# RelatedAmazonProducts.rb
#
# This is a direct port of my RelatedAmazonProduct pyBlosxom implementation. It's a little 
# cludgy because Ruby doesn't have native XML parsing capability as of yet, and I don't want 
# to rely on external libs that may or may not be maintained - I parse the REST response from
# Amazon using some regex trickery instead.
#
# The only difference from my pyBlosxom plugin is that I don't rely on the images response 
# from amazon - I just use Amazon's image server and product ASINs.
# 
# Uses your Amazon Web Services account to query products related to your Folksonomy Tags. This plugin 
# does not require Folksonomy, however if you are using Folksonomy, please get the latest version of it, 
# as there were some potential conlicts over the 'tags' metadata entry.
# 
# Obtains a REST response from the following URL:
# 
# http://webservices.amazon.com/onca/xml?Service=AWSECommerceService
# 	&AWSAccessKeyId=config['aws_access_key']
# 	&AssociateTag=config['aws_associates_id']
# 	&SearchIndex=config['aws_product_type']
# 	&Operation=ItemSearch
# 	&Keywords=<your entries tags>
# 	&Availability=Available
# 	&Merchant=Amazon
# 	[&Publisher=config['aws_favorite_pub']]
# 	[&BrowseNode=config['amazon_browse_node']]
# 
# Specify your AWS Access Key and the type of product to search for in your config.py, 
# like so (defaults are as shown here):
# 
# 	config['aws_access_key']=<my_access_key>	# Your AWS Access key
# 	config['aws_associates_id']=<my_assoc_id>	# Your Amazon Associates ID
# 	config['aws_product_type']=Books			# The type of product to search on.
# 	config['aws_favorite_pub']=<your favorite publish>  # Restricts selections to a particular publisher
# 	config['story_product_count']=4				# The number of books to show on a page.
# 
# I like to use MediumImage and then scale it down using CSS. The SmallImage returns a very low-res
# picture from Amazon, and is useful if you don't have a lot of bandwidth.
# 
# aws_product_type, aws_favorite_pub, and amazon_browse_node can be overridden on a per-story basis by defining 
# metadata with the same key name in your story. So for instance, if your config.py specifies an aws_product_type of "Books", 
# but you just posted about DVDs, you could override this setting for your story by doing:
# 
# 	My Post on DVDs
# 	#tags batman,superman,mighty mouse
# 	#aws_product_type DVD
# 	<p>DVDs are awesome.</p>
# 
# Valid aws_product_type values are [ "Books","Music","DVD","Toys","Video Games","Software","Software Video Games",
# 				    "Electronics","Tools","Sporting Goods","Art Supplies","Kitchen","Gourmet Food",
# 				    "Apparel","PC Hardware","VHS" ]
# 
# aws_product_type defautls to "Books"
# 
# For a list of valid amazon browse nodes, please see:
# 	http://www.amazon.com/gp/aws/sdk/main.html/002-5817805-8594421?s=AWSEcommerceService&v=2005-03-23&p=ApiReference/BrowseNodeValuesArticle
# 
# My site uses BrowseNode 5, which is the code for Computing and Internet under Books in the US.
# 
# Populates a varaible $relatedproducts for use in your story template, that will contain HTML like:
# 
# 	<div id="relatedproducts">
# 		<a class="productlink" href="product url"><img class="productimg" src="product img"/></a>
# 		<a class="productlink" href="product url"><img class="productimg" src="product img"/></a>
# 	</div>
# 
# If you do not want to generate related products for a particular entry, add a "noproducts" metadata entry:
# 
# 	My Entry with No Products
# 	#noproducts

require 'cgi'
require 'net/http'

class Product
	def initialize( asin,title,link,img,relevance )
		@dict = {}
		@dict['asin'] = asin
		@dict['title'] = title
		@dict['link'] = link
		@dict['img'] = img
		@dict['relevance'] = relevance
	end
	
	def []( name )
		@dict[name]
	end
	
	@dict
end

def cb_prepare( app )
	config = app.getConfiguration
	
	return if not app.hasValue?('aws_access_key')
	
	app['story_product_count'] = 5 if not app.hasValue?('story_product_count')

	itemSearchUrl = "http://webservices.amazon.com/"
	itemSearchPath = "/onca/xml?Service=AWSECommerceService&AWSAccessKeyId=" + config['aws_access_key'] + "&Operation=ItemSearch&Merchant=Amazon&ResponseGroup=Small&Sort=relevancerank"
	
	if app.hasValue?('aws_associates_id')
		itemSearchPath = itemSearchPath + "&AssociateTag=" + app.getValue('aws_associates_id')
	end
	
	app['itemsearch'] = itemSearchUrl
	app['itemsearchpath'] = itemSearchPath	
end

# Looks up related products at amazon.com if we're viewing a product directly (i.e. through 
# it's /item/id path).
def cb_story( app )
	entry = app['entry']
	config = app['config']
	pathinfo = app['PATH_INFO']

	return if not app.isPermalink?( pathinfo )	
	return if not app.hasValue?('aws_access_key',entry) or app.getValue('aws_access_key',entry).length == 0
	return if entry.getTags.length == 0 or entry.getMetadata( 'noproducts' )
	
	tags = entry.getTags()

	producttype = app.getValue( 'aws_product_type', entry ) 
	producttype = 'Books' if not producttype

	publisher = app.getValue( 'aws_favorite_pub', entry )
	publisher = "" if not publisher

	browseNode = app.getValue( 'amazon_browse_node', entry )	
	browseNode = "-1" if not browseNode
	
	products = []
	tags.each{ |tag| 
		products.concat( getProductsForTag( app, tag, producttype, publisher, browseNode ) ) 
	}
	
	products = sortProducts( products )

	relatedproductsheader = app.getValue('related_products_header',entry)
	relatedproductsheader = entry['related_products_header'] if entry['related_products_header']

	if products and products.length > 0
		products = products.slice(0, [products.length, app.getValue('story_product_count',entry)].min)
		relatedproducts = "<div id='relatedproducts'>\n<h3>"+relatedproductsheader+"</h3>\n<p>"
		products.each{ |p|
			relatedproducts += "<a class='productlink' href='%s'><img class='productimg' src='%s'/></a>\n" % [ p['link'], p['img'] ]
		}
		relatedproducts += "</p>\n</div>"
		
		app['relatedproducts'] = relatedproducts
	end
end

# This is the meat and potatoes of how we determine which products to 
# show. We first sort based on how many tags a product appeared in. If 
# there is no one product that appeared in more than one tag, we resort
# the whole shebang based on search relevance.
def sortProducts( products )
	ranked = {}
	products.each{ |p|
		if ranked.keys.member?(p['asin'])
			ranked[p['asin']] = [ ranked[p['asin']][0] + 1, p ]
		else
			ranked[p['asin']] = [ 1, p ]
		end
	}
	
	counts = []
	ranked.each{ |asin,pair| counts << pair[0] if pair[0] }
	
	return if counts.min == nil
	
	if (2 > counts.min)
		ranked = {}
		products.each{ |p| ranked[p['asin']] = [ p['relevance'].to_i, p ] }
		ranked = ranked.values.sort{|a,b| a[0] - b[0]}
	else
		ranked = ranked.values.sort{|a,b| b[0] - a[0]}
	end

	products = []
	ranked.each{ |r| products << r[1] }
	
	products
end

def getProductsForTag( app, tag, producttype, publisher, browsenode )
	config = app.getConfiguration
	
	itemSearchUrl = app['itemsearch']  
	itemSearchPath = app['itemsearchpath'] + "&Keywords="+CGI::escape(tag)+"&SearchIndex="+producttype 
	
	if ( publisher.length > 0 ):
		itemSearchPath = itemSearchPath + "&Publisher=" + publisher
	end

	if ( browsenode.to_i >= 0 ):
		itemSearchPath = itemSearchPath + "&BrowseNode=" + browsenode.to_s
	end
	
	url = URI::parse( itemSearchUrl )
	res = Net::HTTP.start(url.host, url.port) {|http| http.get(itemSearchPath)}

	products = []
	relevance = 0
	xmlresp = res.body.gsub( "<Item>", "\n<Item>" ).gsub( "</Item>", "</Item>\n" )

	xmlresp.each{ |line|
		asin = ""
		title = ""
		link = ""
		img = ""
		
		line.scan( /<ASIN>(.*)<\/ASIN>/ ).each{ |x| asin = x[0] }
		line.scan( /<Title>(.*)<\/Title>/ ).each{ |x| title = x[0] }
		line.scan( /<DetailPageURL>(.*)<\/DetailPageURL>/ ).each{ |x| link = x[0] }
		img = "http://images.amazon.com/images/P/" + asin + ".01.TZZZZZZZ.jpg"

		p = Product::new( asin,title,link,img,relevance ) 
		if asin and asin.strip.length > 0
			products << p 
			relevance += 1 
		end
	} if xmlresp
	
	products
end