Identify Which Books Need Rerun of "Modify ePub"

[Rob557]

Greetings.

I use the plugin "Modify ePub" primarily so that my ePub files contain the current cover and current metadata data. [ FWIW, when I periodically run the plugin for recently-acquired books, I input the run date into a user-defined column to identify that I have run the plugin. ]

Over time, however, I have subsequently revised covers and metadata information for previous books but usually without re-running Modify ePub.

Is there a bulk way to identify which books would now need to be rerun using Modify ePub?

I did find an old thread that did not seem to give me the results I was hoping for (data was generated but did not seem to provide a useable distinction):
https://www.mobileread.com/forums/sho...light=identify

Thanks for any help on this.

[chaley]

Originally Posted by Rob557

I did find an old thread that did not seem to give me the results I was hoping for (data was generated but did not seem to provide a useable distinction):
https://www.mobileread.com/forums/sho...light=identify

That template works fine for me. It tells you the number of days between when the epub format was modified and the metadata was modified. A positive number means that the epub format is older than the metadata modification date. The larger the number, the greater the difference. It does not, and can not, tell you why the metadata date was changed.

[Rob557]

Thanks for your comment on that template chaley.

I guess I should note here that the template in that other thread (for purposes of defining a user-created column that is a combination of the information in other columns) is:

{last_modified:'days_between($, select(formats_modtimes('iso'), 'EPUB'))'}

[/QUOTE]
but the numeric value that was produced in my Calibre column today for all my books was the number of days between when the ePub file was last changed and today's date.

The problem seems to be that when I added a user-created column today to test out that template (or whenever I add any temporary column for special sorts or whatever), it results in changing the "metadata modification date" for all my ePubs.

The primary metadata changes that I would mainly care about for rerunning Modify ePub on previous books would be changes to standard non-user-created metadata fields (e.g. author, title, series, index, comments) and the book cover.

But I guess there is no way to distinguish the dates for changes to one or more of those fields, from an overall "metadata modification date" that would change for all books when any user-created column is added, modified or deleted.

That template may be the best thing available, but depending on how an individual works with the various data fields, I guess the template may not be able to provide the information needed to determine with hindsight which ePubs warrant a re-run of the plugin Modify ePub?

[chaley]

Originally Posted by Rob557

That template may be the best thing available, but depending on how an individual works with the various data fields, I guess the template may not be able to provide the information needed to determine with hindsight which ePubs warrant a re-run of the plugin Modify ePub?

The only thing that one can determine from the modified date is that *something* changed. As you are pointing out, the granularity is not fine. Unfortunately there isn't anything else available.

Question: why use modify epub instead of polish books? If your library isn't huge, you might be able to solve your problem by from time to time polishing all the books in your library. The advantage is you can apply polish in batches.

[Rob557]

Originally Posted by chaley

Question: why use modify epub instead of polish books? If your library isn't huge, you might be able to solve your problem by from time to time polishing all the books in your library. The advantage is you can apply polish in batches.

I was using Modify ePub before becoming aware of Polish Books. Generally I only use the "add/replace metadata jacket", "insert or replace jacket", and "update metadata" in Modify ePub. I didn't know whether Polish Books would produce the same results, so stuck with what I was familiar (and Modify ePub seemed to have a wider range of other options that might at some point be useful).

I can run Modify ePub against groups/batches of books. Is that what you meant with regards to applying Polish Books in batches? I couldn't find anything that compares those two tools.

[Rob557]

Originally Posted by chaley

you might be able to solve your problem by from time to time polishing all the books in your library.

I guess you're right, whether it is Polish Books or Modify ePub. A full update can take awhile for a large library, but not excessively so. And now that I have a column set up that contains that template, that should help me identify incremental changes for the time being, at least until I muck around with creating, modifying or deleting entire columns of metadata ... in which case I could again later do a full rerun using Modify ePub or Polish Books if it seems worthwhile.

Thanks for your comments Chaley !

[PatNY]

{last_modified:'days_between($, select(formats_modtimes('iso'), 'EPUB'))'}

I thought this would be very useful to have a custom column like this. So I set one up today. But I am getting negative values in the column? Like -1.5 or -1.4? Is this normal? For books that I change the metatdata within minutes of modifying the epub file, shouldn't it say 0? Maybe I set up the custom column wrong?

--Pat

[chaley]

Originally Posted by PatNY

I thought this would be very useful to have a custom column like this. So I set one up today. But I am getting negative values in the column? Like -1.5 or -1.4? Is this normal? For books that I change the metatdata within minutes of modifying the epub file, shouldn't it say 0? Maybe I set up the custom column wrong?

Nope, not wrong. You found a bug. Negative values are computed incorrectly and are between 0 and 2 days too small (too negative). Using your examples, depending on rounding the -1.5 should be -0.4 or -0.5, and the -1.4 should be -0.5 or -0.6. The value -1.9 should probably be -0.0.

I have submitted a fix.

[PatNY]

Thanks, Charles ... So if I understand you correctly, you are saying even when fixed, a small negative value would be normal in some circumstances?

--Pat

[chaley]

Originally Posted by PatNY

Thanks, Charles ... So if I understand you correctly, you are saying even when fixed, a small negative value would be normal in some circumstances?

--Pat

IIRC updating a format doesn't change the modified date, so the format can be newer than the metadata. In that case the column will show a negative number in tenths of days.