cannot get the span text within the "table", thanks !
from bs4 import BeautifulSoup
import urllib2
url1 = "url"
content1 = urllib2.urlopen(url1).read()
soup = BeautifulSoup(content1,"lxml")
table = soup.findAll("div", {"class" : "iw_component","id":"c1417094965154"})
rows = table.find_all('span',recursive=False)
for row in rows:
print(row.text)
table = soup.findAll("div", {"class" : "iw_component","id":"c1417094965154"})
In the above line, findAll() returns a list.
So, in the next line you are getting the error because its expecting an HTML string.
If you expect only one table, try using the following code. Just replace
rows = table.find_all('span',recursive=False)
with
rows = table[0].find_all('span')
If you expect multiple tables in the page, run a for loop on the table and then run the rest of the statements inside the for loop.
Also, for pretty output, you can replace the tabs with spaces as in the following code:
row = row.get_text()
row = row.replace('\t', '')
print(row)
The final working code for you is:
from bs4 import BeautifulSoup
import urllib2
url1 = "url"
content1 = urllib2.urlopen(url1).read()
soup = BeautifulSoup(content1,"lxml")
table = soup.findAll("div", {"class" : "iw_component","id":"c1417094965154"})
rows = table[0].find_all('span')
for row in rows:
row_str = row.get_text()
row_str = row_str.replace('\t', '')
print(row_str)
Regarding recursive=False parameter, if it's set to false, it will only find in direct children which, in your case will give no result.
Recursive Argument in find()
If you only want Beautiful Soup to consider direct children, you can pass in recursive=False
Here's another approach using lxml instead of beautifulsoup:
import requests
from lxml import html
req = requests.get("<URL>")
raw_html = html.fromstring(req.text)
spans = raw_html.xpath('//div[#id="c1417094965154"]//span/text()')
print("".join([x.replace("\t", "").replace("\r\n","").strip() for x in spans]))
Output: Kranji Mile Day simulcast races, Kranji Racecourse, SINClass 3 Handicap - 1200M TURFSaturday, 26 May 2018Race 1, 5:15 PM
As you see, the output need a little formatting, spans is a list of all spans text, so you can do any processing you need.
You seem to use python 2.x, here is a python 3.x solution, since I do not have a python 2.x environment at the moment :
from bs4 import BeautifulSoup
import urllib.request as urllib
url1 = "<URL>"
# Read the HTML page
content1 = urllib.urlopen(url1).read()
soup = BeautifulSoup(content1, "lxml")
# Find the div (there is only one, so you do not need findAll) -> this is your problem
div = soup.find("div", class_="iw_component", id="c1417094965154")
# Now you retrieve all the span within this div
rows = div.find_all("span")
# You can do what you want with it !
line = ""
for row in rows:
row_str = row.get_text()
row_str = row_str.replace('\t', '')
line += row_str + ", "
print(line)
Related
I am new to data scraping with Beautiful Soup. I would like to get data from pro-football-reference on these stats: https://www.pro-football-reference.com/boxscores/201009090nor.htm#all_pbp
I would like to iterate through every row under the 'Detail Column' under the Full Play-By-Play Table so that if the Detail contains the word "Penalty" I can save that. Any chance anyone knows how I could possibly do this? This table seems different than others.
# Any example of how I extracted another element (Referee Name)
# from the same page but different table
table = soup.select_one('#all_officials').find_next(text=lambda t: isinstance(t, Comment))
table = BeautifulSoup(table, 'html.parser')
for tr in table.select('tr'):
tds = [td.get_text(strip=True) for td in tr.select('td')]
if str(*tds) != "Officials":
referee = str(*tds)
break
The table is commented out. A common and reliable way is to import Comment and handle with for comment in soup.find_all(text=lambda text: isinstance(text, Comment)) as shown here: https://stackoverflow.com/a/60381103.
For this particular instance, I am just removing the comments strings through substitution.
Then I use :-soup-contains to target the appropriate rows, filtering on only those rows within the table where the text Penalty appears in the elements with data-stat attribute having value = detail i.e. the detail column.
I then use pandas to reconstitute the table from the filtered trs html joined and then book-ended by table tags
from bs4 import BeautifulSoup as bs
import pandas as pd
import requests
import re
r = requests.get('https://www.pro-football-reference.com/boxscores/201009090nor.htm#all_pbp',
headers={'User-Agent': 'Mozilla/5.0'})
s = re.sub(r'<!--|-->', '', r.text)
soup = bs(s, 'lxml')
s2 = '<table>' + ''.join([str(i) for i in soup.select(
'#pbp tr:has([data-stat=detail]:-soup-contains("Penalty"))')]) + '</table>'
df = pd.read_html(s2)[0]
df.columns = [i.text for i in soup.select('#pbp thead > tr > th')]
df
As of now, I'm only getting ['1'] as the output of what's being printed with my current code below. I want to grab 1-54 on the Team Batting table in the Rk column on the website https://www.baseball-reference.com/teams/NYY/2019.shtml.
How would I go about modifying colNum so it can print the 1-54 in the Rk column? I'm pointing out the colNum line because I feel the issue lies there but I could be wrong.
import pandas as pd
import requests
from bs4 import BeautifulSoup
page = requests.get('https://www.baseball-reference.com/teams/NYY/2019.shtml')
soup = BeautifulSoup(page.content, 'html.parser') # parse as HTML page, this is the source code of the page
week = soup.find(class_='table_outer_container')
items = week.find("thead").get_text() # grabs table headers
th = week.find("th").get_text() # grabs Rk only.
tbody = week.find("tbody")
tr = tbody.find("tr")
thtwo = tr.find("th").get_text()
colNum = [thtwo for thtwo in thtwo]
print(colNum)
Your mistake was in the last few lines as you mentioned. If I understood right, you wanted a list of all the values in the "Rk" column. In order to get all the rows, you have to use the find_all() function. I tweaked your code a little bit in order to get the text of the first field in each row in the following lines:
import pandas as pd
import requests
from bs4 import BeautifulSoup
page = requests.get('https://www.baseball-reference.com/teams/NYY/2019.shtml')
soup = BeautifulSoup(page.content, 'html.parser')
is the source code of the page
week = soup.find(class_='table_outer_container')
items = week.find("thead").get_text()
th = week.find("th").get_text()
tbody = week.find("tbody")
tr = tbody.find_all("tr")
colnum = [row.find("th").get_text() for row in tr]
print(colnum)
I'm trying to use BeautifulSoup to select the date, url, description, and additional url from table and am having trouble accessing them given the weird white spaces:
So far I've written:
import urllib
import urllib.request
from bs4 import BeautifulSoup
def make_soup(url):
thepage = urllib.request.urlopen(url)
soupdata = BeautifulSoup(thepage, "html.parser")
return soupdata
soup = make_soup('https://www.sec.gov/litigation/litreleases/litrelarchive/litarchive2010.shtml')
test1 = soup.findAll("td", {"nowrap" : "nowrap"})
test2 = [item.text.strip() for item in test1]
With bs4 4.7.1 you can use :has and nth-of-type in combination with next_sibling to get those columns
from bs4 import BeautifulSoup
import requests, re
def make_soup(url):
the_page = requests.get(url)
soup_data = BeautifulSoup(the_page.content, "html.parser")
return soup_data
soup = make_soup('https://www.sec.gov/litigation/litreleases/litrelarchive/litarchive2010.shtml')
releases = []
links = []
dates = []
descs = []
addit_urls = []
for i in soup.select('td:nth-of-type(1):has([href^="/litigation/litreleases/"])'):
sib_sib = i.next_sibling.next_sibling.next_sibling.next_sibling
releases+= [i.a.text]
links+= [i.a['href']]
dates += [i.next_sibling.next_sibling.text.strip()]
descs += [re.sub('\t+|\s+',' ',sib_sib.text.strip())]
addit_urls += ['N/A' if sib_sib.a is None else sib_sib.a['href']]
result = list(zip(releases, links, dates, descs, addit_urls))
print(result)
Unfortunately there is no class or id HTML attribute to quickly identify the table to scrape; after experimentation I found it was the table at index 4.
Next we ignore the header by separating it from the data, which still has table rows that are just separations for quarters. We can skip over these using a try-except block since those only contain one table data tag.
I noticed that the description is separated by tabs, so I split the text on \t.
For the urls, I used .get('href') rather than ['href'] since not every anchor tag has an href attribute from my experience scraping. This avoids errors should that case occur. Finally the second anchor tag does not always appear, so this is wrapped in a try-except block as well.
data = []
table = soup.find_all('table')[4] # target the specific table
header, *rows = table.find_all('tr')
for row in rows:
try:
litigation, date, complaint = row.find_all('td')
except ValueError:
continue # ignore quarter rows
id = litigation.text.strip().split('-')[-1]
date = date.text.strip()
desc = complaint.text.strip().split('\t')[0]
lit_url = litigation.find('a').get('href')
try:
comp_url = complaint.find('a').get('href')
except AttributeError:
comp_ulr = None # complaint url is optional
info = dict(id=id, date=date, desc=desc, lit_url=lit_url, comp_url=comp_url)
data.append(info)
I am currently running the following python script:
import requests
from bs4 import BeautifulSoup
origin= ["USD","GBP","EUR"]
i=0
while i < len(origin):
page = requests.get("https://www.x-rates.com/table/?from="+origin[i]+"&amount=1")
soup = BeautifulSoup(page.content, "html.parser")
tables = soup.findChildren('table')
my_table = tables[0]
rows = my_table.findChildren(['td'])
i = i +1
for rows in rows:
cells = rows.findChildren('a')
for cell in cells:
value = cell.string
print(value)
To scrape data from this HTML:
https://i.stack.imgur.com/DkX83.png
The problem I have is that I'm struggling to only scrape the first column without scraping the second one as well because they are both under tags and in the same table row as each other. The href is the only thing which differentiates between the two tags and I have tried filtering using this but it doesn't seem to work and returns a blank value. Also when i try to sort the data manually the output is amended vertically and not horizontally, I am new to coding so any help would be appreciated :)
There is another way you might wanna try as well to achieve the same:
import requests
from bs4 import BeautifulSoup
keywords = ["USD","GBP","EUR"]
for keyword in keywords:
page = requests.get("https://www.x-rates.com/table/?from={}&amount=1".format(keyword))
soup = BeautifulSoup(page.content, "html.parser")
for items in soup.select_one(".ratesTable tbody").find_all("tr"):
data = [item.text for item in items.find_all("td")[1:2]]
print(data)
It is easier to follow what happens when you print every item you got from the top e.g. in this case from table item. The idea is to go one by one so you can follow.
import requests
from bs4 import BeautifulSoup
origin= ["USD","GBP","EUR"]
i=0
while i < len(origin):
page = requests.get("https://www.x-rates.com/table/?from="+origin[i]+"&amount=1")
soup = BeautifulSoup(page.content, "html.parser")
tables = soup.findChildren('table')
my_table = tables[0]
i = i +1
rows = my_table.findChildren('tr')
for row in rows:
cells = row.findAll('td',class_='rtRates')
if len(cells) > 0:
first_item = cells[0].find('a')
value = first_item.string
print(value)
I am writing a Python script using BeautifulSoup to scrape values from this webpage: https://uk-air.defra.gov.uk/latest/currentlevels
I want to use soup.find() to get values for "Hourly mean Nitrogen dioxide" and "Last updated" from the table row where the "Monitoring site" is "Edinburgh St Leonards".
As I am new to web scraping I am having a bit of trouble so would be grateful for any help on this.
Scrap all the html tables in a list of tables.
The table index may change, then you should not rely on a row/column index.
A part of the folowing script look up for the index of the searched data. Moreover, it prints the header name: so you know want are the data you get.
from bs4 import BeautifulSoup
import urllib.request
import re
with urllib.request.urlopen('https://uk-air.defra.gov.uk/latest/currentlevels?view=region') as response:
htmlData = response.read()
soup = BeautifulSoup(htmlData, 'html5lib')
tables = soup.find_all('table', attrs={'class':'current_levels_table'})
#what you want to check:
Iwant = ['nitrogen', 'update']
about = 'Edinburgh'
for table in tables:
#get header to have the data (we're looking for) column number and table real names
table_head = table.find('thead')
headrows = table_head.find_all('tr')
measures = headrows[1].find_all('th')
for colnum, measure in enumerate(measures):
index.update({colnum: measure.text.strip() for wanted in Iwant if re.search(wanted+'(?iu)', measure.text)})
#get table content and look for Edinburgh
table_body = table.find('tbody')
rows = table_body.find_all('tr')
for row in rows:
cels = row.find_all('td')
rowContent = [cel.text.strip().replace(u'\xa0', u' ').replace(u'\n Timeseries Graph', u'') for cel in cels if cel]
if re.search(about+'(?iu)', rowContent[0]):
for indexwanted, measurewanted in index.items():
print(measurewanted, ':', rowContent[indexwanted])
Making use of the suggestion from d2718nis, you can do it in this way. Of course, many other ways would work too.
First, find the link that has the 'Edinburgh St Leonards' text in it. Then find the grandparent of that link element, which is a tr element. Now identify the td elements in the tr. When you examine the table you see that the columns you want are the 4th and 7th. Get those from all of the td elements as the (0-relative) 3rd and 6th. Finally, display the crude texts of these elements.
You will need to do something clever to extract properly readable strings from these results.
>>> import requests
>>> import bs4
>>> page = requests.get('https://uk-air.defra.gov.uk/latest/currentlevels', headers={'User-Agent': 'Not blank'}).content
>>> soup = bs4.BeautifulSoup(page, 'lxml')
>>> Edinburgh_link = soup.find_all('a',string='Edinburgh St Leonards')[0]
>>> Edinburgh_link
Edinburgh St Leonards
>>> Edinburgh_row = Edinburgh_link.findParent('td').findParent('tr')
>>> Edinburgh_columns = Edinburgh_row.findAll('td')
>>> Edinburgh_columns[3]
<td class="center"><span class="bg_low1 bold">20 (1 Low)</span></td>
>>> Edinburgh_columns[6]
<td>05/08/2017<br/>14:00:00</td>
>>> Edinburgh_columns[3].text
'20\xa0(1\xa0Low)'
>>> Edinburgh_columns[6].text
'05/08/201714:00:00'
you can start with this:
import requests
from bs4 import BeautifulSoup
# Request the page, set headers to prevent 403 Forbidden
page = requests.get(
url='https://uk-air.defra.gov.uk/latest/currentlevels',
headers={'User-Agent': 'Not blank'})
# Get html from page
html = page.text
# BeautifulSoup object
soup = BeautifulSoup(html, 'html5lib')
for table in soup.find_all('table'):
# Print all tables on the page
print(table)