According to Jim Priest:
> I'm trying to setup a htdig search on a new Cold Fusion site.
I must say I was quite surprised to find out a few years ago that there
was a product of that name out on the market. To me, the infamous term
"cold fusion" implied "all hype and no substance", which is not exactly
the sort of connotations you'd want associated with a product, or a web
site for that matter. :-)
> Things are working - but I have one hitch - I'm passing around URL
> tokens:
>
> http://192.168.0.24/index.cfm?CFID=36896&CFTOKEN=6580729
>
> to track the user (can't use cookies per customer request)
>
> Problem is that htdig as it searches - picks up this ID token - any way I
> can get rid of it when I output the search results? It also seems to
> cause htdig to report multiple search results - apparently because the
> url is different (new CFID) Is there anyway to force htdig to use only
> one ID when it indexes the entire site??
You'd need a patch to htdig to allow rewriting URLs. Andy Armstrong's
url_rewrite_rules patch,
ftp://ftp.ccsf.org/htdig-patches/3.1.5/htdig-3.1.5.aarmstrong.tar.gz
adds the capability to do this.
> Also can I customize what gets output in the 'summary' area? RIght
> now the search is picking up (and displaying) info from my header
> (which is common across all pages) and therefore all my results start
> off with:
>
> "... --- NINTENDO 64 --- Rainbow Six Roswell Conspiracies --- DREAMCAST --- Roswell
>Conspiracies --- PC/MAC --- Politika news Rouge Spear PSX Site Live Product News |
>September 11, 2000 The website for the PlayStation version of Rogue Spear is live!
>..."
>
> which is basically the text from a dropdown menu at the top of the
> page - and not very useful in the search results :)
Well, you can get it to display meta descriptions instead of document
excerpts by setting the use_meta_description attribute to true.
However, htsearch usually highlights the search words that are matched in
the excerpts. If you don't want these menu items to appear in excerpts,
you probably also don't want these documents to be picked up as matches
if the search word(s) only appears in the menu and nowhere else, right?
If so, you should arrange to have them excluded. You can do this by
surrounding the dropdown menu code with <noindex> and </noindex> tags.
Since htdig can't follow the links in dropdown menus anyway, all that
this will end up eliminating is the text in the menus. If you don't want
to have to add all these tags to the pages, and the menus are delimited
by unique strings, you can use the "noindex_start" and "noindex_end"
attributes to tell htdig to strip out everything between these strings.
See http://www.htdig.org/FAQ.html#q4.15
--
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]>
Information: http://lists.sourceforge.net/lists/listinfo/htdig-general
FAQ: http://htdig.sourceforge.net/FAQ.html
