def create_from_query_string(qs):
ans = MultiDict()
if ispy3:
qs = as_unicode(qs)
for k, v in iteritems(parse_qs(qs, keep_blank_values=True)):
dict.__setitem__(ans, as_unicode(k), [as_unicode(x) for x in v])
return ans
def get_series(title, authors, timeout=60):
mi = Metadata(title, authors)
if title and title[0] in _ignore_starts:
title = title[1:]
title = re.sub(r'^(A|The|An)\s+', '', title).strip()
if not title:
return mi
if isinstance(title, unicode_type):
title = title.encode('utf-8')
title = quote_plus(title)
author = authors[0].strip()
if not author:
return mi
if ',' in author:
author = author.split(',')[0]
else:
author = author.split()[-1]
url = URL.format(author, title)
br = browser()
try:
raw = br.open_novisit(url, timeout=timeout).read()
except URLError as e:
if isinstance(e.reason, socket.timeout):
raise Exception('KDL Server busy, try again later')
raise
if 'see the full results' not in raw:
return mi
raw = xml_to_unicode(raw)[0]
soup = BeautifulSoup(raw)
searcharea = soup.find('div', attrs={'class': 'searcharea'})
if searcharea is None:
return mi
ss = searcharea.find('div', attrs={'class': 'seriessearch'})
if ss is None:
return mi
a = ss.find('a', href=True)
if a is None:
return mi
href = a['href'].partition('?')[-1]
data = parse_qs(href)
series = data.get('SeriesName', [])
if not series:
return mi
series = series[0]
series = re.sub(r' series$', '', series).strip()
if series:
mi.series = series
ns = ss.nextSibling
if ns.contents:
raw = unicode_type(ns.contents[0])
raw = raw.partition('.')[0].strip()
try:
mi.series_index = int(raw)
except Exception:
pass
return mi
def get_series(title, authors, timeout=60):
mi = Metadata(title, authors)
if title and title[0] in _ignore_starts:
title = title[1:]
title = re.sub(r'^(A|The|An)\s+', '', title).strip()
if not title:
return mi
if isinstance(title, unicode_type):
title = title.encode('utf-8')
title = quote_plus(title)
author = authors[0].strip()
if not author:
return mi
if ',' in author:
author = author.split(',')[0]
else:
author = author.split()[-1]
url = URL.format(author, title)
br = browser()
try:
raw = br.open_novisit(url, timeout=timeout).read()
except URLError as e:
if isinstance(e.reason, socket.timeout):
raise Exception('KDL Server busy, try again later')
raise
if 'see the full results' not in raw:
return mi
raw = xml_to_unicode(raw)[0]
soup = BeautifulSoup(raw)
searcharea = soup.find('div', attrs={'class':'searcharea'})
if searcharea is None:
return mi
ss = searcharea.find('div', attrs={'class':'seriessearch'})
if ss is None:
return mi
a = ss.find('a', href=True)
if a is None:
return mi
href = a['href'].partition('?')[-1]
data = parse_qs(href)
series = data.get('SeriesName', [])
if not series:
return mi
series = series[0]
series = re.sub(r' series$', '', series).strip()
if series:
mi.series = series
ns = ss.nextSibling
if ns.contents:
raw = unicode_type(ns.contents[0])
raw = raw.partition('.')[0].strip()
try:
mi.series_index = int(raw)
except:
pass
return mi
def id_from_url(self, url): # {{{
from polyglot.urllib import urlparse, parse_qs
purl = urlparse(url)
if purl.netloc == 'books.google.com':
q = parse_qs(purl.query)
gid = q.get('id')
if gid:
return 'google', gid[0]
def __init__(self, format):
'''
Create a query object by passing it the url format obtained
from the opensearch Description.
'''
self.format = format
# unpack the url to a tuple
self.url_parts = urlparse(format)
# unpack the query string to a dictionary
self.query_string = parse_qs(self.url_parts[4])
# look for standard macros and create a mapping of the
# opensearch names to the service specific ones
# so q={searchTerms} will result in a mapping between searchTerms and q
self.macro_map = {}
for key,values in self.query_string.items():
# TODO eventually optional/required params should be
# distinguished somehow (the ones with/without trailing ?
macro = values[0].replace('{', '').replace('}', '').replace('?', '')
if macro in Query.standard_macros:
self.macro_map[macro] = key
def __init__(self, format):
'''
Create a query object by passing it the url format obtained
from the opensearch Description.
'''
self.format = format
# unpack the url to a tuple
self.url_parts = urlparse(format)
# unpack the query string to a dictionary
self.query_string = parse_qs(self.url_parts[4])
# look for standard macros and create a mapping of the
# opensearch names to the service specific ones
# so q={searchTerms} will result in a mapping between searchTerms and q
self.macro_map = {}
for key,values in self.query_string.items():
# TODO eventually optional/required params should be
# distinguished somehow (the ones with/without trailing ?
macro = values[0].replace('{', '').replace('}', '').replace('?', '')
if macro in Query.standard_macros:
self.macro_map[macro] = key
def create_from_query_string(qs):
ans = MultiDict()
for k, v in iteritems(parse_qs(qs, keep_blank_values=True)):
dict.__setitem__(ans, k.decode('utf-8'), [x.decode('utf-8') for x in v])
return ans
def create_from_query_string(qs):
ans = MultiDict()
for k, v in iteritems(parse_qs(qs, keep_blank_values=True)):
dict.__setitem__(ans, k.decode('utf-8'),
[x.decode('utf-8') for x in v])
return ans
