2005-11-26  Marco Presi  <zufus@zufus.org>

	* DisplayStatus.py (BaseDisplayStatus.clean): 
	  added code to clean properly self._items after download queue is
	  empty.

	* Download.py (Download.d_fetchers_list): 
	  code refactoring.
	  proposal for function renaming: Download.get() -> Download.request()

	* Pool.py (RemotePool.__init__): 
	* Dists.py (RemoteDists.update): 
	* Files.py (RemoteFiles.upgrade): 
	assegned default name to download queue

2005-11-25  Marco Presi  <zufus@zufus.org>

	* Download.py 
	(Curl.__init__): fixed detection of pycurl version (it was
	detecting an old version even if I have installed 7.15.0)

2005-11-24  Otavio Salvador  <otavio@debian.org>

	* Download.py
	(Curl.__init__): detect buggy curl version;
	(Curl.set_target): avoid to resume when buggy_curl is True;
	(Curl.set_target): avoid to replace full files;

2005-11-23  Otavio Salvador  <otavio@debian.org>

	* Dists.py (MergeDists.__init__): moved code to build __mirrors
	list allowing get_mirrors to work again. This wasn't working since
	the need code was include in merge method only and thus breaking
	any other use of it.

	* Pool.py (MergePool._calc_need_files): implemented to add suport
	for merge cleanup.

	* Backend.py (Backend.clean): moved from MirrorBackend class since
	we now support cleanup of all kinda of mirrors.

2005-11-23  Marco Presi  <zufus@zufus.org>

	* Download.py (Download): fixed attribut name: self.__name

2005-11-22  Marco Presi  <zufus@zufus.org>

	* Download.py: (Download): Added a dictionary containing each downloading
	thread, indexed by name: in this way the main application, can act
	on a single thread by name, once a Donwload instance has been
	called.
	(Curl.set_target): set RESUME_FROM to 0 each time we set a
	target (if a file as been previously resumed with the same curl
	instance, this counter contains arbitrary values..)

2005-11-21  Marco Presi  <zufus@zufus.org>

	* Download.py 
	(Download:): changed namespace fetcher ---> dFetcher, to avoid
	             confusion with terminology. 
		     Removed Download.queue as a global vairable.
	(DownloadFetcher.__init__): Added self.queue = DonwloadQueue, so
	             that each Pool (or Dists), has its own private queue
		     of files to grab from remote servers.

2005-11-19  Otavio Salvador  <otavio@debian.org>

	* Pool.py (MergePool.merge): Fix to avoid to try to link already
	existent files. Also, I removed the evaluate method since there
	should only has hard links of files so it's obivious useless.

	* Download.py (Curl): Fix code to handle non-set filename attribute

2005-11-18  Marco Presi  <zufus@zufus.org>

	* Download.py 
	(Curl.set_target): comment cleaning.
	(Curl.__init__):   moved self._startSize here, with all other
	                   properties for better readability.
	(Curl.set_target): add code to append .partial suffix to files
	                   that are not completely downloaded.
	(Curl.callback):   when files are succesfully downloaded, close
	                   them and remove .partial suffix
	(Curl.close):      file closing is done on the callback

