According to David T. Ashley: > In the future, I'd like to also provide a search engine which rides along as > part of the install, based on ht://Dig. The ht://Dig databases would be > part of the install too--they would not be built on the target machine. My > idea would be that there are one of two scenarios: > > a)The user opens an HTML document on his machine, and *somehow* when he does > a search a windows version of htsearch is invoked that lets him search the > databases corresponding to documents on his machine. > > b)The user fires up a server that serves a page and the server runs > htsearch. > > Any ideas here? > > I don't mind creating a "custom" server that does the searches and so on. > But is there a better way?
Well, the 3.2.0b4 snapshots support file:// URLs, so that might be the best way to build the index. As for the 2 scenarios above, the key to option (a) is what that *somehow* is. If you want the search to be done from an HTML form, the HTML 4.01 standard doesn't leave a lot of options as to what the form's action attribute will be. It says: "User agent behavior for a value other than an HTTP URI is undefined." I've once seen an HTML form that used a "mailto:"; action, so I assume that browsers commonly implement this too, but it's non-standard. I don't see how you have much choice but to run htsearch from an HTTP server. If you want to run htsearch directly from another program, you can do that too but it wouldn't be from an HTML-based form unless you're running that HTML though a custom browser with hooks for running htsearch directly. -- Gilles R. Detillieux E-mail: <[EMAIL PROTECTED]> Spinal Cord Research Centre WWW: http://www.scrc.umanitoba.ca/~grdetil Dept. Physiology, U. of Manitoba Phone: (204)789-3766 Winnipeg, MB R3E 3J7 (Canada) Fax: (204)789-3930 _______________________________________________ htdig-general mailing list <[EMAIL PROTECTED]> To unsubscribe, send a message to <[EMAIL PROTECTED]> with a subject of unsubscribe FAQ: http://htdig.sourceforge.net/FAQ.html
