AdvSearch 1.0.0 RC2 available in MODx Repository

Quote from: iamram at Oct 03, 2011, 08:55 AM

Is there a way to stop AdvSearch / AdvSearchForm from outputting CSS and JavaScript to the <head>, other than commenting out the regClientStartupScript & regClientCss items in advsearch.class.php and advsearchform.class.php?

I found &addJQuery=`0` already, but no option for advsearch.css, advsearchform.min.js and the AdvSearch header.

No It doesn't exist an option to cancel the automatic load of the css and js file. Do you merge & minify all your css and js files ?

Well, still in production, so no minified CSS at the moment.

It’s not too much of an issue, just a bit of a hassle to edit advsearch.class.php and advsearchform.class.php if the script gets updated. Just trying to keep things clean and not include redundant files. Would be a nice to have though.

I don’t use any of the original CSS classes, making the CSS file redundant. My jQuery is referenced at the bottom of the document so JS in the head throws an error when it encounters unavailable jQuery, and I already have scripts to handle default input box text and the like anyway.

Hi

I uploaded the indexation file after editing the line 22 with the folder i wanted to update. But when i run the file in my browser it says "indexation starting...." but nothing seems to be going beyond that. What could be the reason for this

You could have a look at the file at http://expatkey.com/indexation.php

Hi,
how can I use searchstrings with special characters? For german Sites this is important. If I try to search with special characters even on revo.wangba.fr I get "Inconsistent or invalid query".

You can try "über" wich means "over" or "ähnlich" wich means "similar". Even words with accent in french are not possible like "téléphone". Or is it necessary to use an not documented property to search with native utf8.

The content in database is utf8 without html-entitys
Thanks toelva

You are right this an issue. It doesn't occur on windows but exists on linux.
On Linux you get the following message :

iconv_strlen() : Detected an illegal character in input string.

This message is raised inside the Lucene/search/QueryLexer.php file. Line 342. It happens when the query phrase is parsed. Apparently, for some platforms, the default encoding is not UTF-8. So the encoding should be passed inside the parse function.

I just opened a bug on Github. See my comment to know how to fix this issue. I will publish a commit later.
You could also see a new demo page with a search inside French Documents (#c4).

I hope this help you. Thanks for your feedback.

Thank you very much,
this was again a quick response and solution.
toelva

Hi coroico,

I've installed this on a new 2.2 install however I can't get it to return any results,

I have uploaded the zend files to assets/libraries/Zend and using the following calls:

Search Form:


Landing page:



However I only ever get the following result:

There were no search results. Please try using more general terms to get more results.

All documents are searchable.

Thanks

Paulp,

You need to define the ids you want to search in:


You might find Coroico's GetIds handy.

Hope that will help

Hi Romain,

Thanks for that but is there no way for it to search every ID, the client will be adding content himself and this site will have thousands of ID's so I don't want to start having to go in adding ID's every time he adds something

Thanks for the response.

That's why i linked GetIds extra/snippet

At first i was kinda upset of that "feature", but in the end it brings way more granularity!