Decreasing sensitivity of SunONE Portal search engine

When calculating a relevancy score, the SunOne Portal Search Engine (formerly Compass server) calculates relevancy out to 6 significant digits. Apparently due to the floating-point rounding routines, 2 documents with identical data will have slightly different relevancy scores, given the same search parameters. This ordinarily wouldn't be a problem, but for our application of Search, the documents need to show up with an identical score. Has anyone else run into this problem? Is there any way to force the search engine to ignore the last few digits of score, or round up to the nearest integer?

the search config files are stored in /var/opt/SUNWps/https-hostname/portal/config
can you clarify more about the type of search ur doing ..
I did a search and listed and the values only in integer format
HTH ..

Similar Messages

Portal Search Engine

When using the Portals search engine, does it search the descriptions or does it search for the name of the folders, iViews, and pages? It seems that it searches what ever is in the description. For instance, if your search paramenters is "News and Announcements", it will search for all the folders, iViews, and pages that have "News and Announcements" in the description.
Message was edited by:
Francis Cung

Francis,
as per the Help documentation it should search name, id or description of the iViews if correctly configured. See the documentation quoted below:
<b>Searching For Portal Pages and iViews</b>
<i>This function is only available if configured for your system. The system searches for all portal pages and iViews in the roles of the logged-on user.
In the search field in the portal header, enter the name, ID, or description of a portal page or iView. Open the Search Options dialog box and make sure that the checkbox for the corresponding search area or index is active.
In the search results list, you only have to click a hit to navigate to the corresponding portal page.</i>
You can find the whole documentation <a href="http://help.sap.com/saphelp_47x200/helpdata/en/7e/5e115e4a1611d1894c0000e829fbbd/content.htm">HERE</a>
Regards,
Shubhadip

Sending query to third party portal from TREX Search Engine

HI Experts,
we need to implement normal search option in SAP portal.
when we search any thing from portal ,
It has to search in EP & KM and also it has to search
windows sharepoint portal and bring the data back to SAP Portal.
For this purpose one solution is
1) implementing enterpise search,
but my client dont want to go for enterprise search(federated search).
2) The other way
Sending query to third party portal from TREX Search Engine
In this way what i want is ,
how TREX can send query to third party portal SEARCH ENGINE
(in my case share point portal search ENGINE).
There is no problem to search in EP & KM becuse it is default.
To search in Micro soft Share point portal,
TREX should pass the query to share point portal search ENGINE.
Is there any API TO send a query to third party portal like sharepoint?
I searched in SDN and Other sites also but i am not getting exactly what I require.
If any one has ideas or implemented already please guide me.
My client require searching option like this,
we need to provide drop down box in SAP Portal with 3 options like
1) search in Share point portal
2) search in SAP EP & KM
3) Search in both the portals
please provide any code samples in case if you have.
please help me , its urgent
Thanks in advance.
Regards
Bala

Hi Bala,
please check the information on the KM IMS (Index Management Service) in KMC's developer guide. A connection to a 3rd party search is done from KM, then calling TREX and XY in parallel, not serially from TREX.
Here's a paper describing this for an older KM release:
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/5e514b57-0701-0010-3796-deb3636835fa
Regards, Karsten

Sun One Search Engine Integration with the Fatwire Content Server

Hello everyone,
I am presently using Fatwire(Divine) Content Server. I have uploaded certain documents using Fatwire flex attributes(metadata). I want to use portal search engine for implementing Full Text Search of uploaded documents. how can i map metadata(attribute) of fatwire/divine content server with the portal server search engine metadata.
I hope someone might have tried it earlier while integrating content server with the portal. please give me some suggestions in this regard.
thanku
jenni

Hi,
I don't know about the metadata mapping,
but you can definitely index the FatWire local
data directory with the search engine.
Just specify "file:///.../<fatwiredir>" as starting point for your search robot.
Cheers,
Alex :-)
PS: After "Sun Forum Accounts Update" I couldn't login to this forum and at SUN
no one cares - they just ignore my mails. "Thanks a lot" for supporting free community!
(Check my old profile at <http://swforum.sun.com/jive/profile.jspa?userID=3455>)
OK. I have now a new account and I will try to help you out here...
-------------------------------------------------------------------------

Portal search from an external website

We have a requirement to extend the portal search capability to other websites. I'm wondering if anyone else has done this.
Basically the client has an Aqualogic 6.1 portal and also hosts a number of websites on a separate (i.e.: non portal) web server. They are unhappy with the search provided by MS Index Server and would like to use the Portal search from these static websites.
My thinking is that we can crawl the content of these websites into Aqualogic and use the PRC (search API) from these websites to execute the user search. I'm figuring that this should be possible if we install the ALUI API service on these web servers (effectively making then Remote Servers?)
Any thoughts?
thanks in advance
Simon

Let me see if I understand this. The client has a few static HTML websites that have no relation to the ALUI portal, but you want to you the portal search engine to search those sites?
Yes, your proposed solution is certainly possible. You could crawl the contents of those websites into specific folders in the KD and set the security so only a specific user (which you would use to authenticate in the custom search page) can see them. You don't even need to install the ALUI API service on those web servers to do this. The API endpoint URL of the portal needs to be accessible from those web servers, and you just need to reference the portal IDK dll or jar files when you write your custom search page.
I have done something similar to this, but the static web site was actually migrated in to be a part of the portal. Still a separate URL and look and feel, but hosted on the portal. However, I did crawl all the pages in and have a custom search page using PRC to search for documents. It works like a charm.
DJ Dewey | VHA Inc. | [email protected]

Use of KM services and TREX Search Engine From share point portal

Hi All,
We would like to Use of KM services and TREX Search Engine From share point portal.
For this purpose we interoperability between windows sharepoint and SAP Portal.
Please provide necessary documents or suggestions for interoperability for above requirement.
mail id : <b>[email protected]</b>
Thanks in advance
Regards
Swarna B

Hi,
See this this links:
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/9e891378-0601-0010-5386-a3387f1b161b
Regards,
Senthil K.

Indexing portal content by external search engine

Hi All,
I have a question about how to enable external search engine to index portal content.
as I searched in sap help [Indexing Portal Content|http://help.sap.com/saphelp_nw70ehp1/helpdata/en/46/9d1405fa743ef0e10000000a1553f7/frameset.htm]
To enable public search engines, such as Google, to index portal content, provide the search engine with the following URL:
irj/servlet/prt/portal/prtroot/com.sap.portal.utilities.tools.portalspider.SpiderWebInterface?
How should I provide this above url to external search engine?
Many thanks for all reply.
Cheers,
Kanok-on K.
P.S. We use the default KM.

Hi...
Did you get any solution? Is it working now?
I need same Information.... Please share if you have solved the issue.... and your site is indexed by search engine.
Thanks,
PradeeP

What search engines does iPlanet Portal support

iPlanet Portal Server facilitates search services by integrating iPlanet Compass Server with it.
iPlanet Compass Server has built-in search engine and also Verity search engine forms part of it.
For further references, look it in to the following URL:
http://docs.iplanet.com/docs/manuals/compass.html
-SaralaVijay

Portal Page Dynamic URLS & Search Engines

Hi,
I have created a portal site. These portal pages urls are created dynamically.
How can I optimize this dynamic (and long) URL for a search engine?
I am having trouble getting them spidered and ranked in major search engines.
Any suggestions?
Thanks.

Despite many posts on this issue, I am skeptical that there
is a problem
with search engines indexing pages with query strings. I have
a number
of pages with URLs like showdetail.cfm?articleID=25 and they
are always
indexed by Google and other search engines.
It is my understanding it largely depends on how you link to
the detail
pages. If you have direct anchors to the detail pages Google
will index
them. If you use a form or something else, Google is much
less likely
to follow.
For example:
<a href="showdetail.cfm?articleID=25">The best red ball
ever</a>

<form action="showdetail.cfm" method="get">
<select name="articleID">
<option value="25">The best red ball
ever<option>
<option value="26">The best green ball
ever<option>
<option value="27">The best blue ball
ever<option>
</select>
<input type="submit"/>
</form>

This makes some sense, because how can a robot complete a
form even if
it could determine it is for navigation.

Webcenter Portal Application Search Engine Optimization

Hi All,
We have developed Webcenter Portal Appliation(website).
To increase the search engine ranking what are the ways to to do SEO(Search Engine Optimization).
Adding meta-tags in the page is also one of them to improve page ranking level

refer below if helpful,
http://prsync.com/oracle/search-engine-optimization-for-webcenter---seo-372600/
https://blogs.oracle.com/webcenter/entry/social_media_changes_and_challenges

How to change what search engine the adress bar uses when typing in a search to something that isn't Google

I currently left all Google services I used, Gmail, Chrome, etc.. because of the recent selling info to the gov and what not and although it may or may not be true I am being safe and switched to Firefox and a different email provider. Anyway so far I like Firefox I am using DuckDuckGo as my current default search and I would like to use it as my search engine when I type in a search in the address bar, by default it's Google how do I change it to a search provider I want to like DuckDuckGo?

Hello,
If you have installed DuckDuckGo on Firefox, you can follow these steps:
# Type '''about:config''' in the address bar.
# A warning appears, click "I'll be careful, I promise!"
# Type '''browser.search.defaultEngine''' in the search box at the top of the page.
# Double-click the item that appears.
# Set the value to '''DuckDuckGo''' or whatever search engine that you would like to use (case sensitive I think).
(I have attached a screenshot).

3rd Party Search Engine Integration - some property not rendering

Hi,
I have tried to integrate a 3rd Party Search Engine into TREX following the published example.
I have successfully integrated and search results from the 3rd party engine are showing up.
However, some properties like cm_description, cm_contentlength, cm_modifedby are not rendering; their value are null; the search layout are specified correctly.
I only managed to set
1) cm_displayname
2) cm_contenttype
3) cm_id
and are showing up.
The setting of the rest of the properties are done in the same way and put into the propertymap.
Any idea why are the not rendering?? Any help appreciated. Thanks!

Hello Thilo,
I tried to deploy the original package com.customer.search.par, but unfortunately I got also an error with this package.
The disadvantage is, that I don't get an error message during start-up in one of the logger - for cause analysis.
I got the following error message within KM:
java.lang.NullPointerException
     at com.sapportals.wcm.repository.runtime.CmSystem.getResourceImpl (CmSystem.java:774)
     at com.sapportals.wcm.repository.runtime.CmSystem.getResource (CmSystem.java:325)
     at com.sapportals.wcm.repository.runtime.CmAdapter.getResource (CmAdapter.java:81)
     at com.sapportals.wcm.portal.service.KMServiceImpl.afterInit (KMServiceImpl.java:260)
     at com.sapportals.portal.prt.core.broker.PortalServiceItem.__initServiceInstanceStep2(PortalServiceItem.java:736)
What can I do at this point?
Best regards,
Walter

An add-on to force firefox switch back to default search engine in search bar

Hi,
I have a couple search engines installed in firefox's search bar. I find it extremely annoying that each time I change the search engine it remains like that even after restarting firefox. Is there any add-on that would switch back to the one that is on the top of the list (which happens to be my default search engine for navigation bar as well) every time I restart the browser?
Let me give you an example: I use duckduckgo as a default search engine for the nav bar and in the search bar. Let's say I need to find a definition for which I use wikipedia search engine. I switch to wikipedia one then. I turn off the browser and the next time I start firefox the search engine that appears in the search bar is wikipedia and I would prefer it to be duckduckgo.
I hope it isn't to vague.
cheers,
Omen

Easiest is to set the preferred default engine to the preferred search engine via the user.js file in the Firefox profile folder by copying the related user_pref() line to this file.
The user.js file is read each time Firefox is started and initializes preferences to the value specified in this file, so preferences set via user.js can only be changed temporarily for the current session.
You can search the <b>about:config</b> page for browser.search. to locate the pref(s) that you need.
*user_pref("browser.search.defaultenginename", "Google");
You can use this button to go to the currently used Firefox profile folder:
*Help > Troubleshooting Information > Profile Directory: Show Folder (Linux: Open Directory; Mac: Show in Finder)
* Create the chrome folder (lowercase) in the <xxxxxxxx>.default profile folder if this folder doesn't exist
* Use a plain text editor like Notepad to create a (new) user.js file in this folder (the names are case sensitive)
* Make sure that you select "All files" and not "Text files" when you save the file via "Save file as" in the text editor as user.js<br />Otherwise Windows may add a hidden .txt file extension and you end up with a not working user.js.txt file
* http://kb.mozillazine.org/user.js_file

How to make my site top in the search engines for a high ranking?

I am new at this, and I am planning to build a big site on iweb, but my question is where can I get information on how to make my site with high rankings? Would I have to post my site in lots of search engines, or have the hosting company help out? how do i do this?

As SEO (Search Engine Optimization) is important to you, you may want to consider alternatives to iWeb as "iWeb websites are not made to be search engine friendly" — read more here:
SEO For iWeb: How to get your iWeb Websites into Google & Other Major Search Engines
...iWeb is designed for personal sites and, as such, Apple apparently thinks there's no need for Google — you tell grandma the URL of your personal iWeb site, she painstakingly types it in and then gazes admiringly at the photos of her grandchildren.
Apple hasn't changed this approach in three iterations of iWeb but perhaps, with all the feedback it receives on this issue, there's still hope. Meanwhile, web-design apps for business sites tend to integrate search engine optimization — Sandvox Pro is an example:
http://www.karelia.com/sandvox/getyour_site_noticed_andb.html
Another advantage of Sandvox is that, when viewing your site in Safari, the text increase & decrease commands will work — they don't work for an iWeb-created site. ...For business use, it's important that customers don't turn away from your site because they consider your text too small to read.
However, if you still really want to use iWeb, this article may help:
http://www.iwebformusicians.com/SearchEngines/SEO.html
By the way, for more info about SEO in general, this blog of a web-design app (Yola) recently started a multi-part series on the subject — the first is here:
http://blog.yola.com/2009/06/15/seo-101
The rest of the series (in no particular order) can be found here:
http://blog.yola.com/category/seo/
And a video:
_Using Title Tags Effectively For Search Engine Optimization (SEO)_

How to implement search engine for pdf file.

Hi
I am newbie to AIR technolgy .how to implement search engine . when i enter the name of file in search box , it should fetch correspoding the pdf file .
Please help me out
Thanks
Mohan

Hi sekhar,
Your Requirement can be implemented by OVS(Object Value selector) This is the custom search help .So that you can define on what basis the value has to be fetched.
Look at the below link
http://wiki.sdn.sap.com/wiki/display/WDABAP/ABAPWDObjectValueSelector(OVS)
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/606288d6-04c6-2c10-b5ae-a240304c88ea?quicklink=index&overridelayout=true
You need to use the component WDR_OVS.
It has 3 phases
In phase 1
     You will fetch the value from input field to F4 help dialog box.
In phase 2
     You will generate the search results , in this phase look for the input value given in field 2 and accordingly generate search result for the field 3.
In phase 3
     The selected value will be returned to the original screen.
Regards
Karthiheyan M

Decreasing sensitivity of SunONE Portal search engine

Similar Messages

Maybe you are looking for