CODEWITHSUNDEEP
pythonpandascsvextractkeyword
Notna
Notna

Reputation: 525

Extracting keywords from csv in Python

I have a comma-separated csv file with three columns: 

"Date", "URL", "Views"

and I am trying to extract certain rows that contain specific keywords in the column URL, like the word charger for example.

import pandas as pd

keywords = {"charger"}

df = pd.read_csv("original_file.csv", sep=",")

listMatchURL = []

for i in range(len(df.index)):
    if any(x in df['URL'][i] for x in keywords):
        listMatchURL.append(df['URL'][i])

output = pd.DataFrame({'URL': listMatchURL})
output.to_csv("new_file.csv", index=False)

This writes in a new csv file the entire URL row that contains the keyword. But how can I extract and write only the keyword, instead of the entire URL? I don't want to extract the entire http://www.example.com/search/iphone+charger.html but simply charger.

And also, how can I keep the two other corresponding columns Date and Views in the new csv file I'm writing? For now, it extracts only the URL column.

I'm looking to get a new csv file that has the columns: 

"Date", "Keyword", "Views"

Upvotes: 0

Views: 1429

Answers (1)

Martin Evans
Martin Evans

Reputation: 46759

As an alternative, this could be done without Pandas as follows:

import csv

keywords = {"charger"}

with open('original_file.csv', newline='') as f_input, open('new_file.csv', 'w', newline='') as f_output:
    csv_input = csv.reader(f_input)
    csv_output = csv.writer(f_output)

    header = next(csv_input)
    csv_output.writerow(['Date', 'Keyword', 'Views'])

    for date, url, views in csv_input:
        for keyword in keywords:
            if keyword in url:
                csv_output.writerow([date, keyword, views])
                break       # Remove if multiple keywords per url are allowed

Upvotes: 1

Related Questions