assaf-scrapi 1.2.1
Sign up to get free protection for your applications and to get access to all the features.
- data/CHANGELOG +36 -0
- data/MIT-LICENSE +20 -0
- data/README.rdoc +88 -0
- data/Rakefile +33 -0
- data/lib/html/document.rb +64 -0
- data/lib/html/htmlparser.rb +407 -0
- data/lib/html/node.rb +534 -0
- data/lib/html/node_ext.rb +86 -0
- data/lib/html/selector.rb +825 -0
- data/lib/html/tokenizer.rb +105 -0
- data/lib/html/version.rb +11 -0
- data/lib/scraper/base.rb +990 -0
- data/lib/scraper/microformats.rb +93 -0
- data/lib/scraper/reader.rb +240 -0
- data/lib/scrapi.rb +8 -0
- data/lib/tidy/libtidy.dll +0 -0
- data/lib/tidy/libtidy.so +0 -0
- data/test/mock_net_http.rb +54 -0
- data/test/node_ext_test.rb +24 -0
- data/test/reader_test.rb +318 -0
- data/test/scraper_test.rb +804 -0
- data/test/selector_test.rb +637 -0
- metadata +89 -0
@@ -0,0 +1,804 @@
|
|
1
|
+
# ScrAPI toolkit for Ruby
|
2
|
+
#
|
3
|
+
# Copyright (c) 2006 Assaf Arkin, under Creative Commons Attribution and/or MIT License
|
4
|
+
# Developed for http://co.mments.com
|
5
|
+
# Code and documention: http://labnotes.org
|
6
|
+
|
7
|
+
|
8
|
+
require "rubygems"
|
9
|
+
require "time"
|
10
|
+
require "test/unit"
|
11
|
+
require File.join(File.dirname(__FILE__), "mock_net_http")
|
12
|
+
require File.join(File.dirname(__FILE__), "../lib", "scrapi")
|
13
|
+
|
14
|
+
|
15
|
+
class ScraperTest < Test::Unit::TestCase
|
16
|
+
|
17
|
+
DIVS123 = <<-EOF
|
18
|
+
<div id="1"></div>
|
19
|
+
<div id="2"></div>
|
20
|
+
<div id="3"></div>
|
21
|
+
EOF
|
22
|
+
|
23
|
+
DIVS1_23 = <<-EOF
|
24
|
+
<div id="1">
|
25
|
+
<div id="2"></div>
|
26
|
+
<div id="3"></div>
|
27
|
+
</div>
|
28
|
+
EOF
|
29
|
+
|
30
|
+
|
31
|
+
def setup
|
32
|
+
Net::HTTP.reset_on_get
|
33
|
+
end
|
34
|
+
|
35
|
+
def teardown
|
36
|
+
Net::HTTP.reset_on_get
|
37
|
+
end
|
38
|
+
|
39
|
+
|
40
|
+
#
|
41
|
+
# Tests selector methods.
|
42
|
+
#
|
43
|
+
|
44
|
+
def test_define_selectors
|
45
|
+
scraper = new_scraper(DIVS123) do
|
46
|
+
selector :test, "div"
|
47
|
+
end
|
48
|
+
assert_equal 3, scraper.test(scraper.document).size
|
49
|
+
3.times do |i|
|
50
|
+
assert_equal String(i + 1), scraper.test(scraper.document)[i].attributes["id"]
|
51
|
+
end
|
52
|
+
end
|
53
|
+
|
54
|
+
|
55
|
+
def test_selector_blocks
|
56
|
+
scraper = new_scraper(DIVS123) do
|
57
|
+
selector :test, "div" do |elements|
|
58
|
+
return elements[0..-2]
|
59
|
+
elements[0..-2]
|
60
|
+
end
|
61
|
+
end
|
62
|
+
assert_equal 2, scraper.test(scraper.document).size
|
63
|
+
end
|
64
|
+
|
65
|
+
|
66
|
+
def test_array_selectors
|
67
|
+
scraper = new_scraper(DIVS123) do
|
68
|
+
selector :test, "#?", "2"
|
69
|
+
end
|
70
|
+
assert_equal 1, scraper.test(scraper.document).size
|
71
|
+
assert_equal "2", scraper.test(scraper.document)[0].attributes["id"]
|
72
|
+
end
|
73
|
+
|
74
|
+
|
75
|
+
def test_object_selectors
|
76
|
+
scraper = new_scraper(DIVS123) do
|
77
|
+
selector :test, HTML::Selector.new("div")
|
78
|
+
end
|
79
|
+
assert_equal 3, scraper.test(scraper.document).size
|
80
|
+
end
|
81
|
+
|
82
|
+
|
83
|
+
def test_selector_returns_array
|
84
|
+
scraper = new_scraper(DIVS123) do
|
85
|
+
selector :test0, "#4"
|
86
|
+
selector :test1, "#1"
|
87
|
+
selector :test3, "div"
|
88
|
+
end
|
89
|
+
assert_equal 0, scraper.test0(scraper.document).size # No elements (empty)
|
90
|
+
assert_equal 1, scraper.test1(scraper.document).size # One element (array)
|
91
|
+
assert_equal 3, scraper.test3(scraper.document).size # Array of elements
|
92
|
+
end
|
93
|
+
|
94
|
+
|
95
|
+
def test_select_in_document_order
|
96
|
+
scraper = new_scraper(DIVS123) do
|
97
|
+
selector :test, "#2,#1"
|
98
|
+
end
|
99
|
+
assert_equal 2, scraper.test(scraper.document).size
|
100
|
+
assert_equal "1", scraper.test(scraper.document)[0].attributes["id"]
|
101
|
+
assert_equal "2", scraper.test(scraper.document)[1].attributes["id"]
|
102
|
+
end
|
103
|
+
|
104
|
+
|
105
|
+
def test_selecting_first_element
|
106
|
+
scraper = new_scraper(DIVS123) do
|
107
|
+
selector :test, "div"
|
108
|
+
end
|
109
|
+
assert_equal 3, scraper.test(scraper.document).size
|
110
|
+
assert scraper.first_test(scraper.document)
|
111
|
+
assert_equal "1", scraper.first_test(scraper.document).attributes["id"]
|
112
|
+
|
113
|
+
scraper = new_scraper(DIVS123) do
|
114
|
+
selector :test, "div" do |element|
|
115
|
+
element[0].attributes["id"]
|
116
|
+
end
|
117
|
+
end
|
118
|
+
assert scraper.first_test(scraper.document)
|
119
|
+
assert_equal "1", scraper.first_test(scraper.document)
|
120
|
+
end
|
121
|
+
|
122
|
+
|
123
|
+
#
|
124
|
+
# Tests process methods.
|
125
|
+
#
|
126
|
+
|
127
|
+
def test_processing_rule
|
128
|
+
scraper = new_scraper(DIVS123) do
|
129
|
+
process "div" do |element|
|
130
|
+
@count = (@count || 0) + 1
|
131
|
+
end
|
132
|
+
attr :count
|
133
|
+
end
|
134
|
+
scraper.scrape
|
135
|
+
assert_equal 3, scraper.count
|
136
|
+
end
|
137
|
+
|
138
|
+
|
139
|
+
def test_processing_rule_with_array
|
140
|
+
scraper = new_scraper(DIVS123) do
|
141
|
+
process "#?", "1" do |element|
|
142
|
+
@count = (@count || 0) + 1
|
143
|
+
end
|
144
|
+
attr :count
|
145
|
+
end
|
146
|
+
scraper.scrape
|
147
|
+
assert_equal 1, scraper.count
|
148
|
+
end
|
149
|
+
|
150
|
+
|
151
|
+
def test_processing_rule_with_selector
|
152
|
+
scraper = new_scraper(DIVS123) do
|
153
|
+
process HTML::Selector.new("div") do |element|
|
154
|
+
@count = (@count || 0) + 1
|
155
|
+
end
|
156
|
+
attr :count
|
157
|
+
end
|
158
|
+
scraper.scrape
|
159
|
+
assert_equal 3, scraper.count
|
160
|
+
end
|
161
|
+
|
162
|
+
|
163
|
+
def test_extracting_in_code
|
164
|
+
scraper = new_scraper(DIVS123) do
|
165
|
+
process "div" do |element|
|
166
|
+
@concat = (@concat || "") << element.attributes["id"]
|
167
|
+
end
|
168
|
+
attr :concat
|
169
|
+
end
|
170
|
+
scraper.scrape
|
171
|
+
assert_equal "123", scraper.concat
|
172
|
+
end
|
173
|
+
|
174
|
+
|
175
|
+
def test_processing_in_document_order
|
176
|
+
scraper = new_scraper(DIVS123) do
|
177
|
+
process "#2,#1" do |element|
|
178
|
+
@concat = (@concat || "") << element.attributes["id"]
|
179
|
+
end
|
180
|
+
attr :concat
|
181
|
+
end
|
182
|
+
scraper.scrape
|
183
|
+
assert_equal "12", scraper.concat
|
184
|
+
end
|
185
|
+
|
186
|
+
|
187
|
+
def test_process_once_if_skipped
|
188
|
+
scraper = new_scraper(DIVS123) do
|
189
|
+
def prepare(document)
|
190
|
+
@found = []
|
191
|
+
end
|
192
|
+
process("#1") { |element| @found[0] = true }
|
193
|
+
process("#1") { |element| @found[1] = true ; skip element }
|
194
|
+
process("#1") { |element| @found[2] = true }
|
195
|
+
process("#2", :skip=>true){ |element| @found[3] = true }
|
196
|
+
process("#2") { |element| @found[4] = true }
|
197
|
+
attr_reader :found
|
198
|
+
end
|
199
|
+
scraper.scrape
|
200
|
+
assert_equal [true, true, nil, true], scraper.found
|
201
|
+
end
|
202
|
+
|
203
|
+
|
204
|
+
def test_skip_children
|
205
|
+
scraper = new_scraper(DIVS1_23) do
|
206
|
+
process "div" do |element|
|
207
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
208
|
+
skip id2(element)
|
209
|
+
end
|
210
|
+
selector :id2, "#2"
|
211
|
+
attr :concat
|
212
|
+
end
|
213
|
+
scraper.scrape
|
214
|
+
assert_equal "13", scraper.concat
|
215
|
+
end
|
216
|
+
|
217
|
+
|
218
|
+
def test_skip_descendants
|
219
|
+
# Root, child of root, grandchild of root.
|
220
|
+
scraper = new_scraper(DIVS1_23) do
|
221
|
+
process "div" do |element|
|
222
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
223
|
+
end
|
224
|
+
attr :concat
|
225
|
+
end
|
226
|
+
scraper.scrape
|
227
|
+
assert_equal "123", scraper.concat
|
228
|
+
|
229
|
+
# Stop at root.
|
230
|
+
scraper = new_scraper(DIVS1_23) do
|
231
|
+
process "div" do |element|
|
232
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
233
|
+
skip
|
234
|
+
end
|
235
|
+
attr :concat
|
236
|
+
end
|
237
|
+
scraper.scrape
|
238
|
+
assert_equal "1", scraper.concat
|
239
|
+
|
240
|
+
scraper.scrape
|
241
|
+
# Child of root, and child of root's child
|
242
|
+
scraper = new_scraper(DIVS1_23) do
|
243
|
+
process "div>div" do |element|
|
244
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
245
|
+
end
|
246
|
+
attr :concat
|
247
|
+
end
|
248
|
+
scraper.scrape
|
249
|
+
assert_equal "23", scraper.concat
|
250
|
+
|
251
|
+
# Stop at child of root.
|
252
|
+
scraper = new_scraper(DIVS1_23) do
|
253
|
+
process "div>div" do |element|
|
254
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
255
|
+
skip element.next_element
|
256
|
+
end
|
257
|
+
attr :concat
|
258
|
+
end
|
259
|
+
scraper.scrape
|
260
|
+
assert_equal "2", scraper.concat
|
261
|
+
|
262
|
+
# Child of root, the child of child of root.
|
263
|
+
scraper = new_scraper(DIVS1_23) do
|
264
|
+
process "div div" do |element|
|
265
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
266
|
+
end
|
267
|
+
attr :concat
|
268
|
+
end
|
269
|
+
scraper.scrape
|
270
|
+
assert_equal "23", scraper.concat
|
271
|
+
|
272
|
+
# Child of root.
|
273
|
+
scraper = new_scraper(DIVS1_23) do
|
274
|
+
process "div div" do |element|
|
275
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
276
|
+
skip element.next_element
|
277
|
+
end
|
278
|
+
attr :concat
|
279
|
+
end
|
280
|
+
scraper.scrape
|
281
|
+
assert_equal "2", scraper.concat
|
282
|
+
end
|
283
|
+
|
284
|
+
|
285
|
+
def test_skip_from_extractor
|
286
|
+
html = %Q{<div id="1">this</div>"}
|
287
|
+
scraper = new_scraper(html) do
|
288
|
+
process "#1", :this1=>:text
|
289
|
+
process "#1", :this2=>:text
|
290
|
+
end
|
291
|
+
scraper.scrape
|
292
|
+
assert_equal "this", scraper.this1
|
293
|
+
assert_equal "this", scraper.this2
|
294
|
+
|
295
|
+
scraper = new_scraper(html) do
|
296
|
+
process "#1", :this1=>:text, :skip=>false
|
297
|
+
process "#1", :this2=>:text
|
298
|
+
end
|
299
|
+
scraper.scrape
|
300
|
+
assert_equal "this", scraper.this1
|
301
|
+
assert_equal "this", scraper.this2
|
302
|
+
|
303
|
+
scraper = new_scraper(html) do
|
304
|
+
process "#1", :this1=>:text, :skip=>true do
|
305
|
+
false
|
306
|
+
end
|
307
|
+
process "#1", :this2=>:text
|
308
|
+
end
|
309
|
+
scraper.scrape
|
310
|
+
assert_equal "this", scraper.this1
|
311
|
+
assert_equal nil, scraper.this2
|
312
|
+
end
|
313
|
+
|
314
|
+
|
315
|
+
def test_stop
|
316
|
+
scraper = new_scraper(DIVS123) do
|
317
|
+
process "div" do |element|
|
318
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
319
|
+
stop
|
320
|
+
end
|
321
|
+
attr :concat
|
322
|
+
end
|
323
|
+
scraper.scrape
|
324
|
+
assert_equal "1", scraper.concat
|
325
|
+
end
|
326
|
+
|
327
|
+
|
328
|
+
def test_process_first
|
329
|
+
scraper = new_scraper(DIVS123) do
|
330
|
+
process "div" do |element|
|
331
|
+
@all = (@all || 0) + 1
|
332
|
+
end
|
333
|
+
process_first "div" do |element|
|
334
|
+
@first = (@first || 0) + 1
|
335
|
+
end
|
336
|
+
attr_accessor :all, :first
|
337
|
+
end
|
338
|
+
scraper.scrape
|
339
|
+
assert_equal 3, scraper.all
|
340
|
+
assert_equal 1, scraper.first
|
341
|
+
end
|
342
|
+
|
343
|
+
|
344
|
+
def test_accessors
|
345
|
+
time = Time.new.rfc2822
|
346
|
+
Net::HTTP.on_get do |address, path, headers|
|
347
|
+
if path == "/redirect"
|
348
|
+
response = Net::HTTPSuccess.new(Net::HTTP.version_1_2, 200, "OK")
|
349
|
+
response["Last-Modified"] = time
|
350
|
+
response["ETag"] = "etag"
|
351
|
+
[response, <<-EOF
|
352
|
+
<html>
|
353
|
+
<head>
|
354
|
+
<meta http-equiv="content-type" value="text/html; charset=other-encoding">
|
355
|
+
</head>
|
356
|
+
<body>
|
357
|
+
<div id="x"/>
|
358
|
+
</body>
|
359
|
+
</html>
|
360
|
+
EOF
|
361
|
+
]
|
362
|
+
else
|
363
|
+
response = Net::HTTPMovedPermanently.new(Net::HTTP.version_1_2, 300, "Moved")
|
364
|
+
response["Location"] = "http://localhost/redirect"
|
365
|
+
[response, ""]
|
366
|
+
end
|
367
|
+
end
|
368
|
+
scraper = new_scraper(URI.parse("http://localhost/source"))
|
369
|
+
scraper.scrape
|
370
|
+
assert_equal "http://localhost/source", scraper.page_info.original_url.to_s
|
371
|
+
assert_equal "http://localhost/redirect", scraper.page_info.url.to_s
|
372
|
+
assert_equal time, scraper.page_info.last_modified
|
373
|
+
assert_equal "etag", scraper.page_info.etag
|
374
|
+
assert_equal "other-encoding", scraper.page_info.encoding
|
375
|
+
end
|
376
|
+
|
377
|
+
|
378
|
+
def test_scraping_end_to_end
|
379
|
+
Net::HTTP.on_get do |address, path, headers|
|
380
|
+
[Net::HTTPSuccess.new(Net::HTTP.version_1_2, 200, "OK"), <<-EOF
|
381
|
+
<html>
|
382
|
+
<body>
|
383
|
+
<div id="1"/>
|
384
|
+
<div id="2"/>
|
385
|
+
</body>
|
386
|
+
</html>
|
387
|
+
EOF
|
388
|
+
]
|
389
|
+
end
|
390
|
+
scraper = new_scraper(URI.parse("http://localhost/")) do
|
391
|
+
process "div" do |element|
|
392
|
+
@concat = (@concat || "") << (element.attributes["id"] || "")
|
393
|
+
end
|
394
|
+
attr :concat
|
395
|
+
end
|
396
|
+
scraper.scrape
|
397
|
+
assert_equal "12", scraper.concat
|
398
|
+
end
|
399
|
+
|
400
|
+
|
401
|
+
#
|
402
|
+
# Tests extractor methods.
|
403
|
+
#
|
404
|
+
|
405
|
+
def test_extractors
|
406
|
+
html = %Q{<div id="1"></div>}
|
407
|
+
scraper = new_scraper(html) do
|
408
|
+
process "div", extractor(:div_id=>"@id")
|
409
|
+
attr :div_id
|
410
|
+
end
|
411
|
+
scraper.scrape
|
412
|
+
assert_equal "1", scraper.div_id
|
413
|
+
scraper = new_scraper(html) do
|
414
|
+
process "div", :div_id=>"@id"
|
415
|
+
attr :div_id
|
416
|
+
end
|
417
|
+
scraper.scrape
|
418
|
+
assert_equal "1", scraper.div_id
|
419
|
+
end
|
420
|
+
|
421
|
+
|
422
|
+
def test_text_and_element_extractors
|
423
|
+
html = %Q{<div>some text</div>}
|
424
|
+
# Extract the node itself.
|
425
|
+
scraper = new_scraper(html) do
|
426
|
+
process "div", extractor(:value=>:element)
|
427
|
+
attr :value
|
428
|
+
end
|
429
|
+
scraper.scrape
|
430
|
+
assert_equal "div", scraper.value.name
|
431
|
+
# Extract the text value of the node.
|
432
|
+
scraper = new_scraper(html) do
|
433
|
+
process "div", extractor(:value=>:text)
|
434
|
+
attr :value
|
435
|
+
end
|
436
|
+
scraper.scrape
|
437
|
+
assert_equal "some text", scraper.value
|
438
|
+
end
|
439
|
+
|
440
|
+
|
441
|
+
def test_extractors_objects
|
442
|
+
html = <<-EOF
|
443
|
+
<h1 class="header"></h1>
|
444
|
+
<h2 class="header"></h2>
|
445
|
+
EOF
|
446
|
+
# Extract both elements based on class, return the second one.
|
447
|
+
scraper = new_scraper(html) do
|
448
|
+
process ".header", extractor(:header=>:element)
|
449
|
+
attr :header
|
450
|
+
end
|
451
|
+
scraper.scrape
|
452
|
+
assert_equal "h2", scraper.header.name
|
453
|
+
# Extracting a specific element skips the second match.
|
454
|
+
html = <<-EOF
|
455
|
+
<h1 class="header"></h1>
|
456
|
+
<h2 class="header"></h2>
|
457
|
+
EOF
|
458
|
+
scraper = new_scraper(html) do
|
459
|
+
process ".header", extractor(:header=>"h1")
|
460
|
+
attr :header
|
461
|
+
end
|
462
|
+
scraper.scrape
|
463
|
+
assert_equal "h1", scraper.header.name
|
464
|
+
end
|
465
|
+
|
466
|
+
|
467
|
+
def test_attribute_extractors
|
468
|
+
# Extracting the attribute skips the second match.
|
469
|
+
html = <<-EOF
|
470
|
+
<abbr title="foo">bar</div>
|
471
|
+
<abbr>foo</abbr>
|
472
|
+
EOF
|
473
|
+
scraper = new_scraper(html) do
|
474
|
+
process "abbr", extractor(:title=>"@title")
|
475
|
+
attr :title
|
476
|
+
end
|
477
|
+
scraper.scrape
|
478
|
+
assert_equal "foo", scraper.title
|
479
|
+
# Extracting a specific element skips the second match.
|
480
|
+
html = <<-EOF
|
481
|
+
<h1 class="header" id="1"></h1>
|
482
|
+
<h2 class="header" id="2"></h2>
|
483
|
+
EOF
|
484
|
+
scraper = new_scraper(html) do
|
485
|
+
process ".header", extractor(:header=>"h1@id")
|
486
|
+
attr :header
|
487
|
+
end
|
488
|
+
scraper.scrape
|
489
|
+
assert_equal "1", scraper.header
|
490
|
+
end
|
491
|
+
|
492
|
+
|
493
|
+
def test_class_extractors
|
494
|
+
headers = Class.new(Scraper::Base)
|
495
|
+
headers.instance_eval do
|
496
|
+
root_element nil
|
497
|
+
process "h1,h2", :h1=>"h1", :h2=>"h2"
|
498
|
+
attr :h1
|
499
|
+
attr :h2
|
500
|
+
end
|
501
|
+
html = <<-EOF
|
502
|
+
<div>
|
503
|
+
<h1>first</h1>
|
504
|
+
<h2>second</h2>
|
505
|
+
</div>
|
506
|
+
EOF
|
507
|
+
scraper = new_scraper(html) do
|
508
|
+
process "div", extractor(:headers=>headers)
|
509
|
+
attr :headers
|
510
|
+
end
|
511
|
+
scraper.scrape
|
512
|
+
assert scraper.headers
|
513
|
+
assert_equal "h1", scraper.headers.h1.name
|
514
|
+
assert_equal "h2", scraper.headers.h2.name
|
515
|
+
end
|
516
|
+
|
517
|
+
|
518
|
+
def test_array_extractors
|
519
|
+
html = <<-EOF
|
520
|
+
<div>
|
521
|
+
<h1>first</h1>
|
522
|
+
<h1>second</h1>
|
523
|
+
</div>
|
524
|
+
EOF
|
525
|
+
scraper = new_scraper(html) do
|
526
|
+
process "h1", extractor("headers[]"=>:text)
|
527
|
+
attr :headers
|
528
|
+
end
|
529
|
+
scraper.scrape
|
530
|
+
assert scraper.headers.is_a?(Array)
|
531
|
+
assert_equal 2, scraper.headers.size
|
532
|
+
assert_equal "first", scraper.headers[0]
|
533
|
+
assert_equal "second", scraper.headers[1]
|
534
|
+
end
|
535
|
+
|
536
|
+
|
537
|
+
def test_hash_extractors
|
538
|
+
html = <<-EOF
|
539
|
+
<div>
|
540
|
+
<h1 id="1" class="header">first</h1>
|
541
|
+
</div>
|
542
|
+
EOF
|
543
|
+
scraper = new_scraper(html) do
|
544
|
+
process "h1", extractor("header"=>{:id=>"@id", :class=>"@class", :text=>:text})
|
545
|
+
attr :header
|
546
|
+
end
|
547
|
+
scraper.scrape
|
548
|
+
assert scraper.header.is_a?(Hash)
|
549
|
+
assert_equal 3, scraper.header.size
|
550
|
+
assert_equal "1", scraper.header[:id]
|
551
|
+
assert_equal "header", scraper.header[:class]
|
552
|
+
assert_equal "first", scraper.header[:text]
|
553
|
+
end
|
554
|
+
|
555
|
+
|
556
|
+
def test_multi_value_extractors
|
557
|
+
html = <<-EOF
|
558
|
+
<div>
|
559
|
+
<h1 id="1" class="header">first</h1>
|
560
|
+
</div>
|
561
|
+
EOF
|
562
|
+
scraper = new_scraper(html) do
|
563
|
+
process "h1", [:text, :kls]=>Scraper.define {
|
564
|
+
process "*", :text=>:text, :kls=>"@class"
|
565
|
+
}
|
566
|
+
end
|
567
|
+
result = scraper.scrape
|
568
|
+
assert "first", result.text
|
569
|
+
assert "header", result.kls
|
570
|
+
end
|
571
|
+
|
572
|
+
|
573
|
+
def test_conditional_extractors
|
574
|
+
# Look for id attribute (second header only),
|
575
|
+
# if not found look for class attribute (first
|
576
|
+
# two headers), otherwise just get text (third
|
577
|
+
# header).
|
578
|
+
html = <<-EOF
|
579
|
+
<div>
|
580
|
+
<h1 class="foo">first</h1>
|
581
|
+
<h1 class="foo" id="bar">second</h1>
|
582
|
+
<h1>third</h1>
|
583
|
+
</div>
|
584
|
+
EOF
|
585
|
+
scraper = new_scraper(html) do
|
586
|
+
process "h1", extractor("headers[]"=>["@id", "@class", :text])
|
587
|
+
attr :headers
|
588
|
+
end
|
589
|
+
scraper.scrape
|
590
|
+
assert scraper.headers.is_a?(Array)
|
591
|
+
assert_equal 3, scraper.headers.size
|
592
|
+
assert_equal "foo", scraper.headers[0]
|
593
|
+
assert_equal "bar", scraper.headers[1]
|
594
|
+
assert_equal "third", scraper.headers[2]
|
595
|
+
end
|
596
|
+
|
597
|
+
|
598
|
+
DIVS_ST_ND = <<-EOF
|
599
|
+
<div id="1">first</div>
|
600
|
+
<div id="2">second</div>
|
601
|
+
EOF
|
602
|
+
|
603
|
+
def test_accessors_from_extractor
|
604
|
+
scraper = new_scraper(DIVS_ST_ND) do
|
605
|
+
process_first "div", :div_id=>"@id", :div_text=>:text
|
606
|
+
result :div_id
|
607
|
+
end
|
608
|
+
value = scraper.scrape
|
609
|
+
assert_equal "1", value
|
610
|
+
|
611
|
+
scraper = new_scraper(DIVS_ST_ND) do
|
612
|
+
process_first "div", :div_id=>"@id", :div_text=>:text
|
613
|
+
result :div_id, :div_text
|
614
|
+
end
|
615
|
+
value = scraper.scrape
|
616
|
+
assert_equal "1", value.div_id
|
617
|
+
assert_equal "first", value.div_text
|
618
|
+
|
619
|
+
scraper = new_scraper(DIVS_ST_ND) do
|
620
|
+
process_first "div", :div_id=>"@id", :div_text=>:text
|
621
|
+
end
|
622
|
+
value = scraper.scrape
|
623
|
+
assert_equal "1", value.div_id
|
624
|
+
assert_equal "first", value.div_text
|
625
|
+
|
626
|
+
scraper = new_scraper(DIVS_ST_ND) do
|
627
|
+
attr_accessor :div_class
|
628
|
+
process_first "div", :div_id=>"@id", :div_text=>:text
|
629
|
+
result :div_id, :div_class
|
630
|
+
end
|
631
|
+
value = scraper.scrape
|
632
|
+
assert_equal "1", value.div_id
|
633
|
+
assert_raise(NoMethodError) { value.div_text }
|
634
|
+
|
635
|
+
scraper = new_scraper(DIVS_ST_ND) do
|
636
|
+
process "div", "div_ids[]"=>"@id"
|
637
|
+
result :div_ids
|
638
|
+
end
|
639
|
+
value = scraper.scrape
|
640
|
+
assert_equal "1", value[0]
|
641
|
+
assert_equal "2", value[1]
|
642
|
+
end
|
643
|
+
|
644
|
+
|
645
|
+
def test_array_accessors
|
646
|
+
scraper = new_scraper(DIVS_ST_ND) do
|
647
|
+
array :div_id, :div_text
|
648
|
+
process "div", :div_id=>"@id", :div_text=>:text
|
649
|
+
result :div_id, :div_text
|
650
|
+
end
|
651
|
+
value = scraper.scrape
|
652
|
+
assert_equal 2, value.div_id.size
|
653
|
+
assert_equal 2, value.div_text.size
|
654
|
+
assert_equal "1", value.div_id[0]
|
655
|
+
assert_equal "2", value.div_id[1]
|
656
|
+
assert_equal "first", value.div_text[0]
|
657
|
+
assert_equal "second", value.div_text[1]
|
658
|
+
end
|
659
|
+
|
660
|
+
|
661
|
+
#
|
662
|
+
# Root element tests.
|
663
|
+
#
|
664
|
+
|
665
|
+
HTML_EMPTY = <<-EOF
|
666
|
+
<html>
|
667
|
+
<head>
|
668
|
+
</head>
|
669
|
+
<body>
|
670
|
+
</body>
|
671
|
+
</html>
|
672
|
+
EOF
|
673
|
+
|
674
|
+
def test_scrape_body_by_default
|
675
|
+
scraper = Class.new(Scraper::Base).new(HTML_EMPTY)
|
676
|
+
scraper.class.instance_eval do
|
677
|
+
process "head" do |element| @head = element end
|
678
|
+
attr :head
|
679
|
+
process "body" do |element| @body = element end
|
680
|
+
attr :body
|
681
|
+
end
|
682
|
+
scraper.scrape
|
683
|
+
assert scraper.head
|
684
|
+
assert scraper.body
|
685
|
+
end
|
686
|
+
|
687
|
+
|
688
|
+
def test_changing_root_element
|
689
|
+
only_header = new_scraper(HTML_EMPTY) do
|
690
|
+
root_element "head"
|
691
|
+
process "head" do |element| @head = element end
|
692
|
+
attr :head
|
693
|
+
process "body" do |element| @body = element end
|
694
|
+
attr :body
|
695
|
+
end
|
696
|
+
only_body = Class.new(only_header.class).new(HTML_EMPTY)
|
697
|
+
only_body.class.root_element "body"
|
698
|
+
both_parts = Class.new(only_body.class).new(HTML_EMPTY)
|
699
|
+
both_parts.class.root_element nil
|
700
|
+
# We set this scraper to begin with the head element,
|
701
|
+
# so we can see the head element, but not the body.
|
702
|
+
only_header.scrape
|
703
|
+
assert only_header.head
|
704
|
+
assert only_header.body.nil?
|
705
|
+
# Now switch to a scraper that processes the body element,
|
706
|
+
# skipping the header.
|
707
|
+
only_body.scrape
|
708
|
+
assert only_body.head.nil?
|
709
|
+
assert only_body.body
|
710
|
+
# Now switch to a scraper that doesn't specify a root element,
|
711
|
+
# and it will process both header and body.
|
712
|
+
both_parts.scrape
|
713
|
+
assert both_parts.head
|
714
|
+
assert both_parts.body
|
715
|
+
end
|
716
|
+
|
717
|
+
|
718
|
+
# Test prepare/result.
|
719
|
+
|
720
|
+
def test_prepare_and_result
|
721
|
+
# Extracting the attribute skips the second match.
|
722
|
+
scraper = new_scraper(DIVS123) do
|
723
|
+
process("div") { |element| @count +=1 }
|
724
|
+
define_method(:prepare) { @count = 1 }
|
725
|
+
define_method(:result) { @count }
|
726
|
+
end
|
727
|
+
result = scraper.scrape
|
728
|
+
assert_equal 4, result
|
729
|
+
end
|
730
|
+
|
731
|
+
|
732
|
+
def test_changing_document_from_prepare
|
733
|
+
# Extracting the attribute skips the second match.
|
734
|
+
scraper = new_scraper(DIVS123) do
|
735
|
+
selector :divs, "div"
|
736
|
+
define_method :prepare do |document|
|
737
|
+
@document = divs(document)[1]
|
738
|
+
end
|
739
|
+
array :ids
|
740
|
+
process "div", :ids=>"@id"
|
741
|
+
result :ids
|
742
|
+
end
|
743
|
+
result = scraper.scrape
|
744
|
+
assert_equal 1, result.size
|
745
|
+
assert_equal "2", result[0]
|
746
|
+
end
|
747
|
+
|
748
|
+
|
749
|
+
def test_anonymous_scrapers
|
750
|
+
scraper = Scraper.define do
|
751
|
+
array :ids
|
752
|
+
process "div", :ids=>"@id"
|
753
|
+
result :ids
|
754
|
+
end
|
755
|
+
result = scraper.scrape(DIVS123)
|
756
|
+
assert_equal "1", result[0]
|
757
|
+
assert_equal "2", result[1]
|
758
|
+
assert_equal "3", result[2]
|
759
|
+
end
|
760
|
+
|
761
|
+
|
762
|
+
def test_named_rules
|
763
|
+
scraper = Scraper.define do
|
764
|
+
array :ids1, :ids2
|
765
|
+
process :main, "div", :ids1=>"@id"
|
766
|
+
process :main, "div", :ids2=>"@id"
|
767
|
+
result :ids1, :ids2
|
768
|
+
end
|
769
|
+
result = scraper.scrape(DIVS123)
|
770
|
+
assert_equal nil, result.ids1
|
771
|
+
assert_equal 3, result.ids2.size
|
772
|
+
assert_equal "1", result.ids2[0]
|
773
|
+
assert_equal "2", result.ids2[1]
|
774
|
+
assert_equal "3", result.ids2[2]
|
775
|
+
end
|
776
|
+
|
777
|
+
|
778
|
+
protected
|
779
|
+
|
780
|
+
def new_scraper(what, &block)
|
781
|
+
cls = Class.new(Scraper::Base)
|
782
|
+
cls.root_element nil
|
783
|
+
cls.parser :html_parser
|
784
|
+
cls.class_eval &block if block
|
785
|
+
cls.new(what)
|
786
|
+
end
|
787
|
+
|
788
|
+
end
|
789
|
+
|
790
|
+
|
791
|
+
# Repeats the same set of tests, but using Tidy instead of HTMLParser.
|
792
|
+
class ScraperUsingTidyTest < ScraperTest
|
793
|
+
|
794
|
+
protected
|
795
|
+
|
796
|
+
def new_scraper(what, &block)
|
797
|
+
cls = Class.new(Scraper::Base)
|
798
|
+
cls.root_element nil
|
799
|
+
cls.parser :tidy
|
800
|
+
cls.class_eval &block if block
|
801
|
+
cls.new(what)
|
802
|
+
end
|
803
|
+
|
804
|
+
end
|