your libraries to local storage. A library stored in the cloud is a disaster waiting to happen. You may get away with it for years and then hope you have a good backup. One recent thread is found at Library management issues after editing metadata.

Also remember that when you save your epub to disk or send to a device, the metadata will be updated from calibre's metadata.opf file.

[BetterRed]

Originally Posted by DNSB

Also remember that when you save your epub to disk or send to a device, the metadata will be updated from calibre's metadata.opf file.

Pretty sure that's false.

The metadata embedded into the file that is Saved or Sent will be taken from the database library metadata.db file, if you just did bulk update the update of the metadata.opf file could still be in the lazy write queue.

BR

[DNSB]

Originally Posted by BetterRed

Pretty sure that's false.

The metadata embedded into the file that is Saved or Sent will be taken from the database library metadata.db file, if you just did bulk update the update of the metadata.opf file could still be in the lazy write queue.

BR

Perhaps I should have just said the metadata embedded in the epub file that is sent or saved will be updated to match the corresponding metadata.opf -- assuming all lazy writes are completed.

It's a lot easier to look at the metadata.opf compared to browing the database file.

[BetterRed]

Originally Posted by DNSB

It's a lot easier to look at the metadata.opf compared to browing the database file.

For you maybe, but you ain't embedding the data in to the Saved/Sent files. The calibre software has the in memory database at its fingertips

FWIW I find it easier to press 'I' on a book to see all of its metadata.

BR

[DNSB]

Originally Posted by BetterRed

For you maybe, but you ain't embedding the data in to the Saved/Sent files. The calibre software has the in memory database at its fingertips

FWIW I find it easier to press 'I' on a book to see all of its metadata.

BR

But I'm the one looking at the metadata with that old fashioned mark 1 eyeball. AFAIK, calibre doesn't supply a handy tool to show me all the metadata for an ebook in a human readable form. Maybe, it would be better to say, I don't know of such a handy tool in calibre.

Hitting 'I' doesn't show me all the metadata I'm interested in. Such items as are fonts correctly identified and does the titlepage entry have the svg properties set. Yeah, I know. That's what epubcheck is for.

[BetterRed]

I didn't know the metadata.opf had font and other typographical artefacts in it. I mainly buy books for their content rather than their looks so I don't look for that stuff, except to delete it. I say mainly, because I did buy one book based purely on its appearance, it wasn't an ebook, it caught my eye on a remainder table.

BR

[theducks]

Originally Posted by DNSB

But I'm the one looking at the metadata with that old fashioned mark 1 eyeball. AFAIK, calibre doesn't supply a handy tool to show me all the metadata for an ebook in a human readable form. Maybe, it would be better to say, I don't know of such a handy tool in calibre.

Hitting 'I' doesn't show me all the metadata I'm interested in. Such items as are fonts correctly identified and does the titlepage entry have the svg properties set. Yeah, I know. That's what epubcheck is for.

Confusion Reigns
'I' shows CALIBRE (db) metadata in full
content.opf (part of the book) contains the structure as well as metadata. embedded fonts is STRUCTURE.

Note the Quality check PI opens the books content,opf to determine fonts embedded. It does not enumerate which ones

[DNSB]

Originally Posted by theducks

Confusion Reigns
'I' shows CALIBRE (db) metadata in full
content.opf (part of the book) contains the structure as well as metadata. embedded fonts is STRUCTURE.

Note the Quality check PI opens the books content,opf to determine fonts embedded. It does not enumerate which ones

True. The only time I look at metadata.opf is when I am wondering where some odds and sods in the epub's content.opf came from.