def scrape_document_text(link):
'''
This function takes in a link to the page with text for speech/statement/press and returns
just the relevant text
'''
soup = make_soup(link)
text = soup.find("span", {"class":"displaytext"}).text
return text
def scrape_document_text(link):
'''
This function takes in a link to the page with text for speech/statement/press and returns
just the relevant text
'''
soup = make_soup(link)
text = soup.find("span", {"class": "displaytext"}).text
return text
def get_page_text(link):
'''
Given a link to a senate doc, returns relevant text on that get_page_text
'''
soup = custom_utilities.make_soup(link)
text = soup.find("pre", {"class": "styled"}).text
text = re.sub('<[^>]+>', ' ', text)
text = re.sub("[\[].*?[\]]", " ", text)
return text
def get_page_text(link):
'''
Given a link to a senate doc, returns relevant text on that get_page_text
'''
soup = custom_utilities.make_soup(link)
text = soup.find("pre", {"class":"styled"}).text
text = re.sub('<[^>]+>', ' ', text)
text = re.sub("[\[].*?[\]]", " ", text)
return text
def get_text_of_individual_documents(link):
'''
Takes in link to the page with a table of |candidate -- date -- link| and returns a
dictionary of {date: text of link}
'''
rv = {}
soup = make_soup(link)
# Find the table of links to individual documents and pull out the rows
cand_table = soup.findAll("table", {"width":"700", "border":"0", "align":"center"})
rows = cand_table[0].findAll("tr")
rows = rows[1:]
# For each row, find each column. Then scrape the text for the link in that row
# and store in return dictionary
for row in rows:
elems = row.findAll("td")
link = "http://www.presidency.ucsb.edu" + elems[2].findAll("a", href=True)[0]['href'][2:]
rv[elems[1].text] = scrape_document_text(link)
return rv
def collect_suitable_links_for_day(link):
'''
Given a link to a day when there were senate proceedings, returns all links that should
be followed and scraped
'''
links = {}
soup = custom_utilities.make_soup(link)
main_table = soup.findAll("table", {"class":"item_table"})
if len(main_table)==0: # this happens when no senate activity on date or invalid date
return []
# Find all columns in the table
tds = main_table[0].findAll("td")
# For each column, grab the link
for td in tds:
a = td.findAll("a")
if should_follow_link(a[0].text):
links[a[0].text] = {}
links[a[0].text]['link'] = a[0]['href']
return links
def collect_suitable_links_for_day(link):
'''
Given a link to a day when there were senate proceedings, returns all links that should
be followed and scraped
'''
links = {}
soup = custom_utilities.make_soup(link)
main_table = soup.findAll("table", {"class":"item_table"})
if len(main_table)==0: # this happens when no senate activity on date or invalid date
return []
# Find all columns in the table
tds = main_table[0].findAll("td")
# For each column, grab the link
for td in tds:
a = td.findAll("a")
if should_follow_link(a[0].text):
links[a[0].text] = {}
links[a[0].text]['link'] = a[0]['href']
return links
def get_links_to_documents_pages(link):
'''
Given the main link, traverses page and gets all links to the pages with
statements, etc. on them. Stores in dictionary by candidate name
'''
rv = {}
# Get soup for the main page and parse it into the rows
soup = make_soup(link)
rows = soup.findAll('td', {'class': 'doctext'})
# Loop over rows
for row in rows:
elems = row.findAll("span")
# if no span elements, is not a row with links
if len(elems) != 0:
candidate = elems[0].text
rv[candidate] = {}
links = row.findAll("a")
for l in links:
rv[candidate][l.text] = "http://www.presidency.ucsb.edu/" + l['href']
return rv
def get_links_to_documents_pages(link):
'''
Given the main link, traverses page and gets all links to the pages with
statements, etc. on them. Stores in dictionary by candidate name
'''
rv = {}
# Get soup for the main page and parse it into the rows
soup = make_soup(link)
rows = soup.findAll('td', {'class': 'doctext'})
# Loop over rows
for row in rows:
elems = row.findAll("span")
# if no span elements, is not a row with links
if len(elems) != 0:
candidate = elems[0].text
rv[candidate] = {}
links = row.findAll("a")
for l in links:
rv[candidate][
l.text] = "http://www.presidency.ucsb.edu/" + l['href']
return rv
def get_text_of_individual_documents(link):
'''
Takes in link to the page with a table of |candidate -- date -- link| and returns a
dictionary of {date: text of link}
'''
rv = {}
soup = make_soup(link)
# Find the table of links to individual documents and pull out the rows
cand_table = soup.findAll("table", {
"width": "700",
"border": "0",
"align": "center"
})
rows = cand_table[0].findAll("tr")
rows = rows[1:]
# For each row, find each column. Then scrape the text for the link in that row
# and store in return dictionary
for row in rows:
elems = row.findAll("td")
link = "http://www.presidency.ucsb.edu" + elems[2].findAll(
"a", href=True)[0]['href'][2:]
rv[elems[1].text] = scrape_document_text(link)
return rv
