Comprehensive Guide to Python File Operations, Paths, CSV, In-Memory Streams, and Serialization

This tutorial introduces the essential techniques for working with files in Python, organized into seven sections.

1. Opening and Closing Files

The built‑in open() function opens a file and returns a file handle. Example:

<code>f1 = open(r'd:\测试文件.txt', mode='r', encoding='utf-8')
content = f1.read()
print(content)
f1.close()
</code>

Using a context manager automatically closes the file:

<code>with open(r'd:\测试文件.txt', mode='r', encoding='utf-8') as f1:
    content = f1.read()
    print(content)
</code>

Key points: encoding defaults to the OS setting (Windows GBK, Linux/macOS UTF‑8); mode defaults to 'r' . The with statement avoids manual close() and allows handling multiple files in one line.

2. Absolute and Relative Paths

Absolute path: full location, e.g., C:/Users/Python37/python.exe .

Relative path: starts from the current directory, e.g., test.txt , ./test.txt , ../test.txt , demo/test.txt .

Path syntax examples: <code># Windows style file = open('C:\\Users\\Python基础\\xxx.txt') # Raw string file = open(r'C:\Users\Python基础\xxx.txt') # Forward slashes (recommended) file = open('C:/Users/Python基础/xxx.txt') </code>

3. Common File Modes

r – read (default, file must exist).

w – write (creates or truncates).

a – append (creates if missing).

Binary modes: rb , wb , ab .

Combined read/write: r+ , w+ , a+ .

When using r+ , the file pointer starts at the beginning; writing before reading can overwrite existing data.

4. Reading and Writing Data

Reading methods:

read() – reads the whole file (or a specified number of characters/bytes).

readline() – reads one line.

readlines() – returns a list of all lines (may consume much memory).

Iterating over the file object yields lines lazily, saving memory.

<code># Example file content
lucy最帅
lucy很励志
abcdef
哈哈哈

# read whole file
f1 = open('文件操作的读', encoding='utf-8')
content = f1.read()
print(content, type(content))

# read first 6 characters
f1 = open('文件操作的读', encoding='utf-8')
print(f1.read(6))  # lucy最帅

# read line by line
f1 = open('文件操作的读', encoding='utf-8')
print(f1.readline().strip())
print(f1.readline())

# iterate
f1 = open('文件操作的读', encoding='utf-8')
for line in f1:
    print(line.strip())
</code>

Writing modes:

<code># Write text (creates or truncates)
f1 = open('文件操作的写', encoding='utf-8', mode='w')
f1.write('lucy真帅')
f1.close()

# Write binary data (e.g., copy an image)
f1 = open(r'C:\Users\lenovo\Desktop\编码进阶.png', mode='rb')
content = f1.read()
f1.close()

f2 = open('图片.jpg', mode='wb')
f2.write(content)
f2.close()
</code>

Closing a file handle and reopening it in w mode clears its contents.

5. Pointer Positioning

tell() – returns the current offset.

seek(offset, whence) – moves the pointer. whence can be 0 (start), 1 (current), or 2 (end).

<code>f = open('test.txt', 'rb')
print(f.read(3))
print(f.tell())

f.seek(2, 0)   # from start, skip 2 bytes
print(f.read())

f.seek(1, 1)   # from current, skip 1 byte
print(f.read())

f.seek(-4, 2)  # from end, go back 4 bytes
print(f.read())

f.close()
</code>

6. File Copy Example

<code>import os
file_name = input('请输入一个文件路径:')
if os.path.isfile(file_name):
    old_file = open(file_name, 'rb')
    names = os.path.splitext(file_name)
    new_file_name = names[0] + '.bak' + names[1]
    new_file = open(new_file_name, 'wb')
    while True:
        content = old_file.read(1024)
        if not content:
            break
        new_file.write(content)
    new_file.close()
    old_file.close()
else:
    print('您输入的文件不存在')
</code>

7. CSV File Read/Write

CSV stores tabular data as plain text with commas separating fields.

<code># Write CSV
import csv
file = open('test.csv', 'w', newline='')
writer = csv.writer(file)
writer.writerow(['name', 'age', 'score'])
writer.writerows([
    ['zhangsan', '18', '98'],
    ['lisi', '20', '99'],
    ['wangwu', '17', '90'],
    ['jerry', '19', '95']
])
file.close()

# Read CSV
file = open('test.csv', 'r')
reader = csv.reader(file)
for row in reader:
    print(row)
file.close()
</code>

8. In‑Memory Streams

StringIO works like a file object for text data stored in memory:

<code>from io import StringIO
f = StringIO()
f.write('hello\r\n')
f.write('good')
print(f.getvalue())
f.close()
</code>

BytesIO does the same for binary data:

<code>from io import BytesIO
f = BytesIO()
f.write('你好\r\n'.encode('utf-8'))
f.write('中国'.encode('utf-8'))
print(f.getvalue())
f.close()
</code>

9. sys Module for I/O Redirection

<code>import sys
# Redirect stdout to a file
m = open('stdout.txt', 'w', encoding='utf8')
sys.stdout = m
print('hello')
print('yes')
print('good')
m.close()

# Redirect stderr to a file
x = open('stderr.txt', 'w', encoding='utf8')
sys.stderr = x
print(1 / 0)  # raises ZeroDivisionError
x.close()
</code>

10. Serialization and Deserialization

JSON converts Python objects to a portable text format:

<code>import json
names = ['zhangsan', 'lisi', 'wangwu', 'jerry']
json_str = json.dumps(names)  # -> string
with open('names.txt', 'w') as f:
    f.write(json_str)

# Load back
with open('names.txt', 'r') as f:
    data = json.load(f)
print(data)
</code>

Pickle serializes objects to binary, which is Python‑specific:

<code>import pickle
names = ['张三', '李四', '杰克', '亨利']
binary = pickle.dumps(names)
with open('names.bin', 'wb') as f:
    f.write(binary)

# Load back
with open('names.bin', 'rb') as f:
    data = pickle.load(f)
print(data)
</code>

JSON is cross‑platform and human‑readable, but only supports basic data types; pickle preserves full Python objects but is not portable across languages.

Source: https://blog.csdn.net/mall_lucy/article/details/104547365