08ae38b6ce 2010-06-25 c.kworr@b84a3: #!/usr/bin/env python3.1
08ae38b6ce 2010-06-25 c.kworr@b84a3:
08ae38b6ce 2010-06-25 c.kworr@b84a3: import datetime, http.cookiejar, optparse, os, sys, shelve, re, urllib.request
08ae38b6ce 2010-06-25 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: from spacemap import SpaceMap
80f8e3804a 2010-08-20 c.kworr@b84a3:
08ae38b6ce 2010-06-25 c.kworr@b84a3: parser = optparse.OptionParser()
08ae38b6ce 2010-06-25 c.kworr@b84a3: parser.add_option('-v', '--verbose', action = 'store_true', dest = 'verbose', help = 'turns on verbose status notifications', metavar = 'bool', default = False)
08ae38b6ce 2010-06-25 c.kworr@b84a3: parser.add_option('-d', '--dir', action = 'store', dest = 'dir', help = 'specify directory where the files should be stored', metavar = 'string', default = None)
08ae38b6ce 2010-06-25 c.kworr@b84a3: parser.add_option('-r', '--root', action = 'store', dest = 'root', help = 'specify a site from which data should be mirrored', metavar = 'string', default = None)
08ae38b6ce 2010-06-25 c.kworr@b84a3: parser.add_option('-l', '--log', action = 'store', dest = 'log', help = 'specify a log file to process', metavar = 'string', default = None)
38b25713eb 2010-07-26 c.kworr@b84a3: parser.add_option('-e', '--skip-etag', action = 'store_true', dest = 'noetag', help = 'do not process etags', metavar = 'bool', default = False)
80f8e3804a 2010-08-20 c.kworr@b84a3: parser.add_option('-p', '--port', action = 'store', dest = 'port', help = 'listen on this port for incoming connections', metavar = 'integer', default = None)
80f8e3804a 2010-08-20 c.kworr@b84a3: parser.add_option('-n', '--no-update', action = 'store_true', dest = 'noupdate', help = 'do not update already downloaded files', metavar = 'bool', default = 'False')
08ae38b6ce 2010-06-25 c.kworr@b84a3: (options, args) = parser.parse_args()
08ae38b6ce 2010-06-25 c.kworr@b84a3:
38b25713eb 2010-07-26 c.kworr@b84a3: assert options.dir, 'Directory not specified'
38b25713eb 2010-07-26 c.kworr@b84a3: assert options.root, 'Server not specified'
80f8e3804a 2010-08-20 c.kworr@b84a3: assert options.log or options.port, 'Log file or port not specified'
80f8e3804a 2010-08-20 c.kworr@b84a3: assert options.port or os.access(options.log, os.R_OK), 'Log file unreadable'
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: optionsDirWithSep = re.compile('^(.*?)/?$').match(options.dir)
80f8e3804a 2010-08-20 c.kworr@b84a3: if optionsDirWithSep:
80f8e3804a 2010-08-20 c.kworr@b84a3: options.dir = optionsDirWithSep.group(1)
08ae38b6ce 2010-06-25 c.kworr@b84a3:
08ae38b6ce 2010-06-25 c.kworr@b84a3: # this is file index - everything is stored in this file
80f8e3804a 2010-08-20 c.kworr@b84a3: # _parts - list of stored parts of file
80f8e3804a 2010-08-20 c.kworr@b84a3: # _time - last time the file was checked
80f8e3804a 2010-08-20 c.kworr@b84a3: # everything else is just the headers
80f8e3804a 2010-08-20 c.kworr@b84a3: index = shelve.open(options.dir + os.sep + '.index')
38b25713eb 2010-07-26 c.kworr@b84a3: desc_fields = ('Content-Length', 'Pragma', 'Last-Modified')
38b25713eb 2010-07-26 c.kworr@b84a3: ignore_fields = ('Accept-Ranges', 'Age', 'Cache-Control', 'Connection', 'Content-Type', 'Date', 'Expires', 'Server', 'Via', 'X-Cache', 'X-Cache-Lookup', 'X-Powered-By')
38b25713eb 2010-07-26 c.kworr@b84a3:
38b25713eb 2010-07-26 c.kworr@b84a3: if not options.noetag:
38b25713eb 2010-07-26 c.kworr@b84a3: desc_fields += 'ETag',
38b25713eb 2010-07-26 c.kworr@b84a3: else:
38b25713eb 2010-07-26 c.kworr@b84a3: ignore_fields += 'ETag',
38b25713eb 2010-07-26 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: block_size = 4096
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file_name = options.dir + os.sep + '.tmp'
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: '''
80f8e3804a 2010-08-20 c.kworr@b84a3: # later, kqueue would be good but later
80f8e3804a 2010-08-20 c.kworr@b84a3: class Connection:
80f8e3804a 2010-08-20 c.kworr@b84a3: __slots__ = frozenset(('__address', '__input', '__socket', '__status', 'error', 'method', 'url', 'http_version'))
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: def __init__(self, socket, address):
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__address = address
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__input = b''
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__socket = socket
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__status = 0
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: def read(self, kev):
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = self.__socket.recv(kev.data)
80f8e3804a 2010-08-20 c.kworr@b84a3: exhausted = False
80f8e3804a 2010-08-20 c.kworr@b84a3: if len(buffer) == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: eof = True
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__input += buffer
80f8e3804a 2010-08-20 c.kworr@b84a3: while not exhausted:
80f8e3804a 2010-08-20 c.kworr@b84a3: if self.__status == -1:
80f8e3804a 2010-08-20 c.kworr@b84a3: exhausted = True
80f8e3804a 2010-08-20 c.kworr@b84a3: elif self.__status == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: endstring = self.__input.find(b'\n')
80f8e3804a 2010-08-20 c.kworr@b84a3: if endstring > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Processing request line.')
80f8e3804a 2010-08-20 c.kworr@b84a3: line = self.__input[:endstring].decode('ascii')
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__input = self.__input[endstring + 1:]
80f8e3804a 2010-08-20 c.kworr@b84a3: isRequest = re.compile('(GET) ([^ ]+) HTTP/(1\.0)').match(line)
80f8e3804a 2010-08-20 c.kworr@b84a3: if not isRequest:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.error = 'Not a HTTP connection.'
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__status = -1
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.method = isRequest.group(1)
80f8e3804a 2010-08-20 c.kworr@b84a3: self.url = isRequest.group(2)
80f8e3804a 2010-08-20 c.kworr@b84a3: self.http_version = isRequest.group(3)
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__status = 1
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: exhausted = True
80f8e3804a 2010-08-20 c.kworr@b84a3: elif self.__status == 1:
80f8e3804a 2010-08-20 c.kworr@b84a3: endstring = self.__input.find(b'\n')
80f8e3804a 2010-08-20 c.kworr@b84a3: if endstring > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Processing header line.' + repr(self.__input))
80f8e3804a 2010-08-20 c.kworr@b84a3: line = self.__input[:endstring].decode('ascii')
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__input = self.__input[endstring + 1:]
80f8e3804a 2010-08-20 c.kworr@b84a3: isHeader = re.compile('([^:]*): +(.*)').match(line)
80f8e3804a 2010-08-20 c.kworr@b84a3: if not isHeader:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.error = 'Bad header.'
80f8e3804a 2010-08-20 c.kworr@b84a3: return(False)
80f8e3804a 2010-08-20 c.kworr@b84a3: # process header here
80f8e3804a 2010-08-20 c.kworr@b84a3: elif endstring == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.__status = 2
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: exhausted = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: def write(self, kev):
80f8e3804a 2010-08-20 c.kworr@b84a3: pass
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if options.port:
80f8e3804a 2010-08-20 c.kworr@b84a3: import select, socket
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
80f8e3804a 2010-08-20 c.kworr@b84a3: try:
80f8e3804a 2010-08-20 c.kworr@b84a3: sock.bind(('127.0.0.1', int(options.port)))
80f8e3804a 2010-08-20 c.kworr@b84a3: sock.listen(-1)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: kq = select.kqueue()
80f8e3804a 2010-08-20 c.kworr@b84a3: assert kq.fileno() != -1, "Fatal error: can't initialise kqueue."
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: kq.control([select.kevent(sock, select.KQ_FILTER_READ, select.KQ_EV_ADD)], 0)
80f8e3804a 2010-08-20 c.kworr@b84a3: timeout = None
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: connections = {sock.fileno(): None}
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: while True:
80f8e3804a 2010-08-20 c.kworr@b84a3: kevs = kq.control(None, 1, timeout)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: for kev in kevs:
80f8e3804a 2010-08-20 c.kworr@b84a3: if type(connections[kev.ident]) == Connection:
80f8e3804a 2010-08-20 c.kworr@b84a3: print(kev.ident, kev.data, kev.filter, kev.flags)
80f8e3804a 2010-08-20 c.kworr@b84a3: assert kev.data != 0, 'No data available.'
80f8e3804a 2010-08-20 c.kworr@b84a3: if kev.filter == select.KQ_FILTER_READ:
80f8e3804a 2010-08-20 c.kworr@b84a3: connections[kev.ident].read(kev)
80f8e3804a 2010-08-20 c.kworr@b84a3: elif kev.filter == select.KQ_FILTER_WRITE:
80f8e3804a 2010-08-20 c.kworr@b84a3: connections[kev.ident].write(kev)
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: assert kev.filter in (select.KQ_FILTER_READ, select.KQ_FILTER_WRITE), 'Do we support other filters?'
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: (conn, addr) = sock.accept()
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Connection from ' + repr(addr))
80f8e3804a 2010-08-20 c.kworr@b84a3: kq.control([select.kevent(conn, select.KQ_FILTER_READ, select.KQ_EV_ADD)], 0)
80f8e3804a 2010-08-20 c.kworr@b84a3: connections[conn.fileno()] = Connection(conn, addr)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if kev.flags >> 15 == 1:
80f8e3804a 2010-08-20 c.kworr@b84a3: kq.control([select.kevent(kev.ident, select.KQ_FILTER_READ, select.KQ_EV_DELETE)], 0)
80f8e3804a 2010-08-20 c.kworr@b84a3: kq.control([select.kevent(kev.ident, select.KQ_FILTER_WRITE, select.KQ_EV_DELETE)], 0)
80f8e3804a 2010-08-20 c.kworr@b84a3: del(connections[kev.ident])
80f8e3804a 2010-08-20 c.kworr@b84a3: finally:
80f8e3804a 2010-08-20 c.kworr@b84a3: sock.close()
80f8e3804a 2010-08-20 c.kworr@b84a3: '''
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if options.port:
80f8e3804a 2010-08-20 c.kworr@b84a3: import http.server
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: class MyRequestHandler(http.server.BaseHTTPRequestHandler):
80f8e3804a 2010-08-20 c.kworr@b84a3: def __process(self):
80f8e3804a 2010-08-20 c.kworr@b84a3: # reload means file needs to be reloaded to serve request
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = False
80f8e3804a 2010-08-20 c.kworr@b84a3: # recheck means file needs to be checked, this also means that if file hav been modified we can serve older copy
80f8e3804a 2010-08-20 c.kworr@b84a3: recheck = False
80f8e3804a 2010-08-20 c.kworr@b84a3: # file_stat means file definitely exists
80f8e3804a 2010-08-20 c.kworr@b84a3: file_stat = None
80f8e3804a 2010-08-20 c.kworr@b84a3: # requested_ranges holds data about any range requested
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges = None
80f8e3804a 2010-08-20 c.kworr@b84a3: # records holds data from index locally, should be written back upon successfull completion
80f8e3804a 2010-08-20 c.kworr@b84a3: record = None
80f8e3804a 2010-08-20 c.kworr@b84a3: info = 'Checking file: ' + self.path
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: proxy_ignored = ('Accept', 'Accept-Encoding',
80f8e3804a 2010-08-20 c.kworr@b84a3: 'Cache-Control', 'Connection',
80f8e3804a 2010-08-20 c.kworr@b84a3: 'Host',
80f8e3804a 2010-08-20 c.kworr@b84a3: 'User-Agent',
80f8e3804a 2010-08-20 c.kworr@b84a3: 'Via',
80f8e3804a 2010-08-20 c.kworr@b84a3: 'X-Forwarded-For',
80f8e3804a 2010-08-20 c.kworr@b84a3: )
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Command:', self.command)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: for header in self.headers:
80f8e3804a 2010-08-20 c.kworr@b84a3: if header in proxy_ignored:
80f8e3804a 2010-08-20 c.kworr@b84a3: pass
80f8e3804a 2010-08-20 c.kworr@b84a3: elif header in ('Range'):
80f8e3804a 2010-08-20 c.kworr@b84a3: isRange = re.compile('bytes=(\d+)-(\d+)').match(self.headers[header])
80f8e3804a 2010-08-20 c.kworr@b84a3: if isRange:
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges = SpaceMap({int(isRange.group(1)): int(isRange.group(2)) + 1})
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: return()
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Unknown header - ', header, ': ', self.headers[header], sep='')
80f8e3804a 2010-08-20 c.kworr@b84a3: return()
80f8e3804a 2010-08-20 c.kworr@b84a3: print(header, self.headers[header])
80f8e3804a 2010-08-20 c.kworr@b84a3: print(self.path)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # creating empty placeholder in index
80f8e3804a 2010-08-20 c.kworr@b84a3: # if there's no space map and there's no file in real directory - we have no file
80f8e3804a 2010-08-20 c.kworr@b84a3: # if there's an empty space map - file is full
80f8e3804a 2010-08-20 c.kworr@b84a3: # space map generally covers every bit of file we don't posess currently
80f8e3804a 2010-08-20 c.kworr@b84a3: if not self.path in index:
80f8e3804a 2010-08-20 c.kworr@b84a3: info += '\nThis one is new.'
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3: record = {'_parts': None}
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: record = index[self.path]
80f8e3804a 2010-08-20 c.kworr@b84a3: if '_parts' in index[self.path]:
80f8e3804a 2010-08-20 c.kworr@b84a3: print(record['_parts'])
80f8e3804a 2010-08-20 c.kworr@b84a3: if index[self.path]['_parts'] == {0: -1}:
80f8e3804a 2010-08-20 c.kworr@b84a3: index[self.path]['_parts'] = None
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # creating file name from self.path
80f8e3804a 2010-08-20 c.kworr@b84a3: file_name = options.dir + os.sep + re.compile('%20').sub(' ', self.path)
80f8e3804a 2010-08-20 c.kworr@b84a3: # partial file or unfinished download
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_name = options.dir + os.sep + '.parts' + re.compile('%20').sub(' ', self.path)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # forcibly checking file if no file present
80f8e3804a 2010-08-20 c.kworr@b84a3: if os.access(file_name, os.R_OK):
80f8e3804a 2010-08-20 c.kworr@b84a3: file_stat = os.stat(file_name)
80f8e3804a 2010-08-20 c.kworr@b84a3: elif '_parts' in record and os.access(temp_name, os.R_OK):
80f8e3804a 2010-08-20 c.kworr@b84a3: file_stat = os.stat(temp_name)
80f8e3804a 2010-08-20 c.kworr@b84a3: elif not reload:
80f8e3804a 2010-08-20 c.kworr@b84a3: info += '\nFile not found or inaccessible.'
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # forcibly checking file if file size doesn't match with index data
80f8e3804a 2010-08-20 c.kworr@b84a3: if not reload:
80f8e3804a 2010-08-20 c.kworr@b84a3: if '_parts' in record and record['_parts'] == SpaceMap():
80f8e3804a 2010-08-20 c.kworr@b84a3: if 'Content-Length' in record and file_stat and file_stat.st_size != int(record['Content-Length']):
80f8e3804a 2010-08-20 c.kworr@b84a3: info += '\nFile size is {} and stored file size is {}.'.format(file_stat.st_size, record['Content-Length'])
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # forcibly checking file if index holds Pragma header
80f8e3804a 2010-08-20 c.kworr@b84a3: if not reload and 'Pragma' in record and record['Pragma'] == 'no-cache':
80f8e3804a 2010-08-20 c.kworr@b84a3: info +='\nPragma on: recheck imminent.'
80f8e3804a 2010-08-20 c.kworr@b84a3: recheck = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # skipping file processing if there's no need to recheck it and we have checked it at least 4 hours ago
80f8e3804a 2010-08-20 c.kworr@b84a3: if not recheck and not reload and '_time' in record and (datetime.datetime.now() - datetime.timedelta(hours = 4) - record['_time']).days < 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: recheck = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: print(info)
80f8e3804a 2010-08-20 c.kworr@b84a3: if reload or recheck:
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: try:
80f8e3804a 2010-08-20 c.kworr@b84a3: request = options.root + self.path
80f8e3804a 2010-08-20 c.kworr@b84a3: if requested_ranges != None:
80f8e3804a 2010-08-20 c.kworr@b84a3: if '_parts' in record and record['_parts'] != None:
80f8e3804a 2010-08-20 c.kworr@b84a3: needed = record['_parts'] & requested_ranges
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: needed = requested_ranges
80f8e3804a 2010-08-20 c.kworr@b84a3: ranges = ()
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Requesting ranges:', ranges)
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Not stored ranges:', record['_parts'])
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Requested ranges:', requested_ranges)
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Needed ranges:', needed)
80f8e3804a 2010-08-20 c.kworr@b84a3: needed.rewind()
80f8e3804a 2010-08-20 c.kworr@b84a3: while True:
80f8e3804a 2010-08-20 c.kworr@b84a3: range = needed.pop()
80f8e3804a 2010-08-20 c.kworr@b84a3: if range[0] == None:
80f8e3804a 2010-08-20 c.kworr@b84a3: break
80f8e3804a 2010-08-20 c.kworr@b84a3: ranges += '{}-{}'.format(range[0], range[1] - 1),
80f8e3804a 2010-08-20 c.kworr@b84a3: request = urllib.request.Request(request, headers = {'Range': 'bytes=' + ','.join(ranges)})
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: with urllib.request.urlopen(request) as source:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record = {}
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record['_parts'] = record['_parts']
80f8e3804a 2010-08-20 c.kworr@b84a3: headers = source.info()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # stripping unneeded headers (XXX make this inplace?)
80f8e3804a 2010-08-20 c.kworr@b84a3: for header in headers:
80f8e3804a 2010-08-20 c.kworr@b84a3: if header in desc_fields:
80f8e3804a 2010-08-20 c.kworr@b84a3: #if header == 'Pragma' and headers[header] != 'no-cache':
80f8e3804a 2010-08-20 c.kworr@b84a3: print(header, headers[header])
80f8e3804a 2010-08-20 c.kworr@b84a3: if header == 'Content-Length':
80f8e3804a 2010-08-20 c.kworr@b84a3: if 'Content-Range' not in headers:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record[header] = headers[header]
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record[header] = headers[header]
80f8e3804a 2010-08-20 c.kworr@b84a3: elif header == 'Content-Range':
80f8e3804a 2010-08-20 c.kworr@b84a3: range = re.compile('^bytes (\d+)-(\d+)/(\d+)$').match(headers[header])
80f8e3804a 2010-08-20 c.kworr@b84a3: if range:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record['Content-Length'] = range.group(3)
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: assert False, 'Content-Range unrecognized.'
80f8e3804a 2010-08-20 c.kworr@b84a3: elif not header in ignore_fields:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Undefined header "', header, '": ', headers[header], sep='')
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if new_record['_parts'] == None:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record['_parts'] = SpaceMap({0: int(new_record['Content-Length'])})
80f8e3804a 2010-08-20 c.kworr@b84a3: print(new_record)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # comparing headers with data found in index
80f8e3804a 2010-08-20 c.kworr@b84a3: # if any header has changed (except Pragma) file is fully downloaded
80f8e3804a 2010-08-20 c.kworr@b84a3: # same if we get more or less headers
80f8e3804a 2010-08-20 c.kworr@b84a3: old_keys = set(record.keys())
80f8e3804a 2010-08-20 c.kworr@b84a3: old_keys.discard('_time')
80f8e3804a 2010-08-20 c.kworr@b84a3: old_keys.discard('Pragma')
80f8e3804a 2010-08-20 c.kworr@b84a3: more_keys = set(new_record.keys()) - old_keys
80f8e3804a 2010-08-20 c.kworr@b84a3: more_keys.discard('Pragma')
80f8e3804a 2010-08-20 c.kworr@b84a3: less_keys = old_keys - set(new_record.keys())
80f8e3804a 2010-08-20 c.kworr@b84a3: if len(more_keys) > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: if not len(old_keys) == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('More headers appear:', more_keys)
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3: elif len(less_keys) > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Less headers appear:', less_keys)
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: for key in record.keys():
80f8e3804a 2010-08-20 c.kworr@b84a3: if key[0] != '_' and key != 'Pragma' and not record[key] == new_record[key]:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Header "', key, '" changed from [', record[key], '] to [', new_record[key], ']', sep='')
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if reload:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Reloading.')
80f8e3804a 2010-08-20 c.kworr@b84a3: if os.access(temp_name, os.R_OK):
80f8e3804a 2010-08-20 c.kworr@b84a3: os.unlink(temp_name)
80f8e3804a 2010-08-20 c.kworr@b84a3: if os.access(file_name, os.R_OK):
80f8e3804a 2010-08-20 c.kworr@b84a3: os.unlink(file_name)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # downloading file or segment
80f8e3804a 2010-08-20 c.kworr@b84a3: if 'Content-Length' in new_record:
80f8e3804a 2010-08-20 c.kworr@b84a3: if requested_ranges == None:
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges = new_record['_parts']
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: if len(requested_ranges) > 1:
80f8e3804a 2010-08-20 c.kworr@b84a3: print("Multipart requests currently not supported.")
80f8e3804a 2010-08-20 c.kworr@b84a3: assert False, 'Skip this one for now.'
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: assert False, 'No Content-Length or Content-Range header.'
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if reload:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record['_time'] = datetime.datetime.now()
80f8e3804a 2010-08-20 c.kworr@b84a3: if self.command not in ('HEAD'):
80f8e3804a 2010-08-20 c.kworr@b84a3: # file is created at temporary location and moved in place only when download completes
80f8e3804a 2010-08-20 c.kworr@b84a3: if not os.access(temp_name, os.R_OK):
80f8e3804a 2010-08-20 c.kworr@b84a3: empty_name = options.dir + os.sep + '.tmp'
80f8e3804a 2010-08-20 c.kworr@b84a3: with open(empty_name, 'w+b') as some_file:
80f8e3804a 2010-08-20 c.kworr@b84a3: pass
80f8e3804a 2010-08-20 c.kworr@b84a3: os.renames(empty_name, temp_name)
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file = open(temp_name, 'r+b')
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges.rewind()
80f8e3804a 2010-08-20 c.kworr@b84a3: while True:
80f8e3804a 2010-08-20 c.kworr@b84a3: (start, end) = requested_ranges.pop()
80f8e3804a 2010-08-20 c.kworr@b84a3: if start == None:
80f8e3804a 2010-08-20 c.kworr@b84a3: break
80f8e3804a 2010-08-20 c.kworr@b84a3: stream_last = start
80f8e3804a 2010-08-20 c.kworr@b84a3: old_record = new_record
80f8e3804a 2010-08-20 c.kworr@b84a3: if end - start < block_size:
80f8e3804a 2010-08-20 c.kworr@b84a3: req_block_size = end - start
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: req_block_size = block_size
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = source.read(req_block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: print(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: length = len(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: while length > 0 and stream_last < end:
80f8e3804a 2010-08-20 c.kworr@b84a3: stream_pos = stream_last + length
80f8e3804a 2010-08-20 c.kworr@b84a3: assert not stream_pos > end, 'Received more data then requested: pos:{} start:{} end:{}.'.format(stream_pos, start, end)
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Writing', length, 'bytes to temp file at position', stream_last)
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file.seek(stream_last)
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file.write(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: new_record['_parts'] = new_record['_parts'] - SpaceMap({stream_last: stream_pos})
80f8e3804a 2010-08-20 c.kworr@b84a3: print(new_record)
80f8e3804a 2010-08-20 c.kworr@b84a3: index[self.path] = old_record
80f8e3804a 2010-08-20 c.kworr@b84a3: index.sync()
80f8e3804a 2010-08-20 c.kworr@b84a3: old_record = new_record
80f8e3804a 2010-08-20 c.kworr@b84a3: stream_last = stream_pos
80f8e3804a 2010-08-20 c.kworr@b84a3: if end - stream_last < block_size:
80f8e3804a 2010-08-20 c.kworr@b84a3: req_block_size = end - stream_last
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = source.read(req_block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: print(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: length = len(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: print(new_record)
80f8e3804a 2010-08-20 c.kworr@b84a3: index[self.path] = new_record
80f8e3804a 2010-08-20 c.kworr@b84a3: index.sync()
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file.close()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # moving downloaded data to real file
80f8e3804a 2010-08-20 c.kworr@b84a3: if new_record['_parts'] == SpaceMap():
80f8e3804a 2010-08-20 c.kworr@b84a3: if type(request) != str:
80f8e3804a 2010-08-20 c.kworr@b84a3: # just moving
80f8e3804a 2010-08-20 c.kworr@b84a3: # drop old dirs XXX
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Moving temporary file to new destination.')
80f8e3804a 2010-08-20 c.kworr@b84a3: os.renames(temp_name, file_name)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: except urllib.error.HTTPError as error:
80f8e3804a 2010-08-20 c.kworr@b84a3: # in case of error we don't need to do anything actually,
80f8e3804a 2010-08-20 c.kworr@b84a3: # if file download stalls or fails the file would not be moved to it's location
80f8e3804a 2010-08-20 c.kworr@b84a3: print(error)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if self.command == 'HEAD':
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_response(200)
80f8e3804a 2010-08-20 c.kworr@b84a3: if 'Content-Length' in index[self.path]:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Content-Length', index[self.path]['Content-Length'])
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Accept-Ranges', 'bytes')
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Content-Type', 'application/octet-stream')
80f8e3804a 2010-08-20 c.kworr@b84a3: if 'Last-Modified' in index[self.path]:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Last-Modified', index[self.path]['Last-Modified'])
80f8e3804a 2010-08-20 c.kworr@b84a3: self.end_headers()
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: if index[self.path]['_parts'] != SpaceMap():
80f8e3804a 2010-08-20 c.kworr@b84a3: file_name = temp_name
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: with open(file_name, 'rb') as real_file:
80f8e3804a 2010-08-20 c.kworr@b84a3: file_stat = os.stat(file_name)
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_response(200)
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Last-Modified', index[self.path]['Last-Modified'])
80f8e3804a 2010-08-20 c.kworr@b84a3: if requested_ranges != None:
80f8e3804a 2010-08-20 c.kworr@b84a3: ranges = ()
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges.rewind()
80f8e3804a 2010-08-20 c.kworr@b84a3: while True:
80f8e3804a 2010-08-20 c.kworr@b84a3: pair = requested_ranges.pop()
80f8e3804a 2010-08-20 c.kworr@b84a3: if pair[0] == None:
80f8e3804a 2010-08-20 c.kworr@b84a3: break
80f8e3804a 2010-08-20 c.kworr@b84a3: ranges += '{}-{}'.format(pair[0], str(pair[1] - 1)),
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Content-Range', 'bytes ' + ','.join(ranges) + '/' + index[self.path]['Content-Length'])
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Content-Length', str(file_stat.st_size))
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges = SpaceMap({0: file_stat.st_size})
80f8e3804a 2010-08-20 c.kworr@b84a3: self.send_header('Content-Type', 'application/octet-stream')
80f8e3804a 2010-08-20 c.kworr@b84a3: self.end_headers()
80f8e3804a 2010-08-20 c.kworr@b84a3: if self.command in ('GET'):
80f8e3804a 2010-08-20 c.kworr@b84a3: requested_ranges.rewind()
80f8e3804a 2010-08-20 c.kworr@b84a3: (start, end) = requested_ranges.pop()
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Seeking file to position', start)
80f8e3804a 2010-08-20 c.kworr@b84a3: real_file.seek(start)
80f8e3804a 2010-08-20 c.kworr@b84a3: if block_size > end - start:
80f8e3804a 2010-08-20 c.kworr@b84a3: req_block_size = end - start
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: req_block_size = block_size
80f8e3804a 2010-08-20 c.kworr@b84a3: print('block_size is', req_block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = real_file.read(req_block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: length = len(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: while length > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: self.wfile.write(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: start += len(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: if req_block_size > end - start:
80f8e3804a 2010-08-20 c.kworr@b84a3: req_block_size = end - start
80f8e3804a 2010-08-20 c.kworr@b84a3: if req_block_size == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: break
80f8e3804a 2010-08-20 c.kworr@b84a3: print('block_size is', req_block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = real_file.read(req_block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: length = len(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: def do_HEAD(self):
80f8e3804a 2010-08-20 c.kworr@b84a3: return self.__process()
80f8e3804a 2010-08-20 c.kworr@b84a3: def do_GET(self):
80f8e3804a 2010-08-20 c.kworr@b84a3: return self.__process()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: server = http.server.HTTPServer(('127.0.0.1', int(options.port)), MyRequestHandler)
80f8e3804a 2010-08-20 c.kworr@b84a3: server.serve_forever()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: while True:
80f8e3804a 2010-08-20 c.kworr@b84a3: unchecked_files = set()
80f8e3804a 2010-08-20 c.kworr@b84a3: checked_files = 0
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # reading log and storing found urls for processing
80f8e3804a 2010-08-20 c.kworr@b84a3: # check file mtime XXX
80f8e3804a 2010-08-20 c.kworr@b84a3: with open(options.log, 'r') as log_file:
80f8e3804a 2010-08-20 c.kworr@b84a3: log_line = re.compile('^[^ ]+ - - \[.*] "(GET|HEAD) (.*?)(\?.*)? HTTP/1.1" (\d+) \d+ "(.*)" "(.*)"$')
80f8e3804a 2010-08-20 c.kworr@b84a3: for line in log_file:
80f8e3804a 2010-08-20 c.kworr@b84a3: this_line = log_line.match(line.strip())
80f8e3804a 2010-08-20 c.kworr@b84a3: if this_line:
80f8e3804a 2010-08-20 c.kworr@b84a3: unchecked_files.add(this_line.group(2))
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: for url in unchecked_files:
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = False
80f8e3804a 2010-08-20 c.kworr@b84a3: recheck = False
80f8e3804a 2010-08-20 c.kworr@b84a3: info = 'Checking file: ' + url
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # creating empty placeholder in index
80f8e3804a 2010-08-20 c.kworr@b84a3: if not url in index:
80f8e3804a 2010-08-20 c.kworr@b84a3: info += '\nThis one is new.'
80f8e3804a 2010-08-20 c.kworr@b84a3: index[url] = {}
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # creating file name from url
80f8e3804a 2010-08-20 c.kworr@b84a3: file_name = options.dir + re.compile('%20').sub(' ', url)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # forcibly checking file if no file present
80f8e3804a 2010-08-20 c.kworr@b84a3: if not reload and not os.access(file_name, os.R_OK):
80f8e3804a 2010-08-20 c.kworr@b84a3: info += '\nFile not found or inaccessible.'
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # forcibly checking file if file size doesn't match with index data
80f8e3804a 2010-08-20 c.kworr@b84a3: elif not reload and 'Content-Length' in index[url] and os.stat(file_name).st_size != int(index[url]['Content-Length']):
80f8e3804a 2010-08-20 c.kworr@b84a3: info += '\nFile size is ' + os.stat(file_name).st_size + ' and stored file size is ' + index[url]['Content-Length'] + '.'
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # forcibly checking file if index hods Pragma header
80f8e3804a 2010-08-20 c.kworr@b84a3: if not reload and 'Pragma' in index[url] and index[url]['Pragma'] == 'no-cache':
80f8e3804a 2010-08-20 c.kworr@b84a3: info +='\nPragma on: recheck imminent.'
80f8e3804a 2010-08-20 c.kworr@b84a3: recheck = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # skipping file processing if there's no need to recheck it and we have checked it at least 4 hours ago
80f8e3804a 2010-08-20 c.kworr@b84a3: if not recheck and not reload and (options.noupdate or ('_time' in index[url] and (datetime.datetime.now() - datetime.timedelta(hours = 4) - index[url]['_time']).days < 0)):
80f8e3804a 2010-08-20 c.kworr@b84a3: if options.verbose:
80f8e3804a 2010-08-20 c.kworr@b84a3: print(info)
80f8e3804a 2010-08-20 c.kworr@b84a3: continue
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: print(info)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: try:
80f8e3804a 2010-08-20 c.kworr@b84a3: with urllib.request.urlopen(options.root + url) as source:
80f8e3804a 2010-08-20 c.kworr@b84a3: new_headers = {}
80f8e3804a 2010-08-20 c.kworr@b84a3: headers = source.info()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # stripping unneeded headers (XXX make this inplace?)
80f8e3804a 2010-08-20 c.kworr@b84a3: for header in headers:
80f8e3804a 2010-08-20 c.kworr@b84a3: if header in desc_fields:
80f8e3804a 2010-08-20 c.kworr@b84a3: if header == 'Pragma' and headers[header] != 'no-cache':
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Pragma:', headers[header])
80f8e3804a 2010-08-20 c.kworr@b84a3: new_headers[header] = headers[header]
80f8e3804a 2010-08-20 c.kworr@b84a3: elif not header in ignore_fields:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Undefined header "', header, '": ', headers[header], sep='')
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # comparing headers with data found in index
80f8e3804a 2010-08-20 c.kworr@b84a3: # if any header has changed (except Pragma) file is fully downloaded
80f8e3804a 2010-08-20 c.kworr@b84a3: # same if we get more or less headers
80f8e3804a 2010-08-20 c.kworr@b84a3: old_keys = set(index[url].keys())
80f8e3804a 2010-08-20 c.kworr@b84a3: old_keys.discard('_time')
80f8e3804a 2010-08-20 c.kworr@b84a3: old_keys.discard('Pragma')
80f8e3804a 2010-08-20 c.kworr@b84a3: more_keys = set(new_headers.keys()) - old_keys
80f8e3804a 2010-08-20 c.kworr@b84a3: more_keys.discard('Pragma')
80f8e3804a 2010-08-20 c.kworr@b84a3: less_keys = old_keys - set(new_headers.keys())
80f8e3804a 2010-08-20 c.kworr@b84a3: if len(more_keys) > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: if not len(old_keys) == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('More headers appear:', more_keys)
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3: elif len(less_keys) > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Less headers appear:', less_keys)
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: for key in index[url].keys():
80f8e3804a 2010-08-20 c.kworr@b84a3: if key[0] != '_' and key != 'Pragma' and not index[url][key] == new_headers[key]:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Header "', key, '" changed from [', index[url][key], '] to [', new_headers[key], ']', sep='')
80f8e3804a 2010-08-20 c.kworr@b84a3: reload = True
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # downloading file
80f8e3804a 2010-08-20 c.kworr@b84a3: if reload:
80f8e3804a 2010-08-20 c.kworr@b84a3: if 'Content-Length' in headers:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Downloading', headers['Content-Length'], 'bytes [', end='')
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('Downloading [', end='')
80f8e3804a 2010-08-20 c.kworr@b84a3: sys.stdout.flush()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # file is created at temporary location and moved in place only when download completes
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file = open(options.dir + os.sep + '.tmp', 'wb')
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = source.read(block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: megablocks = 0
80f8e3804a 2010-08-20 c.kworr@b84a3: blocks = 0
80f8e3804a 2010-08-20 c.kworr@b84a3: megs = 0
80f8e3804a 2010-08-20 c.kworr@b84a3: while len(buffer) > 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file.write(buffer)
80f8e3804a 2010-08-20 c.kworr@b84a3: buffer = source.read(block_size)
80f8e3804a 2010-08-20 c.kworr@b84a3: blocks += 1
80f8e3804a 2010-08-20 c.kworr@b84a3: if blocks > 102400/block_size:
80f8e3804a 2010-08-20 c.kworr@b84a3: megablocks += 1
80f8e3804a 2010-08-20 c.kworr@b84a3: if megablocks > 10:
80f8e3804a 2010-08-20 c.kworr@b84a3: megablocks = megablocks - 10
80f8e3804a 2010-08-20 c.kworr@b84a3: megs += 1
80f8e3804a 2010-08-20 c.kworr@b84a3: print('{}Mb'.format(megs), end='')
80f8e3804a 2010-08-20 c.kworr@b84a3: else:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('.', end='')
80f8e3804a 2010-08-20 c.kworr@b84a3: blocks = blocks - 102400/block_size
80f8e3804a 2010-08-20 c.kworr@b84a3: sys.stdout.flush()
80f8e3804a 2010-08-20 c.kworr@b84a3: temp_file.close()
80f8e3804a 2010-08-20 c.kworr@b84a3: print(']')
80f8e3804a 2010-08-20 c.kworr@b84a3: os.renames(options.dir + os.sep + '.tmp', file_name)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: checked_files += 1
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # storing new time mark and storing new headers
80f8e3804a 2010-08-20 c.kworr@b84a3: new_headers['_time'] = datetime.datetime.now()
80f8e3804a 2010-08-20 c.kworr@b84a3: index[url] = new_headers
80f8e3804a 2010-08-20 c.kworr@b84a3: index.sync()
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: except urllib.error.HTTPError as error:
80f8e3804a 2010-08-20 c.kworr@b84a3: # in case of error we don't need to do anything actually,
80f8e3804a 2010-08-20 c.kworr@b84a3: # if file download stalls or fails the file would not be moved to it's location
80f8e3804a 2010-08-20 c.kworr@b84a3: print(error)
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: if options.verbose:
80f8e3804a 2010-08-20 c.kworr@b84a3: print('[', len(unchecked_files), '/', checked_files, ']')
80f8e3804a 2010-08-20 c.kworr@b84a3:
80f8e3804a 2010-08-20 c.kworr@b84a3: # checking if there were any files downloaded, if yes - restarting sequence
80f8e3804a 2010-08-20 c.kworr@b84a3: if checked_files == 0:
80f8e3804a 2010-08-20 c.kworr@b84a3: break