Lines of
samesite.py
from check-in 53dcfdb8f7
that are changed by the sequence of edits moving toward
check-in 083ec707ea:
1: #!/usr/bin/env python3.1
2:
3: import datetime, http.cookiejar, optparse, os, sys, shelve, re, urllib.request
4:
5: parser = optparse.OptionParser()
6: parser.add_option('-v', '--verbose', action = 'store_true', dest = 'verbose', help = 'turns on verbose status notifications', metavar = 'bool', default = False)
7: parser.add_option('-d', '--dir', action = 'store', dest = 'dir', help = 'specify directory where the files should be stored', metavar = 'string', default = None)
8: parser.add_option('-r', '--root', action = 'store', dest = 'root', help = 'specify a site from which data should be mirrored', metavar = 'string', default = None)
9: parser.add_option('-l', '--log', action = 'store', dest = 'log', help = 'specify a log file to process', metavar = 'string', default = None)
10: (options, args) = parser.parse_args()
11:
12: if not options.dir:
13: print('Directory not specified')
14: exit(1)
15:
16: if not options.root:
17: print('Server not specified')
18: exit(1)
19:
20: if not options.log:
21: print('Log file not specified')
22: exit(1)
23:
24: if not os.access(options.log, os.R_OK):
25: print('Log file unreadable')
26: exit(1)
27:
28: # this is file index - everything is stored in this file
29: index = shelve.open(options.dir + '/.index')
30: desc_fields = ('Content-Length', 'ETag', 'Pragma', 'Last-Modified')
31: ignore_fields = ('Accept-Ranges', 'Age', 'Cache-Control', 'Connection', 'Content-Type', 'Date', 'Expires', 'Server', 'Via', 'X-Cache', 'X-Cache-Lookup')
32:
33: block_size = 32768
34:
35: while True:
36: unchecked_files = set()
37: checked_files = 0
38:
39: # reading log and storing found urls for processing
40: # check file mtime XXX
41: with open(options.log, 'r') as log_file:
42: log_line = re.compile('^[^ ]+ - - \[.*] "(GET|HEAD) (.*?)(\?.*)? HTTP/1.1" (\d+) \d+ "(.*)" "(.*)"$')
43: for line in log_file:
44: this_line = log_line.match(line.strip())
45: if this_line:
46: unchecked_files.add(this_line.group(2))
47:
48: for url in unchecked_files:
49:
50: # creating empty placeholder in index
51: if not url in index:
52: index[url] = {}
53dcfdb8f7 2010-07-05 53: reload = False
54:
55: # creating file name from url
56: file_name = options.dir + re.compile('%20').sub(' ', url)
53dcfdb8f7 2010-07-05 57: print('Checking file:', url)
58:
59: # forcibly checking file if no file present
53dcfdb8f7 2010-07-05 60: if not os.access(file_name, os.R_OK):
61: print('File not found or inaccessible.')
62: reload = True
63:
64: # forcibly checking file if file size doesn't match with index data
53dcfdb8f7 2010-07-05 65: elif 'Content-Length' in index[url] and os.stat(file_name).st_size != int(index[url]['Content-Length']):
53dcfdb8f7 2010-07-05 66: print('File size is', os.stat(file_name).st_size, 'and stored file size is', index[url]['Content-Length'])
67: reload = True
68:
69: # forcibly checking file if index hods Pragma header
53dcfdb8f7 2010-07-05 70: if 'Pragma' in index[url] and index[url]['Pragma'] == 'no-cache':
53dcfdb8f7 2010-07-05 71: print('Pragma on: recheck iminent.')
53dcfdb8f7 2010-07-05 72: reload = True
73:
74: # skipping file processing if there's no need to recheck it and we have checked it at least 4 hours ago
53dcfdb8f7 2010-07-05 75: if not reload and '__time__' in index[url] and (datetime.datetime.now() - datetime.timedelta(hours = 4) - index[url]['__time__']).days < 0:
76: continue
77: try:
53dcfdb8f7 2010-07-05 78: print('Contacting website.')
79: with urllib.request.urlopen(options.root + url) as source:
80: new_headers = {}
81: headers = source.info()
82:
83: # stripping unneeded headers (XXX make this inplace?)
84: for header in headers:
85: if header in desc_fields:
86: if header == 'Pragma' and headers[header] != 'no-cache':
87: print('Pragma:', headers[header])
88: new_headers[header] = headers[header]
89: elif not header in ignore_fields:
90: print('Undefined header "', header, '": ', headers[header], sep='')
91:
92: # comparing headers with data found in index
93: # if any header has changed (except Pragma) file is fully downloaded
94: # same if we get more or less headers
95: old_keys = set(index[url].keys())
96: old_keys.discard('__time__')
97: old_keys.discard('Pragma')
98: more_keys = set(new_headers.keys()) - old_keys
99: more_keys.discard('Pragma')
100: less_keys = old_keys - set(new_headers.keys())
101: if len(more_keys) > 0:
53dcfdb8f7 2010-07-05 102: if len(old_keys) == 0:
53dcfdb8f7 2010-07-05 103: print('No data on that file yet.')
53dcfdb8f7 2010-07-05 104: else:
105: print('More headers appear:', more_keys)
106: reload = True
107: elif len(less_keys) > 0:
108: print('Less headers appear:', less_keys)
109: reload = True
110: else:
111: for key in index[url].keys():
112: if key not in ('__time__', 'Pragma') and not index[url][key] == new_headers[key]:
113: print('Header "', key, '" changed from [', index[url][key], '] to [', new_headers[key], ']', sep='')
114: reload = True
115:
116: # downloading file
117: if reload:
118: if 'Content-Length' in headers:
119: print('Downloading', headers['Content-Length'], 'bytes [', end='')
120: else:
121: print('Downloading [', end='')
122: sys.stdout.flush()
123:
124: # file is created at temporary location and moved in place only when download completes
125: temp_file = open(options.dir + '/.tmp', 'wb')
126: buffer = source.read(block_size)
127: blocks = 0
128: megs = 0
129: while len(buffer) > 0:
130: temp_file.write(buffer)
131: print('.', end='')
132: sys.stdout.flush()
133: buffer = source.read(block_size)
134: blocks += 1
135: if blocks > 1024*1024/block_size:
136: blocks = blocks - 1024*1024/block_size
137: megs += 1
138: print('{}Mb'.format(megs), end='')
139: temp_file.close()
140: print(']')
141: os.renames(options.dir + '/.tmp', file_name)
142:
143: checked_files += 1
144:
145: # storing new time mark and storing new headers
146: new_headers['__time__'] = datetime.datetime.now()
147: index[url] = new_headers
148: index.sync()
149:
150: except urllib.error.HTTPError as error:
151: # in case of error we don't need to do anything actually,
152: # if file download stalls or fails the file would not be moved to it's location
153: print(error)
154:
155: print('[', len(unchecked_files), '/', checked_files, ']')
156:
157: # checking if there were any files downloaded, if yes - restarting sequence
158: if checked_files == 0:
159: break