def get_player_url_from_br(player_name):
index_url = get_index_url_from_name(player_name)
souped_index = soupify(index_url)
link_finder = player_link_finder(player_name)
player_link_element = souped_index.find(link_finder)
if player_link_element:
link = player_link_element['href']
return link
else:
return None
def get_week_results ( year, week_number ) :
"""
Build the request URL, and make the request. Return the HTML result as soup
"""
url = build_results_url ( year, week_number )
print ( 'requesting {}'.format(url) )
soup = soupify ( url )
if soup is None:
print ('Error: page request unsuccessful. Exiting script.')
exit(1)
else:
return soup
def getRandomWikiURL(url):
bsobj = soupify(url)
new_url = getRandomURL(bsobj)
while new_url.startswith('/wiki') is not True:
new_url = getRandomURL(bsobj)
checks = [':', '#', '?']
while any(char in new_url for char in checks):
new_url = getRandomWikiURL(new_url)
return 'https://en.wikipedia.org' + new_url
def parse(url, model=model):
data = soupify.soupify(url)
# mine your data here!
model['url'] = url
model['ID'] = soupify.getID(url)
model['extrDate'] = datetime.date.today().isoformat()
# check for redirect:
if not data:
model['error'] = 'Redirected'
return model
else:
try:
# format arguments en masse as {'model.key':['tag',{'attr':'value','attr':'value',...}],...}
userData = {
'user': ['a', {
"class": "orange nou"
}],
'name': ['span', {
"class": "fn"
}],
'url': ['a', {
'rel': 'nofollow me'
}],
'aim': ['a', {
'rel': 'me',
'class': 'url'
}],
'isPrinted': ['a', {
'href': '/alumniclub'
}]
}
# process arguments en masse
for k, v in userData.iteritems():
model[k] = data.findAll(v[0], v[1])
# if no data was returned, do nothing. Else, grab data contents.
if model[k]:
model[k] = model[k][0].contents[0]
# field-specific post-processing
# convert isPrinted to boolean: if we gathered alumniclub data, user has been printed. Else, not.
if model['isPrinted']:
model['isPrinted'] = True
else:
model['isPrinted'] = False
# use regex to parse unmarked string for data we want. send data as {'model.key':'regex',...}
userData = data.findAll(
'span', {"style": "font-size:12px;"})[0].contents[2]
# 'data' will mark the group we need.
userRegex = {
'age': 'is a (?P<data>.*) year',
'numDesSco': 'scored (?P<data>.*) submissions',
'numScoPri': 'helping (?P<data>.*) designs',
'avgSco': 'of (?P<data>.*), helping',
'joinDate': 'since (?P<data>.*), has',
'gender': '(?P<data>girl|boy)'
}
for k, v in userRegex.iteritems():
v = re.search(v, userData)
if v:
model[k] = v.group('data')
else:
model[k] = u''
# field-specific post-processing
# eliminate commas, turning "1,000" into "1000"
for i in ['numDesSco', 'numScoPri']:
if model[i]:
model[i] = model[i].replace(',', '')
# convert age into timedelta, and subtract it from today to get an approximate birth date
if model['age']:
model['age'] = (datetime.datetime.today() - datetime.timedelta(
float(model['age']) * 365.25)).isoformat()
# convert joinDate to a datetime object
if model['joinDate']:
date = model['joinDate'].replace(',', '').split(' ')
date[0] = soupify.month[date[0]]
model['joinDate'] = datetime.date(int(date[2]), date[0],
int(date[1])).isoformat()
# convert from unicode so the resultant csv isn't full of u''
model['avgSco'] = float(model['avgSco'])
for k in model.keys():
if type(model[k]) is unicode:
model[k] = str(model[k].encode('ascii', 'ignore'))
return model
except:
model['issue'] = 'Issues'
print model['ID']
raise
return model
def parse(self, url):
# mine your data here!
# get data. Anything that requires no post-processing is collected here.
model = soupify.soupify(url, self.model)
# get some easy values
model['url'] = url
model['ID'] = soupify.getID(url)
model['extrDate'] = datetime.date.today().isoformat()
# check for redirect:
if model['ID'] not in model['url']:
model['error'] = 'Redirected'
return model
else:
# if no redirect, commence post-processing
try:
# convert isPrinted to boolean: if we gathered alumniclub data, user has been printed. Else, not.
if model['isPrinted']:
model['isPrinted'] = True
else:
model['isPrinted'] = False
# recall that 'age' stores the html containing data for several other fields
if model['age']:
if len(model['age'][0].contents) > 1:
userData = model['age'][0].contents[2]
else:
userData = model['age'][0].contents[0]
else:
model['issue'] = "Empty"
print "Empty results at ID " + model['ID']
return model
# use regex to parse that html for data we want. send data as {'model.key':'regex',...}
# group 'data' will mark the group we need.
userRegex = {
'age': 'is a (?P<data>.*) year',
'numDesSco': 'scored (?P<data>.*) submissions',
'numScoPri': 'helping (?P<data>.*) designs',
'avgSco': 'of (?P<data>.*), helping',
'joinDate': 'since (?P<data>.*), has',
'gender': '(?P<data>girl|boy)'
}
for k, v in userRegex.iteritems():
v = re.search(v, userData)
if v:
model[k] = v.group('data')
else:
model[k] = u''
# format returned values
# eliminate commas, turning "1,000" into "1000"
for i in ['numDesSco', 'numScoPri']:
if model[i]:
model[i] = model[i].replace(',', '')
# convert age into timedelta, and subtract it from today to get an approximate birth date
if model['age']:
model['age'] = (
datetime.datetime.today() - datetime.timedelta(
float(model['age']) * 365.25)).isoformat()
# convert joinDate to a datetime object
if model['joinDate']:
date = model['joinDate'].replace(',', '').split(' ')
date[0] = soupify.month[date[0]]
model['joinDate'] = datetime.date(int(
date[2]), date[0], int(date[1])).isoformat()
# convert from unicode so the resultant csv isn't full of u''
model['avgSco'] = float(model['avgSco'])
for k in model.keys():
if type(model[k]) is unicode:
model[k] = str(model[k].encode('ascii', 'ignore'))
print "Successfully scraped ID " + model['ID']
return model
except:
model['issue'] = 'Issues'
print "Issues encountered at ID " + model['ID']
raise
return model
def get_curry_soup():
return soupify('http://www.basketball-reference.com/players/c/curryst01.html')
def parse(self,url):
# mine your data here!
# get data. Anything that requires no post-processing is collected here.
model = soupify.soupify(url,self.model)
# get some easy values
model['url'] = url
model['ID'] = soupify.getID(url)
model['extrDate'] = datetime.date.today().isoformat()
# check for redirect:
if model['ID'] not in model['url']:
model['error'] = 'Redirected'
return model
else:
# if no redirect, commence post-processing
try:
# convert isPrinted to boolean: if we gathered alumniclub data, user has been printed. Else, not.
if model['isPrinted']:
model['isPrinted'] = True
else:
model['isPrinted'] = False
# recall that 'age' stores the html containing data for several other fields
if model['age']:
if len(model['age'][0].contents) > 1:
userData = model['age'][0].contents[2]
else:
userData = model['age'][0].contents[0]
else:
model['issue'] = "Empty"
print "Empty results at ID "+model['ID']
return model
# use regex to parse that html for data we want. send data as {'model.key':'regex',...}
# group 'data' will mark the group we need.
userRegex = {'age': 'is a (?P<data>.*) year',
'numDesSco':'scored (?P<data>.*) submissions',
'numScoPri':'helping (?P<data>.*) designs',
'avgSco':'of (?P<data>.*), helping',
'joinDate':'since (?P<data>.*), has',
'gender':'(?P<data>girl|boy)'}
for k,v in userRegex.iteritems():
v = re.search(v,userData)
if v:
model[k] = v.group('data')
else:
model[k] = u''
# format returned values
# eliminate commas, turning "1,000" into "1000"
for i in ['numDesSco','numScoPri']:
if model[i]:
model[i] = model[i].replace(',','')
# convert age into timedelta, and subtract it from today to get an approximate birth date
if model['age']:
model['age'] = (datetime.datetime.today() - datetime.timedelta(float(model['age'])*365.25)).isoformat()
# convert joinDate to a datetime object
if model['joinDate']:
date = model['joinDate'].replace(',','').split(' ')
date[0] = soupify.month[date[0]]
model['joinDate'] = datetime.date(int(date[2]),date[0],int(date[1])).isoformat()
# convert from unicode so the resultant csv isn't full of u''
model['avgSco'] = float(model['avgSco'])
for k in model.keys():
if type(model[k]) is unicode:
model[k] = str(model[k].encode('ascii','ignore'))
print "Successfully scraped ID "+model['ID']
return model
except:
model['issue'] = 'Issues'
print "Issues encountered at ID "+model['ID']
raise
return model
def get_player_soup(player_name):
url = get_player_url(player_name)
soup = soupify(url)
return soup
def parse(url,model=model):
data = soupify.soupify(url)
# mine your data here!
model['url'] = url
model['ID'] = soupify.getID(url)
model['extrDate'] = datetime.date.today().isoformat()
# check for redirect:
if not data:
model['error'] = 'Redirected'
return model
else:
try:
# format arguments en masse as {'model.key':['tag',{'attr':'value','attr':'value',...}],...}
userData = {'user':['a',{"class":"orange nou"}],
'name':['span',{"class":"fn"}],
'url':['a',{'rel':'nofollow me'}],
'aim':['a',{'rel':'me','class':'url'}],
'isPrinted':['a',{'href':'/alumniclub'}]}
# process arguments en masse
for k,v in userData.iteritems():
model[k] = data.findAll(v[0],v[1])
# if no data was returned, do nothing. Else, grab data contents.
if model[k]:
model[k] = model[k][0].contents[0]
# field-specific post-processing
# convert isPrinted to boolean: if we gathered alumniclub data, user has been printed. Else, not.
if model['isPrinted']:
model['isPrinted'] = True
else:
model['isPrinted'] = False
# use regex to parse unmarked string for data we want. send data as {'model.key':'regex',...}
userData = data.findAll('span',{"style":"font-size:12px;"})[0].contents[2]
# 'data' will mark the group we need.
userRegex = {'age': 'is a (?P<data>.*) year',
'numDesSco':'scored (?P<data>.*) submissions',
'numScoPri':'helping (?P<data>.*) designs',
'avgSco':'of (?P<data>.*), helping',
'joinDate':'since (?P<data>.*), has',
'gender':'(?P<data>girl|boy)'}
for k,v in userRegex.iteritems():
v = re.search(v,userData)
if v:
model[k] = v.group('data')
else:
model[k] = u''
# field-specific post-processing
# eliminate commas, turning "1,000" into "1000"
for i in ['numDesSco','numScoPri']:
if model[i]:
model[i] = model[i].replace(',','')
# convert age into timedelta, and subtract it from today to get an approximate birth date
if model['age']:
model['age'] = (datetime.datetime.today() - datetime.timedelta(float(model['age'])*365.25)).isoformat()
# convert joinDate to a datetime object
if model['joinDate']:
date = model['joinDate'].replace(',','').split(' ')
date[0] = soupify.month[date[0]]
model['joinDate'] = datetime.date(int(date[2]),date[0],int(date[1])).isoformat()
# convert from unicode so the resultant csv isn't full of u''
model['avgSco'] = float(model['avgSco'])
for k in model.keys():
if type(model[k]) is unicode:
model[k] = str(model[k].encode('ascii','ignore'))
return model
except:
model['issue'] = 'Issues'
print model['ID']
raise
return model
