def google_light_parse(gresult):
gresult = parse_util.x_to_soup(gresult)
parse_dict = dict()
resultStats = input.find(name='div', attrs={'id': 'resultStats'})
if resultStats:
parse_dict['_resultStats'] = google.parse_number_of_results(
resultStats)
def element_presence_dict(input):
input = putil.x_to_soup(input)
html = input.renderContents()
text = input.get_text(' ')
d = dict()
# search_term_redirect (when google shows results for a different search term than the entered--should be in _spell)
d['isa_search_term_redirect'] = bool(
re.search('(Showing results for)|(Including results for)', text))
# did_you_mean
d['did_you_mean'] = bool(re.search('Did you mean', text))
# number of results
d['number_of_results'] = re.findall('About (' + RE_NUMBER + ') results',
text)
if len(d['number_of_results']) > 0:
d['isa_number_of_results'] = True
else:
if bool(re.search('Your search .* did not match any documents.')):
d['number_of_results'] = 0
else:
d['number_of_results'] = -1
# number of center_col_elements
center_col = input.find('div', {'id': 'center_col'})
if center_col:
d['num_of_center_col_children'] = len([x for x in center_col.children])
else:
d['num_of_center_col_children'] = 0
def num_of_results_soup01(soup):
"""
num_of_results using Beautifulsoup
"""
soup = util.x_to_soup(soup)
assert isinstance(soup,BeautifulSoup), "hey, I'm expecting soup!"
t = get_section(soup,[{'id':'appbar'},{'id':'topabar'}])
t = re.findall("("+RE_NUMBER+").*?result",t.text)
return int(re.sub(RE_NUM_SEP,"",t[-1]))
def num_of_results_soup01(soup):
"""
num_of_results using Beautifulsoup
"""
soup = util.x_to_soup(soup)
assert isinstance(soup, BeautifulSoup), "hey, I'm expecting soup!"
t = get_section(soup, [{'id': 'appbar'}, {'id': 'topabar'}])
t = re.findall("(" + RE_NUMBER + ").*?result", t.text)
return int(re.sub(RE_NUM_SEP, "", t[-1]))
def google_webResults_soup(source):
"""
google_webResults using Beautifulsoup
"""
soup = util.x_to_soup(source)
list_of_li = soup.findAll('li', attrs={'class':'g'})
return [
{'pos':i+1, 'title': li.find('a'), 'text': li.find('span', attrs={'class':'st'})}
for (i,li) in enumerate(list_of_li)
]
def get_pois_near_hotel_location(html):
html = parse_util.x_to_soup(html)
html = html.find('div', attrs={'id': "location-distances"}).renderContents()
t = html.split('<br/>')
t = ['. '+x for x in t]
# print len(t)
# return [re.search(pois_near_hotel_exp, x) for x in t]
return [x.group(0).strip() for x in
[re.search(pois_near_hotel_exp_0, x) for x in t]
if x]
def rh_ads(input):
"""
right hand side ads
"""
output = []
input = util.x_to_soup(input)
rhs_block_element = rhs_block(input)
if len(rhs_block_element)!=0:
ol_section = rhs_block_element[0].findAll('ol')
if len(ol_section):
return [str(li) for li in ol_section[0].findAll('li')] # TODO: do we want string, unicode, or the tag itself as is?
def google_webResults_soup(source):
"""
google_webResults using Beautifulsoup
"""
soup = util.x_to_soup(source)
list_of_li = soup.findAll('li', attrs={'class': 'g'})
return [{
'pos': i + 1,
'title': li.find('a'),
'text': li.find('span', attrs={'class': 'st'})
} for (i, li) in enumerate(list_of_li)]
def rh_ads(input):
"""
right hand side ads
"""
output = []
input = util.x_to_soup(input)
rhs_block_element = rhs_block(input)
if len(rhs_block_element) != 0:
ol_section = rhs_block_element[0].findAll('ol')
if len(ol_section):
return [
str(li) for li in ol_section[0].findAll('li')
] # TODO: do we want string, unicode, or the tag itself as is?
def element_presence_dict(input):
input = putil.x_to_soup(input)
html = input.renderContents()
text = input.get_text(' ')
d = dict()
# search_term_redirect (when google shows results for a different search term than the entered--should be in _spell)
d['isa_search_term_redirect'] = bool(re.search('(Showing results for)|(Including results for)',text))
# did_you_mean
d['did_you_mean'] = bool(re.search('Did you mean',text))
# number of results
d['number_of_results'] = re.findall('About ('+RE_NUMBER+') results',text)
if len(d['number_of_results']) > 0:
d['isa_number_of_results'] = True
else:
if bool(re.search('Your search .* did not match any documents.')):
d['number_of_results'] = 0
else:
d['number_of_results'] = -1
# number of center_col_elements
center_col = input.find('div',{'id':'center_col'})
if center_col:
d['num_of_center_col_children'] = len([x for x in center_col.children])
else:
d['num_of_center_col_children'] = 0
def google_light_parse(gresult):
gresult = parse_util.x_to_soup(gresult)
parse_dict = dict()
resultStats = input.find(name='div',attrs={'id':'resultStats'})
if resultStats:
parse_dict['_resultStats'] = google.parse_number_of_results(resultStats)
def get_info_dict_01(input):
input = putil.x_to_soup(input)
def top_elements(input):
"""
the top ads, hotel finder and other elements coming before webResults
"""
input = util.x_to_soup(input)
return input.findAll('span', attrs={'id':'taw'})
def root_dict(soup):
soup = util.x_to_soup(soup)
return extract_tag_dict_from_node(soup, 'root')
def top_elements(input):
"""
the top ads, hotel finder and other elements coming before webResults
"""
input = util.x_to_soup(input)
return input.findAll('span', attrs={'id': 'taw'})
def mk_gresult_tag_dict(input):
"""
mk_result_dict(input)
takes a soup, html string, or filename of google result htmls as an input
and returns a dict containing components of the html we're interested in
"""
input = util.x_to_soup(input)
d = dict()
# number of results
resultStats = input.find(name='div', attrs={'id': 'resultStats'})
if resultStats: d['_resultStats'] = resultStats
# center_col
center_col = input.find(name='div', attrs={'id': 'center_col'})
if center_col:
d['_center_col'] = center_col
# tads
tads = d['_center_col'].find(name='div', attrs={'id': 'tads'})
if tads:
d['_tads'] = tads
# top_ads
top_ads = d['_tads'].findAll('li')
if top_ads:
d['_top_ads_list'] = top_ads
# res
res = d['_center_col'].find(name='div', attrs={'id': 'res'})
if res:
d['_res'] = res
# searchInstead
topstuff = d['_res'].find(name='div', attrs={'id': 'topstuff'})
if topstuff:
d['_topstuff'] = topstuff
# spell
spell = d['_topstuff'].find(name='a', attrs={'class': 'spell'})
if spell: d['_spell'] = spell
# search
search = d['_res'].find(name='div', attrs={'id': 'search'})
if search:
d['_search'] = search
# ires
ires = d['_search'].find(name='div', attrs={'id': 'ires'})
if ires:
d['_ires'] = ires
# organicResults
organic_results = d['_ires'].findAll('li')
if organic_results:
d['_organic_results_list'] = organic_results
# related_search
after_res = d['_res'].nextSibling
if after_res:
related_search = after_res.find('table')
if related_search:
d['_related_search'] = related_search
# rhs_block
rhs_block = input.find(name='div', attrs={'id': 'rhs_block'})
if rhs_block:
d['_rhs_block'] = rhs_block
rhs_ads = [] # initializing
# rhs_ads from mbEnd
mbEnd = d['_rhs_block'].find(name='div', attrs={'id': 'mbEnd'})
if mbEnd:
d['_mbEnd'] = mbEnd
rhs_ads = rhs_ads + mbEnd.findAll('li')
# rhs_ads from nobr
nobr = d['_rhs_block'].find(name='ol', attrs={'class': 'nobr'})
if nobr:
d['_nobr'] = nobr
rhs_ads = rhs_ads + nobr.findAll('li')
# puting rhs_ads in the dict
if rhs_ads:
d['_rhs_ads_list'] = rhs_ads
# Okay, no more parsing wishes, return the dict d
return d
def mk_gresult_tag_dict(input):
"""
mk_result_dict(input)
takes a soup, html string, or filename of google result htmls as an input
and returns a dict containing components of the html we're interested in
"""
input = putil.x_to_soup(input)
d = dict()
# number of results
resultStats = input.find(name='div',attrs={'id':'resultStats'})
if resultStats: d['_resultStats'] = resultStats
# center_col
center_col = input.find(name='div',attrs={'id':'center_col'})
if center_col:
d['_center_col'] = center_col
# tads (center_col.taw.tvcap.tads)
tads = d['_center_col'].find(name='div',attrs={'id':'tads'})
if tads:
d['_tads'] = tads
# top_ads
top_ads = d['_tads'].findAll('li')
if top_ads:
d['_top_ads_list'] = top_ads
# c (center_col.taw.tvcap.c)
c = d['_center_col'].find(name='div',attrs={'class':'c'})
if c:
d['_c'] = c
# c_list
c_list = d['_c'].findAll('li')
if c_list:
d['_c_list'] = c_list
# res
res = d['_center_col'].find(name='div',attrs={'id':'res'})
if res:
d['_res'] = res
# searchInstead
topstuff = d['_res'].find(name='div',attrs={'id':'topstuff'})
if topstuff:
d['_topstuff'] = topstuff # used to contain spell,
# but then realized the spell appeared in other places some times, so moved it to child of _center_col
# spell
spell = d['_center_col'].find(name='a',attrs={'class':'spell'})
if spell:
d['_spell'] = spell
# search
search = d['_res'].find(name='div',attrs={'id':'search'})
if search:
d['_search'] = search
# ires
ires = d['_search'].find(name='div',attrs={'id':'ires'})
if ires:
d['_ires'] = ires
# organicResults
organic_results = d['_ires'].findAll('li')
if organic_results:
d['_organic_results_list'] = organic_results
# # related_search
# extrares = d['_center_col'].find(name='div',attrs={'id':'extrares'})
# if extrares:
# related_search = extrares.find('table')
# if related_search:
# d['_related_search'] = related_search
# related_search
after_res = d['_res'].nextSibling
if after_res:
related_search = after_res.find('table')
if related_search:
d['_related_search'] = related_search
# rhs_block
rhs_block = input.find(name='div',attrs={'id':'rhs_block'})
if rhs_block:
d['_rhs_block'] = rhs_block
# lu_pinned_rhs (where some hotel finder, maps, specific hotels might be)
lu_pinned_rhs = d['_rhs_block'].find(name='div',attrs={'id':'lu_pinned_rhs'})
if lu_pinned_rhs:
d['_lu_pinned_rhs'] = lu_pinned_rhs
# knop (another place where some hotel finder, maps, specific hotels might be)
knop = d['_rhs_block'].find(name='div',attrs={'id':'knop'})
if knop:
d['_knop'] = knop
rhs_ads = [] # initializing
# rhs_ads from mbEnd
mbEnd = d['_rhs_block'].find(name='div',attrs={'id':'mbEnd'})
if mbEnd:
d['_mbEnd'] = mbEnd
rhs_ads = rhs_ads + mbEnd.findAll('li')
# rhs_ads from nobr
nobr = d['_rhs_block'].find(name='ol',attrs={'class':'nobr'})
if nobr:
d['_nobr'] = nobr
rhs_ads = rhs_ads + nobr.findAll('li')
# puting rhs_ads in the dict
if rhs_ads:
d['_rhs_ads_list'] = rhs_ads
# Okay, no more parsing wishes, return the dict d
return d
def get_info_dict_01(input):
input = putil.x_to_soup(input)
def root_dict(soup):
soup = util.x_to_soup(soup)
return extract_tag_dict_from_node(soup,'root')
def mk_gresult_tag_dict(input):
"""
mk_result_dict(input)
takes a soup, html string, or filename of google result htmls as an input
and returns a dict containing components of the html we're interested in
"""
input = putil.x_to_soup(input)
d = dict()
# number of results
resultStats = input.find(name='div', attrs={'id': 'resultStats'})
if resultStats: d['_resultStats'] = resultStats
# center_col
center_col = input.find(name='div', attrs={'id': 'center_col'})
if center_col:
d['_center_col'] = center_col
# tads (center_col.taw.tvcap.tads)
tads = d['_center_col'].find(name='div', attrs={'id': 'tads'})
if tads:
d['_tads'] = tads
# top_ads
top_ads = d['_tads'].findAll('li')
if top_ads:
d['_top_ads_list'] = top_ads
# c (center_col.taw.tvcap.c)
c = d['_center_col'].find(name='div', attrs={'class': 'c'})
if c:
d['_c'] = c
# c_list
c_list = d['_c'].findAll('li')
if c_list:
d['_c_list'] = c_list
# res
res = d['_center_col'].find(name='div', attrs={'id': 'res'})
if res:
d['_res'] = res
# searchInstead
topstuff = d['_res'].find(name='div', attrs={'id': 'topstuff'})
if topstuff:
d['_topstuff'] = topstuff # used to contain spell,
# but then realized the spell appeared in other places some times, so moved it to child of _center_col
# spell
spell = d['_center_col'].find(name='a', attrs={'class': 'spell'})
if spell:
d['_spell'] = spell
# search
search = d['_res'].find(name='div', attrs={'id': 'search'})
if search:
d['_search'] = search
# ires
ires = d['_search'].find(name='div', attrs={'id': 'ires'})
if ires:
d['_ires'] = ires
# organicResults
organic_results = d['_ires'].findAll('li')
if organic_results:
d['_organic_results_list'] = organic_results
# # related_search
# extrares = d['_center_col'].find(name='div',attrs={'id':'extrares'})
# if extrares:
# related_search = extrares.find('table')
# if related_search:
# d['_related_search'] = related_search
# related_search
after_res = d['_res'].nextSibling
if after_res:
related_search = after_res.find('table')
if related_search:
d['_related_search'] = related_search
# rhs_block
rhs_block = input.find(name='div', attrs={'id': 'rhs_block'})
if rhs_block:
d['_rhs_block'] = rhs_block
# lu_pinned_rhs (where some hotel finder, maps, specific hotels might be)
lu_pinned_rhs = d['_rhs_block'].find(name='div',
attrs={'id': 'lu_pinned_rhs'})
if lu_pinned_rhs:
d['_lu_pinned_rhs'] = lu_pinned_rhs
# knop (another place where some hotel finder, maps, specific hotels might be)
knop = d['_rhs_block'].find(name='div', attrs={'id': 'knop'})
if knop:
d['_knop'] = knop
rhs_ads = [] # initializing
# rhs_ads from mbEnd
mbEnd = d['_rhs_block'].find(name='div', attrs={'id': 'mbEnd'})
if mbEnd:
d['_mbEnd'] = mbEnd
rhs_ads = rhs_ads + mbEnd.findAll('li')
# rhs_ads from nobr
nobr = d['_rhs_block'].find(name='ol', attrs={'class': 'nobr'})
if nobr:
d['_nobr'] = nobr
rhs_ads = rhs_ads + nobr.findAll('li')
# puting rhs_ads in the dict
if rhs_ads:
d['_rhs_ads_list'] = rhs_ads
# Okay, no more parsing wishes, return the dict d
return d
def parse_hotel_info_page(html):
html = parse_util.x_to_soup(html)
d = dict()
# hotel name
d = parse_bsoup.add_text_to_parse_dict(soup=html, parse_dict=d,
key='hotel_name', name='h1', attrs={'property': 'v:name'}, text_transform=parse_util.strip_spaces)
# hotel address
tag = html.find(name='p', attrs={'id': 'property-address'})
if tag:
d['hotel_address'] = pstr_trans.strip(tag.text)
d = parse_bsoup.add_text_to_parse_dict(soup=tag, parse_dict=d,
key='hotel_street_address', name='span', attrs={'property': "v:street-address"},
text_transform=parse_util.strip_spaces)
d = parse_bsoup.add_text_to_parse_dict(soup=tag, parse_dict=d,
key='hotel_locality', name='span', attrs={'property': "v:locality"},
text_transform=parse_util.strip_spaces)
# average price
d = parse_bsoup.add_text_to_parse_dict(soup=html, parse_dict=d,
key='currency', name='span', attrs={'id': 'currency-symbol'}, text_transform=parse_util.strip_spaces)
avgPriceEl0 = html.find(name='span', attrs={'id': 'avgPriceEl0'})
avgPriceDecimals = html.find(name='sup', attrs={'id': 'avgPriceDecimals'})
if avgPriceEl0:
d['average_price'] = avgPriceEl0.text
if avgPriceDecimals:
d['average_price'] = d['average_price'] + avgPriceDecimals.text
d['average_price'] = float(d['average_price'])
# facebook likes
d = parse_bsoup.add_text_to_parse_dict(soup=html, parse_dict=d,
key='facebook_likes', name='span', attrs={'class': 'pluginCountTextDisconnected'}, text_transform=float)
# num_of_photos
tag = html.find(name='div', attrs={'id': 'photo_gallery'})
if tag:
d['num_of_photos'] = len(tag.findAll(name='li'))
# hotel description
d = parse_bsoup.add_text_to_parse_dict(soup=html, parse_dict=d,
key='hotel_description', name='div', attrs={'id': 'hotel-description-body'}, text_transform=parse_util.strip_spaces)
# average_venere_rating
tag = html.find(name='div', attrs={'id': 'avg_guest_rating'})
if tag:
d['average_venere_rating'] = float(tag.find(name='b', attrs={'property': 'v:rating'}).text)
# facilities
tag = html.find(name='div', attrs={'id': 'facilities'})
if tag:
facilities = tag.findAll(name='li')
if facilities:
d['facilities'] = [parse_util.strip_spaces(x.text) for x in facilities]
# alternate names
tag = html.find(name='div', attrs={'id': 'also_known_as'})
if tag:
tag = tag.find(name='p')
if tag:
t = [parse_util.strip_spaces(x) for x in tag.renderContents().split('<br>')]
t = [parse_util.strip_tags(x) for x in t]
d['alternate_names'] = t
# overview_reviews
tag = html.find(name='div', attrs={'id': 'reviews-overview-hbar-box'})
if tag:
tagg = tag.findAll(name='div', attrs={'class': 'reviews-overview-horizzontalbar'})
if tagg:
d['overview_reviews'] = dict()
for t in tagg:
d['overview_reviews'][t.find(name='p').text] = float(t.find(name='b').text)
# location_distances
tag = html.find(name='div', attrs={'id': 'location-distances'})
if tag:
t = re.sub("^[^<]+<h2>.+</h2>","", tag.renderContents()).split('<br/>')
tt = [re.findall(pois_near_hotel_exp, x) for x in t]
tt = [x[0] for x in tt if x]
d['poi_and_distances'] = [{'poi': parse_util.strip_spaces(x[0].replace('"', '')), 'km': float(x[1]), 'mi': float(x[2])} for x in tt]
return d
def mk_gresult_tag_dict(input):
"""
mk_result_dict(input)
takes a soup, html string, or filename of google result htmls as an input
and returns a dict containing components of the html we're interested in
"""
input = util.x_to_soup(input)
d = dict()
# number of results
resultStats = input.find(name='div',attrs={'id':'resultStats'})
if resultStats: d['_resultStats'] = resultStats
# center_col
center_col = input.find(name='div',attrs={'id':'center_col'})
if center_col:
d['_center_col'] = center_col
# tads
tads = d['_center_col'].find(name='div',attrs={'id':'tads'})
if tads:
d['_tads'] = tads
# top_ads
top_ads = d['_tads'].findAll('li')
if top_ads:
d['_top_ads_list'] = top_ads
# res
res = d['_center_col'].find(name='div',attrs={'id':'res'})
if res:
d['_res'] = res
# searchInstead
topstuff = d['_res'].find(name='div',attrs={'id':'topstuff'})
if topstuff:
d['_topstuff'] = topstuff
# spell
spell = d['_topstuff'].find(name='a',attrs={'class':'spell'})
if spell: d['_spell'] = spell
# search
search = d['_res'].find(name='div',attrs={'id':'search'})
if search:
d['_search'] = search
# ires
ires = d['_search'].find(name='div',attrs={'id':'ires'})
if ires:
d['_ires'] = ires
# organicResults
organic_results = d['_ires'].findAll('li')
if organic_results:
d['_organic_results_list'] = organic_results
# related_search
after_res = d['_res'].nextSibling
if after_res:
related_search = after_res.find('table')
if related_search:
d['_related_search'] = related_search
# rhs_block
rhs_block = input.find(name='div',attrs={'id':'rhs_block'})
if rhs_block:
d['_rhs_block'] = rhs_block
rhs_ads = [] # initializing
# rhs_ads from mbEnd
mbEnd = d['_rhs_block'].find(name='div',attrs={'id':'mbEnd'})
if mbEnd:
d['_mbEnd'] = mbEnd
rhs_ads = rhs_ads + mbEnd.findAll('li')
# rhs_ads from nobr
nobr = d['_rhs_block'].find(name='ol',attrs={'class':'nobr'})
if nobr:
d['_nobr'] = nobr
rhs_ads = rhs_ads + nobr.findAll('li')
# puting rhs_ads in the dict
if rhs_ads:
d['_rhs_ads_list'] = rhs_ads
# Okay, no more parsing wishes, return the dict d
return d
