Dan Mahoney, System Admin wrote: > Assume I have a site that I want to create a static mirror of. Normally > this site is database driven, but I figure if I spider the entire site, > and map all the GET URLS to static urls I can have a full mirror. Has > anyone known of this being successfully done? How would I get apache to > "see" the page names as full names (for example a page named > exec.pl?name=blah&foo=bar actually being a file rather than a command?)
Wget should already do what you want (provided that the file system where you will be mirroring the results can handle things like "?", "=", and "&" in a file name). Wget does not care how Apache processes a URL; it only cares that when it does a GET of a URL that some object is returned. The issue for you will be making sure that all the things you want to mirror are referenced as links on the site. How does a person visiting your site know that "blah" is a valid value for name or that "bar" is a valid value for foo? If they "learn" this by clicking on a link, then everything should work as you want. However, if the user must supply the value for name and foo (perhaps by entering them in a form) then there is no way for wget to know those values. If that is the case, you will have to construct your own list of URLs with all the combinations of name and foo that you want to mirror. HTH. Tony
