Hey there, let's dig into why your Brotli decompression is throwing that error. I've run into similar issues before, so here's what's going on and how to fix it:

最可能的原因：Requests自动帮你处理了压缩解压 #

The big gotcha here is that Requests library automatically handles decompression by default. When you set Accept-Encoding: gzip, deflate, br in your headers and the server returns Brotli-compressed content, Requests quietly unpacks it for you. That means response.content is already the uncompressed raw byte data—not the Brotli-compressed bytes you're expecting. Trying to run brotli.decompress() on already-unpacked data will definitely throw that BrotliDecompress failed error.

Here's a simplified, fixed version of your code that leverages Requests' built-in handling:

import requests

headers = {
    'Accept': "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
    'Accept-Encoding': "gzip, deflate, br",
    'Host': "book.douban.com",
    'Referer': "https://book.douban.com/",  # Fixed: Use full URL instead of partial domain
    'Sec-Fetch-Dest': "document",
    'Sec-Fetch-Mode': "navigate",
    'Upgrade-Insecure-Requests': "1"
}

s = requests.Session()
url = "https://book.douban.com/tag/%E5%B0%8F%E8%AF%B4"

try:
    response = s.get(url, headers=headers)
    response.raise_for_status()  #主动抛出HTTP errors like 403/500 instead of silently ignoring
except Exception as e:
    print(f"Request failed: {str(e)}")
    return ""

if response.status_code == 200:
    print(response.headers)
    # Requests already handled decompression—just use response.text directly
    return response.text
else:
    return ""

If you really need to handle Brotli manually #

If you have a specific reason to unpack the content yourself (like debugging), you need to disable Requests' automatic decompression. Do this by enabling stream=True and turning off decode_content on the raw response:

import brotli
import requests

headers = {
    'Accept': "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9",
    'Accept-Encoding': "gzip, deflate, br",
    'Host': "book.douban.com",
    'Referer': "https://book.douban.com/",
    'Sec-Fetch-Dest': "document",
    'Sec-Fetch-Mode': "navigate",
    'Upgrade-Insecure-Requests': "1"
}

s = requests.Session()
url = "https://book.douban.com/tag/%E5%B0%8F%E8%AF%B4"

try:
    # Enable stream mode to access raw compressed content
    response = s.get(url, headers=headers, stream=True)
    response.raw.decode_content = False  # Disable automatic decompression
    response.raise_for_status()
except Exception as e:
    print(f"Request failed: {str(e)}")
    return ""

if response.status_code == 200:
    print(response.headers)
    if response.headers.get('Content-Encoding') == 'br':
        # Read raw compressed bytes and decompress manually
        raw_compressed_data = response.raw.read()
        uncompressed_data = brotli.decompress(raw_compressed_data)
        return uncompressed_data.decode('utf-8')
    else:
        return response.text
else:
    return ""

Extra tips to avoid future issues #

• Upgrade your Brotli version: Brotli 1.0.9 is pretty old (released in 2019). Newer versions fix compatibility issues with modern Brotli compression formats. Run this command to upgrade:

  pip install --upgrade brotli
  

• Fix your Referer header: Your original code used book.douban.com instead of the full URL https://book.douban.com/—this might trigger anti-scraping measures from Douban.
• Avoid early returns in try blocks: Your original code had a return "" inside the try block that skipped all the status code checking logic. Use raise or print to handle errors instead of bailing early.

内容的提问来源于stack exchange，提问作者xin.chen