[Metadata Source Plugin] Goodreads

[kiwidude]

Changes in this release:

• Handle change to Goodreads website which prevented title/author results returning

Goodreads have been "fiddling" with their website again it seems, this should allow title/author type searches to start matching again.

EDIT: I forgot to update the version in the zip - please download again if you were the first couple of users and plugin updater keeps telling you you have 1.0.5.

[hakan42]

Search for a book with only goodreads id does not work, you have to add at least the title.

One of my older books had the title "Dark Imperium", and "Unknown" as author. I used the "Link to goodreads..." menu item to add the id "goodreads:588334" to that book, then went into the "Edit Metadata" dialog and tried to download metadata. The Goodreads Metadata plugin failed to return any data:

****************************** Goodreads ****************************** 
Request extra headers: [('User-agent', 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:2.0.1) Gecko/20100101 Firefox/4.0.1')] 
Found 0 results 
Downloading from Goodreads took 0.551999807358 
Goodreads book url: u'http://www.goodreads.com/book/show/588334'
Could not find title/authors/goodreads id for u'http://www.goodreads.com/book/show/588334'
Goodreads: u'588334' Title: 'Dark Imperium (Warhammer 40,000 Novels\n          \n            (Warhammer 40,000)' Authors: None 

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

What I don't really understand is, according to the plugin source (version 1.0.6), __init__.py, the plugin should not perform a search if the goodreads_id is available and good:

        goodreads_id = identifiers.get('goodreads', None)
        isbn = check_isbn(identifiers.get('isbn', None))
        br = self.browser
        if goodreads_id:
            matches.append('%s/book/show/%s'%(Goodreads.BASE_URL, goodreads_id))
        else:

[kiwidude]

@hakan42 - I just tried exactly your described scenario and it worked perfectly fine for me. I did the following:

Created an empty book with title "Dark Imperium", author "Unknown"
Set the goodreads id to goodreads:588334
In the Edit Metadata->Configure download metadata screen, had only the Goodreads plugin enabled, with all fields to download checked.
Clicked the Download metadata buton

Starting download 
Query: title:Dark Imperium; authors:Unknown; goodreads:588334 
Running identify query with parameters: 
{u'authors': None, u'identifiers': {u'goodreads': u'588334'}, u'timeout': 30, u'title': u'Dark Imperium'} 
Using plugins: Goodreads 
The log from individual plugins is below 

****************************** Goodreads ****************************** 
Request extra headers: [('User-agent', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/525.19 (KHTML, like Gecko) Chrome/0.2.153.1 Safari/525.19')] 
Found 1 results 
Downloading from Goodreads took 1.53900003433 


--- 
Title               : Dark Imperium (Warhammer 40,000 Novels (Warhammer 40,000)
Author(s)           : Marc Gascoigne & Andy Jones
Publisher           : Games Workshop
Rating              : 3.17
Published           : 2001-01-30T00:00:00+00:00
Identifiers         : isbn:9780743411622, goodreads:588334
Comments            : 
In the war-torn 41st millennium humanity stands on the shores of damnation, their only saviour is the Immortal God Emperor and the massed armies of the Imperium, in this searing anthology of Warhammer 40,000 stories torn from the pages of Inferno! magazine.
 
Goodreads book url: u'http://www.goodreads.com/book/show/588334' 

******************************************************************************** 
The identify phase took 1.60 seconds 
The longest time (1.539000) was taken by: Goodreads 
Merging results from different sources and finding earliest publication dates 
We have 1 merged results, merging took: 0.00 seconds

As you can see it found and parsed the result.

I am running with the latest Calibre source code, though with only one plugin enabled the changes Kovid has made should not make a difference

[hakan42]

Yes, but try an empty title with empty authors and only the goodreads id set. The book had in my case a very broken title:

'Dark Imperium (Warhammer 40,000 Novels\n          \n            (Warhammer 40,000)'

so I removed title and authors in the "edit metadata" dialog, saved, and tried the "download metadata" step. What I would have expected from reading the source is that now I should get correct metadata without any searching involved...

Anyway, setting the title to something sensible works, so my problem might only be a corner case that does not happen in the normal flow of work... Back to

[kiwidude]

As I said above that is with no authors set, you can see it in the log where it says:

Running identify query with parameters: 
{u'authors': None...

As per the log you originally posted the plugin did the right thing by connecting directly to the url for the goodreads id as per this part of your log:

Goodreads book url: u'http://www.goodreads.com/book/show/588334'

The line underneath where it says "Could not find title/authors/goodreads id" is indicating that in attempting to scrape that URL above it could not find sufficient of title/authors/goodreads id to qualify it as a valid title. Perhaps you just had a blip on the internet which gave a dodgy web response back. Certainly it is all working fine for me so unless someone can post a scenario to replicate I'll let this one slide.

[Noughty]

I was wondering if it is possible for a book linked to goodreads (link is valid, shows correct book) to not download any matadata? Downloading metadata simple fails. What could be the reason for this?

When using link to goodreads it finds book fine for linking. I tried adding ISBN still no results. Not getting results isn't strange, strange is that book is linked with goodreads from which I download metadata.

[kiwidude]

Use the Edit Metadata window to do the download, so you can see the results (to see if Calibre is choosing a book from a different ISBN pool). Also you can take a look at the log to see what it says. And finally if you absolutely want to guarantee that it downloads from Goodreads with a linked book, only enable the Goodreads plugin.

[Noughty]

All done. No success. I can view linked book but metadata downlaod fails. Only goodreads metadata is on.

Log:
Error parsing title and series for url: u'http://www.goodreads.com/book/show/[book number]'
Traceback (most recent call last):
File "calibre_plugins.goodreads.worker", line 108, in parse_details
File "calibre_plugins.goodreads.worker", line 227, in parse_title_series
ValueError: invalid literal for float(): 2008 Jan.

Could not find title/authors/goodreads id for u'http://www.goodreads.com/book/show/[book number]'

[kiwidude]

@Noughty - unfortunately you have not posted the book details. If you don't want to publicly display them just send me a PM - it looks like you have found some kind of problem with scraping the the data for that particular book.

[kiwidude]

Changes in this release:

• Ensure a "close but not quite" series # does not throw an error within the plugin.

This release addresses a problem where the series name/number specified on Goodreads does not match a valid calibre format. So for instance:
MyTitle (MySeries #2008 Jan)

In this circumstance the series information will be treated as part of the title.

[Noughty]

Thanks. Also a late thanks for plugin linking with goodreads on metadata download. I barely need to use Goodreads sync link to goodreads book. I really like how it works now

[jersysman]

I must be doing something wrong. When I download Metadata, I see the tags that are downloaded (I have it set for 5). I select the cover and go to the Basic Metadata screen. I select OK, but on the main Calibre screen, my book has way more than the 5 tags I wanted to download. What am I doing incorrectly?

[DoctorOhh]

Originally Posted by jersysman

I must be doing something wrong. When I download Metadata, I see the tags that are downloaded (I have it set for 5). I select the cover and go to the Basic Metadata screen. I select OK, but on the main Calibre screen, my book has way more than the 5 tags I wanted to download. What am I doing incorrectly?

The metadata download will not remove any existing tags, it will just add to it. I remove all tags prior to downloading metadata from Goodreads.

Also I went and configured the Google plugin and the Fantastic Fiction plugin so tags from these sources were not downloaded. I only leave Goodreads configured to download tags.

[jersysman]

Thanks, Walt. I had previously removed the tags from being sent by Google. Okay... so I get rid of the tags first and then search. I thought it would overwrite it. Thanks for the info!!

EDIT: Worked great. Thanks!

[Noughty]

I wanted to ask if it's possible to fix authors of anthologies. When I download metadata of anthologies with a lot of authors I get something like this

Author1, Author2, Author3, Author4...More Author5...less

This gives 'new' authors, like Author4..more and Author5...less (can't remember for sure, since I just fixed all these)
This happens if in goodreads you have to press more to see all authors. Since the same is with book description and it's possible to download correctly this probably is possible too.