使用Pandas read_html导出多表至CSV遇AttributeError问题求助

Hey there! Let's break down why you're hitting that AttributeError: 'list' object has no attribute 'to_csv' error and how to fix it.

Why the Error Happens #

The pandas.read_html() function doesn't return a single DataFrame—it returns a list of DataFrames, one for each table found in your example.html file. When you try calling .to_csv() directly on that list, Python throws an error because lists don't have that method.

Two Common Solutions for Exporting Multiple Tables to One CSV #

Depending on how you want your final CSV structured, use one of these approaches:

1. Combine Tables with the Same Structure (Vertical Stack)

If all your HTML tables have identical columns and you want to stack them into one big table:

import pandas as pd

# Read all tables from the HTML file
table_list = pd.read_html('example.html')

# Combine all DataFrames into one (ignore_index resets row numbers)
combined_table = pd.concat(table_list, ignore_index=True)

# Export to CSV
combined_table.to_csv('combined_output.csv', index=False)

• ignore_index=True ensures your final table has continuous row numbers instead of keeping the original indices from each individual table.
• index=False prevents pandas from adding an extra index column to your CSV.

2. Write Tables Separately (With Headers/Blank Lines)

If your tables have different structures and you want to keep them distinct in the CSV (e.g., each table with its own header, separated by blank lines):

import pandas as pd

# Read all tables
table_list = pd.read_html('example.html')

# Open the CSV file in write mode
with open('separate_tables_output.csv', 'w', newline='', encoding='utf-8') as csv_file:
    for idx, table in enumerate(table_list):
        # Only write the header for the first table (adjust if you want headers for all)
        table.to_csv(csv_file, index=False, header=idx == 0)
        # Add a blank line between tables for readability
        csv_file.write('\n')

• header=idx == 0 tells pandas to write the column headers only once (for the first table). Change this to header=True if you want each table to show its own headers.
• newline='' avoids extra blank lines in the CSV on Windows systems.

Quick Debug Tip #

Before exporting, you can check how many tables were read and inspect their structure:

print(f"Number of tables found: {len(table_list)}")
# Inspect the first table's columns
print(table_list[0].columns)

内容的提问来源于stack exchange，提问作者statsguy