basic_to_solr.rb

# Example of pushing stuff to solr with solrj in jruby

require 'rubygems'

# Load any .jar files your want with "require '../path/to/jarfile.jar'"
# For both marc4j4r and jruby_streaming_update_solr_server, if you load the
# appropriate jar first that's the version that will be used. If not,
# we fall back on the one shipped with the gem

# require '../jars/myjavacode.jar'

require 'jruby_streaming_uddate_solr_server'
require 'threach'
# require 'marc4j4r' # if that's your thing

########## config #########

# URL to solr
solrURL = 'http://localhost:8983/solr' 

# Size of the suss queue 
sussQueueSize = 20

# Number of threads to use to push stuff to solr
sussThreads = 1 

# Number of threads to use to consume stuff and push it into the suss
consumerThreads = 2

##### end config #########

suss = StreamingUpdateSolrServer.new(solrURL,sussQueueSize,sussThreads)

# If you want to use threach (threaded each) you just need some sort of a feed
# object that implements Enumerable. Otherwise, you can do whatever you wanna do.

# we'll pretend we're getting tab-delimited data from a file.

File.open("myfile.txt") do |f|
  f.threach(consumerThreads) do |line|
    data = line.split("\t")
    doc = SolrInputDocument.new
    doc['name'] = data[0]
    doc['age'] = data[1]
    
    # "siblings", we'll say, is comma-delimited within the column
    # We can add them all at once by using #add and passing an 
    # array of values
    doc.add('sibling', data[2].split(","))

    # Can also call add multiple times; the effect is to append 
    # new values
    data[3].split(',').each do |kid|
      doc.add('kid', kid)
    end

    # You can merge in a hash
    doc.merge! somethingThatReturnsAHash(data)
 
    # Send it to the suss
    suss << doc
  end
end

suss.commit
suss.optimize

# That example used the doc object directly. You can also add a hash to the suss.
# The whole loop could be:

mything.each do |r|
  suss << createAHashFromMyData(r)
end
suss.commit