如何用Pandas将DataFrame中phone1-phone7列的前置0替换为91

Hey there, let's solve this phone number formatting problem step by step. The goal is to replace the single leading 0 in columns phone1 through phone7 with 91, while keeping numbers with multiple leading zeros (like 00601731) unchanged.

Step 1: Identify Target Columns #

First, let's define the list of phone columns we need to process:

import pandas as pd

# Assuming your DataFrame is already loaded as df
phone_columns = [f"phone{i}" for i in range(1, 8)]

Step 2: Process the Columns with Regex Replacement #

We'll use a regular expression to target only numbers that start with a single 0 (not followed by another 0). Here's the cleanest approach:

# Convert columns to string type to handle numeric values properly
df[phone_columns] = df[phone_columns].astype(str)

# Use regex to replace single leading 0 with 91
df[phone_columns] = df[phone_columns].replace(r"^0(?!0)", "91", regex=True)

Regex Explanation:

• ^0: Matches a 0 at the start of the string
• (?!0): Negative lookahead, ensures the next character is not another 0
  This way, only numbers starting with exactly one 0 get modified—values like 00601731 are left untouched.

Step 3: Handle NaN Values (Optional) #

If your DataFrame has NaN values, converting to string will turn them into the literal string "NaN". To preserve actual NaN values, use this adjusted code:

for col in phone_columns:
    # Only process non-null values
    mask = df[col].notna()
    df.loc[mask, col] = df.loc[mask, col].astype(str).replace(r"^0(?!0)", "91", regex=True)

Original vs. Expected Output #

Original Data:

kac  play_id    phone1       phone2       phone3 phone4 phone5 phone6 phone7
0   0  5004490  20002075  0900031349   090891349    NaN    NaN    NaN    NaN
1   1  5003807  00601731   088235311        NaN    NaN    NaN    NaN    NaN
2   2  5003808  00601731   088235311        NaN    NaN    NaN    NaN    NaN

After Processing:

kac  play_id    phone1        phone2        phone3 phone4 phone5 phone6 phone7
0   0  5004490  20002075  91900031349  9190891349    NaN    NaN    NaN    NaN
1   1  5003807  00601731   9188235311         NaN    NaN    NaN    NaN    NaN
2   2  5003808  00601731   9188235311         NaN    NaN    NaN    NaN    NaN

内容的提问来源于stack exchange，提问作者Tilo