Created
August 27, 2009 20:33
-
-
Save CobyR/176539 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
require 'nokogiri' | |
namespace :import do | |
desc "Import XML file" | |
task :xml => :environment do | |
xml_file = File.dirname(__FILE__) + "/../../db/one.xml" | |
rtf_directory = File.dirname(__FILE__) + "/../../db/txt/" | |
puts "DESTROYING! " | |
File.unlink "db/development.sqlite3" rescue nil | |
system "rake db:migrate" | |
system "rake db:fixtures:load" | |
f = File.open(xml_file) | |
n = Nokogiri::XML.parse(f) | |
thoughts = n.xpath('*/thought') | |
puts "Importing: #{xml_file}" | |
puts "#{thoughts.size} thoughts to process." | |
print "Processing" | |
STDOUT.flush | |
# Loop through all of the thoughts and create the initial record set | |
thoughts.each_with_index do | thought, x | | |
t = Thought.new | |
t.original_id = thought.attributes['id'].to_s.to_i | |
t.name = thought.xpath('name').children[0].to_s | |
t.forgotten = thought.attributes['forgotten'].to_s | |
t.private = thought.attributes['private'].to_s | |
t.created_at = thought.attributes['createtimestamp'].to_s | |
t.updated_at = thought.attributes['modtimestamp'].to_s | |
if f = File.open("#{rtf_directory}#{t.original_id}.txt") | |
c = f.read | |
#c.gsub!('\par','<p>') | |
#c.gsub!('\r\n', "<br/>") | |
t.rtf = c | |
end | |
t.save | |
if x % 100 == 0 | |
print '.' | |
STDOUT.flush | |
end | |
end | |
puts "\nAll thoughts imported into the database." | |
# Loop through a second time to establish relations | |
puts "Connecting the thoughts via their relations." | |
print "Processing" | |
STDOUT.flush | |
thoughts.each_with_index do | thought, x | | |
relations = thought.xpath('relations') | |
puts "==================================" | |
puts "#{x} - Relations #{relations.size}" | |
t = Thought.find_by_original_id(thought.attributes['id'].to_s.to_i) | |
relations.children.each_with_index do | relation, y | | |
unless relation.name == "text" | |
puts "\t#{y} - #{relation.name}" | |
rt = Thought.find_by_original_id(relation.attributes['id'].to_s.to_i) | |
r = Relation.new | |
r.thought_id = t.id | |
r.related_to_thought_id = rt.id | |
if relation.name == "jump" | |
r.relation_type_id = 1 | |
elsif relation.name == "child" | |
r.relation_type_id = 2 | |
elsif relation.name == "parent" | |
r.relation_type_id = 3 | |
end | |
r.save | |
end | |
end | |
if x % 100 == 0 | |
print '.' | |
STDOUT.flush | |
end | |
end | |
f.close | |
end | |
end |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment