Python中unicodecsv与csv包的区别：打印带u字符原因解析

Great question! Let's break down the key differences between unicodecsv and Python's standard library csv module, especially since you're seeing that distinct u'...' notation in your output:

1. Core Purpose & Python Version Fit #

• unicodecsv was built specifically to fix a critical limitation of Python 2's standard csv module: it didn't natively support Unicode characters. In Python 3, this gap was closed entirely—the standard csv module handles Unicode seamlessly, so unicodecsv is mostly unnecessary for modern Python versions.

2. String Type Returned (The u'...' Difference) #

• When using unicodecsv in Python 2, every value read from the CSV is returned as a Unicode object (hence the u prefix when printed). This is intentional: it ensures characters outside the ASCII range are properly decoded into text, not stored as raw bytes.
• The standard csv module in Python 2, by contrast, returns byte strings (Python 2's str type, not Unicode). These don't show the u prefix because they're not Unicode objects—they're just sequences of raw bytes from the file.

3. Encoding Handling Simplification #

• unicodecsv takes the hassle out of encoding/decoding: you can pass an encoding parameter (like encoding='utf-8') when creating a reader/writer, and it automatically converts file bytes to Unicode strings (and vice versa for writing).
• In Python 2's standard csv, you have to manually manage encoding. For example, you'd need to open the file with an encoding-aware tool like codecs.open() or decode byte strings after reading, otherwise non-ASCII characters will likely throw errors or appear as garbled text.

Quick Note on Your Code Example #

Your unicodecsv output shows {u'age': u'1'} because every key and value is a Unicode object. If you ran the same logic with Python 3's standard csv module, you'd get output like {'age': '1'}—no u prefix, since Python 3's str type is equivalent to Python 2's Unicode object.

If you're stuck on Python 2 and want to hide the u prefix when printing, you could convert Unicode objects to byte strings (e.g., print({k: str(v) for k, v in i.items()})), but this isn't recommended if your CSV contains non-ASCII characters (they may break or display incorrectly).

内容的提问来源于stack exchange，提问作者Learner