def merge_multithreaded_download_parts(self):
# merging parts
with open(self.filepath, 'wb') as wfd:
for f in range(ConfigHandler.get_threads(self)):
tempfilepath = ConfigHandler.get_temp_dir(
self) + "/temp" + str(f)
with open(tempfilepath, "rb") as fd:
shutil.copyfileobj(fd, wfd)
# delete copied segment
FileHandler.delete_file(self, tempfilepath)
def download(self):
ConfigHandler.parse_config(self)
if ConfigHandler.get_proxy(self):
self.http = urllib3.ProxyManager(ConfigHandler.get_proxy(self))
else:
self.http = urllib3.PoolManager()
# make sure that download path and temp directory exists
FileHandler.create_dir(self, ConfigHandler.get_download_dir(self))
FileHandler.create_dir(self, ConfigHandler.get_temp_dir(self))
# extracting filename from URL
self.filename = os.path.basename(self.url.replace("%20", "_"))
# getting complete filepath
self.filepath = ConfigHandler.get_download_dir(
self) + "/" + self.filename
#making an initial request to get header information
resp = self.make_request()
# extracting the size of file to be downloaded in bytes, from header
self.filesize = int(resp.headers['Content-Length'])
# if server supports segmented download
if self.range_download_support(resp):
# get ranges for download for each thread
ranges_list = self.get_download_ranges_list()
# perform multithreaded download on single system
self.multithreaded_download(ranges_list)
# merge multithreaded download parts
self.merge_multithreaded_download_parts()
else:
print('''Server doesn't support multithreaded downloads!
Download will be performed using single thread, on master system.''')
self.seg_handler(self.filepath, 0, self.filesize - 1)
# perform final cleaning after download completion
self.final_clean(interrupted=False)
def multithreaded_download(self, ranges_list):
# downloading each segment
for f in range(ConfigHandler.get_threads(self)):
# calling seg_handler() for each thread
t = threading.Thread(target=self.seg_handler,
kwargs={
'tempfilepath':
ConfigHandler.get_temp_dir(self) +
"/temp" + str(f),
'range_left':
ranges_list[f][0],
'range_right':
ranges_list[f][1]
})
t.setDaemon(True)
t.start()
# except main_thread, calling join() for each thread
# it ensures that merging of parts occur only after each thread has completed downloading
main_thread = threading.current_thread()
for t in threading.enumerate():
if t is main_thread:
continue
t.join()
def final_clean(self, interrupted=False):
FileHandler.delete_dir(self, ConfigHandler.get_temp_dir(self))
if interrupted == True:
''' delete the partially downloaded file if user interrupted
the download '''
FileHandler.delete_file(self, self.filepath)
