class ArticleItemLoader(ItemLoader):
default_item_class = ArticleItem
default_output_processor = TakeFirst()
title_out = Compose(TakeFirst(), Net39ArticleTitle())
content_out = Compose(Join(''), Net39ArticleContent())
class ArticleLoader(XPathItemLoader):
"""
Used for easier construction of ArticleItem
"""
def is_string(string):
if isinstance(string, str) or isinstance(string, unicode):
if string.strip() != "":
#log.msg("returning string: "+ unicode(string.strip()))
return string.strip()
#log.msg("returning None for string: "+ unicode(string))
return None
def separate_tags(tags_string):
return tags_string.replace(";", ",").split(",")
default_input_processor = MapCompose(is_string)
default_output_processor = TakeFirst()
publishers_in = MapCompose(is_string)
publishers_out = Identity()
title_in = MapCompose(is_string, unicode.title)
title_out = TakeFirst()
time_published_in = MapCompose(is_string)
time_published_out = Identity()
summary_in = MapCompose(is_string)
summary_out = TakeFirst()
tags_in = MapCompose(is_string, separate_tags)
tags_out = Identity()
class CompraLineaItem(Item):
cantidad = Field(output_processor=TakeFirst())
unidad_medida = Field(output_processor=TakeFirst())
importe = Field(output_processor=lambda x: parse_money(x[0]))
importe_total = Field(output_processor=lambda x: parse_money(x[0]))
detalle = Field(output_processor=TakeFirst())
anio = Field(output_processor=TakeFirst())
def process_item(self, task_id):
report = self.db.loadScrapedFullReport(task_id)
if report is None:
return
text = report['full_report_body']
text = "".join(chr(min(ord(c),127)) for c in text)
t = TextResponse (url=report['full_report_url'], body=text.encode('utf-8')) #must have utf-8 here
l = XPathItemLoader(NrcParsedReport(), response=t)
l.add_value('reportnum', task_id)
patterns = self.compile_patterns ()
for p in patterns:
l.add_value(p[0], text, TakeFirst(), unicode.strip, re=p[1])
county = l.get_output_value('county')
pattern = self.get_area_code_pattern(county)
if pattern:
l.add_value ('areaid', county)
l.add_value('blockid', text, TakeFirst(), unicode.strip, re="%s[\s]+(?:BLOCK[\s]+)?([\d]+)" % pattern)
l.add_value('blockid', text, TakeFirst(), unicode.strip, re="BLOCK[\s]+([\d]+)")
item = l.load_item()
yield item
self.item_completed(task_id)
class ZhiHuU_T(Item):
'''
Zhihu user topic relationship
'''
crawled_from = Field(output_processor=TakeFirst())
user_url = Field(output_processor=TakeFirst())
topic_url = Field(output_processor=TakeFirst())
def test_get_xpath(self):
l = TestItemLoader(response=self.response)
self.assertEqual(l.get_xpath('//p/text()'), [u'paragraph'])
self.assertEqual(l.get_xpath('//p/text()', TakeFirst()), u'paragraph')
self.assertEqual(l.get_xpath('//p/text()', TakeFirst(), re='pa'), u'pa')
self.assertEqual(l.get_xpath(['//p/text()', '//div/text()']), [u'paragraph', 'marta'])
class VkItem(Item):
id = Field(output_processor=TakeFirst())
name = Field(output_processor=TakeFirst())
text = Field(input_processor=MapCompose(remove_tags),
output_processor=TakeFirst())
date = Field(output_processor=TakeFirst())
words = Field()
def parse(self, response):
"""
Default callback used by Scrapy to process download response
Testing contracts:
@url http://www.livingsocial.com/cities/15-san-francisco
@returns items 1
@scrapes title link
:param response:
:return:
"""
selector = HtmlXPathSelector(response)
# iterate over deals
for deal in selector.select(self.deals_list_xpath):
loader = XPathItemLoader(LivingSocialDeal(), selector=deal)
# define processors
loader.default_input_processor = TakeFirst()
loader.default_input_processor = MapCompose(unicode.strip)
loader.default_input_processor = Join()
loader.defalut_output_processor = TakeFirst()
# iterate over fields and add xpaths to the loader
for field, xpath in self.item_fields.iteritems():
loader.add_xpath(field, xpath)
yield loader.load_item()
class Product(Item):
id = Field()
name = Field(input_processor=Compose(TakeFirst(), unicode.strip))
price = Field(input_processor=Compose(TakeFirst(), unicode.strip,
remove_comma, float))
cat = Field()
avgStars = Field(input_processor=Compose(only_elem_or_default, float))
nReviews = Field(
input_processor=Compose(only_elem, unicode.strip, remove_comma, int))
salesRank = Field(
input_processor=Compose(unicode.strip, remove_comma, int))
subCatRank = Field(input_processor=Compose(
only_elem_or_default, unicode.strip, remove_comma, int))
subCat = Field(
input_processor=Compose(only_elem_or_default, unicode.strip))
manufact = Field(
input_processor=Compose(only_elem_or_default, unicode.strip))
referrer = Field()
@property
def export_filename(self):
return 'product'
@property
def key(self):
return self._values['id']
def parse_item(self, response):
loader = self.get_product_item_loader_with_default_values(response)
loader.brand_in = lambda x: x[0][14:] if x else 'no brand'
loader.brand_out = TakeFirst()
loader.description_out = JoinExcludingEmptyValues('\n')
loader.sale_price_out = TakeFirst()
reviews = self.parse_review(response)
loader.add_value('reviews', reviews)
loader.add_value('url', response.meta['url'])
loader.add_value('product_number', response.meta['product_number'])
loader.add_xpath('brand', '//a[@class="brandstore"]/text()')
loader.add_xpath(
'title', '//div[@id="divCaption"]/h1[@class="captionText"]/text()')
loader.add_xpath('description', '//div[@id="divPromosPDetail"]')
loader.add_xpath('description', '//div[@id="divingredientsPDetail"]')
loader.add_xpath('original_price', '//span[@class="rowMSRP"]/s/text()')
loader.add_xpath('sale_price', '//div[@id="productprice"]/span/text()')
loader.add_xpath(
'sizes',
'//div[@id="divCaption"]//span[@class="captionSizeText"]/text()')
# images
for sel in response.xpath('//div[@id="divPImage"]'):
image_loader = ProductImageLoader(response=response, selector=sel)
image_loader.add_value('thumbnail', response.meta['thumbnail'])
image_loader.add_xpath('normal_size', 'a/img/@src')
image_loader.add_xpath('zoomed', 'a/img/@src')
loader.add_value('images', image_loader.load_item())
yield loader.load_item()
def parse_item(self, response):
loader = self.get_product_item_loader_with_default_values(response)
loader.original_price_out = TakeFirst()
loader.sale_price_out = TakeFirst()
values_from_list = response.meta.get('values_from_list', {})
for key, value in values_from_list.iteritems():
loader.add_value(key, value)
loader.add_xpath('product_number', '//div[@id="swatchContent"]/div[@id="productNumber"]/text()', re='#(.*)')
loader.add_xpath('title', '//div[@id="productNameText"]/span[@class="productName"]/text()')
loader.add_xpath('description', '//div[@id="tabWindow"]//text()')
loader.add_xpath('original_price', '//div[@id="selectionContent"]/span[@id="priceText"]/strike/text()')
loader.add_xpath('original_price', '//div[@id="selectionContent"]/span[@id="priceText"]/text()')
loader.add_xpath('sale_price', '//div[@id="selectionContent"]/span[@id="priceText"]/span[@class="salePrice"]/text()')
loader.add_xpath('sale_price', '//div[@id="selectionContent"]/span[@id="priceText"]/text()')
loader.add_xpath('sizes', '//div[@id="productContentRight"]/div[@id="swatchContent"]/div[@id="sizeDimensionSwatchContent"]/div[@id="sizeDimension1SwatchContent"]/div[@id="sizeDimension1Swatches"]/button/text()')
loader.add_xpath('default_color', '//div[@id="selectionContent"]/span[@id="selectionConfirmText"]/text()')
#colors
for selector in response.xpath('//div[@id="swatchContent"]/div[@id="colorSwatchContent"]/input'):
color_loader = ProductColorLoader(response=response, selector=selector)
color_loader.add_xpath('name', '@alt', re='(.*) product image$')
color_loader.add_xpath('swatch_image', '@src')
loader.add_value('colors', color_loader.load_item())
#images
images_data = response.meta.get('images_data', {})
if images_data.get('P01'):
image_loader = ProductImageLoader(response=response)
image_loader.add_value('thumbnail', images_data.get('T'))
image_loader.add_value('normal_size', images_data.get('P01'))
image_loader.add_value('zoomed', images_data.get('Z'))
loader.add_value('images', image_loader.load_item())
num = 1
while num < 9:
av_num = 'AV%s' % num
if images_data.get(av_num):
image_loader = ProductImageLoader(response=response)
image_loader.add_value('thumbnail', images_data.get('%s_T' % av_num))
image_loader.add_value('normal_size', images_data.get(av_num))
image_loader.add_value('zoomed', images_data.get('%s_Z' % av_num))
loader.add_value('images', image_loader.load_item())
num += 1
#reviews
for selector in response.xpath('//div[@id="BVRRContainer"]//ol[contains(@class,"bv-content-list")]/li[contains(@class,"bv-content-item")]'):
review_loader = ProductReviewLoader(response=response, selector=selector)
review_loader.body_out = JoinExcludingEmptyValues('\n')
review_loader.add_xpath('author', 'div[@class="bv-author-profile"]/div[@class="bv-inline-profile"]/div[@class="bv-author-avatar"]/div[@class="bv-author-avatar-nickname"]/div[@class="bv-content-author-name"]/span/h3/text()')
review_loader.add_xpath('title', 'div/div[@class="bv-content-container"]//h4[@class="bv-content-title"]/text()')
review_loader.add_xpath('date', 'div/div[@class="bv-content-container"]//div[@class="bv-content-datetime"]/meta[@itemprop="dateCreated"]/@content',
MapCompose(Date('%Y-%m-%d')))
review_loader.add_xpath('body', 'div/div[@class="bv-content-container"]//div[contains(@class,"bv-content-summary-body-text")]/p/text()')
review_loader.add_xpath('max_stars', 'div/div[@class="bv-content-container"]//span[contains(@class,"bv-content-rating")]/meta[@itemprop="bestRating"]/@content')
review_loader.add_xpath('stars', 'div/div[@class="bv-content-container"]//span[contains(@class,"bv-content-rating")]/meta[@itemprop="ratingValue"]/@content')
loader.add_value('reviews', review_loader.load_item())
yield loader.load_item()
class LazyTweetAnswer(Item):
question_id = Field(input_processor=MapCompose(lambda x: int(x)),
output_processor=TakeFirst())
answer_content = Field(input_processor=MapCompose(remove_entities,
unicode.strip),
output_processor=Join())
answerer = Field(output_processor=TakeFirst())
answer_id = Field()
class ZhiHuQ(Item):
title = Field(input_processor=MapCompose(remove_entities, unicode.strip),
output_processor=Join())
content = Field(input_processor=MapCompose(remove_entities, unicode.strip),
output_processor=Join())
id = Field(output_processor=TakeFirst())
user = Field(output_processor=TakeFirst())
num = Field(output_processor=TakeFirst())
class RakutenItem(Item):
# define the fields for your item here like:
# name = scrapy.Field()
handbag_price = Field(output_processor=TakeFirst())
handbag_url = Field(output_processor=TakeFirst())
handbag_brand = Field(output_processor=TakeFirst())
handbag_image_urls = Field(output_processor=TakeFirst())
def parse_product(self, response):
hxs = HtmlXPathSelector(response)
base_url = get_base_url(response)
name_xpath = '//div[@id="product-details"]/h1/span/text()'
names = hxs.select('//h1[@id="product_title"]/text()').extract()
if names and len(names) > 0:
name = names[0].strip()
else:
# product not found. Just continue
self.log('WARNING: Product not found => %s' % response.url)
return
quantity = hxs.select('//p[@id="stock_status"]/text()').extract()
if quantity and "In Stock" in quantity.pop():
quantity = None
else:
quantity = 0
category = hxs.select(
'//ul[@id="crumbs"]/li[@class="last"]/a/text()').extract()
brand = hxs.select(
'//div[@id="product_title_container"]/span[@class="secondary"]/text()'
).extract()
loader = ProductLoader(response=response, item=Product())
loader.add_value('url', urljoin(base_url, response.url))
loader.add_value('name', name)
loader.add_xpath('image_url', '//img[@id="main_image"]/@src',
TakeFirst(), Compose(lambda v: urljoin(base_url, v)))
loader.add_xpath(
'price',
'//div[@class="product_price"]/span[@class="price"]/text()',
TakeFirst(),
re="([.0-9]+)")
if not loader.get_output_value('price'):
loader.add_value('price', 0)
if category:
loader.add_value('category', category[0].strip())
loader.add_value('sku', name, TakeFirst(), re='(\d\d\d+)\s*$')
if brand:
loader.add_value('brand', brand[0].strip())
identifier = hxs.select('//input[@name="ProductID"]/@value').extract()
if not identifier:
identifier = hxs.select('//li[@itemprop="id"]/text()').extract()
loader.add_value('identifier', identifier[0])
if quantity == 0:
loader.add_value('stock', 0)
yield loader.load_item()
class RPostItemsLoader(ItemLoader):
default_item_class = RpostResultsItem
default_output_processor = Compose(TakeFirst(), unicode, unicode.strip)
racename_out = Compose(Join(), unicode, unicode.strip)
racetime_out= Compose(Join(),unicode, unicode.strip)
rpOR_out = Compose(TakeFirst(), unicode, unicode.strip, processOR)
rpTS_out = Compose(TakeFirst(), unicode, unicode.strip, processTS)
prizemoney_out =Compose(TakeFirst(), unicode, unicode.strip, toascii)
rphorseurl_out = Compose(TakeFirst(), unicode, unicode.strip)
def test_get_css(self):
l = TestItemLoader(response=self.response)
self.assertEqual(l.get_css('p::text'), [u'paragraph'])
self.assertEqual(l.get_css('p::text', TakeFirst()), u'paragraph')
self.assertEqual(l.get_css('p::text', TakeFirst(), re='pa'), u'pa')
self.assertEqual(l.get_css(['p::text', 'div::text']), [u'paragraph', 'marta'])
self.assertEqual(l.get_css(['a::attr(href)', 'img::attr(src)']),
[u'http://www.scrapy.org', u'/images/logo.png'])
class HospitalItem(Item):
_hospitalName = Field(output_processor=TakeFirst(), )
grade = Field(output_processor=TakeFirst(), )
feature = Field(
input_processor=MapCompose(lambda v: v.strip()),
output_processor=TakeFirst(),
)
city = Field(output_processor=TakeFirst(), )
area = Field(output_processor=TakeFirst(), )
class CommentItemLoader(ItemLoader):
default_item_class = CommentItem
default_input_processor = MapCompose(lambda x: x.strip())
default_output_processor = Compose(TakeFirst(), lambda x: x.strip())
default_selector_class = Selector
textpost_out = Compose(Join(" "), lambda x: x.strip())
comments_out = Compose(TakeFirst(), get_comments_count, lambda x: x.strip())
upvoted_out = Compose(TakeFirst(), get_upvoted, lambda x: x.strip())
comment_out = Compose(Join(" "), lambda x: x.strip())
class ZhiHuA(Item):
id = Field(input_processor=MapCompose(lambda x: int(x)),
output_processor=TakeFirst())
qid = Field(output_processor=TakeFirst())
asr = Field(output_processor=TakeFirst())
content = Field(input_processor=MapCompose(remove_entities, unicode.strip),
output_processor=Join())
score = Field(input_processor=MapCompose(lambda x: int(x)),
output_processor=TakeFirst())
def parse_brand_list(self, response):
hxs = HtmlXPathSelector(response)
# products
product_items = hxs.select('//div[@class="productGrid"]/ul/li/div[@class="item"]')
category_items = hxs.select('//h1[@class="categoryLandingPageTitle_heading"]/a/text()').extract()
category = category_items[0] if category_items else ''
brand_name = get_brand_from_url(response.url)
def get_full_image_url(url):
return get_full_url(response, url)
for product_item in product_items:
image_url = product_item.select(u'div[@class="prodimg"]/a/img/@src').extract()
if image_url:
image_url = get_full_url(response, image_url[0])
ploadr = ProductLoader(item=Product(), selector=product_item, response=response)
ploadr.add_xpath('name',
'div[@class="prodname"]/a/text()',
TakeFirst(), Compose(unicode.strip))
ploadr.add_xpath('url', 'div[@class="prodname"]/a/@href',
TakeFirst(), Compose(unicode.strip), Compose(get_full_image_url))
ploadr.add_value('category', category)
ploadr.add_value('image_url', image_url)
price = ploadr.get_xpath('div[@class="proddetails"]//div[@class="prodnowprice"]/span/text()',
TakeFirst(), Compose(extract_price))
price_excl_vat = Decimal(price)
ploadr.add_value('price', price_excl_vat)
ploadr.add_value('shipping_cost', Decimal('5.00') if price_excl_vat < 50 else Decimal('0.0'))
ploadr.add_xpath('sku',
'div[@class="proddetails"]//div[@class="proditemcode"]/a/span/following-sibling::text()',
TakeFirst(), Compose(unicode.strip))
ploadr.add_value('identifier', ploadr.get_output_value('sku'))
stock_info = product_item.select(u'div[@class="proddetails"]/div/div/span[contains(@class, "instock")]/@class').extract()
buy_button = product_item.select(u'div[@class="proddetails"]/div[@class="prodquickbuy"]/a[@class="primaryBtn"]').extract()
ploadr.add_value('brand', brand_name)
ploadr.add_value('stock', 1 if stock_info or buy_button else 0)
item = ploadr.load_item()
tmp = ''.join(product_item.select("//div[@class='proditemcode']//text()").extract())
item['metadata'] = {'product_code': tmp.split(':')[-1].strip()}
if not ploadr.get_output_value('brand'):
yield Request(item['url'], meta={'item': item}, callback=self.parse_brand)
else:
yield item
def test_get_value(self):
il = NameItemLoader()
self.assertEqual(u'FOO', il.get_value([u'foo', u'bar'], TakeFirst(), unicode.upper))
self.assertEqual([u'foo', u'bar'], il.get_value([u'name:foo', u'name:bar'], re=u'name:(.*)$'))
self.assertEqual(u'foo', il.get_value([u'name:foo', u'name:bar'], TakeFirst(), re=u'name:(.*)$'))
il.add_value('name', [u'name:foo', u'name:bar'], TakeFirst(), re=u'name:(.*)$')
self.assertEqual([u'foo'], il.get_collected_values('name'))
il.replace_value('name', u'name:bar', re=u'name:(.*)$')
self.assertEqual([u'bar'], il.get_collected_values('name'))
def parse(self, response):
for e in response.xpath(
'//table[@id="tbl_proxy_list"]//tr[count(td)=6]'):
l = ItemLoader(ProxyHunterItem(), selector=e)
l.add_value('prot', 'http')
l.add_xpath('ip', 'td[1]', TakeFirst(), remove_tags, unicode.strip)
l.add_xpath('port', 'td[2]', TakeFirst(), remove_tags,
unicode.strip)
yield l.load_item()
def _set_loader(self, response, xs, item):
if not xs:
self.from_detail_page = True
item = response.request.meta['item']
self.loader = XPathItemLoader(item=item, response=response)
self.loader.default_output_processor = TakeFirst()
else:
self.from_detail_page = False
self.loader = XPathItemLoader(item=item, selector=xs)
self.loader.default_output_processor = TakeFirst()
class StackOverflowAnswer(Item):
answer_id = Field(input_processor=MapCompose(lambda x: int(x)),
output_processor=TakeFirst())
answer_content = Field(input_processor=MapCompose(remove_entities,
unicode.strip),
output_processor=Join())
answerer = Field(output_processor=TakeFirst())
marks = Field(input_processor=MapCompose(lambda x: int(x)),
output_processor=TakeFirst())
is_best_answer = Field(output_processor=TakeFirst())
class YelpItem(Item):
source = Field(output_processor=TakeFirst(), )
source_link = Field(output_processor=TakeFirst(), )
name = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
rating = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
category = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=Join(','),
)
reviews = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
price = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
city = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
address = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=Join(),
)
owner_website = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
phone = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
longitude_latitude = Field(
default='',
input_processor=MapCompose(unquote_markup, strip_space),
output_processor=TakeFirst(),
)
last_crawl = Field()
def test_replace_css_multi_fields(self):
l = TestItemLoader(response=self.response)
l.add_css(None, 'div::text', TakeFirst(), lambda x: {'name': x})
self.assertEqual(l.get_output_value('name'), [u'Marta'])
l.replace_css(None, 'p::text', TakeFirst(), lambda x: {'name': x})
self.assertEqual(l.get_output_value('name'), [u'Paragraph'])
l.add_css(None, 'a::attr(href)', TakeFirst(), lambda x: {'url': x})
self.assertEqual(l.get_output_value('url'), [u'http://www.scrapy.org'])
l.replace_css(None, 'img::attr(src)', TakeFirst(), lambda x: {'url': x})
self.assertEqual(l.get_output_value('url'), [u'/images/logo.png'])
def _convert(data):
if t not in ['join', 'list'] and isinstance(data, list):
data = TakeFirst()(data)
if type(data) in [str, unicode]:
data = data.strip()
elif type(data) in [int, float, datetime]:
data = str(data)
else:
return data
if t == 'join':
sep = inf.get('sep', u' ')
return Join(sep)(data)
elif t == 'list':
sep = inf.get('sep', u' ')
return remove_tags(Join(sep)(data)).strip()
elif t == 'text':
return remove_tags(data).strip()
elif t == 'clean':
cleaner = Cleaner(style=True,
scripts=True,
javascript=True,
links=True,
meta=True)
return cleaner.clean_html(data)
elif t == 'unesc':
return HTMLParser().unescape(data)
elif t == 'base64':
return base64.decodestring(data)
elif t == 'sub':
frm = inf.get('from')
to = inf.get('to')
return re.sub(frm, to, data)
elif t == 'jpath':
qs = inf.get('query')
return jsonpath.jsonpath(json.loads(data), qs)
elif t == 'map':
m = inf.get('map')
d = inf.get('default')
return m.get(data, d)
elif t == 'int':
return int(float(data))
elif t == 'float':
return float(data)
elif t == 'date':
fmt = inf.get('fmt', 'auto')
tz = inf.get('tz', '+00:00')
return parse_date(data, fmt, tz)
elif t == 'cst':
fmt = inf.get('fmt', 'auto')
return parse_date(data, fmt, '+08:00')
else:
return data
class ActiveDoctorItem(Item):
_name = Field(output_processor=TakeFirst(), )
hospital = Field(output_processor=TakeFirst(), )
city = Field(output_processor=TakeFirst(), )
area = Field(output_processor=TakeFirst(), )
specialty = Field(output_processor=TakeFirst(), )
title = Field(output_processor=TakeFirst(), )
count_ReplyInTwoWeeks = Field(output_processor=TakeFirst(), )
count_ReplyTotal = Field(output_processor=TakeFirst(), )
count_Calls = Field(output_processor=TakeFirst(), )
external_id = Field(output_processor=TakeFirst(), )
comment = Field(output_processor=Join(), )
class ReviewLoader(XPathItemLoader):
date_in = MapCompose(unicode, unicode.strip, extract_date, date_format='%d/%m/%Y')
date_out = TakeFirst()
rating_in = MapCompose(unicode, extract_rating)
rating_out = TakeFirst()
full_text_in = MapCompose(unicode, unicode.strip, remove_entities)
full_text_out = Join()
url_in = MapCompose(unicode, unicode.strip)
url_out = TakeFirst()
def _convert(data):
if t not in ['join', 'list'] and isinstance(data, list):
data = TakeFirst()(data)
if type(data) in [str, unicode]:
data = data.strip()
elif type(data) in [int, float, datetime]:
data = str(data)
else:
return data
if t=='join':
sep = inf.get('sep', u' ')
return Join(sep)(data)
elif t=='list':
sep = inf.get('sep', u' ')
return remove_tags(Join(sep)(data)).strip()
elif t=='text':
return remove_tags(data).strip()
elif t=='clean':
cleaner = Cleaner(style=True, scripts=True, javascript=True, links=True, meta=True)
return cleaner.clean_html(data)
elif t=='unesc':
return HTMLParser().unescape(data)
elif t=='base64':
return base64.decodestring(data)
elif t=='sub':
frm = inf.get('from')
to = inf.get('to')
return re.sub(frm, to, data)
elif t=='jpath':
qs = inf.get('query')
return jsonpath.jsonpath(json.loads(data), qs)
elif t=='map':
m = inf.get('map')
d = inf.get('default')
return m.get(data, d)
elif t=='int':
return int(float(data))
elif t=='float':
return float(data)
elif t=='date':
fmt = inf.get('fmt', 'auto')
tz = inf.get('tz', '+00:00')
return parse_date(data, fmt, tz)
elif t=='cst':
fmt = inf.get('fmt', 'auto')
return parse_date(data, fmt, '+08:00')
else:
return data