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

[JSWolf]

Originally Posted by Nicolas F

This is really good! I didn't know it was possible to get a high res cover this easily!

Since I don't set ASIN for the books in my Calibre library I played around and made a version that use the title and author name to search for the ASIN on amazon (I used the amazon metadata plugin from Calibre). I will put it on github.

It will probably need testing, but it's working for me so far!

I also changed the name because since it was too long the cover size wasn't showing.

Thanks for your work

Can you please attach your version of the plugin to a message in this thread? Thanks.

[lbschenkel]

Originally Posted by Nicolas F

This is really good! I didn't know it was possible to get a high res cover this easily!

Since I don't set ASIN for the books in my Calibre library I played around and made a version that use the title and author name to search for the ASIN on amazon (I used the amazon metadata plugin from Calibre). I will put it on github.

It will probably need testing, but it's working for me so far!

I also changed the name because since it was too long the cover size wasn't showing.

Thanks for your work

Hi Nicolas. Thanks so much for your work. As I suspected, once you add scraping to the equation it becomes 80% of the code. :-)

I was planning to do this later because honestly I won't be the audience for this feature — when I download covers, I already determined the ASIN and I want the cover for that specific edition, and that edition only. But since you were kind enough to contribute with the implementation, I can incorporate it now.

Is it OK with you if I incorporate your code in the plug-in? Let me know how you want attribution.

I will improve upon this to scrape Goodreads as well since Amazon will not always show you old all existing Kindle editions, but Goodreads usually does. I will make it configurable so you can enable/disable them individually.

How does that sound?

[JSWolf]

Scrape Goodreads for the ASIN or scrape Goodreads for the cover?

[JSWolf]

@Nicolas, the plugin is working rather well. Thank you for your changes to it.

[Nicolas F]

Originally Posted by lbschenkel

Hi Nicolas. Thanks so much for your work. As I suspected, once you add scraping to the equation it becomes 80% of the code. :-)

I was planning to do this later because honestly I won't be the audience for this feature — when I download covers, I already determined the ASIN and I want the cover for that specific edition, and that edition only. But since you were kind enough to contribute with the implementation, I can incorporate it now.

Is it OK with you if I incorporate your code in the plug-in? Let me know how you want attribution.

I will improve upon this to scrape Goodreads as well since Amazon will not always show you old all existing Kindle editions, but Goodreads usually does. I will make it configurable so you can enable/disable them individually.

How does that sound?

Yes, I was surprised as how little code you used to download the covers (there might be a shorter way to code what I did since I'm no professional)!

Of course you can incorporate it, I made a pull request on github if you're OK with my additions.
For the attribution, you can just put my name as a contributor
Also I didn't add a license since there was none, but you should add one (GPL v3 is commonly used)

A setting would be a good idea if you add more sources (I already added one to choose the number of covers to download). And amazon might not even be needed if the goodreads database is enough (if I remember correctly goodreads is now owned by amazon)

Scrape Goodreads for the ASIN or scrape Goodreads for the cover?

For the ASIN, if you want to get the covers the Goodreads metadata plugin already does this.
And thanks for the feedback!

[milady133]

Just a question about the plugin, when there's no ASIN in the book metadata and searchs for it based on title & author, what does the plugin do when there's more than one?
I'll probably use it only setting the ASIN previously to make sure I get the cover I want in case there's more than one edition, but I'm curious.
Thanks

[Nicolas F]

Originally Posted by milady133

Just a question about the plugin, when there's no ASIN in the book metadata and searchs for it based on title & author, what does the plugin do when there's more than one?
I'll probably use it only setting the ASIN previously to make sure I get the cover I want in case there's more than one edition, but I'm curious.
Thanks

The way it's done for now, the plugin search amazon.com for the title + the first author, and it gets the books in the order they appear on the search page (if the books have the good author). You can get multiple covers this way, the good one is not necessarily the first, there is a setting to select the number of covers to download.

[milady133]

Thanks, I've yet to try it, i haven't been at home this weekend, so I haven't installed it.

[Ravensknight]

I've added this plugin and started using it. Now, when I add amazon:B00E257WXW to the identifiers field, I'll get an amazon cover that is equivalent to the highest google image. But when I go to the website you reference and input that same asin, I get a cover that is almost twice the resolution. I've just tested it once with the above asin, not any others. Should I try some others?

Any idea why this might be happening?

[jackie_w]

Originally Posted by Ravensknight

Any idea why this might be happening?

Have a look at your Tweak setting in Preferences > Change calibre behaviour > Tweaks - 'The maximum width and height for covers saved in the calibre library'

e.g I have mine set at

maximum_cover_size = (1500, 2100)

I believe calibre's Edit metadata Download cover option uses any limit you have set here.

[Ravensknight]

Originally Posted by jackie_w

Have a look at your Tweak setting in Preferences > Change calibre behaviour > Tweaks - 'The maximum width and height for covers saved in the calibre library'

e.g I have mine set at

maximum_cover_size = (1500, 2100)

I believe calibre's Edit metadata Download cover option uses any limit you have set here.

Thanks! That was it. Once I changed that setting, the higher resolution covers were available. Man, learning something new every day with Calibre...

[lbschenkel]

I have published a new version for testing, 0.3.0, that automatically finds ASINs via goodreads.com (Amazon disabled in this version). You can compare it with version 0.2.0 (the one from "Nicolas F") and compare the results. I'm trying to determine the difference in results from both sites and if it's necessary to scrape both.

As usual, any feedback is more than welcome.

[Nicolas F]

Running the two versions side by side, every time the goodreads version could at least find the same covers or found more. So it doesn't seem necessary to scrape both (as stated above I believe that since amazon bought goodreads the database are at least partially interconnected)

It could be interesting to scrape the other languages amazon websites since goodreads have a very good database of english titles but isn't has good for foreign editions (searching for french titles in my case). But I haven't included it in what I coded. It could be done with an option to chose which stores to use.

Edit: all my books have a goodreads id, so the plugin is searching using this id.

[Dopedangel]

Originally Posted by Nicolas F

This is really good! I didn't know it was possible to get a high res cover this easily!

Since I don't set ASIN for the books in my Calibre library I played around and made a version that use the title and author name to search for the ASIN on amazon (I used the amazon metadata plugin from Calibre). I will put it on github.

It will probably need testing, but it's working for me so far!

I also changed the name because since it was too long the cover size wasn't showing.

Thanks for your work

Thanks it works for most books though I have to say google image gives me better results it has improved a lot as out of 10 books I tried google found 9 covers amazon found 8 but 1 of those was not in english.

[Nicolas F]

Originally Posted by Dopedangel

Thanks it works for most books though I have to say google image gives me better results it has improved a lot as out of 10 books I tried google found 9 covers amazon found 8 but 1 of those was not in english.

I don't know wich version you tried, the one in the post you cited is not the latest. lbschenkel made a new one that scrapes goodreads to find the ASIN. It can find more covers (but it will also find more covers in the wrong language)