I am creating a very simple site search using Verity. When outputting
the search results (using #summary#), the various variable names used in
the text of the pages are getting output as their variable names, not as
their values. Is there some trick I'm missing?
--
~~~ Ray Champagne ::
Did you index the pages as is? If so it makes sense. Verity will read
in the file, not parse it like CF does. If you want Verity to parse
the _result_ of the page you want to use the local spidering feature.
On 5/24/06, Ray Champagne [EMAIL PROTECTED] wrote:
I am creating a very simple site
Do you mind expanding on that a little? I'm not sure what the local
spidering feature is. This is my first foray into the Verity world
Raymond Camden wrote:
Did you index the pages as is? If so it makes sense. Verity will read
in the file, not parse it like CF does. If you want Verity to
On 5/24/06, Ray Champagne [EMAIL PROTECTED] wrote:
I am creating a very simple site search using Verity. When outputting
the search results (using #summary#), the various variable names used in
the text of the pages are getting output as their variable names, not as
their values. Is there
Yea, now that I think it over, that makes sense.
I'll check this link out. Thanks for the info.
Jim Wright wrote:
On 5/24/06, Ray Champagne [EMAIL PROTECTED] wrote:
I am creating a very simple site search using Verity. When outputting
the search results (using #summary#), the various
Verity can either index the source files, or it can index the
browser-product version.
Do you want to index the content of the cfm as it sits on disk, or do
you want to index the content of the cfm as it appears to the browser?
To index the source files (.htm, .pdf), use Verity the normal way.
Hrm...that sucks, looks like this won't work in a shared environment,
eh? Or, I should say that it won't work unless I get my host to make
the changes for me. Which we all know will never happen.
I can't wait until we get our own server. 2 months and counting!
Jim Wright wrote:
On 5/24/06,
Crow,
You should be able to fake it by building your own spider (just do a
directory listing, turn the filename into a url, and pass the url to
the indexer, one beautiful page at a time).
Yes, I've done this (last summer) to index PDFs stored on Federal
Court websites.
On 5/24/06, Crow T. Robot
Hrm...that sucks, looks like this won't work in a shared environment,
eh? Or, I should say that it won't work unless I get my host to make
the changes for me. Which we all know will never happen.
Well, just to throw out another way...if you have a good list of the
files that you want indexed
I just had a thought, to spoof a spidering the list of files from
the directory or the file-based cfindex may not be what you need...you
need the output from an actual spider, which would include all of your
dynamic files.. (index.cfm?foo=1, index.cfm?foo=2..etc...) You could
use a linkchecker
So, instead of passing the entire directory as the key, I'd pass each
page as a URL?
I could prolly do that, but I want to make sure that is what you mean
before I tackle it.
Jerry Johnson wrote:
Crow,
You should be able to fake it by building your own spider (just do a
directory listing,
Well, in this particular case, there are no pages with query strings,
luckily. Just plain ole .cfm pages.
I think that Jerry's idea might just work as a patch until I get my new
dedicated server up and running here.
Jim Wright wrote:
I just had a thought, to spoof a spidering the list of
OK, done that, or I should say this is running in the background right now.
One thing I'm confused about though. In the following snippet of code,
isn't the collection veritysearch just constantly reindexing itself?
So, it would only reindex the last spidered page?
cfdirectory action=list
Here is what happens when you don't look at the code first.
Our system pulls down the pdf files into a local folder, then runs the
index, but
We cheat.
We use a local Key
key=g:\production\opinions\mi\coaunpub\20060523_C256194_50_256194O.ON.PDF
and an external URL
You still may be able to do something like this...see modifications below...
cfdirectory action=list directory=#ExpandPath('.')#
name=directory_list filter=*.cfm
cfloop query=directory_list
cfhttp url=#webaddress##name# method=GET/cfhttp
cfindex action=UPDATE
I've since looked into Zoom. Pretty cool, does the job. It won't index
it every x days/hours/whatever, but it's just a brochure site anyways.
Content will change about once a year.
http://www.wrensoft.com/zoom/
No worries, thanks for the attempt.
Jerry Johnson wrote:
Here is what happens
We ship the Verity vspider executable with CFMX and you are licensed to use it
on localhost URLs. We even document it.
http://livedocs.macromedia.com/coldfusion/7/htmldocs/1781.htm
If this doesn't cut it, you can buy a license from Verity explicitly for CF
customers to expand vspider to
Tom, is the vspider installed with CF by default, as in:
Do I have the ability to use VSpider on a shared hosting solution
(say, CrystalTech)?
Do I need to share the 250,000 documents with the other people in the
shared environment?
On 5/24/06, Tom Jordahl [EMAIL PROTECTED] wrote:
We ship the
Tom, is the vspider installed with CF by default, as in:
Do I have the ability to use VSpider on a shared hosting
solution (say, CrystalTech)?
Do I need to share the 250,000 documents with the other
people in the shared environment?
I'm no Tom Jordahl - trust me - but the answers to your
19 matches
Mail list logo
