[http] Improve download mechanism when Range HTTP header is ignored

This commit is contained in:
Sergey M․ 2014-04-27 09:32:01 +07:00
parent 7c360e3a04
commit df297c8794

View File

@ -14,6 +14,8 @@
class HttpFD(FileDownloader): class HttpFD(FileDownloader):
_TEST_FILE_SIZE = 10241
def real_download(self, filename, info_dict): def real_download(self, filename, info_dict):
url = info_dict['url'] url = info_dict['url']
tmpfilename = self.temp_name(filename) tmpfilename = self.temp_name(filename)
@ -28,8 +30,10 @@ def real_download(self, filename, info_dict):
basic_request = compat_urllib_request.Request(url, None, headers) basic_request = compat_urllib_request.Request(url, None, headers)
request = compat_urllib_request.Request(url, None, headers) request = compat_urllib_request.Request(url, None, headers)
if self.params.get('test', False): is_test = self.params.get('test', False)
request.add_header('Range', 'bytes=0-10240')
if is_test:
request.add_header('Range', 'bytes=0-%s' % str(self._TEST_FILE_SIZE - 1))
# Establish possible resume length # Establish possible resume length
if os.path.isfile(encodeFilename(tmpfilename)): if os.path.isfile(encodeFilename(tmpfilename)):
@ -100,6 +104,15 @@ def real_download(self, filename, info_dict):
return False return False
data_len = data.info().get('Content-length', None) data_len = data.info().get('Content-length', None)
# Range HTTP header may be ignored/unsupported by a webserver
# (e.g. extractor/scivee.py, extractor/bambuser.py).
# However, for a test we still would like to download just a piece of a file.
# To achieve this we limit data_len to _TEST_FILE_SIZE and manually control
# block size when downloading a file.
if is_test and data_len > self._TEST_FILE_SIZE:
data_len = self._TEST_FILE_SIZE
if data_len is not None: if data_len is not None:
data_len = int(data_len) + resume_len data_len = int(data_len) + resume_len
min_data_len = self.params.get("min_filesize", None) min_data_len = self.params.get("min_filesize", None)
@ -118,7 +131,7 @@ def real_download(self, filename, info_dict):
while True: while True:
# Download and write # Download and write
before = time.time() before = time.time()
data_block = data.read(block_size) data_block = data.read(block_size if not is_test else min(block_size, data_len - byte_counter))
after = time.time() after = time.time()
if len(data_block) == 0: if len(data_block) == 0:
break break
@ -162,6 +175,9 @@ def real_download(self, filename, info_dict):
'speed': speed, 'speed': speed,
}) })
if is_test and byte_counter == data_len:
break
# Apply rate limit # Apply rate limit
self.slow_down(start, byte_counter - resume_len) self.slow_down(start, byte_counter - resume_len)