Pandas技术实现：将横向多产品账户数据转置为纵向格式并重复账户关联信息

This is a classic wide-to-long data reshaping task, and pandas has a built-in function melt() that makes this super straightforward. Here's how to transform your data into the exact format you need for Salesforce:

Step-by-Step Breakdown #

1. Set Up Pandas: First, ensure pandas is installed and imported.
2. Replicate Your Input: Let's start by building the sample DataFrame from your provided input.
3. Melt the Wide Data: Use pd.melt() to pivot all product columns into a single Product column, keeping the account fields as fixed identifiers.
4. Clean Up Empty Entries: Drop any rows where the Product value is blank (since those accounts don't own that product).
5. Reset the Index: Adjust the index to match your expected sequential output.

Full Code Implementation #

import pandas as pd

# Create the original wide-format DataFrame
data = {
    'Account Name': ['csld solutions', 'global dynamics', 'running inc.'],
    'Acct ID': ['1234', '0256', '2564'],
    'crest': ['', 'crest', ''],
    'ptx': ['ptx', '', ''],
    'gtchange': ['', 'gtchange', 'gtchange'],
    'gtfire': ['gtfire', '', 'gtfire'],
    'dfdata': ['', 'dfdata', '']
}
df = pd.DataFrame(data)

# Reshape to long format using melt()
melted_df = pd.melt(
    df,
    id_vars=['Account Name', 'Acct ID'],  # Columns to retain as identifiers
    value_vars=['crest', 'ptx', 'gtchange', 'gtfire', 'dfdata'],  # Product columns to pivot
    value_name='Product'  # Name for the new consolidated product column
)

# Remove rows with empty product values and reset index
final_df = melted_df[melted_df['Product'] != ''].reset_index(drop=True)

print(final_df)

Resulting Output #

Account Name Acct ID    Product
0   csld solutions    1234        ptx
1   csld solutions    1234     gtfire
2  global dynamics    0256      crest
3  global dynamics    0256  gtchange
4  global dynamics    0256     dfdata
5     running inc.    2564  gtchange
6     running inc.    2564     gtfire

This output matches exactly the structure you need—each product gets its own row paired with the full account details, ready for upload to Salesforce.

内容的提问来源于stack exchange，提问作者mtm1186