[Metadata Source Plugin] Kindle (Amazon) High-res Covers

[davidfor]

Originally Posted by JSWolf

What would the error be then? Can it be fixed fairly easily?

No idea as I don't know what "calibre_plugins.dummy1.__init__" and if it is the plugin, then line 31 does not make sense for that error.

[DNSB]

Perhaps Jon could remove the plugin and re-install? The plugin works here so it seems somewhat logical that the issue is with his installation.

[JSWolf]

Originally Posted by DNSB

Perhaps Jon could remove the plugin and re-install? The plugin works here so it seems somewhat logical that the issue is with his installation.

Thanks. The reinstall works perfectly.

[Tenome]

How easy would it be to add other sites? I want to add Booklive for Japanese books since they have a consistent cover URL scheme. Maybe Amazon JP too.

Booklive example:
https://booklive.jp/product/index/ti...859/vol_no/001

The high-res cover's URL is:
https://res.booklive.jp/331859/001/thumbnail/X.jpg

So the format would be:
https://res.booklive.jp/<title_id>/<vol_no>/thumbnail/X.jpg

[lbschenkel]

Originally Posted by Tenome

How easy would it be to add other sites? I want to add Booklive for Japanese books since they have a consistent cover URL scheme. Maybe Amazon JP too.

It depends on how hard scraping the site would be and how good the search capabilities are.

However, I'm not inclined to add more (non-Amazon) sites to this plug-in: it's called "Kindle high-res covers" after all, and I want to keep it simple and laser-focused. That said, the plug-in is open source and it shouldn't be too hard for someone motivated enough to clone it for a different site.

[Lily7]

Hello -

This plugin seems to have stopped working for me. I uninstalled it and installed it again, and no change.

This is what shows when I view log, but these are the same results I get for any book I try to use this plugin with:

Failed to download valid cover
Took 3.095991849899292 seconds
ASIN not present in metadata
Goodreads id present in metadata: 16066971
Fetching book details: https://www.goodreads.com/book/show/16066971
ISBN present in metadata: %s 9781250037633
Searching Goodreads for book: https://www.goodreads.com/search?q=9781250037633
Fetching book details: https://www.goodreads.com/book/show/16066971-scarlet

Can anyone help?

[DNSB]

Originally Posted by Lily7

Hello -

This plugin seems to have stopped working for me. I uninstalled it and installed it again, and no change.

This is what shows when I view log, but these are the same results I get for any book I try to use this plugin with:

Failed to download valid cover
Took 3.095991849899292 seconds
ASIN not present in metadata
Goodreads id present in metadata: 16066971
Fetching book details: https://www.goodreads.com/book/show/16066971
ISBN present in metadata: %s 9781250037633
Searching Goodreads for book: https://www.goodreads.com/search?q=9781250037633
Fetching book details: https://www.goodreads.com/book/show/16066971-scarlet

Can anyone help?

You need to have an ASIN present in the metadata for the plugin to function. No ASIN, no images.

You can try downloading the metadata from Amazon and if that succeeds the ASIN will be automatically populated to the Ids file or you can manually search the book on Amazon and copy the ASIN using mobi-asin: or amazon: format in the Ids field.

The plugin is based off the code used in the How to Get High Resolution Cover Image of Any Amazon Kindle eBooks webpage. As you will notice, that page does two types of search but both are keyed off the ASIN and as far as I have been able to tell in limited testing, the second portal does not seem to return any results other than access denied.

[Lily7]

Originally Posted by DNSB

You need to have an ASIN present in the metadata for the plugin to function. No ASIN, no images.

You can try downloading the metadata from Amazon and if that succeeds the ASIN will be automatically populated to the Ids file

That fixed it. Thank you!

[Tenome]

Would it be possible to add an option to automatically reject images that are smaller than the current cover's resolution?

[lbschenkel]

Originally Posted by Tenome

Would it be possible to add an option to automatically reject images that are smaller than the current cover's resolution?

Maybe that is technically possible, I would need to take a look at the Calibre plug-in API. But that said: the results are already sorted from largest to smallest. Doesn't that work out for you?

I feel that such a setting would be way more useful in the Calibre side, since it would apply to all plug-ins. I don't see filtering inside specific plug-ins as the optimal choice, as others will still populate with their own results. Even if all plug-ins implement such functionality, it's an inefficient duplication of effort. If done in the Calibre side, it could be much better from a GUI perspective: it could present the smaller covers grayed out, or marked in a different way, or even completely hidden as you want.

[JSWolf]

This plugin used to work without needing to put the ASIN in the identifier. Can it be fixed to once again work the same way?

[Tenome]

Originally Posted by lbschenkel

Maybe that is technically possible, I would need to take a look at the Calibre plug-in API. But that said: the results are already sorted from largest to smallest. Doesn't that work out for you?

I feel that such a setting would be way more useful in the Calibre side, since it would apply to all plug-ins. I don't see filtering inside specific plug-ins as the optimal choice, as others will still populate with their own results. Even if all plug-ins implement such functionality, it's an inefficient duplication of effort. If done in the Calibre side, it could be much better from a GUI perspective: it could present the smaller covers grayed out, or marked in a different way, or even completely hidden as you want.

Hm, this probably is a request for Caliber itself, but I don't think they'll implement it. Specifically, this is for the bulk metadata download feature, which only grabs one cover. I just asked here since I'm grabbing Kindle covers specifically.

[DNSB]

Originally Posted by JSWolf

This plugin used to work without needing to put the ASIN in the identifier. Can it be fixed to once again work the same way?

AFAIR, the ASIN was inserted into the identifier list when I downloaded metadata from Amazon which is getting iffier these days with the anti-DDOS precautions most web sites seem to be using. I'm finding myself having to use a web browser to grab the ASINs and manually insert them before the metadata download step.

[DNSB]

Originally Posted by Tenome

Hm, this probably is a request for Caliber itself, but I don't think they'll implement it. Specifically, this is for the bulk metadata download feature, which only grabs one cover. I just asked here since I'm grabbing Kindle covers specifically.

Personally, I avoid bulk metadata download like a plague. It's one of the best ways of getting my IP temporarily banned. See my previous message and anti-DDOS.

As for returning multiple covers? I seem to remember the person who asked about that previously was recommended to use the individual metadata download.

[lbschenkel]

I took a quick look at the API and unless I missed something it doesn't seem possible. Calibre doesn't pass the size of the existing cover when invoking metadata plug-ins.

To make that possible Calibre would need to extend its API. But as I explained, if a change is needed in Calibre to enable this feature than it's actually better to implement the feature itself in Calibre: the user experience can potentially be much better than doing it on the plug-in side, which doesn't have control of the UI and how results are presented.