[Metadata Source Plugin] Barnes & Noble

[Caleb666]

Hey, you're the one doing the hard work!

By the way, merely looking at the HTML code of the cover page, I spotted:

<a href="imageviewer.asp?ean=9781593080778&amp;imId=7 1155241"><img src="http://images.barnesandnoble.com/presources/images/btn_front_off.gif" width="41" height="25" border="0" title="Front" alt="Front"></a>

Edit: Ok I think I got it. Basically, look at the HREF attribute, and take out the imId parameter, which is "71155241".
Then generate a link as follows:
http://images.barnesandnoble.com/images/<first-4-digits-of-imId>0000/<imId>.JPG

And that will be the front page image.

[Caleb666]

I've got another bug to report!

Certain book title are improperly parsed, for example: ISBN: 9780471741725 http://search.barnesandnoble.com/Cui.../9780471741725

In those kind of books, for some reason it grabs only "Edition 2", instead of taking the whole title.

[tweebee]

Hi, thanks for a nice plug-in! I just noticed something and wondering if it is by design. Sorry, I'm new with calibre.

When downloading metadata and cover by right-clicking on a book, it doesn't show the id for B&N in the book details. If I'm in edit metadata for a book and then click on Download metadata, then the details will have B&N in ids. Is this by design that if I want B&N id, I would have to do the latter process?

[kiwidude]

@Caleb666 - I will take a look later into that, thx again for reporting it.

@tweebee - when you do a download via the right-click, you are leaving it up to Calibre to decide which version of the book results to pick from, in the situation where multiple results come back and it decides that they are not in the same ISBN pool. In that circumstance, it could well be that the decision Calibre makes means that it picks results from other metatadata sources, and discards the B&N result (hence why you see no B&N id). The same thing would happen if you use the Edit Metadata dialog in that particular circumstance, except you are probably manually picking the B&N result in the situation where there are multiple to match.

So in direct answer to your question - if Calibre picks the metadata result which came from B&N, you will see the B&N id immediately appear in the book details panel.

You can usually fix this by doing a second metadata fetch, as the first metadata fetch would most likely have changed the ISBN associated with the book, which *might* result in getting a different match from B&N that is in the same ISBN pool (set of ISBNs identified as representing the same book). However it is usually best to do this via the edit metadata dialog, or else if again there are multiple results it will be left to Calibre to once again decide for itself and you may end up with nothing changed.

[tweebee]

Originally Posted by kiwidude

@Caleb666 - I will take a look later into that, thx again for reporting it.

@tweebee - when you do a download via the right-click, you are leaving it up to Calibre to decide which version of the book results to pick from, in the situation where multiple results come back and it decides that they are not in the same ISBN pool. In that circumstance, it could well be that the decision Calibre makes means that it picks results from other metatadata sources, and discards the B&N result (hence why you see no B&N id). The same thing would happen if you use the Edit Metadata dialog in that particular circumstance, except you are probably manually picking the B&N result in the situation where there are multiple to match.

So in direct answer to your question - if Calibre picks the metadata result which came from B&N, you will see the B&N id immediately appear in the book details panel.

You can usually fix this by doing a second metadata fetch, as the first metadata fetch would most likely have changed the ISBN associated with the book, which *might* result in getting a different match from B&N that is in the same ISBN pool (set of ISBNs identified as representing the same book). However it is usually best to do this via the edit metadata dialog, or else if again there are multiple results it will be left to Calibre to once again decide for itself and you may end up with nothing changed.

Thank you for the further explanations on this! They are helpful in helping me understand this plug-in and how calibre works.

[kiwidude]

Changes in this release:

• Grab the front cover when there are multiple covers available
• Support change to website where wgt-ProductTitle class titles no longer inside a <span>

[Caleb666]

Thank you very much kiwidude, everything's dandy now!

[Caleb666]

For some reason it doesn't seem to show the cover of this ISBN 9781400078974
Any idea why? Because there is a high quality cover on the book's page: http://www.barnesandnoble.com/w/rubi...00078974&itm=2

A similar thing has happened to me with other books that have a good cover, but they don't appear in the results.

[theducks]

Originally Posted by Caleb666

For some reason it doesn't seem to show the cover of this ISBN 9781400078974
Any idea why? Because there is a high quality cover on the book's page: http://www.barnesandnoble.com/w/rubi...00078974&itm=2

A similar thing has happened to me with other books that have a good cover, but they don't appear in the results.

When you don't get the cover you want: Use the Search the Internet Plugin

or simply drag the cover you found onto the Books cover in the Book Browser area or Metadata edit window

[kiwidude]

Changes in this release:

• Change logic for determining image directory to handle smaller numbered images

Thanks Caleb666 for the example where it was going wrong meaning for some books images would not be available for download.

[BradV]

Kiwidude, I have just realized that I have not been getting metadata results from BN for the past few days. Do you know if anybody else is encountering issues?

Thanks!
BradV

[kiwidude]

@BradV - thx for pointing this out - looks like B&N have had another website revamp, sigh. I'll have to modify and push out a new version.

[kiwidude]

Changes in this release:

• Rewritten to support new B&N website for non textbooks

As discussed above B&N have changed their book details page website layout for fiction books.

On the plus side, this should make cover retrieval slightly faster, as they don't do that stupid separate popup window for larger images (so one less round trip for me to get and parse the image).

On the minus side, they really have changed a lot of "stuff". I've done my best and verified it with a few books, but the way some of their HTML is generated is not very "scrape" friendly so there may well be other changes needed.

So if you find a book that you don't get certain details through to Calibre from B&N that you can see on their page, post the link for the B&N page and what you are missing and I can take a look.

[BradV]

Thanks for the update!

[BradV]

While I was thinking of it, I have a feature request for this plugin.

On children's books that have an age range listed, such as:
http://www.barnesandnoble.com/w/i-wa...ri=jon+klassen
and
http://search.barnesandnoble.com/w/h...ica/1100361460

Could you turn the age range into a series of tags?
So that a book that says Age range: 4 - 8 Years
would get tags of:
Age4, Age5, Age6, Age7, Age8

This allows a user to quickly and easily select a tag that corresponds to the age of a child.

I currently do this manually for the children's books in my library (although I "convert" to grade level instead of age).

Anyway, just a request, and thanks again for your work on the plugin.

BradV