[GUI Plugin] Import List

[tagehring]

Originally Posted by tagehring

I have a CSV file with columns for UUID, title, author, and author_sort for each of the books in my library that I've manually corrected in Excel from catalog output. I've tried using Import List to import the corrections via CSV file but have run into a few problems.

1. There doesn't seem to be a way to correct the author_sort value this way at all.
   
2. If the author field has significantly changed, Import List returns an error even if the title is identical. When I use the plugin to match the CSV row to the book, it doesn't overwrite the correct author information from the CSV file.
3. To solve #2 another way, despite manually matching the "ID" column of my CSV file to the ID:uuid identifier, I cannot get it to index using uuid. Every row returns an error. (See attached.)

I am at my wit's end trying to get this to work. Any help is appreciated.

Adding screenshots of the problem after fixing #2 above. I'm trying to map the author_sort to a user-created Author Sort column, I know it's not the default author_sort metadata tag. But I would prefer overwriting the default author_sort via CSV if possible.

[chaley]

Originally Posted by tagehring

Adding screenshots of the problem after fixing #2 above. I'm trying to map the author_sort to a user-created Author Sort column, I know it's not the default author_sort metadata tag. But I would prefer overwriting the default author_sort via CSV if possible.

I suspect that it isn't possible without changing the plugin. The problem: depending on the API used, changing the author causes calibre to recompute the author sort. If the plugin sets author after author_sort then the recomputation will overwrite what was set. If my guess is correct then the fix would be for the plugin to ensure that author_sort is always set after author.

The most reliable way to do this is an expansion of what you are already doing.

1. Clear all values from the sort custom column using bulk metadata edit.
2. Run Import list.
3. Select all books now having values in the custom column using #column_key:true
4. Use bulk metadata edit / search and replace to copy the custom column to the author_sort column.

You could use the Action Chains plugin to do all the above steps in one chain.

[tagehring]

Just as a friendly note to kiwidude, there was a crucial step I was missing and didn't find it anywhere in the documentation.

There's no mention of the "Update Metadata" button in Step 2. As I discovered, unless you select all of the books you want to commit changes to and then click this button and select which fields you want updated, nothing happens. From the UI it wasn't immediately obvious what it did or that it was necessary until I hovered over it with my mouse and saw the rollover text.

It might help future users having the same problem to add this scenario (call it Replacing Existing Metadata, maybe) as a usage example on the wiki.

[tagehring]

Originally Posted by chaley

I suspect that it isn't possible without changing the plugin. The problem: depending on the API used, changing the author causes calibre to recompute the author sort. If the plugin sets author after author_sort then the recomputation will overwrite what was set. If my guess is correct then the fix would be for the plugin to ensure that author_sort is always set after author.

The most reliable way to do this is an expansion of what you are already doing.

1. Clear all values from the sort custom column using bulk metadata edit.
2. Run Import list.
3. Select all books now having values in the custom column using #column_key:true
4. Use bulk metadata edit / search and replace to copy the custom column to the author_sort column.

You could use the Action Chains plugin to do all the above steps in one chain.

Now that I've got import working reliably and overwriting data the way I want, I think I'm going to just replace the Author column and then tweak the author sort algorithm to include some of the prefixes/suffixes/copywords it's missing and then recompute the values.

[kiwidude]

Release Notes
https://github.com/kiwidude68/calibr...rt_list-v1.9.2

[thinktwice2]

Hello

Created a CSV file with a book using the Bookshelf app. Loaded into "Import List" (tab "CSV File"). The book is displayed under "Contents". Title and author recognised correctly.

"Match Method" cannot be changed from Title/Author to ISDN.

"Step 2"
No Match


Greetings

[Wolgan]

Is there any way to turn off the tags column on the Step 2 window or prevent it from auto-sizing width? If there are a lot of tags on a book it can get huge, and runs way off the screen. If the columns are custom columns, it seems like they always populate to the right of that column, so I either have to do a lot of scrolling, or spend time shrinking that column down one screen at a time until I can have all the columns visible.

[kiwidude]

Originally Posted by thinktwice2

Hello

Created a CSV file with a book using the Bookshelf app. Loaded into "Import List" (tab "CSV File"). The book is displayed under "Contents". Title and author recognised correctly.

"Match Method" cannot be changed from Title/Author to ISDN.

"Step 2"
No Match


Greetings

Can you send me the CSV file? Can you also provide screenshots of the CSV File page so I can see what your import page looks like?

You are saying that you definitely have the specified book in the CSV file in calibre and it is stored with the same title, that is why no match is surprising you?

And if all of that is true has this worked for you in the past/have you tried with calibre 6.x to confirm it is not an issue specific to calibre 7?

[kiwidude]

Originally Posted by Wolgan

Is there any way to turn off the tags column on the Step 2 window or prevent it from auto-sizing width? If there are a lot of tags on a book it can get huge, and runs way off the screen. If the columns are custom columns, it seems like they always populate to the right of that column, so I either have to do a lot of scrolling, or spend time shrinking that column down one screen at a time until I can have all the columns visible.

Hi - are you talking about the right-hand side grid in Step 2. That currently has Title, Authors, Series, Tags, Publisher columns?

The only column to the right of Tags is Publisher - are you finding the need to resize columns just so you can see the Publisher column? That sems a strange thing to want to need - I would normally expect just Title/Authors/Series to be the columns you actually want to look at and Tags is to the right of those so it doesn't matter how wide it gets?

Perhaps you are seeing something different - picture worth a thousand words and all that - attach or send me a screenshot of what you mean so I can understand how this is a problem. Certainly I could restrict the column on the right to a fixed width so it doesnt autosize by default but perhaps you are talking about somewhere else...

[Wolgan]

Originally Posted by kiwidude

Hi - are you talking about the right-hand side grid in Step 2. That currently has Title, Authors, Series, Tags, Publisher columns?

The only column to the right of Tags is Publisher - are you finding the need to resize columns just so you can see the Publisher column? That sems a strange thing to want to need - I would normally expect just Title/Authors/Series to be the columns you actually want to look at and Tags is to the right of those so it doesn't matter how wide it gets?

Perhaps you are seeing something different - picture worth a thousand words and all that - attach or send me a screenshot of what you mean so I can understand how this is a problem. Certainly I could restrict the column on the right to a fixed width so it doesnt autosize by default but perhaps you are talking about somewhere else...

Nope, that's the exact spot I'm referring to!

I use a few custom columns, in the attached example I'm using #collections to group things like the various ebook Humble Bundles I pick up.

As you can see the tags column is quite wide, and my custom column is the one at the end. Changing the order of columns in the csv seems to have no effect on the order, any custom column populates to the right of the mandatory columns in that table.

[kiwidude]

@Wogan - thanks for the screenshot. Without looking at this plugin for so long I didnt realise it is possible to have custom columns appear on the right. Certainly agree it makes sense to restrict default width of that column, I had just wanted to make sure we were talking about the same place. Will take a look later…

[kiwidude]

Release Notes
https://github.com/kiwidude68/calibr...rt_list-v1.9.3

This incorporates a number of changes to that Step 2 page to not auto-size the matched book columns, but also adding a splitter between the two grids to make it easier for resizing the screen to allow more flexibility.

[Wolgan]

Originally Posted by kiwidude

Release Notes
https://github.com/kiwidude68/calibr...rt_list-v1.9.3

This incorporates a number of changes to that Step 2 page to not auto-size the matched book columns, but also adding a splitter between the two grids to make it easier for resizing the screen to allow more flexibility.

Just tested with a few of my different imports and this works great! I love the splitter between the grids, that's a great idea as well!

Thank you!

[kiwidude]

Release Notes
https://github.com/kiwidude68/calibr...rt_list-v1.9.4

Fix for calibre 7.