1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288 | """
Crawls movie data from iTunes website.
Crawls the webpages in iTunes Store (http://itunes.apple.com) to retreive the
movie data. All webpages under the genres of our interest are crawled.
Usage:
scrapy crawl itunes [options]
Examples:
# Crawl all movies in Play.
scrapy crawl itunes
# Crawl all movies in Play and write them crawl movie items to
# itunes.json in JSON format.
scrapy crawl itunes -o itunes.json -t json
# Test single movie items.
scrapy crawl itunes -a test=movie_item
# Test few movie listing pages.
scrapy crawl itunes -a test=listing
"""
import errno
import os
import re
from scrapy.contrib.loader import XPathItemLoader
from scrapy.contrib.loader.processor import MapCompose
from scrapy.http import Request
from scrapy.selector import HtmlXPathSelector
from scrapy.spider import BaseSpider
from movies.items import ItunesMovieItem
DOMAIN = 'itunes.apple.com'
SITE_DOMAIN = 'https://' + DOMAIN
SITE_PREFIX = SITE_DOMAIN + '/us/movie/'
SAMPLE_MOVIE_ITEMS = (
'https://itunes.apple.com/us/movie/the-fault-in-our-stars/id899849704',
'https://itunes.apple.com/us/movie/the-descendants/id485148916',
'https://itunes.apple.com/us/movie/drive/id492170756',
)
SAMPLE_MOVIE_LISTING = (
'https://itunes.apple.com/us/genre/movies-action-adventure/id4401?letter=I&page=2#page',
)
SITE_GENRE = SITE_DOMAIN + '/us/genre/movies-'
GENRES = (
SITE_GENRE + 'action-adventure/id4401',
SITE_GENRE + 'anime/id4402',
SITE_GENRE + 'bollywood/id4431',
SITE_GENRE + 'classics/id4403',
SITE_GENRE + 'comedy/id4404',
SITE_GENRE + 'documentary/id4405',
SITE_GENRE + 'drama/id4406',
SITE_GENRE + 'foreign/id4407',
SITE_GENRE + 'holiday/id4420',
SITE_GENRE + 'horror/id4408',
SITE_GENRE + 'independent/id4409',
SITE_GENRE + 'kids-family/id4410',
SITE_GENRE + 'music-documentaries/id4423',
SITE_GENRE + 'music-feature-films/id4424',
SITE_GENRE + 'musicals/id4411',
SITE_GENRE + 'regional-indian/id4432',
SITE_GENRE + 'romance/id4412',
SITE_GENRE + 'sci-fi-fantasy/id4413',
SITE_GENRE + 'special-interest/id4415',
SITE_GENRE + 'sports/id4417',
SITE_GENRE + 'thriller/id4416',
SITE_GENRE + 'urban/id4419',
SITE_GENRE + 'western/id4418',
)
def MkDir(path):
"""Creates a directory with the given path.
Create the directory (possibly recursively) and ignore if it already exists.
Args:
path: A valid path to create a directory.
"""
try:
os.makedirs(path)
except OSError as exc:
if exc.errno == errno.EEXIST and os.path.isdir(path):
pass
else: raise
def GetID(url):
"""Fetches the id of the movie from the url.
Args:
url: The url of the webpage for the movie.
Returns:
The id of the movie. For example:
If url is https://itunes.apple.com/us/movie/the-fault-in-our-stars/id899849704
id = 899849704
"""
m = re.match('.*id(\d*)', url)
return m.group(1)
class ItunesSpider(BaseSpider):
"""The Spider class which crawls iTunes movie data.
Attributes:
name: Name of the spider.
allowed_domains: The domains this spider is allowed to crawl.
mode: Mode of operation of the spider. 'normal' crawls the entire site.
'test_movie_item' tests crawling a few movie items. 'test_listing'
tests crawling a few movie listing pages.
seen: Dictionary of movie-ids already crawled.
"""
name = "itunes"
allowed_domains = [DOMAIN]
def __init__(self, *args, **kwargs):
"""Constructor.
Initializes iTunes spider with normal mode or test mode.
Args:
args: Ignored.
kwargs: If 'test' key is set, mode is changed to:
'test_' + kwargs['test']
"""
super(ItunesSpider, self).__init__(*args, **kwargs)
self.seen = set()
if kwargs.has_key('test'):
self.mode = 'test_' + kwargs['test']
else:
self.mode = 'normal'
def start_requests(self):
"""Returns intial seed URLs to start crawling.
Starts requesting pages under each genre in alphabetical order.
Yields:
Request() objects for all genres, which are processed by
self.parse_pages() method.
"""
if self.mode == 'test_movie_item':
for url in SAMPLE_MOVIE_ITEMS:
yield Request(url, callback=self.parse_item)
elif self.mode == 'test_listing':
for url in SAMPLE_MOVIE_LISTING:
yield Request(url, callback=self.parse_listing)
elif self.mode == 'test_genre':
#genre = SITE_GENRE + 'classics/id4403'
genre = SITE_GENRE + 'action-adventure/id4401'
letters = map(chr, xrange(ord('A'), ord('A') + 26)) + [ '*' ]
for letter in letters:
letter_url = genre + '?letter=' + letter
yield Request(letter_url, callback=self.parse_pages)
elif self.mode == 'normal':
for url in GENRES:
letters = map(chr, xrange(ord('A'), ord('A') + 26)) + [ '*' ]
for letter in letters:
letter_url = url + '?letter=' + letter
yield Request(letter_url, callback=self.parse_pages)
def parse_pages(self, response):
"""Parse the genre page.
Parse the genre page and return Request()'s for all pages under a
particular alphabet.
Args:
response: Response object for the genre page. response.body has the
text of the response.
Yields:
Request() objects for all the pages in that particular movie genre,
which are processed by self.parse_listing() method.
"""
self.dump('genre', response.url, response.body)
hxs = HtmlXPathSelector(response)
atags = hxs.select('//div[@id="selectedgenre"]/ul[@class="list paginate"][1]/li/a/@href')
if len(atags) > 0:
for url in atags.extract():
yield Request(url, callback=self.parse_listing)
else:
# All entries in one page.
for item in self.parse_listing(response):
yield item
def parse_listing(self, response):
"""Parse all movies within a particular page.
Args:
response: Response object for the movie listing page. response.body
has the text of the response.
Yields:
Request() objects for all the movies in that movie listing page,
which are processed by self.parse_item() method.
"""
self.dump('listing', response.url, response.body)
hxs = HtmlXPathSelector(response)
atags = hxs.select('//div[@id="selectedcontent"]//ul/li/a/@href')
for url in atags.extract():
movie_id = GetID(url)
if movie_id not in self.seen:
self.seen.add(movie_id)
yield Request(url, callback=self.parse_item)
def parse_item(self, response):
"""Parse all the necessary information from a movie page.
Args:
response: Response object for the movie page. response.body has the
text of the response.
Yields:
ItunesMovieItem() object with all the movie attributes parsed.
"""
self.dump('item', response.url, response.body)
hxs = HtmlXPathSelector(response)
loader = XPathItemLoader(item=ItunesMovieItem(), selector=hxs)
loader.add_value('id',
GetID(hxs.select('//head/link[@rel="canonical"]/@href').extract()[0]))
loader.add_xpath('name', '//div[@id="title"]/div/h1[1]/text()')
loader.add_xpath('description', '//div[@class="plot-summary"]/p/text()')
loader.add_xpath('actors', '//div[@metrics-loc="Titledbox_Actors"]/ul/li/a/text()')
loader.add_xpath('directors', '//div[@metrics-loc="Titledbox_Director"]/ul/li/a/text()')
loader.add_xpath('writers', '//div[@metrics-loc="Titledbox_Screenwriter"]/ul/li/a/text()')
loader.add_xpath('producers', '//div[@metrics-loc="Titledbox_Producers"]/ul/li/a/text()')
loader.add_xpath('price', '//span[@class="price"]/text()')
loader.add_xpath('content_rating', '//span[@class="content-rating"]/text()')
loader.add_xpath('genre', '//li[@class="genre"]/a/text()')
loader.add_xpath('year', '//li[@class="release-date"]/text()')
loader.add_xpath('copyright', '//li[@class="copyright"]/text()')
loader.add_value('stars',
len(hxs.select('//div[@id="left-stack"]//span[@class="rating-star"]'))*1.0 +
len(hxs.select('//div[@id="left-stack"]//span[@class="rating-star half"]'))*0.5)
loader.add_xpath('star_ratings',
'//div[@id="left-stack"]//span[@class="rating-count"]/text()',
re='(.*) Ratings')
loader.add_xpath('rotten_tomatoes_tomatometer',
'//div[@class="tomatometer"]/span[@class="percent"]/text()',
MapCompose(unicode.strip))
loader.add_xpath('rotten_tomatoes_reviews',
'//div[@class="movie-review-left"]//li[@class="total-reviews"]/text()',
MapCompose(unicode.strip))
loader.add_xpath('rotten_tomatoes_fresh',
'//div[@class="movie-review-left"]//li[@class="fresh-reviews"]/text()',
MapCompose(unicode.strip))
loader.add_xpath('rotten_tomatoes_rotten',
'//div[@class="movie-review-left"]//li[@class="rotten-reviews"]/text()',
MapCompose(unicode.strip))
loader.add_xpath('rotten_tomatoes_average_rating',
'//div[@class="movie-review-left"]//li[@class="average-reviews"]/text()',
MapCompose(unicode.strip), re='(.*)/10')
loader.add_xpath('similar_movies', '//div[@metrics-loc="Titledbox_Viewers Also Bought"]//a[@class="name"]/text()')
loader.add_xpath('similar_movies_id',
'//div[@metrics-loc="Titledbox_Viewers Also Bought"]//a[@class="name"]/@href',
MapCompose(GetID))
return loader.load_item()
def dump(self, type, url, body):
""" Dump the HTML content in the corresponding directories.
Args:
type: Type of the URL.
url: URL of the page.
body: Contents of the URL.
"""
directory = 'data/{0}/{1}'.format(self.name, type)
MkDir(directory)
filename = '{0}/{1}'.format(directory, self.dump_filename(url))
with open(filename, 'w') as dump:
dump.write(body)
def dump_filename(self, url):
"""Formats the filename before dumping.
Args:
url: URL of the HTML page.
Returns:
Sane path from the URL replacesing '/', '?', '&', and '=' with '_'.
"""
url = url[len(SITE_DOMAIN)+1:].replace('/', '_').replace('?',
'_').replace('&', '_').replace('=', '_')
return url + '.html'
|