我开始使用 Python 学习网络抓取。目前，我想下载日本饮食的视频。 (https://www.shugiintv.go.jp/jp/index.php?ex=VL&deli_id=40124&media_type=)
视频似乎有一种机制，可以从playlist.m3u8中调用chunklist.m3u8，然后按顺序调用chunklist.m3u8中描述的ts文件。
我想先从playlist.m3u8 URL下载内容，然后调用chunklist.m3u8按顺序并连接下载ts文件。
但是，我尝试下载 Playlist.m3u8，但它没有生成我期望的文本。
另外，playlist.m3u8的示例URL在这里↓
http://hlsvod.shugiintv.go.jp/vod/_definst_/amlst:2011/2011-1207-0900-12/playlist.m3u8
代码：
import requests
url = "http://hlsvod.shugiintv.go.jp/vod/_definst_/amlst:2011/2011-1207-0900-12/playlist.m3u8"
res = requests.get(url)
print(res.text)
例外文字：
#EXTM3U
#EXT-X-VERSION:3
#EXT-X-STREAM-INF:BANDWIDTH=564000,NAME="500k",RESOLUTION=640x360
chunklist_w60346572_b564000_t64NTAwaw==.m3u8
实际文字：
<html><head><title>Wowza Streaming Engine 4 Perpetual Bundle Unlimited Edition 4.7.7 build20181108145350</title></head><body>Wowza Streaming Engine 4 Perpetual Bundle Unlimited Edition 4.7.7 build20181108145350</body></html>
我认为URL中的冒号有问题，但我没有明确的解决方案。我想知道如何避免 URL 问题并成功下载 playlist.m3u8 中的文本。谢谢。
版本：
Python 3.7.9
请求2.25.1
您的网址有问题：
>>> url = "http://hlsvod.shugiintv.go.jp/vod/_definst_/amlst:2011/2011-1207-0900-12/playlist.m3u8"
>>> res = requests.get(url)
>>> res.request.url
'https://hlsvod.shugiintv.go.jp/vod/_definst_/amlst:2011/2011-1207-0900-12/playlist.m3u8%20'
看到最后的“%20”了吗？
我不太确定你是怎么弄错的，但是复制粘贴应该可以：
url = 'https://hlsvod.shugiintv.go.jp/vod/_definst_/amlst:2011/2011-1207-0900-12/playlist.m3u8'
您可以尝试下载 m3u8、任何加密密钥和 ts 文件部分。
import sys
import os
import requests
OUTPUT_FOLDER = 'output' # default output folder name
HEADERS = {
"User-Agent": "Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:57.0)"
" Gecko/20100101 Firefox/57.0",
"Accept": "*/*",
"Accept-Language": "en-US,en;q=0.5",
"Accept-Encoding": "gzip, deflate, br",
"DNT": "1",
"Connection": "keep-alive",
"Pragma": "no-cache",
"Cache-Control": "no-cache"
}
def downloadM3u8(m3u8url, headers=HEADERS, depth=0):
""" recursively download m3u8 files"""
if not os.path.isdir(OUTPUT_FOLDER):
os.mkdir(OUTPUT_FOLDER)
base_url = '/'.join(m3u8url.split('/')[0:-1]) + '/' # get the base url
print('processing: {}'.format(m3u8url))
m3u8 = requests.get(m3u8url, headers=HEADERS) # get the m3u8 file
folder = m3u8url.split('/')[-2] # get the filename
parent_folder = None
if depth > 0:
parent_folder = m3u8url.split('/')[-3]
filename = m3u8url.split('/')[-1].split('?')[0] # get the filename
path_parts = list(filter(lambda x : x is not None, [
OUTPUT_FOLDER,
parent_folder,
folder,
]))
target_path = os.path.join(*path_parts, filename)
if not os.path.isdir(os.path.join(*path_parts)):
os.mkdir(os.path.join(*path_parts))
with open(target_path, 'wb') as f:
print('writing file to {}'.format(target_path))
f.write(m3u8.content)
# Download encrypted key files
key_urls = extractKeyUrls(m3u8)
print('key_urls', key_urls)
for key_url in key_urls:
key_filename = key_url.split('/')[-1].split('?')[0]
key_file = requests.get(base_url + key_url, headers=HEADERS)
with open(os.path.join(*path_parts, key_filename), 'wb') as f:
f.write(key_file.content)
ts_urls = extractTsUrls(m3u8) # get all the .ts urls
print('ts_urls', ts_urls)
# list the .ts files if they exist in the dir
# list contents of the directory
ts_target_dir = os.path.join(*path_parts)
ts_files = set(filter(lambda x: '.ts' in x, os.listdir(ts_target_dir)))
print('all ts files existing: {}'.format(ts_files))
if len(ts_files) > 0:
ts_urls = list(filter(lambda x: x.split('?')[0] not in ts_files, ts_urls))
for ts in ts_urls:
ts_url = base_url + ts
print('downloading: {}'.format(ts_url))
ts_filename = ts.split('?')[0]
ts_file = requests.get(ts_url, headers=HEADERS)
with open(os.path.join(*path_parts, ts_filename), 'wb') as f:
f.write(ts_file.content)
child_urls = extractM3u8Urls(m3u8) # get all the urls in the m3u8 file
all_urls = []
print('child_urls', child_urls)
for child in child_urls:
new_url = base_url + child
all_urls.append(new_url)
subchildren = downloadM3u8(new_url, headers=HEADERS, depth=depth + 1)
print('subchildren', subchildren)
all_urls.extend(subchildren)
return all_urls
def extractTsUrls(m3):
""" get a list of .ts urls from the m3u8 file """
lines = m3.text.split('\n')
urls = []
for line in lines:
if '.ts' in line:
urls.append(line)
return urls
def extractM3u8Urls(m3):
""" get a list of m3u8 urls from the m3u8 file """
lines = m3.text.split('\n')
urls = []
for line in lines:
if '.m3u8' in line:
urls.append(line)
return urls
def extractKeyUrls(m3):
""" get a list of key urls from the m3u8 file """
lines = m3.text.split('\n')
urls = []
for line in lines:
match = re.search(r'URI="([^"]+)"', line)
if match:
urls.append(match.group(1))
return urls
if __name__ == "__main__":
downloadM3u8(sys.argv[0], headers=HEADERS)
print('done')