Downloading Google CodeIn tasks without an API Key.

gci_instances.py

"""
Note : This approach is an alternative to using the API for fetching instance info by 
saving the webpages in ~/Downloads/ instead and using BeautifulSoup for parsing the data.
Go to https://codein.withgoogle.com/dashboard/task-instances/?sp-order=name&sp-my_tasks=false&sp-page_size=100 
Iterate over all pages (1, 2, 3...) and save them.
"""
from glob import glob
from bs4 import BeautifulSoup as soup
import pandas as pd

all_rows = []
for fname in glob("/Users/vishalgupta/Downloads/Task instances _ Google Code-in *.htm"):
    with open(fname) as f:
        cont = f.read()
    s = soup(cont)
    rows = s.select('md-table-container tbody tr')
    for row in rows:
        vals = [i.text.strip() for i in row.select('td') if i.text.strip()]
        all_rows.append(vals)

col_names = [i.text.strip() for i in s.select('md-table-container th') if i.text.strip()]
df = pd.DataFrame(all_rows,columns=col_names)
df.to_csv("GCI_instance_dump.csv")