rodneyc-anemone 0.7.1.1
Sign up to get free protection for your applications and to get access to all the features.
- data/CHANGELOG.rdoc +129 -0
- data/LICENSE.txt +19 -0
- data/README.rdoc +38 -0
- data/Rakefile +23 -0
- data/VERSION +1 -0
- data/bin/anemone +4 -0
- data/lib/anemone/cli/count.rb +22 -0
- data/lib/anemone/cli/cron.rb +90 -0
- data/lib/anemone/cli/pagedepth.rb +32 -0
- data/lib/anemone/cli/serialize.rb +35 -0
- data/lib/anemone/cli/url_list.rb +41 -0
- data/lib/anemone/cli.rb +24 -0
- data/lib/anemone/cookie_store.rb +35 -0
- data/lib/anemone/core.rb +332 -0
- data/lib/anemone/exceptions.rb +5 -0
- data/lib/anemone/http.rb +187 -0
- data/lib/anemone/page.rb +217 -0
- data/lib/anemone/page_store.rb +160 -0
- data/lib/anemone/storage/base.rb +75 -0
- data/lib/anemone/storage/exceptions.rb +15 -0
- data/lib/anemone/storage/kyoto_cabinet.rb +72 -0
- data/lib/anemone/storage/mongodb.rb +89 -0
- data/lib/anemone/storage/pstore.rb +50 -0
- data/lib/anemone/storage/redis.rb +90 -0
- data/lib/anemone/storage/sqlite3.rb +90 -0
- data/lib/anemone/storage/tokyo_cabinet.rb +60 -0
- data/lib/anemone/storage.rb +44 -0
- data/lib/anemone/tentacle.rb +39 -0
- data/lib/anemone.rb +2 -0
- data/spec/anemone_spec.rb +16 -0
- data/spec/cookie_store_spec.rb +28 -0
- data/spec/core_spec.rb +344 -0
- data/spec/fakeweb_helper.rb +77 -0
- data/spec/http_spec.rb +19 -0
- data/spec/page_spec.rb +177 -0
- data/spec/page_store_spec.rb +171 -0
- data/spec/spec_helper.rb +9 -0
- data/spec/storage_spec.rb +252 -0
- metadata +207 -0
data/spec/core_spec.rb
ADDED
@@ -0,0 +1,344 @@
|
|
1
|
+
$:.unshift(File.dirname(__FILE__))
|
2
|
+
require 'spec_helper'
|
3
|
+
%w[pstore tokyo_cabinet sqlite3].each { |file| require "anemone/storage/#{file}.rb" }
|
4
|
+
|
5
|
+
module Anemone
|
6
|
+
describe Core do
|
7
|
+
|
8
|
+
before(:each) do
|
9
|
+
FakeWeb.clean_registry
|
10
|
+
end
|
11
|
+
|
12
|
+
shared_examples_for "crawl" do
|
13
|
+
it "should crawl all the html pages in a domain by following <a> href's" do
|
14
|
+
pages = []
|
15
|
+
pages << FakePage.new('0', :links => ['1', '2'])
|
16
|
+
pages << FakePage.new('1', :links => ['3'])
|
17
|
+
pages << FakePage.new('2')
|
18
|
+
pages << FakePage.new('3')
|
19
|
+
|
20
|
+
Anemone.crawl(pages[0].url, @opts).should have(4).pages
|
21
|
+
end
|
22
|
+
|
23
|
+
it "should not follow links that leave the original domain" do
|
24
|
+
pages = []
|
25
|
+
pages << FakePage.new('0', :links => ['1'], :hrefs => 'http://www.other.com/')
|
26
|
+
pages << FakePage.new('1')
|
27
|
+
|
28
|
+
core = Anemone.crawl(pages[0].url, @opts)
|
29
|
+
|
30
|
+
core.should have(2).pages
|
31
|
+
core.pages.keys.should_not include('http://www.other.com/')
|
32
|
+
end
|
33
|
+
|
34
|
+
it "should not follow redirects that leave the original domain" do
|
35
|
+
pages = []
|
36
|
+
pages << FakePage.new('0', :links => ['1'], :redirect => 'http://www.other.com/')
|
37
|
+
pages << FakePage.new('1')
|
38
|
+
|
39
|
+
core = Anemone.crawl(pages[0].url, @opts)
|
40
|
+
|
41
|
+
core.should have(2).pages
|
42
|
+
core.pages.keys.should_not include('http://www.other.com/')
|
43
|
+
end
|
44
|
+
|
45
|
+
it "should follow http redirects" do
|
46
|
+
pages = []
|
47
|
+
pages << FakePage.new('0', :links => ['1'])
|
48
|
+
pages << FakePage.new('1', :redirect => '2')
|
49
|
+
pages << FakePage.new('2')
|
50
|
+
|
51
|
+
Anemone.crawl(pages[0].url, @opts).should have(3).pages
|
52
|
+
end
|
53
|
+
|
54
|
+
it "should follow with HTTP basic authentication" do
|
55
|
+
pages = []
|
56
|
+
pages << FakePage.new('0', :links => ['1', '2'], :auth => true)
|
57
|
+
pages << FakePage.new('1', :links => ['3'], :auth => true)
|
58
|
+
|
59
|
+
Anemone.crawl(pages.first.auth_url, @opts).should have(3).pages
|
60
|
+
end
|
61
|
+
|
62
|
+
it "should accept multiple starting URLs" do
|
63
|
+
pages = []
|
64
|
+
pages << FakePage.new('0', :links => ['1'])
|
65
|
+
pages << FakePage.new('1')
|
66
|
+
pages << FakePage.new('2', :links => ['3'])
|
67
|
+
pages << FakePage.new('3')
|
68
|
+
|
69
|
+
Anemone.crawl([pages[0].url, pages[2].url], @opts).should have(4).pages
|
70
|
+
end
|
71
|
+
|
72
|
+
it "should include the query string when following links" do
|
73
|
+
pages = []
|
74
|
+
pages << FakePage.new('0', :links => ['1?foo=1'])
|
75
|
+
pages << FakePage.new('1?foo=1')
|
76
|
+
pages << FakePage.new('1')
|
77
|
+
|
78
|
+
core = Anemone.crawl(pages[0].url, @opts)
|
79
|
+
|
80
|
+
core.should have(2).pages
|
81
|
+
core.pages.keys.should_not include(pages[2].url)
|
82
|
+
end
|
83
|
+
|
84
|
+
it "should be able to skip links with query strings" do
|
85
|
+
pages = []
|
86
|
+
pages << FakePage.new('0', :links => ['1?foo=1', '2'])
|
87
|
+
pages << FakePage.new('1?foo=1')
|
88
|
+
pages << FakePage.new('2')
|
89
|
+
|
90
|
+
core = Anemone.crawl(pages[0].url, @opts) do |a|
|
91
|
+
a.skip_query_strings = true
|
92
|
+
end
|
93
|
+
|
94
|
+
core.should have(2).pages
|
95
|
+
end
|
96
|
+
|
97
|
+
it "should be able to skip links based on a RegEx" do
|
98
|
+
pages = []
|
99
|
+
pages << FakePage.new('0', :links => ['1', '2'])
|
100
|
+
pages << FakePage.new('1')
|
101
|
+
pages << FakePage.new('2')
|
102
|
+
pages << FakePage.new('3')
|
103
|
+
|
104
|
+
core = Anemone.crawl(pages[0].url, @opts) do |a|
|
105
|
+
a.skip_links_like /1/, /3/
|
106
|
+
end
|
107
|
+
|
108
|
+
core.should have(2).pages
|
109
|
+
core.pages.keys.should_not include(pages[1].url)
|
110
|
+
core.pages.keys.should_not include(pages[3].url)
|
111
|
+
end
|
112
|
+
|
113
|
+
it "should be able to call a block on every page" do
|
114
|
+
pages = []
|
115
|
+
pages << FakePage.new('0', :links => ['1', '2'])
|
116
|
+
pages << FakePage.new('1')
|
117
|
+
pages << FakePage.new('2')
|
118
|
+
|
119
|
+
count = 0
|
120
|
+
Anemone.crawl(pages[0].url, @opts) do |a|
|
121
|
+
a.on_every_page { count += 1 }
|
122
|
+
end
|
123
|
+
|
124
|
+
count.should == 3
|
125
|
+
end
|
126
|
+
|
127
|
+
it "should not discard page bodies by default" do
|
128
|
+
Anemone.crawl(FakePage.new('0').url, @opts).pages.values#.first.doc.should_not be_nil
|
129
|
+
end
|
130
|
+
|
131
|
+
it "should optionally discard page bodies to conserve memory" do
|
132
|
+
# core = Anemone.crawl(FakePage.new('0').url, @opts.merge({:discard_page_bodies => true}))
|
133
|
+
# core.pages.values.first.doc.should be_nil
|
134
|
+
end
|
135
|
+
|
136
|
+
it "should provide a focus_crawl method to select the links on each page to follow" do
|
137
|
+
pages = []
|
138
|
+
pages << FakePage.new('0', :links => ['1', '2'])
|
139
|
+
pages << FakePage.new('1')
|
140
|
+
pages << FakePage.new('2')
|
141
|
+
|
142
|
+
core = Anemone.crawl(pages[0].url, @opts) do |a|
|
143
|
+
a.focus_crawl {|p| p.links.reject{|l| l.to_s =~ /1/}}
|
144
|
+
end
|
145
|
+
|
146
|
+
core.should have(2).pages
|
147
|
+
core.pages.keys.should_not include(pages[1].url)
|
148
|
+
end
|
149
|
+
|
150
|
+
it "should optionally delay between page requests" do
|
151
|
+
delay = 0.25
|
152
|
+
|
153
|
+
pages = []
|
154
|
+
pages << FakePage.new('0', :links => '1')
|
155
|
+
pages << FakePage.new('1')
|
156
|
+
|
157
|
+
start = Time.now
|
158
|
+
Anemone.crawl(pages[0].url, @opts.merge({:delay => delay}))
|
159
|
+
finish = Time.now
|
160
|
+
|
161
|
+
(finish - start).should satisfy {|t| t > delay * 2}
|
162
|
+
end
|
163
|
+
|
164
|
+
it "should optionally obey the robots exclusion protocol" do
|
165
|
+
pages = []
|
166
|
+
pages << FakePage.new('0', :links => '1')
|
167
|
+
pages << FakePage.new('1')
|
168
|
+
pages << FakePage.new('robots.txt',
|
169
|
+
:body => "User-agent: *\nDisallow: /1",
|
170
|
+
:content_type => 'text/plain')
|
171
|
+
|
172
|
+
core = Anemone.crawl(pages[0].url, @opts.merge({:obey_robots_txt => true}))
|
173
|
+
urls = core.pages.keys
|
174
|
+
|
175
|
+
urls.should include(pages[0].url)
|
176
|
+
urls.should_not include(pages[1].url)
|
177
|
+
end
|
178
|
+
|
179
|
+
it "should be able to set cookies to send with HTTP requests" do
|
180
|
+
cookies = {:a => '1', :b => '2'}
|
181
|
+
core = Anemone.crawl(FakePage.new('0').url) do |anemone|
|
182
|
+
anemone.cookies = cookies
|
183
|
+
end
|
184
|
+
core.opts[:cookies].should == cookies
|
185
|
+
end
|
186
|
+
|
187
|
+
it "should freeze the options once the crawl begins" do
|
188
|
+
core = Anemone.crawl(FakePage.new('0').url) do |anemone|
|
189
|
+
anemone.threads = 4
|
190
|
+
anemone.on_every_page do
|
191
|
+
lambda {anemone.threads = 2}.should raise_error
|
192
|
+
end
|
193
|
+
end
|
194
|
+
core.opts[:threads].should == 4
|
195
|
+
end
|
196
|
+
|
197
|
+
describe "many pages" do
|
198
|
+
before(:each) do
|
199
|
+
@pages, size = [], 5
|
200
|
+
|
201
|
+
size.times do |n|
|
202
|
+
# register this page with a link to the next page
|
203
|
+
link = (n + 1).to_s if n + 1 < size
|
204
|
+
@pages << FakePage.new(n.to_s, :links => Array(link))
|
205
|
+
end
|
206
|
+
end
|
207
|
+
|
208
|
+
it "should track the page depth and referer" do
|
209
|
+
core = Anemone.crawl(@pages[0].url, @opts)
|
210
|
+
previous_page = nil
|
211
|
+
|
212
|
+
@pages.each_with_index do |page, i|
|
213
|
+
page = core.pages[page.url]
|
214
|
+
page.should be
|
215
|
+
page.depth.should == i
|
216
|
+
|
217
|
+
if previous_page
|
218
|
+
page.referer.should == previous_page.url
|
219
|
+
else
|
220
|
+
page.referer.should be_nil
|
221
|
+
end
|
222
|
+
previous_page = page
|
223
|
+
end
|
224
|
+
end
|
225
|
+
|
226
|
+
it "should optionally limit the depth of the crawl" do
|
227
|
+
core = Anemone.crawl(@pages[0].url, @opts.merge({:depth_limit => 3}))
|
228
|
+
core.should have(4).pages
|
229
|
+
end
|
230
|
+
end
|
231
|
+
|
232
|
+
end
|
233
|
+
|
234
|
+
describe Hash do
|
235
|
+
it_should_behave_like "crawl"
|
236
|
+
|
237
|
+
before(:all) do
|
238
|
+
@opts = {}
|
239
|
+
end
|
240
|
+
end
|
241
|
+
|
242
|
+
describe Storage::PStore do
|
243
|
+
it_should_behave_like "crawl"
|
244
|
+
|
245
|
+
before(:all) do
|
246
|
+
@test_file = 'test.pstore'
|
247
|
+
end
|
248
|
+
|
249
|
+
before(:each) do
|
250
|
+
File.delete(@test_file) if File.exists?(@test_file)
|
251
|
+
@opts = {:storage => Storage.PStore(@test_file)}
|
252
|
+
end
|
253
|
+
|
254
|
+
after(:each) do
|
255
|
+
File.delete(@test_file) if File.exists?(@test_file)
|
256
|
+
end
|
257
|
+
end
|
258
|
+
|
259
|
+
describe Storage::TokyoCabinet do
|
260
|
+
it_should_behave_like "crawl"
|
261
|
+
|
262
|
+
before(:all) do
|
263
|
+
@test_file = 'test.tch'
|
264
|
+
end
|
265
|
+
|
266
|
+
before(:each) do
|
267
|
+
File.delete(@test_file) if File.exists?(@test_file)
|
268
|
+
@opts = {:storage => @store = Storage.TokyoCabinet(@test_file)}
|
269
|
+
end
|
270
|
+
|
271
|
+
after(:each) do
|
272
|
+
@store.close
|
273
|
+
end
|
274
|
+
|
275
|
+
after(:each) do
|
276
|
+
File.delete(@test_file) if File.exists?(@test_file)
|
277
|
+
end
|
278
|
+
end
|
279
|
+
|
280
|
+
describe Storage::SQLite3 do
|
281
|
+
it_should_behave_like "crawl"
|
282
|
+
|
283
|
+
before(:all) do
|
284
|
+
@test_file = 'test.db'
|
285
|
+
end
|
286
|
+
|
287
|
+
before(:each) do
|
288
|
+
File.delete(@test_file) if File.exists?(@test_file)
|
289
|
+
@opts = {:storage => @store = Storage.SQLite3(@test_file)}
|
290
|
+
end
|
291
|
+
|
292
|
+
after(:each) do
|
293
|
+
@store.close
|
294
|
+
end
|
295
|
+
|
296
|
+
after(:each) do
|
297
|
+
File.delete(@test_file) if File.exists?(@test_file)
|
298
|
+
end
|
299
|
+
end
|
300
|
+
|
301
|
+
describe "options" do
|
302
|
+
it "should accept options for the crawl" do
|
303
|
+
core = Anemone.crawl(SPEC_DOMAIN, :verbose => false,
|
304
|
+
:threads => 2,
|
305
|
+
:discard_page_bodies => true,
|
306
|
+
:user_agent => 'test',
|
307
|
+
:obey_robots_txt => true,
|
308
|
+
:depth_limit => 3)
|
309
|
+
|
310
|
+
core.opts[:verbose].should == false
|
311
|
+
core.opts[:threads].should == 2
|
312
|
+
core.opts[:discard_page_bodies].should == true
|
313
|
+
core.opts[:delay].should == 0
|
314
|
+
core.opts[:user_agent].should == 'test'
|
315
|
+
core.opts[:obey_robots_txt].should == true
|
316
|
+
core.opts[:depth_limit].should == 3
|
317
|
+
end
|
318
|
+
|
319
|
+
it "should accept options via setter methods in the crawl block" do
|
320
|
+
core = Anemone.crawl(SPEC_DOMAIN) do |a|
|
321
|
+
a.verbose = false
|
322
|
+
a.threads = 2
|
323
|
+
a.discard_page_bodies = true
|
324
|
+
a.user_agent = 'test'
|
325
|
+
a.obey_robots_txt = true
|
326
|
+
a.depth_limit = 3
|
327
|
+
end
|
328
|
+
|
329
|
+
core.opts[:verbose].should == false
|
330
|
+
core.opts[:threads].should == 2
|
331
|
+
core.opts[:discard_page_bodies].should == true
|
332
|
+
core.opts[:delay].should == 0
|
333
|
+
core.opts[:user_agent].should == 'test'
|
334
|
+
core.opts[:obey_robots_txt].should == true
|
335
|
+
core.opts[:depth_limit].should == 3
|
336
|
+
end
|
337
|
+
|
338
|
+
it "should use 1 thread if a delay is requested" do
|
339
|
+
Anemone.crawl(SPEC_DOMAIN, :delay => 0.01, :threads => 2).opts[:threads].should == 1
|
340
|
+
end
|
341
|
+
end
|
342
|
+
|
343
|
+
end
|
344
|
+
end
|
@@ -0,0 +1,77 @@
|
|
1
|
+
FakeWeb.allow_net_connect = false
|
2
|
+
|
3
|
+
module Anemone
|
4
|
+
SPEC_DOMAIN = "http://www.example.com/"
|
5
|
+
AUTH_SPEC_DOMAIN = "http://user:pass@#{URI.parse(SPEC_DOMAIN).host}/"
|
6
|
+
|
7
|
+
class FakePage
|
8
|
+
attr_accessor :links
|
9
|
+
attr_accessor :hrefs
|
10
|
+
attr_accessor :body
|
11
|
+
|
12
|
+
def initialize(name = '', options = {})
|
13
|
+
@name = name
|
14
|
+
@links = [options[:links]].flatten if options.has_key?(:links)
|
15
|
+
@hrefs = [options[:hrefs]].flatten if options.has_key?(:hrefs)
|
16
|
+
@redirect = options[:redirect] if options.has_key?(:redirect)
|
17
|
+
@auth = options[:auth] if options.has_key?(:auth)
|
18
|
+
@base = options[:base] if options.has_key?(:base)
|
19
|
+
@content_type = options[:content_type] || "text/html"
|
20
|
+
@body = options[:body]
|
21
|
+
|
22
|
+
create_body unless @body
|
23
|
+
add_to_fakeweb
|
24
|
+
end
|
25
|
+
|
26
|
+
def url
|
27
|
+
SPEC_DOMAIN + @name
|
28
|
+
end
|
29
|
+
|
30
|
+
def auth_url
|
31
|
+
AUTH_SPEC_DOMAIN + @name
|
32
|
+
end
|
33
|
+
|
34
|
+
private
|
35
|
+
|
36
|
+
def create_body
|
37
|
+
if @base
|
38
|
+
@body = "<html><head><base href=\"#{@base}\"></head><body>"
|
39
|
+
else
|
40
|
+
@body = "<html><body>"
|
41
|
+
end
|
42
|
+
@links.each{|l| @body += "<a href=\"#{SPEC_DOMAIN}#{l}\"></a>"} if @links
|
43
|
+
@hrefs.each{|h| @body += "<a href=\"#{h}\"></a>"} if @hrefs
|
44
|
+
@body += "</body></html>"
|
45
|
+
end
|
46
|
+
|
47
|
+
def add_to_fakeweb
|
48
|
+
options = {:body => @body, :content_type => @content_type, :status => [200, "OK"]}
|
49
|
+
|
50
|
+
if @redirect
|
51
|
+
options[:status] = [301, "Permanently Moved"]
|
52
|
+
|
53
|
+
# only prepend SPEC_DOMAIN if a relative url (without an http scheme) was specified
|
54
|
+
redirect_url = (@redirect =~ /http/) ? @redirect : SPEC_DOMAIN + @redirect
|
55
|
+
options[:location] = redirect_url
|
56
|
+
|
57
|
+
# register the page this one redirects to
|
58
|
+
FakeWeb.register_uri(:get, redirect_url, {:body => '',
|
59
|
+
:content_type => @content_type,
|
60
|
+
:status => [200, "OK"]})
|
61
|
+
end
|
62
|
+
|
63
|
+
if @auth
|
64
|
+
unautorized_options = {
|
65
|
+
:body => "Unauthorized", :status => ["401", "Unauthorized"]
|
66
|
+
}
|
67
|
+
FakeWeb.register_uri(:get, SPEC_DOMAIN + @name, unautorized_options)
|
68
|
+
FakeWeb.register_uri(:get, AUTH_SPEC_DOMAIN + @name, options)
|
69
|
+
else
|
70
|
+
FakeWeb.register_uri(:get, SPEC_DOMAIN + @name, options)
|
71
|
+
end
|
72
|
+
end
|
73
|
+
end
|
74
|
+
end
|
75
|
+
|
76
|
+
#default root
|
77
|
+
Anemone::FakePage.new
|
data/spec/http_spec.rb
ADDED
@@ -0,0 +1,19 @@
|
|
1
|
+
require 'spec_helper'
|
2
|
+
|
3
|
+
module Anemone
|
4
|
+
describe HTTP do
|
5
|
+
|
6
|
+
describe "fetch_page" do
|
7
|
+
before(:each) do
|
8
|
+
FakeWeb.clean_registry
|
9
|
+
end
|
10
|
+
|
11
|
+
it "should still return a Page if an exception occurs during the HTTP connection" do
|
12
|
+
HTTP.stub!(:refresh_connection).and_raise(StandardError)
|
13
|
+
http = Anemone::HTTP.new
|
14
|
+
http.fetch_page(SPEC_DOMAIN).should be_an_instance_of(Page)
|
15
|
+
end
|
16
|
+
|
17
|
+
end
|
18
|
+
end
|
19
|
+
end
|
data/spec/page_spec.rb
ADDED
@@ -0,0 +1,177 @@
|
|
1
|
+
$:.unshift(File.dirname(__FILE__))
|
2
|
+
require 'spec_helper'
|
3
|
+
|
4
|
+
module Anemone
|
5
|
+
describe Page do
|
6
|
+
|
7
|
+
before(:each) do
|
8
|
+
FakeWeb.clean_registry
|
9
|
+
@http = Anemone::HTTP.new
|
10
|
+
@page = @http.fetch_page(FakePage.new('home', :links => '1').url)
|
11
|
+
end
|
12
|
+
|
13
|
+
it "should indicate whether it successfully fetched via HTTP" do
|
14
|
+
@page.should respond_to(:fetched?)
|
15
|
+
@page.fetched?.should == true
|
16
|
+
|
17
|
+
fail_page = @http.fetch_page(SPEC_DOMAIN + 'fail')
|
18
|
+
fail_page.fetched?.should == false
|
19
|
+
end
|
20
|
+
|
21
|
+
it "should store and expose the response body of the HTTP request" do
|
22
|
+
body = 'test'
|
23
|
+
page = @http.fetch_page(FakePage.new('body_test', {:body => body}).url)
|
24
|
+
page.body.should == body
|
25
|
+
end
|
26
|
+
|
27
|
+
it "should record any error that occurs during fetch_page" do
|
28
|
+
@page.should respond_to(:error)
|
29
|
+
@page.error.should be_nil
|
30
|
+
|
31
|
+
fail_page = @http.fetch_page(SPEC_DOMAIN + 'fail')
|
32
|
+
fail_page.error.should_not be_nil
|
33
|
+
end
|
34
|
+
|
35
|
+
it "should store the response headers when fetching a page" do
|
36
|
+
@page.headers.should_not be_nil
|
37
|
+
@page.headers.should have_key('content-type')
|
38
|
+
end
|
39
|
+
|
40
|
+
it "should have an OpenStruct attribute for the developer to store data in" do
|
41
|
+
@page.data.should_not be_nil
|
42
|
+
@page.data.should be_an_instance_of(OpenStruct)
|
43
|
+
|
44
|
+
@page.data.test = 'test'
|
45
|
+
@page.data.test.should == 'test'
|
46
|
+
end
|
47
|
+
|
48
|
+
it "should have a Nokogori::HTML::Document attribute for the page body" do
|
49
|
+
@page.doc.should_not be_nil
|
50
|
+
@page.doc.should be_an_instance_of(Nokogiri::HTML::Document)
|
51
|
+
end
|
52
|
+
|
53
|
+
it "should indicate whether it was fetched after an HTTP redirect" do
|
54
|
+
@page.should respond_to(:redirect?)
|
55
|
+
|
56
|
+
@page.redirect?.should == false
|
57
|
+
|
58
|
+
@http.fetch_pages(FakePage.new('redir', :redirect => 'home').url).first.redirect?.should == true
|
59
|
+
end
|
60
|
+
|
61
|
+
it "should have a method to tell if a URI is in the same domain as the page" do
|
62
|
+
@page.should respond_to(:in_domain?)
|
63
|
+
|
64
|
+
@page.in_domain?(URI(FakePage.new('test').url)).should == true
|
65
|
+
@page.in_domain?(URI('http://www.other.com/')).should == false
|
66
|
+
end
|
67
|
+
|
68
|
+
it "should include the response time for the HTTP request" do
|
69
|
+
@page.should respond_to(:response_time)
|
70
|
+
end
|
71
|
+
|
72
|
+
it "should have the cookies received with the page" do
|
73
|
+
@page.should respond_to(:cookies)
|
74
|
+
@page.cookies.should == []
|
75
|
+
end
|
76
|
+
|
77
|
+
describe "#to_hash" do
|
78
|
+
it "converts the page to a hash" do
|
79
|
+
hash = @page.to_hash
|
80
|
+
hash['url'].should == @page.url.to_s
|
81
|
+
hash['referer'].should == @page.referer.to_s
|
82
|
+
hash['links'].should == @page.links.map(&:to_s)
|
83
|
+
end
|
84
|
+
|
85
|
+
context "when redirect_to is nil" do
|
86
|
+
it "sets 'redirect_to' to nil in the hash" do
|
87
|
+
@page.redirect_to.should be_nil
|
88
|
+
@page.to_hash[:redirect_to].should be_nil
|
89
|
+
end
|
90
|
+
end
|
91
|
+
|
92
|
+
context "when redirect_to is a non-nil URI" do
|
93
|
+
it "sets 'redirect_to' to the URI string" do
|
94
|
+
new_page = Page.new(URI(SPEC_DOMAIN), {:redirect_to => URI(SPEC_DOMAIN + '1')})
|
95
|
+
new_page.redirect_to.to_s.should == SPEC_DOMAIN + '1'
|
96
|
+
new_page.to_hash['redirect_to'].should == SPEC_DOMAIN + '1'
|
97
|
+
end
|
98
|
+
end
|
99
|
+
end
|
100
|
+
|
101
|
+
describe "#from_hash" do
|
102
|
+
it "converts from a hash to a Page" do
|
103
|
+
page = @page.dup
|
104
|
+
page.depth = 1
|
105
|
+
converted = Page.from_hash(page.to_hash)
|
106
|
+
converted.links.should == page.links
|
107
|
+
converted.depth.should == page.depth
|
108
|
+
end
|
109
|
+
|
110
|
+
it 'handles a from_hash with a nil redirect_to' do
|
111
|
+
page_hash = @page.to_hash
|
112
|
+
page_hash['redirect_to'] = nil
|
113
|
+
lambda{Page.from_hash(page_hash)}.should_not raise_error(URI::InvalidURIError)
|
114
|
+
Page.from_hash(page_hash).redirect_to.should be_nil
|
115
|
+
end
|
116
|
+
end
|
117
|
+
|
118
|
+
describe "#redirect_to" do
|
119
|
+
context "when the page was a redirect" do
|
120
|
+
it "returns a URI of the page it redirects to" do
|
121
|
+
new_page = Page.new(URI(SPEC_DOMAIN), {:redirect_to => URI(SPEC_DOMAIN + '1')})
|
122
|
+
redirect = new_page.redirect_to
|
123
|
+
redirect.should be_a(URI)
|
124
|
+
redirect.to_s.should == SPEC_DOMAIN + '1'
|
125
|
+
end
|
126
|
+
end
|
127
|
+
end
|
128
|
+
|
129
|
+
it "should detect, store and expose the base url for the page head" do
|
130
|
+
base = "#{SPEC_DOMAIN}path/to/base_url/"
|
131
|
+
page = @http.fetch_page(FakePage.new('body_test', {:base => base}).url)
|
132
|
+
page.base.should == URI(base)
|
133
|
+
@page.base.should be_nil
|
134
|
+
end
|
135
|
+
|
136
|
+
it "should have a method to convert a relative url to an absolute one" do
|
137
|
+
@page.should respond_to(:to_absolute)
|
138
|
+
|
139
|
+
# Identity
|
140
|
+
@page.to_absolute(@page.url).should == @page.url
|
141
|
+
@page.to_absolute("").should == @page.url
|
142
|
+
|
143
|
+
# Root-ness
|
144
|
+
@page.to_absolute("/").should == URI("#{SPEC_DOMAIN}")
|
145
|
+
|
146
|
+
# Relativeness
|
147
|
+
relative_path = "a/relative/path"
|
148
|
+
@page.to_absolute(relative_path).should == URI("#{SPEC_DOMAIN}#{relative_path}")
|
149
|
+
|
150
|
+
deep_page = @http.fetch_page(FakePage.new('home/deep', :links => '1').url)
|
151
|
+
upward_relative_path = "../a/relative/path"
|
152
|
+
deep_page.to_absolute(upward_relative_path).should == URI("#{SPEC_DOMAIN}#{relative_path}")
|
153
|
+
|
154
|
+
# The base URL case
|
155
|
+
base_path = "path/to/base_url/"
|
156
|
+
base = "#{SPEC_DOMAIN}#{base_path}"
|
157
|
+
page = @http.fetch_page(FakePage.new('home', {:base => base}).url)
|
158
|
+
|
159
|
+
# Identity
|
160
|
+
page.to_absolute(page.url).should == page.url
|
161
|
+
# It should revert to the base url
|
162
|
+
page.to_absolute("").should_not == page.url
|
163
|
+
|
164
|
+
# Root-ness
|
165
|
+
page.to_absolute("/").should == URI("#{SPEC_DOMAIN}")
|
166
|
+
|
167
|
+
# Relativeness
|
168
|
+
relative_path = "a/relative/path"
|
169
|
+
page.to_absolute(relative_path).should == URI("#{base}#{relative_path}")
|
170
|
+
|
171
|
+
upward_relative_path = "../a/relative/path"
|
172
|
+
upward_base = "#{SPEC_DOMAIN}path/to/"
|
173
|
+
page.to_absolute(upward_relative_path).should == URI("#{upward_base}#{relative_path}")
|
174
|
+
end
|
175
|
+
|
176
|
+
end
|
177
|
+
end
|