如何在Python DataFrame中拆分单元格合并数据至对应列？

Let's break down what's going wrong and fix it step by step:

1. Why Your Current Code Isn't Working #

• Professional Missing Spaces: When you split the extracted string with str.split("\s", expand=True), you're grabbing date_err[0] (Jenn) and date_err[1] (Blossoms) and concatenating them directly—no space in between, hence "JennBlossoms" instead of "Jenn Blossoms".
• Empty/Incomplete Description: You're only taking date_err[2] (which is "Telephone"), but the full description includes everything after "Jenn Blossoms", not just a single word.
• SettingWithCopyWarning: This pops up because your date_err is created from a slice of the original DataFrame (dftopdata[s]["Date"]), and pandas is warning you that you might be modifying a copy instead of the original data.

2. Corrected Solution Code #

import pandas as pd

# Recreate your sample DataFrame for reference
data = {
    "Date": [
        "2019-12-19 00:00:00",
        "2019-12-20 00:00:00",
        "2019-12-27 00:00:00",
        "2019-12-27 00:00:00",
        "2019-12-30 00:00:00",
        "12-30-2019 Jenn Blossoms Telephone Call to A. Bell return her multiple voicemails."
    ],
    "Professional": ["Katie Cool", "Jenn Blossoms", "Jenn Blossoms", "Jenn Blossoms", "Jenn Blossoms", pd.NA],
    "Description": ["Travel to Space ...", "Review stuff; prepare cancellations of ...", "Review lots of stuff/o...", "Draft email to world leader...", "Review this thing.", pd.NA]
}
dftopdata = pd.DataFrame(data)

# Step 1: Identify rows where Date parsing fails
s = pd.to_datetime(dftopdata['Date'], errors='coerce').isna()

# Step 2: Extract components from malformed Date strings with precise regex
# Captures 3 groups: date, full professional name, full description
pattern = r'^(\d{2}-\d{2}-\d{4})\s+(\w+\s\w+)\s+(.*)$'
extracted = dftopdata.loc[s, 'Date'].str.extract(pattern)

# Step 3: Update original DataFrame using .loc to avoid warnings
dftopdata.loc[s, 'Date'] = extracted[0]
dftopdata.loc[s, 'Professional'] = extracted[1]
dftopdata.loc[s, 'Description'] = extracted[2]

# Verify the fixed result
print(dftopdata)

3. Key Improvements Explained #

• Precise Regex: The pattern r'^(\d{2}-\d{2}-\d{4})\s+(\w+\s\w+)\s+(.*)$' captures exactly what we need in one go:
  • Group 1: The standalone date (12-30-2019)
  • Group 2: The full professional name (with space included: Jenn Blossoms)
  • Group 3: The complete description text after the name
• Eliminated Copy Warnings: By using dftopdata.loc[s, 'Date'] directly for extraction, we ensure we're modifying the original DataFrame, not a slice copy.
• Full Description Capture: Group 3 grabs every part of the text after the professional name, so your description will show the full message instead of a single word.

4. Handling Edge Cases #

If you have professional names with more than two words (e.g., "Mary Ann Smith"), adjust the regex pattern to capture multi-word names: use (\w+(?:\s\w+)+) instead of (\w+\s\w+) to match one or more word groups separated by spaces.

内容的提问来源于stack exchange，提问作者Harrison Alley