Unable to download metadata without ISBN present

[Overquoted]

Windows 11 - Calibre version 9.8.0

When attempting to download metadata with just the title and author, it searches for approximately 1-3 seconds, then returns an error. If I input the ISBN, it returns results. Issue was not occurring as of a day or two ago. I am unsure when I updated to 9.8.0.

• Occurs with multiple metadata sources.
• Tried adding a source I had not been using, issue remained.
• Occurs with multiple different books.
• Tried with my normal 10-seconds for first match found and with 30 seconds.
• Both Calibre and plugins are fully up-to-date.
• Tried restarting PC.
• Confirmed Calibre is allowed through Windows Defender.
• Tried the removal of metadata-sources-cache.json and restarting calibre.
• Tried reinstalling a previous Calibre version.

Ultimately, if I remove the entire Calibre configuration folder, I can download metadata again. But I'd rather not do this, if possible, as I've been using Calibre for years and have quite a lot of customization of settings and plugins.


Error:

Running identify query with parameters:
{'title': 'Confessions of a Funeral Director', 'authors': ['Caleb Wilde'], 'identifiers': {}, 'timeout': 30}
Using plugins: Goodreads (1, 8, 5), Amazon.com (1, 3, 18)
The log from individual plugins is below

****************************** Goodreads (1, 8, 5) ******************************
Found 0 results
Downloading from Goodreads took 0.23729872703552246
identify - start. title=Confessions of a Funeral Director, authors=['Caleb Wilde'], identifiers={}
Querying: https://www.goodreads.com/search?sea...e=books&search[query]=Confessions+of+a+Funeral+Director+Caleb+Wilde
No goodreads id via ISBN
Failed to get raw result for query: 'https://www.goodreads.com/search?search_type=books&search[query]=Confessions+of+a+Funeral+Director+Caleb+Wilde'

************************************************** ******************************

****************************** Amazon.com (1, 3, 18) ******************************
Found 0 results
Downloading from Amazon.com took 1.9340929985046387
User-agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/146.0.0.0 Safari/537.36
Server: auto
Making google query: https://www.google.com/search?q=Conf...www.amazon.com
Failed to find any results on results page, with title: Confessions of Funeral Director Caleb Wilde site:www.amazon.com - Google Search
No search engine results for terms: Confessions of Funeral Director Caleb Wilde
Trying the bing search engine instead
Making bing query: https://www.bing.com/search?q=Confes...www.amazon.com
Bing returned no results
No search engine results for terms: Confessions of Funeral Director Caleb Wilde
No matches found with query: ['Confessions', 'of', 'Funeral', 'Director', 'Caleb', 'Wilde']

************************************************** ******************************
The identify phase took 2.00 seconds
The longest time (1.934093) was taken by: Amazon.com
Merging results from different sources
We have 0 merged results, merging took: 0.00 seconds

[kovidgoyal]

Customizr the amaon plugin to download from amazon.com not a search engine since all search engines block calibre now adays

[Overquoted]

Originally Posted by kovidgoyal

Customizr the amaon plugin to download from amazon.com not a search engine since all search engines block calibre now adays

What about GoodReads? It happened with Barnes and Noble, too, but I don't typically use it.

Thank you for helping, sir.

[Overquoted]

I think I fixed it, thanks!!

[getthething]

Originally Posted by Overquoted

I think I fixed it, thanks!!

Were you able to fix the Goodreads plugin or did you just switch to the Amazon one?

[toomuchreading]

Originally Posted by Overquoted

I think I fixed it, thanks!!

What did you do to fix it? Inquiring minds would like to know!

[mj52]

I am having the same issue (no metadata returned when ISBN is blank) and across multiple plugin sources. Here is some further detail:

75% of the time I get the same error as above.
10% of the time I can get a book via Amazon, tho it is always the audio book metadata which is not what I have. Author field contains multiple names.
10% of the time I also get two unrelated books (same ones each time) plus the one I am looking for.
5% of the time I get only those two unrelated books.

I've tried Amazon (Amazon plugin is configured to go to Amazon), Goodreads (really wants an ISBN), Google (error says too many requests tho I was only doing a single book), Fantastic Fiction (forbidden error there), and Storygraph plugins. Only Amazon seems to return any metadata if it gets through as per the above stats.

What is odd, is that when I imported several books, none of them had an ISBN number - that is very unusual. 99% of books I've imported into Calibre contain an ISBN or ASIN number - it looks almost like they were stripped out?

Plus even if I manually add an ISBN, the Calibre plugin still can't find the metadata, and returns the error.

Like the OP, I have customizations (user columns, some rules).

Last, I was on v9.6 and had the error. I upgraded to v9.8 and to the 2 updated Goodreads plugins, thinking it was me. Alas, I continue to have the error.

Any ideas on how to fix?

[Rellwood]

I use Goodreads Sync to link a book to Goodreads to get their identifier so that the isbn can be added as well. When you link to Goodreads there is an option to get the isbn if there isn't one in Calibre.

Not perfect - but a workaround

[mj52]

Originally Posted by Rellwood

I use Goodreads Sync to link a book to Goodreads to get their identifier so that the isbn can be added as well. When you link to Goodreads there is an option to get the isbn if there isn't one in Calibre.

Not perfect - but a workaround

Thanks Rellwood! That is a good workaround idea. I found the ISBN extractor plugin which also pulls in an ISBN from the book itself if it it there. Both are alternatives.

It is odd that Calibre can no longer locate a book without an identifier, and sometimes even with an identifier it fails.

That said, this is fairly recent and it seems others running into the same issue.

Hopeully it's a simple bug Kovid will be able to track down so at least a basic author/title search will work again when doing a metadata download.

Thanks again for the idea!

[kovidgoyal]

There's no bug, metadata download is less reliable these days because most websites have started blocking bots because they want to try to keep their contents safe from AI.

[toomuchreading]

Originally Posted by kovidgoyal

There's no bug, metadata download is less reliable these days because most websites have started blocking bots because they want to try to keep their contents safe from AI.

Dreaming :


Book sites and Calibre come to an agreement to somehow just let Calibre thru without any $$ being spent. I can understand not wanting AI to get a hold of the data. I know that probably won't happen since sites want to make $ any way they can but one can dream.


Book publishers also need to ensure decent metadata as well. I've seen brand new books with complete metadata to no metadata and everything in between. That would mean we don't have to rely on metadata downloads.


Now that I know the reason for it not working just dreaming out loud...

[theducks]

Originally Posted by toomuchreading

Dreaming :


Book sites and Calibre come to an agreement to somehow just let Calibre thru without any $$ being spent. I can understand not wanting AI to get a hold of the data. I know that probably won't happen since sites want to make $ any way they can but one can dream.

Book publishers also need to ensure decent metadata as well. I've seen brand new books with complete metadata to no metadata and everything in between. That would mean we don't have to rely on metadata downloads.

Now that I know the reason for it not working just dreaming out loud...

Metadata is kept in the OPF. Calibre normally(unless YOU changed the Add Books setting) reads that during an 'add'. Thus there is no requirement that you need to download.
Even if you had used filename as the metadata source, the book in the MDE can be forced to supply the metadata and / or cover (assumes you have NOT used Calibre tools the embed the Calibre version) . The 2 icons on the left side of the formats block do this.

None of this supplies good metadata that was not there (like series) or Authors of stories in Anthologies (a few do, but many collections don't, and online metadata may only have the names in the comments )
I get the feeling the some folk think that if they keep hammering these source sites, that the data will magically improve . IMHO hand review / adjustment is the only way to approach 'perfect'

You can edit the left side of the ctrl-D 'get metadata' or use the arrow next to the field to use YOUR current value in that field.

[ownedbycats]

For what it matters, for non-ISBN books (self-pub or older physical) can manually look up edition on GR and paste the ID, it then downloads. Annoyance but works.