# This is a example w3mir config file. The file and options are fully # documented in the w3mir man page, and a lot is spelled out in the # w3mir-HOWTO. This file might not be completely updated, but the man # page should be. # # Comment lines start with '#'. Comments are not allowed at end of # lines. The keywords are not casesensitive, but some settings are, # notably fetch/ignore/apply settings, as well as the urls given. # # Available options: recurse, no-date-check, only-nonexistent, list-urls, # lowercase. It is not default to recurse. Options must be the # first directive to appear in the config file. # The options directive needs to be the first directive in the config file # and there can only be one. Options: recurse # Multipele options: # Options: recurse, only-nonexistent # This directive was previously named 'Fetch-options' # # Start here: URL: http://www.stud.ifi.uio.no/~janl/ # # # How much noise to make, one of quiet, brief or chatty. Default is brief #Verbosity: brief # # Initial referer. Some servers are very picky.... #Initial-Referer: http://foo.bar/ # # Headers to disable for anonymitys sake: #Disable-headers: referer, user # # Nicenes, robotrules can be true or false #Robot-Rules: true # Default pause between http connections is now 0 seconds # Pause: 0 # Pause between retries, in seconds. If w3mir runs as a once-a-day thing # this can and should be high. I suggest at least one hour (3600 seconds). #Retry-Pause: 600 # How many times a failed retrival is retried. #Retries: 3 # # URL selection. Order is significant, first Fetch/Ignore rule overrides # later rules. Supply as many as you want. If not matched by ignore rules # a doc is gotten. #Fetch: *.html # # Ignore urls we can't copy anyway. Actually w3mir can copy # many cgi things with no problems. But not always. Ignore: *.cgi Ignore: *-cgi Ignore: *.map # # RE variants are full perl REs. Mix freely with the others. These # must be legal perl REs, including the starting m and any # options you want, like 'i'. # #Fetch-RE: m/jpeg/ #Ignore-RE: m/\.gif$/ # # URL manipulation. _All_ rules are applied in order. If lowercasing # options is given that is done first. This must be a legal perl # substitution RE. Including the starting s and any options you want, # like 'g' and 'i'. # # This changes the name of all .gif files to .fif #Apply: s/\.gif$/.fif/ # # HTTP Proxy server: #HTTP-Proxy: www-cache.uio.no:3128 # # HTTP Proxy options: # - 'revalidate': Force proxy to revalidate cached documents. # - 'refresh': Force proxy to get cached documents from server _again_. # - 'no-store': Use if the contents of the documents is sensitive. # - 'no-pragma': Force use of HTTP/1.1 'Cache-Control:' header. # If only 'refresh' is specified the default is to use the 'Pragma: no-cache' # HTTP/1.0 header, this can be overruled with the 'no-pragma' option. # You can be fairly sure any proxy will understand 'refresh', ONLY HTTP/1.1 # proxies will understand 'revalidate' and 'no-store'. #proxy-options: refresh # # HTTP Proxy authentication. The mechanism is simpleminded, we either need # to authenticate ourselves with the proxy or not, and there can only be # one user/password tuple, domain/realm information is ignored... # HTTP-Proxy-user: foo # HTTP-Proxy-passwd: bar # # What to do with files, one of save, stdout or forget. #File-disposition: save # # Debug level #Debug: 0|1|2 # # Umask, if not the default #Umask: 077 # # HTML document editing options #Header: Mirrored from http://www.ifi.uio.no/~janl/ts/ #Remove-nomirror: true # # Authentication: #Auth-domain: server:port/realm #Auth-user: me #Auth-passwd: my_password # # You can use '*' in place of server:port and/or realm: #Auth-domain: */* #Auth-user: otherme #Auth-passwd: otherpassword # # 'Fixup' is done by the external w3mfix program. If you want a # mirror that's browseable in a filesystem (as opposed to via a http # server) just put this in: Fixup: run # Related is this, what is the name of the file the http server serves # when a directory URL is requested? w3mir default is index.html. #Index-name: welcome.html