|
|
Rank: Fanatic
Joined: 8/9/2007
Posts: 284
Location: Kentucky, USA
|
Because I need to index binary files (PDFs and MS Office) and because my site will be larger than 1000 nodes, I am trying to replace XSLTSearch with Ismail's version of Umbraco Search. But I keep getting a yellow screen when trying to view the search results. I can post more details, but judging by the age of this code, I'm wondering if this is even supposed to work with umbraco 3.x? Here's the error I get from dotnet: Server Error in '/' Application.
Method not found: 'System.Xml.Xsl.XsltArgumentList umbraco.macro.AddXsltExtensions()'. I got the code and instructions from this URL: http://ismail.umbraco.net/2006/12/4/multimedia-indexing.aspxDoes this ring a bell with anyone?
Pete Koutoulas • Fayette County Public Schools • Lexington, Kentucky
|
|
Rank: Fanatic
Joined: 3/15/2007
Posts: 378
Location: Cary, NC USA
|
I've successfully used a variant of the umbraco extensions search to index pdfs and office files: http://www.codeplex.com/umbracoextIf I remember correctly, I think I also had to locate and compile some other projects to get it to work though... I can probably help you out if needed though...
|
|
Rank: Umbracoholic
Joined: 7/20/2006
Posts: 1,074
Location: Charleston, West Virginia, United States
|
Hi Pete, That is an old error. Basically, the original version of Lucene Search used some built in methods from the macros. The code was changed to make that method not static, thus, it can not be used in the situation above. You basically have to copy the method over to the lucene search code to get it to work. The Codeplex link above should take care of you, as it is written against the current version.
• 2007/2008 MVP • 2008/2009 MVP • Certified • Licensing • Support • Development • Hosting •
|
|
Rank: Addict
Joined: 7/19/2006
Posts: 649
Location: Preston, UK
|
Pete, The link to my blog is v2 version i have done quite a few updates since then. As per Casey's post you will be better off using the codeplex umbSearch. however you could try this version this also has a hacked autoform in there so if you are using autoform you may get conflicts. The new version i have is all search stuff only but its still work in progress im hoping to demo it at codegarden. More info here about it. Regards Ismail
Level 2 certified. If it aint broke dont fix.
|
|
Rank: Fanatic
Joined: 8/9/2007
Posts: 284
Location: Kentucky, USA
|
Thanks to everyone who replied. Sounds like as of right now there is no easy, sure-fire solution. Ismail, your advanced search sounds great. Since my site won't roll for another 4-6 weeks, I'm gonna wait until after you finish it and then give it a try. Thanks for your help.
Pete Koutoulas • Fayette County Public Schools • Lexington, Kentucky
|
|
Rank: Addict
Joined: 7/19/2006
Posts: 649
Location: Preston, UK
|
Pete,
Are you coming to codegarden? If so then bring your site on your laptop and we can during the open session stick it on.
Also just to update you I have started to implement this on a new site we are working on should go live in 2 weeks. The spell checker has worked better than i thought. I did search for word lighting but i mis spelt it to lihting it then gave a few suggested words one of which as expected was lighting but i also found ligting so i clicked on that and i got 3 pdf docs. I searched in the docs and looks like whoever did the pdfs mis spelt lighting!
The advanced search is mint as is the find similar intially i was testing with very small index but this site has over 200 pages and 50+ pdfs and so far so good.
Also how much content do you think you will have on this site?
Regards
Ismail
Level 2 certified. If it aint broke dont fix.
|
|
Rank: Fanatic
Joined: 8/9/2007
Posts: 284
Location: Kentucky, USA
|
Unfortunately, I won't make CG -- I can't seem to sneak a purchase order for a trip to Europe past my boss, Ismail. ;-) Our site will probably be in the neighborhood of 4000 pages of text and binary files (PDFs and MS Office files). Initially, it will be much smaller but should grow rather quickly as content is moved from an existing site.
Pete Koutoulas • Fayette County Public Schools • Lexington, Kentucky
|
|
Rank: Addict
Joined: 7/19/2006
Posts: 649
Location: Preston, UK
|
Pete, Nice. The stuff that im working on only really comes into its own with big sites, its basically overkill for anything small. I have also put in code to make wordnet calls although im not making those calls yet but with really large sites getting more results via conceptual-semantic and lexical relations ( mouthful two words meaning same thing ) would be really cool. In fact can you not blag your boss and say there is potential to get some really cool search stuff for free? 1. ajax search 2. google like search 3. related search 4. wordnet Also when it comes closer to time to implement drop me an msn via the forum i'm more than happy to help out if you have any issues. Regards Ismaill
Level 2 certified. If it aint broke dont fix.
|
|
Rank: Fanatic
Joined: 8/9/2007
Posts: 284
Location: Kentucky, USA
|
Thank you, Ismail. I will contact you after CG.
Pete Koutoulas • Fayette County Public Schools • Lexington, Kentucky
|
|
|
Guest |