Download metadata and covers

[fvdham]

I frequently use the Download metadata and Covers option (ctrl+D) to update metadata.
I recently noticed that the Tag data on the right view is always copied to the left tag field.
Is this a bug or do I do something wrong?

[theducks]

Download (ctrl-D) always copies if blank on the left

Identifiers has a Merge option on the arrow if right clicked
BTW you can also edit any entry on the Left as long as you don't leave it blank (which triggers the copy )

[fvdham]

I am still confused. There is no option to discard the tags on the original field. The tags will always be copied to the downloaded tag field no matter whether that field has an entry or not.
By the way, what is the purpose of the X in the original (right) fields? Clicking on it does not do anything.

[ownedbycats]

I wonder if it was disabled, but not hidden for some reason. It's odd enough that I'll open a ticket.

EDIT: https://bugs.launchpad.net/calibre/+bug/2024366

[fvdham]

The above mentioned problem not only applies to tag's but also to identifiers. I hope that Ownedbycats opened ticket will look into this as well

[fvdham]

I have just updated calibre and indeed the X's are gone now.
But the problem is still there. Tag's and/or Identifiers are always merged from the right to the left field, no matter the left field is blank or not. There is no option to suppress this. It looks like the merge option is always active.

[fvdham]

I still have problems with "Download metadata (ctrl-D)". Is there an option to suppress the setting: "If a downloaded value is blank or unknown, the original value is used" ?

Because:
Tags and Identifiers are always added from the original field to the downloaded field no matter what's in the downloaded field. (Option add to downloaded always on?)
Ratings are always copied to "not rated"
Published date is always copied to "Undefined"

After updating a couple of books I have to do a lot of cleanup work of unwanted or wrong data.

[theducks]

??? Copied is normally: (I always use the Review option on the ctrl-D results)

Downloaded < Existing only if blank
You can use the < to force it to keep your existing FIELD value. Identifiers and Tags has an option (right click the < ) to merge

Also, those results are affected by the settings of Configure Metatadta Download (and the individual sources)

[fvdham]

Originally Posted by theducks

??? Copied is normally: (I always use the Review option on the ctrl-D results)

Downloaded < Existing only if blank
You can use the < to force it to keep your existing FIELD value. Identifiers and Tags has an option (right click the < ) to merge

Also, those results are affected by the settings of Configure Metatadta Download (and the individual sources)

I agree with most of your comments.
However using the review option the Tag and Identifier field are ALWAYS merged to the downloaded fields, which means that the merge to field option is always active.

Is this a bug or do I do something wrong

[theducks]

You can't stop a copy to an empty field. Is this what you are calling 'Merge'?
The Merge option ADDS to the existing VALUES

[fvdham]

I will try to explain what I mean.
When both the original tag field AND the download tag field contains data, the original data is ALWAYS merged to the download field WITHOUT using any of the 2 (right click the <) options. It should only copy data when the download field is empty.
The same applies to the Identifier fields.

[theducks]

It must be a Linux version issue. It does not happen on Windows x64

[fvdham]

I am using Windows 11 Home X64. Calibre 6.23
As far as I remember the problem started a few month ago. I had no problem with it in the past. Are you shure it works ok on your pc? If so, maybe Kovid Goyal can have a look at it.

[theducks]

Originally Posted by fvdham

I am using Windows 11 Home X64. Calibre 6.23
As far as I remember the problem started a few month ago. I had no problem with it in the past. Are you shure it works ok on your pc? If so, maybe Kovid Goyal can have a look at it.

I thought the excess tags came during the Add Books phase (from embedded metadata.
Unlike Get Metadata, you do not get to select those to ignore: [like user]Rating)
Now you got me worried OTOH I am not seeing obvious Identifiers bloat (but I will now watch)

[kovidgoyal]

as far as I know, that has always been the case. Why would you prefer losing existing tags/identifiers when downloading more? For instance, download metadata one book at a time or in bulk without review will also merge these.