+++ /dev/null
-#! /usr/local/bin/python -O
-"""
- Check URL - subprocess
-
- Written by BroytMann, Mar 1999 - Feb 2000. Copyright (C) 1999-2000 PhiloSoft Design
-"""
-
-
-import sys, os, stat, string, time
-import urllib, www_util
-
-import cPickle
-pickle = cPickle
-from subproc import RecordFile
-
-from md5wrapper import md5wrapper
-
-
-ftpcache_key = None
-def myftpwrapper(user, passwd, host, port, dirs):
- global ftpcache_key
- ftpcache_key = (user, host, port, string.joinfields(dirs, '/'))
- return _ftpwrapper(user, passwd, host, port, dirs)
-
-_ftpwrapper = urllib.ftpwrapper
-urllib.ftpwrapper = myftpwrapper
-
-def get_welcome():
- global ftpcache_key
- _welcome = urllib._urlopener.ftpcache[ftpcache_key].ftp.welcome
- ftpcache_key = None # I am assuming there are no duplicate ftp URLs in db.
- # If there are - ftpcache_key in prev line is invalid.
- return _welcome
-
-
-class RedirectException(Exception):
- reloc_dict = {
- 301: "perm",
- 302: "temp"
- }
- def __init__(self, errcode, newurl):
- Exception.__init__(self, "(%s.) to %s" % (self.reloc_dict[errcode], newurl))
-
-
-class MyURLopener(urllib.URLopener):
- # Error 302 -- relocated (temporarily)
- def http_error_302(self, url, fp, errcode, errmsg, headers, data=None):
- if headers.has_key('location'):
- newurl = headers['location']
- elif headers.has_key('uri'):
- newurl = headers['uri']
- else:
- newurl = "Nowhere"
- raise RedirectException(errcode, newurl)
-
- # Error 301 -- also relocated (permanently)
- http_error_301 = http_error_302
-
- # Error 401 -- authentication required
- def http_error_401(self, url, fp, errcode, errmsg, headers, data=None):
- raise IOError, ('http error', errcode, "Authentication required ", headers)
-
-
-def get_error(msg):
- if type(msg) == type(""):
- return msg
-
- else:
- s = []
- for i in msg:
- s.append("'%s'" % string.join(string.split(str(i), "\n"), "\\n"))
- return "(%s)" % string.join(s)
-
-def check_url(record):
- try:
- now = str(int(time.time()))
- url_type, url_rest = urllib.splittype(record["URL"])
- url_host, url_path = urllib.splithost(url_rest)
- url_path, url_tag = urllib.splittag(url_path)
-
- tempfname = record["TEMPFILE"]
- del record["TEMPFILE"]
-
- fname, headers = urllib.urlretrieve("%s://%s%s" % (url_type, url_host, url_path), tempfname)
-
- last_modified = None
- record["Size"] = str(os.stat(tempfname)[stat.ST_SIZE])
-
- if headers:
- try:
- last_modified = headers["Last-Modified"]
- except KeyError:
- last_modified = None
-
- if last_modified:
- last_modified = www_util.parse_time(last_modified)
-
- if last_modified:
- last_modified = str(int(last_modified))
- else:
- last_modified = record["LastVisit"]
-
- record["LastModified"] = last_modified
-
- md5 = md5wrapper()
- if url_type == "ftp": # Pass welcome message through MD5
- md5.update(get_welcome())
-
- md5.md5file(tempfname)
- record["MD5"] = str(md5)
-
- except IOError, msg:
- if (msg[0] == "http error") and (msg[1] == -1):
- record["NoError"] = "The server did not return any header - it is not an error, actually"
- else:
- record["Error"] = get_error(msg)
-
- except EOFError:
- record["Error"] = "Unexpected EOF (FTP server closed connection)"
-
- except RedirectException, msg:
- record["Moved"] = str(msg)
-
- # Mark this even in case of error
- record["LastTested"] = now
-
-
-def run():
- urllib._urlopener = MyURLopener()
-
- # Some sites allow only Mozilla-compatible browsers; way to stop robots?
- server_version = "Mozilla/3.0 (compatible; Python-urllib/%s)" % urllib.__version__
- urllib._urlopener.addheaders[0] = ('User-agent', server_version)
-
- rec_in = RecordFile(sys.stdin)
- rec_out = RecordFile(sys.stdout)
-
- while 1:
- record = pickle.loads(rec_in.read_record())
- check_url(record)
- rec_out.write_record(pickle.dumps(record))
-
-
-if __name__ == '__main__':
- run()