wikio 0.0.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/bin/wikireducer +48 -0
- data/lib/wikio.rb +41 -0
- metadata +46 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 346b20f9174cb1d24c10850f5688fd6d916ddcbf
|
4
|
+
data.tar.gz: bdbc0f5e08ff93d15e570237ce9ded592b76b067
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 39fdd792e2f4c8f9b9dd3242d4ae896606fe8203e9eaaed8c8caea6172f073f91233d44d988c3219d89c813eed26a28b44a7d0a0df9ed9f8418590a06dcd43d6
|
7
|
+
data.tar.gz: 47cbea5777dec4da685ffd82462fbb72b75ab76967cd0ef3a257fc5d168c7e5c7091fd10eb408dffcf033c8c9407a93f2dc195b54fd8d834fb9fd4b49ada198c
|
data/bin/wikireducer
ADDED
@@ -0,0 +1,48 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
require 'optparse'
|
4
|
+
require 'set'
|
5
|
+
|
6
|
+
require 'wikio'
|
7
|
+
|
8
|
+
# Walk the path of first links on wikipedia pages
|
9
|
+
# @param src @String - Wikipedia Url
|
10
|
+
# @param dst @string - Wikipedia Url
|
11
|
+
def walk(src, dst)
|
12
|
+
visited = Set.new([src])
|
13
|
+
current = src
|
14
|
+
count = 0
|
15
|
+
|
16
|
+
while true
|
17
|
+
if current == dst
|
18
|
+
puts "#{src} -> #{dst} in #{count} steps"
|
19
|
+
break
|
20
|
+
end
|
21
|
+
STDERR.puts "#{src} -> #{current}"
|
22
|
+
current = Wikio.get_first_link(current)
|
23
|
+
if visited.include?(current)
|
24
|
+
puts "Cycle detected for #{src} at node #{current}"
|
25
|
+
break
|
26
|
+
end
|
27
|
+
visited.add(current)
|
28
|
+
count += 1
|
29
|
+
end
|
30
|
+
end
|
31
|
+
|
32
|
+
options = {dst: 'Philosophy'}
|
33
|
+
OptionParser.new do |opts|
|
34
|
+
opts.banner = "Usage: wikio --dst=<destination_article> <src1> <src2> ..."
|
35
|
+
|
36
|
+
opts.on('-dst', '--dst DESTINATION', 'Destination article') do |dst|
|
37
|
+
options[:dst] = Wikio.get_wiki_url(dst)
|
38
|
+
end
|
39
|
+
end.parse!
|
40
|
+
|
41
|
+
puts "Searching for #{options[:dst]}"
|
42
|
+
ARGV.map do |term|
|
43
|
+
Wikio.get_wiki_url(term)
|
44
|
+
end.map do |url|
|
45
|
+
Thread.new { walk(url, options[:dst]) }
|
46
|
+
end.map do |thread|
|
47
|
+
thread.join
|
48
|
+
end
|
data/lib/wikio.rb
ADDED
@@ -0,0 +1,41 @@
|
|
1
|
+
require 'json'
|
2
|
+
require 'net/http'
|
3
|
+
require 'nokogiri'
|
4
|
+
require 'uri'
|
5
|
+
|
6
|
+
# Resources
|
7
|
+
# - https://stackoverflow.com/questions/27457977/searching-wikipedia-using-api
|
8
|
+
|
9
|
+
module Wikio
|
10
|
+
|
11
|
+
WIKIPEDIA_API_URL = 'https://en.wikipedia.org/w/api.php'
|
12
|
+
WIKIPEDIA_DOMAIN = 'https://en.wikipedia.org'
|
13
|
+
def self.get_wiki_url(term)
|
14
|
+
params = {
|
15
|
+
action: 'opensearch',
|
16
|
+
search: term,
|
17
|
+
limit: 1,
|
18
|
+
namespace: 0,
|
19
|
+
format: 'json'
|
20
|
+
}
|
21
|
+
uri = URI(WIKIPEDIA_API_URL)
|
22
|
+
uri.query = URI.encode_www_form(params)
|
23
|
+
res = Net::HTTP.get(uri)
|
24
|
+
return JSON.parse(res)[3][0]
|
25
|
+
end
|
26
|
+
|
27
|
+
def self.get_first_link(wiki_url)
|
28
|
+
uri = URI(wiki_url)
|
29
|
+
body = Net::HTTP.get(uri)
|
30
|
+
doc = Nokogiri::HTML(body)
|
31
|
+
subdoc = doc.xpath(
|
32
|
+
"//div[contains(@class, 'mw-parser-output')]/p/a"\
|
33
|
+
" | //div[contains(@class, 'mw-parser-output')]/p/i/a"
|
34
|
+
).each do |node|
|
35
|
+
href = node.attr('href')
|
36
|
+
if href =~ /\A\/wiki/
|
37
|
+
return WIKIPEDIA_DOMAIN + href
|
38
|
+
end
|
39
|
+
end
|
40
|
+
end
|
41
|
+
end
|
metadata
ADDED
@@ -0,0 +1,46 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: wikio
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.0.1
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Daniel Cardoza
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2018-01-30 00:00:00.000000000 Z
|
12
|
+
dependencies: []
|
13
|
+
description: Library containing helper functions for wikipedia
|
14
|
+
email: danielpcardoza@gmail.com
|
15
|
+
executables:
|
16
|
+
- wikireducer
|
17
|
+
extensions: []
|
18
|
+
extra_rdoc_files: []
|
19
|
+
files:
|
20
|
+
- bin/wikireducer
|
21
|
+
- lib/wikio.rb
|
22
|
+
homepage: http://rubygems.org/gems/wikio
|
23
|
+
licenses:
|
24
|
+
- MIT
|
25
|
+
metadata: {}
|
26
|
+
post_install_message:
|
27
|
+
rdoc_options: []
|
28
|
+
require_paths:
|
29
|
+
- lib
|
30
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
31
|
+
requirements:
|
32
|
+
- - ">="
|
33
|
+
- !ruby/object:Gem::Version
|
34
|
+
version: '0'
|
35
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
36
|
+
requirements:
|
37
|
+
- - ">="
|
38
|
+
- !ruby/object:Gem::Version
|
39
|
+
version: '0'
|
40
|
+
requirements: []
|
41
|
+
rubyforge_project:
|
42
|
+
rubygems_version: 2.6.11
|
43
|
+
signing_key:
|
44
|
+
specification_version: 4
|
45
|
+
summary: Library for extracting information from wikipedia pages
|
46
|
+
test_files: []
|