12. String Function In Pandas

880B

string dataset.csv

Open

If you are working on a dataset, there are scenarios where you need to perform string operations / manipulation.

String Function In Pandas

To demonstrate, consider this sample dataframe:

import pandas as pd
df = pd.read_csv('string dataset.csv')
df

Output (DataFrame):

	CustomerID	FirstName	LastName	Address	Feedback	PhoneNumber
0	101	RaHul	Sharma	221B baker street, delhi	Product was good but DELIVERY was late	98765432
1	102	Priya	Singh	mg road, bengaluru	Packing was excellent!!	9988776655
2	103	Anil	Kapoor	sector 22, noida	Quality not same as shown	123456789012
3	104	sneha	Agarwal	park street, kolkata	Delay of 5 days. not satisfied	90807060
4	105	Rohan	Mehta	andheri east, mumbai	Product damaged. refund in 7 days?	9090909090
5	106	Meera	Verma	baner road, pune	Delivered early!! very happy	8008008008
6	107	amit	Kumar	gomti nagar, lucknow	Color was different from the website	9876543210987
7	108	SwaTi	Joshi	anna salai, chennai	Double payment deducted	88112
8	109	Ravi	Shukla	ring road, ahmedabad	Wrong item delivered	9999999999
9	110	Nidhi	Gupta	civil lines, jaipur	Refund amount incorrect by 150 rupees	989898989

To use string functions on columns, use the str accessor. Just like dt brings datetime functions, str brings string functions.

strip function

strip removes white spaces from the left and right side of the string.

Example:

df['FirstName'] = df['FirstName'].str.strip()
df

Output (DataFrame):

	CustomerID	FirstName	LastName	Address	Feedback	PhoneNumber
0	101	RaHul	Sharma	221B baker street, delhi	Product was good but DELIVERY was late	98765432
1	102	Priya	Singh	mg road, bengaluru	Packing was excellent!!	9988776655
2	103	Anil	Kapoor	sector 22, noida	Quality not same as shown	123456789012
3	104	sneha	Agarwal	park street, kolkata	Delay of 5 days. not satisfied	90807060
4	105	Rohan	Mehta	andheri east, mumbai	Product damaged. refund in 7 days?	9090909090
5	106	Meera	Verma	baner road, pune	Delivered early!! very happy	8008008008
6	107	amit	Kumar	gomti nagar, lucknow	Color was different from the website	9876543210987
7	108	SwaTi	Joshi	anna salai, chennai	Double payment deducted	88112
8	109	Ravi	Shukla	ring road, ahmedabad	Wrong item delivered	9999999999
9	110	Nidhi	Gupta	civil lines, jaipur	Refund amount incorrect by 150 rupees	989898989

You can strip multiple columns:

df['LastName'] = df['LastName'].str.strip()
df['Address'] = df['Address'].str.strip()
df['Feedback'] = df['Feedback'].str.strip()
df

Output (DataFrame) — same structure with trimmed whitespace.

lower / upper / swapcase / title / capitalize

These methods change case or capitalization.

Example sequence:

df['Address'] = df['Address'].str.title()
df['Address'] = df['Address'].str.capitalize()
df['Address'] = df['Address'].str.swapcase()
df['Address'] = df['Address'].str.lower()
df

Output (DataFrame): addresses shown in lower case.

len function

Counts the number of characters in a value.

df['count'] = df['FirstName'].str.len()
df

Output includes new column count with character lengths of FirstName.

For phone numbers, first cast to string, then measure length:

df['PhoneNumber'] = df['PhoneNumber'].astype(str)
df['Phone Length'] = df['PhoneNumber'].str.len()
df

Output includes Phone Length column.

replace

Used to replace substrings or characters.

df['Address'] = df['Address'].str.replace('street','st.')
df

And removing punctuation from feedback:

df['Feedback'] = df['Feedback'].str.replace('!','').str.replace('?','')
df

Output: punctuation removed in Feedback, street replaced by st. in Address.

split(expand=True)

Split values by a delimiter.

df['Address'].str.split(',')

This returns a Series of lists. To expand into separate columns:

df['Address'].str.split(',', expand=True)

Output (expanded):

	0	1
0	221b baker st.	delhi
1	mg road	bengaluru
2	sector 22	noida
3	park st.	kolkata
4	andheri east	mumbai
5	baner road	pune
6	gomti nagar	lucknow
7	anna salai	chennai
8	ring road	ahmedabad
9	civil lines	jaipur

Assign to new columns:

df['Area'] = df['Address'].str.split(',', expand=True)[0]
df['City'] = df['Address'].str.split(',', expand=True)[1]
df

Output includes Area and City columns.

contains()

Returns True if string contains the keyword (supports regex). Useful with loc to filter rows.

Examples:

df.loc[df['Feedback'].str.contains('Refund', case=False)]

Matches rows where Feedback contains "Refund" (case-insensitive).

Combine conditions:

df.loc[df['Feedback'].str.contains("refund", case=False) | df['Feedback'].str.contains("not satisfied", case=False)]

Or use a regex alternation:

df.loc[df['Feedback'].str.contains('Refund|not satisfied', case=False)]

To require multiple keywords (AND):

df.loc[df['Feedback'].str.contains('damaged', case=False) & df['Feedback'].str.contains('Refund', case=False)]

Note: there's no shortcut operator inside a single contains call for logical AND — combine expressions with &.

startswith()

Checks if a string starts with the given keyword. There's no case parameter for startswith, so standardize case first if needed.

df['Address'] = df['Address'].str.lower()
df.loc[df['Address'].str.startswith('park')]

endswith()

Checks if a string ends with the given keyword. Also has no case parameter; standardize case first.

df['Address'] = df['Address'].str.lower()
df.loc[df['Address'].str.endswith('kolkata')]

string concatenation

You can concatenate strings from columns:

df['full name'] = df['FirstName'] + ' ' + df['LastName']
df

Output includes full name column.

indexing

Access characters by position:

df['area code'] = df['Address'].str[0]
df

Adds area code column with first character of Address.

slicing

Slice substrings using Python slice notation:

df['short name'] = df['FirstName'].str[0:3]
df

Adds short name column.

You can combine operations to create more complex codes:

df['unique code'] = df['FirstName'].str[0:3] + '-' + df['CustomerID'].astype(str) + '-' + df['City'].str[::-1]
df

Output includes unique code column, e.g. RaH-101-ihled.

---

Assignments

STRING FUNCTION ASSIGNMENTS

---

⭐ Basic Cleaning

1. Whitespace Cleanup

• Remove leading/trailing spaces from FirstName, LastName, Address, and Feedback.

2. Case Standardization

• Convert FirstName to proper case (first letter capital).

• Convert LastName to UPPERCASE.

• Convert Address to lowercase.

• Convert Feedback to swapcase().

---

⭐ Length & Validation Tasks

3. Character Count

• Create a column NameLength = length of FirstName + LastName combined.

• Create a column PhoneLength = number of digits in PhoneNumber.

4. Identify Invalid Phone Numbers

• Return all rows where PhoneNumber length is not equal to 10.

• Return all rows where PhoneNumber contains any non-numeric characters.

---

⭐ Replace Tasks

5. Address Cleaning

• Replace “street” → “st.”

6. Remove Punctuation

• Remove all !, ?, ., and , from Feedback.

7. Mask Phone Number

• Show only last 4 digits, mask the rest with *. Hint : You can use function in pandas here

---

⭐ Split & Extract Tasks

8. Address Splitting

Split Address into:

• Area

• City

Create two new columns.

9. Extract City Initial

• Create a column CityCode = first 3 letters of the city.

---

⭐ contains(), startswith(), endswith()

10. Filter on Feedback

Return rows where Feedback:

• contains “refund” OR “damaged”

• contains BOTH words “not” and “satisfied”

• contains the word “delivered” but not “late”

11. Filter on Address

• Show addresses starting with ‘park’.

• Show addresses ending with ‘mumbai’.

• Show addresses that contain a number.

---

⭐ Concatenation Tasks

12. Full Name Creation

Create a column:

• FullName = FirstName + " " + LastName

13. Create a Customer Code

Format:

<First 2 letters of FirstName>-<last 2 digits of CustomerID>-<reverse City>

---

⭐ Indexing & Slicing

14. Extract Codes

• Create AreaCode = first character of Address.

• Create ShortName = first 3 characters of FirstName.

15. Reverse Manipulations

• Reverse LastName.

• Reverse full Address string.