Python pandas reading table data but commas is removed

how do I have the original data for column "Winning No". How to have the comma not removed?

Result should be 1,45,13,12,13 instead of 145131213

Here my code:

import requests
import pandas as pd

pd.options.display.width = 0
pd.options.display.max_rows = 1000

url = 'https://en.lottolyzer.com/history/singapore/toto'
html = requests.get(url).content
df_list = pd.read_html(html)



df = df_list[-1]
print(df)
df.to_csv('my data.csv')

By inspecting the read_html documentation I found out that the parameters decimal and thousands could be helpful in these cases, but they do not work in this specific one. With that being said, I propose a customized solution which makes use also of the BeautifulSoup library in order to parse the HTML result in an easier way.

import requests
from bs4 import BeautifulSoup
import pandas as pd

url = 'https://en.lottolyzer.com/history/singapore/toto'
html = requests.get(url).content
soup = BeautifulSoup(html, 'html.parser')

# Extract the initial DataFrame from the HTML table.
# Note that read_html returns a list of DataFrames (one for each table retrieved from the HTML), 
# so we need to select the first element (in the webpage there is only one table).
df_initial = pd.read_html(html)[0]

# Find the HTML table containing the winning numbers
table = soup.find('table', class_='responsive-table')

# Extract the "Winning No" column and add commas
# The last row of this block is used to preserve the format of the original HTML table,
# which reports the column label also into the last line.
winning_no = table.find_all('td', class_='sum-p1')[1::3]
winning_no = [num.text for num in winning_no]
winning_no.append(winning_no.pop(0))

# Create a DataFrame with the "Winning No" column
df = pd.DataFrame({'Winning No.': [num for num in winning_no]})

# Replacing the "Winning No" column in the initial DataFrame with the new parsed column.
df_initial['Winning No.'] = df['Winning No.']
print(df_initial)

TL;DR:

the line winning_no = table.find_all('td', class_='sum-p1')[1::3] finds all the table rows with class sum-p1 which by inspecting the results give this output:

    <td class="sum-p1">Date</td>, 
    <td class="sum-p1">Winning No.</td>, 
    <td class="sum-p1">Addl No.</td>, 
    <td class="sum-p1">2023-05-08</td>, 
    <td class="sum-p1">1,3,6,11,23,34</td>, 
    <td class="sum-p1">39</td>, 
    <td class="sum-p1">2023-05-04</td>, 
    <td class="sum-p1">16,21,25,28,37,44</td>, 
    <td class="sum-p1">24</td>, 
    ...
]

As you can see the Winning No. elements starts from the element of index 1 of this array and are reachable by accessing the array starting from that index to the end with step 3 ([1::3]).

This line instead winning_no.append(winning_no.pop(0)) is used to remove the Header, i.e. the string Winning No. from the head of the list and put it in the last position. In this way when in the next step we'll create back the dataframe we will preserve the original structure of the table, having the label of the column both on the top of the table and in its last row.