Wikitable scrapping using python

user_15888985 0
Viewed 46

I was trying to scrape all the tables in the following link of Wikipedia in general to get the episode number and name. But it stops near the first table and doesn't move around with the second one. I need some light on it.

wiki_link : https://en.wikipedia.org/wiki/One_Piece_(season_20)#Episode_list

But the given data in table looks like this : enter image description here

Basically I am trying to fetch the data in the rows with respect to the columns [ No.Overall [n1] & Title[n2] ]

[i.e 892 The Land of Wano! To the Samurai Country where Cherry Blossoms Flutter! ]

*Required output in CSV like:

enter image description here

the code:

from bs4 import BeautifulSoup
from pandas.plotting import table
import requests
 
url = "https://en.wikipedia.org/wiki/One_Piece_(season_20)#Episode_list"

response = requests.get(url)

soup = BeautifulSoup(response.text,'lxml')

table = soup.find('table',{'class':'wikitable plainrowheaders wikiepisodetable'}).tbody
rows = table.find_all('tr')
columns = [v.text.replace('\n','') for v in rows[0].find_all('th')]
#print(len(rows))
#print(table)

df = pd.DataFrame(columns=columns)
print(df)

for i in range(1,len(rows)):
    tds = rows[i].find_all('td')
    print(len(tds))
    if len(tds)==6:
        values1 = [tds[0].text, tds[1].text, tds[2].text, tds[3].text.replace('\n',''.replace('\xa0',''))]
        epi = values1[0]
        title = values1[1].split('Transcription:')
        titles = title[0]
        print(f'{epi}|{titles}')
    else:
        values2 = [td.text.replace('\n',''.replace('\xa0','')) for td in tds]
pandas web-scraping beautifulsoup python
2 Answers

I did a little bit of research regarding the webpage, and you don't need to go far when it comes to obtaining the tables. It seems like the data is split in 3 tables which we can concanate:

df = pd.concat(pd.read_html("https://en.wikipedia.org/wiki/One_Piece_(season_20)#Episode_list")[1:4]).reset_index()

With a little bit of data manipulation, we can extract the useful information:

episodes = pd.to_numeric(df['No.overall [n 1]'],errors='coerce').dropna()[:-2]
names = df.loc[(episodes.index+1),'No.overall [n 1]'].dropna()

Finally, we can create a new dataframe out of it:

output = pd.DataFrame({"Episode No.":episodes.values,
                       "Summary":names.values})

EDIT:

Given I originally considered the wrong column, getting the episode number and title is even easier:

df = pd.concat(pd.read_html("https://en.wikipedia.org/wiki/One_Piece_(season_20)#Episode_list")[1:4])

df['Episode'] = pd.to_numeric(df['No.overall [n 1]'],errors='coerce')
df = df.dropna(subset='Episode').rename(columns={'Title [n 2]':'Title'})[['Episode','Title']]

Returning:

   Episode                                              Title
1      892  "The Land of Wano! To the Samurai Country wher...
3      893  "Otama Appears! Luffy vs. Kaido's Army!"Transc...
5      894  "He'll Come! The Legend of Ace in the Land of ...
1      895  "Side Story! The World's Greatest Bounty Hunte...
3      896  "Side Story! Clash! Luffy vs. the King of Carb...
..     ...                                                ...
269   1027  "Defend Luffy! Zoro and Law's Sword Technique!...
271   1028  "Surpass the Emperor of the Sea! Luffy Strikes...
274   1029  "A Faint Memory! Luffy and Red-Haired's Daught...
275   1030  "The Oath of the New Era! Luffy and Uta"Transc...
278   1031  "Nami's Scream - The Desperate Death Race!"Tra...
user_118970070

If you like to go with your BeautifulSoup approach, select all the <tr> with class vevent and iterate the ResultSet to create a list of dicts that you can use to create a dataframe, ...:

[
    {
        'No.overall':r.th.text,
        'Title':r.select('td:nth-of-type(2)')[0].text.split('Transcription:')[0]
    } 
    for r in soup.select('tr.vevent')
]

Example

from bs4 import BeautifulSoup
import requests
 
url = "https://en.wikipedia.org/wiki/One_Piece_(season_20)#Episode_list"
soup = BeautifulSoup(requests.get(url).text)

pd.DataFrame(
    [
        {
            'No.overall':r.th.text,
            'Title':r.select('td:nth-of-type(2)')[0].text.split('Transcription:')[0]
        } 
        for r in soup.select('tr.vevent')
    ]
)

Output

No.overall Title
892 "The Land of Wano! To the Samurai Country where Cherry Blossoms Flutter!"
893 "Otama Appears! Luffy vs. Kaido's Army!"
894 "He'll Come! The Legend of Ace in the Land of Wano!"
895 "Side Story! The World's Greatest Bounty Hunter, Cidre!"
896 "Side Story! Clash! Luffy vs. the King of Carbonation!"

...

user_144608240
Related