Excluding certain terms form TREX Search results

Hi everybody,
I am using a NW04 Portal with KMC installed, and a TREX 6.1 as search engine.
Is it possible to exclude a certain list of specified terms from the search result?
E.g. I search for "blocked Term" and, although some documents may contain this term, i get nothing displayed as search result.
Thanks and best regards,
Jochen

Hi Jochen,
It is not possible to exclude certain terms from search, neither through KM API nor on TREX side (unless you want to try with python scripting).
You would actually need to avoid indexing documents with certain terms - this is not possible either. Could you please describe shortly your scenario (why index documents that should not be found?), there is probably another way to exclude certain documents from being indexed or found (another criteria for filtering).
best regards,
Bettina

Similar Messages

TREX Search Result Layout: Can you change it?

Hi,
Is it possible to have different flavors for the TREX Search Result Layout?
Ideally would be through the use of parameters in the typical TREX URL. I know that "Navigation.xml" file is hard-coded.....
Thanks.
Dick

Hi Thilo,
Thanks.
But I want to use my new LayoutSet using the URL-based interface to TREX:
https://xxxx.xxxx.com/irj/servlet/prt/portal/prtroot/com.sap.km.cm.service?startpage=searchpage&configfilename=navigation.xml&resourcelisttype=com.sapportals.wcm.searchresultlist&searchtype=ctrlpers&selectedsearchin=from_here&selectedsearchfromhere=/room_extensions/cm_stores/pm_storage/workspaces/d4919b6d-ff00-0010-439b-9a59dec17c56&querystring=dog&searchvisible=false&showoptions=false&allowedsortprops=false&selectedsearchaction=fuzzy&selectedmatchesperpage=5&enablesearchsorting=false.
Dick

Search form and search results page

Hi! I really need some help with my search and search results page. I'm a newbie and am trying to get my site built. My search page will be the most important page and is quite complex. It will have 11 drop down boxes and 8 radio buttons. I have my form set up on the search page. When I try to run it now it returns all results. Because I have so many fields I didn't want to post it all but, here's what my code looks like for my first drop down box:
<select name="area_search[]" multiple="multiple" id="area_search" form="search" >
            <option selected="selected" value="" <?php if (!(strcmp("", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Any</option>
            <option value="Downtown" <?php if (!(strcmp("Downtown", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Downtown</option>
            <option value="Jackson County East" <?php if (!(strcmp("Jackson County East", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Jackson County East</option>
            <option value="Johnson County" <?php if (!(strcmp("Johnson County", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Johnson County</option>
            <option value="Johnson County North" <?php if (!(strcmp("Johnson County North", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Johnson County North</option>
            <option value="Johnson County South" <?php if (!(strcmp("Johnson County South", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Johnson County South</option>
<option value="Midtwon" <?php if (!(strcmp("Midtwon", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Midtown</option>
<option value="North Kansas City" <?php if (!(strcmp("North Kansas City", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>North Kansas City</option>
<option value="Northland" <?php if (!(strcmp("Northland", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Northland</option>
            <option value="Plaza" <?php if (!(strcmp("Plaza", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>Plaza</option>
            <option value="South Kansas City" <?php if (!(strcmp("South Kansas City", $row_rs_searchpropertyinfo['Area']))) {echo "selected=\"selected\"";} ?>>South Kansas City</option>
</select>
This is the code for a radio button:
<input <?php if (!(strcmp($row_rs_searchpropertyinfo['Washer_dryer_in_unit'],"yes"))) {echo "checked=\"checked\"";} ?> name="radio_washerdryerinunit" type="radio" id="radio" form="search" value="yes"/>
The searchresults.php page:
<?php require_once('/Connections/Property_Info.php'); ?>
<?php
if (!function_exists("GetSQLValueString")) {
function GetSQLValueString($theValue, $theType, $theDefinedValue = "", $theNotDefinedValue = "")
if (PHP_VERSION < 6) {
    $theValue = get_magic_quotes_gpc() ? stripslashes($theValue) : $theValue;
$theValue = function_exists("mysql_real_escape_string") ? mysql_real_escape_string($theValue) : mysql_escape_string($theValue);
switch ($theType) {
    case "text":
      $theValue = ($theValue != "") ? "'" . $theValue . "'" : "NULL";
      break;
    case "long":
    case "int":
      $theValue = ($theValue != "") ? intval($theValue) : "NULL";
      break;
    case "double":
      $theValue = ($theValue != "") ? doubleval($theValue) : "NULL";
      break;
    case "date":
      $theValue = ($theValue != "") ? "'" . $theValue . "'" : "NULL";
      break;
    case "defined":
      $theValue = ($theValue != "") ? $theDefinedValue : $theNotDefinedValue;
      break;
return $theValue;
$var_area_search_rs_searchpropertyinfo = "%";
if (isset($_POST['area_search'])) {
$var_area_search_rs_searchpropertyinfo = $_POST['area_search'];
mysql_select_db($database_Property_Info, $Property_Info);
$query_rs_searchpropertyinfo = sprintf("SELECT property_info.Area FROM property_info WHERE property_info.Area LIKE %var_area_search%", GetSQLValueString($var_area_search_rs_searchpropertyinfo, "text"));
$rs_searchpropertyinfo = mysql_query($query_rs_searchpropertyinfo, $Property_Info) or die(mysql_error());
$row_rs_searchpropertyinfo = mysql_fetch_assoc($rs_searchpropertyinfo);
$totalRows_rs_searchpropertyinfo = mysql_num_rows($rs_searchpropertyinfo);
?>
                  <?php
$rs_searchpropertyinfo_endRow = 0;
$rs_searchpropertyinfo_columns = 5; // number of columns
$rs_searchpropertyinfo_hloopRow1 = 0; // first row flag
do {
    if($rs_searchpropertyinfo_endRow == 0 && $rs_searchpropertyinfo_hloopRow1++ != 0) echo "<tr>";
   ?>
                  <td><table width="330px" border="0" cellspacing="7" cellpadding="2">
                    <tr>
                      <td width="300" height="275" bgcolor="#000000"><p><img src="../Properties/Thumbnail_image/<?php echo $row_rs_searchpropertyinfo['Images']; ?>" alt="" width="290" height="290" /></p></td>
                    </tr>
                    <tr style="text-align: center; font-size: 22px; color: #940000;">
                      <td width="300" height="75" bgcolor="#000000"><?php echo $row_rs_searchpropertyinfo['Area']; ?>
                        <p> </p>
                        <p><?php echo $row_rs_searchpropertyinfo['Property_Name']; ?></p>
                        </td>
                    </tr>
                    <tr bgcolor="#F7080C" style="text-align: center; font-size: 18px; color: #050505;">
                      <td height="45" bgcolor="#940000" style="font-size: 14px"><img src="/images/Detail button.jpg" alt="Property Details" onClick="MM_openBrWindow('<?php echo $row_rs_searchpropertyinfo['KCRC_detail_page']; ?>','detail','width=600,height=600')"/></td>
                    </tr>
                    <tr bgcolor="#F7080C" style="text-align: center; font-size: 18px; color: #FFFFFF;">
                      <td height="29" bgcolor="#940000" style="font-size: 14px"><a href="/checkavailability.php" target="new"><img src="/images/Check availability button.jpg" alt="Check availability" onClick="MM_openBrWindow('./checkavailability.php','availability','width=600,height=600') "/></a></td>
                    </tr>
                    <tr bgcolor="#F7080C" style="text-align: center; font-size: 18px; color: #FFFFFF;">
                      <td height="29" bgcolor="#940000" style="font-size: 14px"><a href="/scheduletour.php" target="new"><img src="/images/Schedule tour button.jpg" alt="Schedule tour" onClick="MM_openBrWindow('./scheduletour.php','schedule','width=600,height=600')"/></a></ td>
                    </tr>
                    <tr bgcolor="#F7080C" style="text-align: center; font-size: 18px; color: #FFFFFF;">
                      <td height="29" bgcolor="#940000" style="font-size: 14px"><a href="/contactus.php" target="new"><img src="http://www.kcrentalcentral.com/images/Contact agent button.jpg" alt="Contact a Rental Agent to answer any of your questions" onClick="MM_openBrWindow('./contactus.php','contact','width=600,height=600')"/></a></td>
                    </tr>
                  </table></td>
                  <?php $rs_searchpropertyinfo_endRow++;
if($rs_searchpropertyinfo_endRow >= $rs_searchpropertyinfo_columns) {
?>
                </tr>
                <?php
$rs_searchpropertyinfo_endRow = 0;
} while ($row_rs_searchpropertyinfo = mysql_fetch_assoc($rs_searchpropertyinfo));
if($rs_searchpropertyinfo_endRow != 0) {
while ($rs_searchpropertyinfo_endRow < $rs_browsepages_columns) {
    echo("<td> </td>");
    $rs_searchpropertyinfo_endRow++;
echo("</tr>");
}?>
As I said, I'm a newbie. Any help or guidance would be very much appreciated. THanks in advance!

Thanks for that example, but that is the design based on different technology than mine (ADF BC, and I am using ADF, EJB3/JPA). The link I provided above nowhere presents <methodIterator> nor <methodAction> and these are generated automatically in the page definition file by JDev when you drag-drop collections onto the page. So obviously that manual (online help) does not reflect the internals of current JDev 10.1.3.2 operation. As the matter of fact complete XML schema format seems to be different (although similar to certain extent) than the schema format created by JDev 10.1.3.2

Unable to retreive custom properties for WPC pages in TREX search result

Hi All,
We have created couple of custom properties "custom_news_heading" and "custom_news_abstract" which are indexable and made it available for all folders and documents.
In web pages folder in one of the WPC web site we have created few pages. In details screen for those pages we have updated those custom propeties with some text.
We have created an Index out of that web pages folder and able to see indexed count as same as the pages inside the web pages folder.
For that index the "Display Index Details" screen in TREX monitor is not showing up the custom properties - means TREX is not considering those custom properties even if have flagged it as Indexable.
Search result is returning the properties like "display name", contentlength etc. But we are not able to reterive the custom properties by customizing the Collection Renderer's Diaplayed properties.
Any advice please?
Thanks

Hi Dhaya/Scotts,
Please upload an image of ur layout setting and also the collection renderer in the
[Wiki|https://wiki.sdn.sap.com/wiki/x/IACnB] page that I have created for this thread...
Regards
BP

What character splits search terms (for different search result tabs), searching from the search box.

When we type in a search term, in the search box, the results can be shown in a tab. But how can you split search terms while entering them all at once, each search term providing results in a new tab.
"Search Term|Search Term Two", using |;\ etc only sends the term in entirety beyond. What does this box handle and how to make it open two tabs when given two separated search terms.
So what is this special character or setting?
Thank You

Thank u so Bruce , I finally found an answer to my question.
The solution u provided looks complete except that i am confused by the meaning of strField , may i didnt provide more details in my post .
The strField var i provided earlier was not meant to be placed in my code , the correct one is shown below .However , the strField means according to me is the field against which the search operation should take place based on the ( search Term )
the user provided .
Here is the code after applying the solution u suggested :
'Information collected from the user
strWhereToSearch = IIf(.framSearchIn.Value = 1, "[quknotSubject]", "[quknotBody] ")
strCondition = strWhereToSearch & " Like '* " & strTextSearch & " *' OR " & _
strWhereToSearch & " Like '" & strTextSearch & " *' OR " & _
strWhereToSearch & "Like '* " & strTextSearch & "'"
'Fill result search Recordset
strSql = " Select * from qryExtraTool_QuickNoteSubCatMainCat   "
strSql = strSql & " WHERE   " & strCondition & " ORDER BY quknotDate"
'Control Source for the text box to display matches. ( Highlight the search word )
txtQuickNoteBody = SearchHilight(strWhereToSearch, strTextSearch)
Tell me if my implementation to your code is right ?
Thank u again Bruce ...

Exclude deleted items from the search results

My users state that documents that they deleted days ago are still showing in the search results; how do I exclude all the items in all of the farm's Recycling Bins from being displayed in the search results?
Also, I have heard complaints that the search is displaying results that match their query, but they don't have permissions to the files; how can I exclude items that the user doesn't have permissions for?

Take a look at the following article. There are several policies that cover how long items remain in the index if they are missing from an individual crawl. By default they tend to be set fairly high (I think 30 days is about the average). YOu
can change these values to modify the policies to remove content more quickly. Just remember missing from a crawl isn't always due to an item being deleted. So if you decrease these policies you run the risk or removing existing content along with
deleted content.
http://geekswithblogs.net/simonh/archive/2013/12/09/search-crawl-deletion-policies-sharepoint-2013.aspx
Paul Stork SharePoint Server MVP
Principal Architect: Blue Chip Consulting Group
Blog: http://dontpapanic.com/blog
Twitter: Follow @pstork
Please remember to mark your question as "answered" if this solves your problem.

Apply Thumbnail ResourceRender to TREX Search Result

Hi Guys,
My Scenario is,
I have to show the Thumbnail images in Search Result (TREX). I indexed my image gallery folder in KM. TREX works fine.
And I Customize existing search result's ResourceRender for my need.
I want to show the Thumbnail images (left side) in search result if the result is image category.
Is it possible? any other way to solve this?
Thank you,
Regards,
Devasarathy Pandi

Thank you Guys,
Solved Throguh Applying this display property in Search Result Renderer Settings.
rnd:thumbnail(width=100/height=100/alignLeft/contentLink),
Thank you,
Regards,
Devasarathy Pandi.

TREX search results

Hi,
TREX search is working fine and is fetching results. I link to hide the links "Open Folder", "details", & "see also" & "HTML version" etc. How do I do it?
I tried playing with the Search options but it did not help me.
Please let me know how I can hide these links. Thanks in advance.
Janvi

Hi Janvi,
Go to
Content Management->User Interface->Settings->Collection Renderer Settings
Select StandardSearchResultRenderer and select advanced options. Go to Displayed Properties
The Default value is - rnd:icon(space-right=10/alignTop/alignLeft),rnd:displayname(contentLink/[1;2-3]/alignTop/space-right=30) + rnd:action(/alignTop), rnd:thumbnail(/noIcons/alignTop/height=30), rnd:rankvalue([2;1]/alignLeft/alignTop/noColumnSpacing/space-right=10),rnd:documentabstract([2;2-9]/alignTop/primaryTextStyle),col:rating([3;1]/noColumnSpacing/alignTop/alignLeft/space-right=10),modifiedby([3;2]/alignTop/alignLeft/secondaryTextStyle/space-right=10) + contentlength(secondaryTextStyle/alignTop/space-right=10) + modified(secondaryTextStyle/alignTop/space-right=10) + rnd:highlighted_contentlink(space-right=10/alignTop/secondaryTextStyle),rnd:categories([4;2-9]/alignTop/space-right=10)
Remove the rnd:action and rnd:highlighted_contentlink
Change it to
rnd:icon(space-right=10/alignTop/alignLeft),rnd:displayname(contentLink/[1;2-3]/alignTop/space-right=30), rnd:thumbnail(/noIcons/alignTop/height=30), rnd:rankvalue([2;1]/alignLeft/alignTop/noColumnSpacing/space-right=10),rnd:documentabstract([2;2-9]/alignTop/primaryTextStyle),col:rating([3;1]/noColumnSpacing/alignTop/alignLeft/space-right=10),modifiedby([3;2]/alignTop/alignLeft/secondaryTextStyle/space-right=10) + contentlength(secondaryTextStyle/alignTop/space-right=10) + modified(secondaryTextStyle/alignTop/space-right=10),rnd:categories([4;2-9]/alignTop/space-right=10)
Regards
Prakash

Transform TREX search results to RSS-feed

Is it possible to transform the results of a search query to a rss-feed or an xml file that can be used by a feed reader?
Basic idea is to create an information service that runs a scheduled search on defined topics and provides the results in a list that can be sent by email or combined with search results from other sources, e.g. Google News.
Is using the Open Search API the right way?
Any ideas are greatly appreciated!
Thanks,
Matthias
Edited by: Matthias Killenberg on Sep 9, 2009 12:47 PM

Hi Richard,
I think, you will mostly have to discuss this with Google. Just like TREX has ways to assume other engines results as TREX' own, Google may have these ways. But integration will mostly have to come "top down" from Google in that case.
And it might be easier to find a way to make your pages (it's a catalog, I assume?) accessible to the Google Crawlers, rather than trying to handover the index.
Regards, Karsten

Trex Search Result List

Hello,
i would change the Search Results from Trex.
Goal: I have some Files that have the name "_de". When i found one of this file i would display in the Search Result List the name of the Folder where this file was found.
So i want to add a function like this:
String nameForDisplaying = displayname;
if (displayname.equals("_de)) then
nameForDisplaying = getNameOfRootFolder(displayname);
I am not sure what is the best place to store this function. Is to write an own Ressource Renderer a good way?
Best regards,
Patrick

Hey Patrick,
did you find any answer to this question? I have a similar problem, i would like to change the displayed name for some results.
Many thanks in advance
Steffi

Exposing TREX Search Results to Google?

Hi,
We have a CRM 4.0 E-Commerce solution and want to take the TREX search engine results (indexes?) and make them available to Google/Altavista/msn/etc index externally.
The idea being a search on the web would hit an entry in our online shop and direct the user to the product when clicked on.
I see that there is a Google Connector but this looks to me to be including Google in TREX rather than the other way around.
Is this possible and if so how ?
Regards
Richard

Hi Richard,
I think, you will mostly have to discuss this with Google. Just like TREX has ways to assume other engines results as TREX' own, Google may have these ways. But integration will mostly have to come "top down" from Google in that case.
And it might be easier to find a way to make your pages (it's a catalog, I assume?) accessible to the Google Crawlers, rather than trying to handover the index.
Regards, Karsten

Exclude certain URLs form ICM cache?

Hey community,
does anybody know if and how it is possible to exclude certain files from the ICM cache?
We have a web app running on the web AS (with Portal 7.3), and all html-files from this web app should not be cached in the ICM.
In the web application interface I can see that these files are cached in the server cache. So I need to exclude them.
In case of necessity it would be ok if I exclude all html-files from the cache.
Thanks for your help,
Steffi

Hi Fabio,
thanks for your answer!
I don't want to invalidate the cache manually. I want to maintain something like:
"exclude all html files from web app YXZ from cache". These files should never be cached.
By invalidating the cache the problem is solved only until the next html-file is called.
Thanks,
Steffi

Anonymous user can't access WPC pages in TREX search result list

An anonymous user can use TREX service to enter query string in search iView and to get the list of results.
But when he tries to open a page (web_paragraph or web_article ) generated by WPC (by clicking on the page link in result list) a logon screen appears in a popup window.
The URL of the window is /irj/servlet/prt/prtroot/... rather than irj/go/km/u2026 .
How to work around this?
At the same time he can see Word, Excel documents and XML Forms (News).
We use light framework.

Hi
You need to change PCD Permission for the objects the anonymous used.
And change the parametrer Authentication Scheme to Anonymous.
Good Luck
Eduardo Grilo

Object Link to CRM business object in Trex search result list

Hello Everyone,
has anybody sucessfully configured the functionality "Object Link Properties" which is described in the Document <a href="http://service.sap.com/~form/sapnet?_SHORTKEY=01100035870000647974&_SCENARIO=01100035870000000112&_OBJECT=011000358700005796662005E">Configuring the Business Package for SAP CRM 5.0 SP01 (EN)</a> on Page 169? I'm not the Knowledge Management specialist and would get some help on this topic. Has anybody detailed steps how to configure this?
It shounds interresting as the result would be that when the User does a full text search for Attached Documents he will also get a link to the original CRM business object like Activity, Account or Opportunity.
Regards
Gregor

Hi Gregor,
Thanks for your reply.
using windows explorer mechanism i can see the folders that i created in the CRM config.
Now in Portals KM i did the following steps:
a) Created an HTTP Data Source : with URL = the URL that the service crm_prt_km_dav generates.
http://abrusux156h.pch.chem.corp.local:8086/sap/crm/crm_prt_km_dav
b)Created a memory cache
c)Created a WebDAV respository manager with the following details:
Description :WebDAV
Prefix : /PRD (PRD is the BO Hierarchy name that i have set in the CRM config)
System ID : T86 (Sames as what was set in HTTP Data Source)
Repository Service : svc_acl
ACL Manager cache : ca_cm
Security manager : ACLSecurityManager
Memory Cache :webDAV(same as what i had set in step b)
I restarted the portal server. and then tried doing the test by going to
Content Admin ->KM Content ->root
But i dont see the /PRD folder out there.
Any clues on that???
Thanks,
Anand

Content Snippets not displayed in TREX Search results

Hi Community,
next Problem with my Repository Manager. I used the tutorial from Thilo Brandt as start.
I can index documents from a third party content management system, but when i search for a word form the content it won't be found. When i search for a word from the title or the filename, the document appears in the result list. But, there is no content snippet.
I had this problem a few days ago. It was the contentlegth-attribute, which was 0. I fixed this an everything works fine. But now i have the same problem and the contentlegth attribute is correct.
Any ideas?
Greetz Marcus

Solved my Problem. The Security Manager had an error. It denied the access to the content for the portaluser, which is used for indexing the content.
Marcus

Excluding certain terms form TREX Search results

Similar Messages

Maybe you are looking for