<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	
	>
<channel>
	<title>Comments on: Property Extraction in FS4SP</title>
	<atom:link href="http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/</link>
	<description>A blog about Search as THE solution</description>
	<lastBuildDate>Mon, 26 Oct 2015 18:07:52 +0000</lastBuildDate>
		<sy:updatePeriod>hourly</sy:updatePeriod>
		<sy:updateFrequency>1</sy:updateFrequency>
	<generator>https://wordpress.org/?v=3.9.40</generator>
	<item>
		<title>By: John O'Reilly</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-46</link>
		<dc:creator><![CDATA[John O'Reilly]]></dc:creator>
		<pubDate>Mon, 14 Mar 2011 01:28:29 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-46</guid>
		<description><![CDATA[Hi, thanks for posting!  I had been using dictionary extractors with custom people and locations lists, this is great.

I did run across a strange behaviour, though, when I try to have a refiner for a property mapped to wholewordsx on the same refinement panel as one mapped to a pipeline extensiblity extracted property.  It seems unbelievable but I reproduced it in two completely different environments.  I can have my three refiners mapped to the wholewords dictionary extractors OR I can have my 6 refiners mapped to custom properties extracted with an extensibility component.  The data is there in both cases, and it fails if I have just one of each kind.  

Just wondering if you&#039;ve seen anything like that?

Many thanks, 

John.]]></description>
		<content:encoded><![CDATA[<p>Hi, thanks for posting!  I had been using dictionary extractors with custom people and locations lists, this is great.</p>
<p>I did run across a strange behaviour, though, when I try to have a refiner for a property mapped to wholewordsx on the same refinement panel as one mapped to a pipeline extensiblity extracted property.  It seems unbelievable but I reproduced it in two completely different environments.  I can have my three refiners mapped to the wholewords dictionary extractors OR I can have my 6 refiners mapped to custom properties extracted with an extensibility component.  The data is there in both cases, and it fails if I have just one of each kind.  </p>
<p>Just wondering if you&#8217;ve seen anything like that?</p>
<p>Many thanks, </p>
<p>John.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steen Westergaard</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-45</link>
		<dc:creator><![CDATA[Steen Westergaard]]></dc:creator>
		<pubDate>Thu, 03 Mar 2011 17:31:04 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-45</guid>
		<description><![CDATA[It&#039;s a little bit tricky to see whether a crawled property is multi-valued as the multi-valued field in the GUI is set to false even the property has a multi-valued value. Looking at the CrawledProperty Class, it does not contain a multi-valued property. Retrieving a crawled property through PowerShell you can see that it has an IsMultiValued property that is read only but it is still set to false even the content is multi-valued. The safe way to see if crawled property is multi-valued is to check the content and see if it contains a separator char i.e. a 0x2029]]></description>
		<content:encoded><![CDATA[<p>It&#8217;s a little bit tricky to see whether a crawled property is multi-valued as the multi-valued field in the GUI is set to false even the property has a multi-valued value. Looking at the CrawledProperty Class, it does not contain a multi-valued property. Retrieving a crawled property through PowerShell you can see that it has an IsMultiValued property that is read only but it is still set to false even the content is multi-valued. The safe way to see if crawled property is multi-valued is to check the content and see if it contains a separator char i.e. a 0&#215;2029</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mikael Svenson</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-44</link>
		<dc:creator><![CDATA[Mikael Svenson]]></dc:creator>
		<pubDate>Thu, 03 Mar 2011 11:33:36 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-44</guid>
		<description><![CDATA[I agree that it&#039;s not an easy find. But if you read the remarks on TechNet it says: This property must also be set to [true] to include all values from a multi-valued crawled property. If set to [false] it would only include the first value from a multi-valued crawled property.

I learned it myself by looking at the config for location and people managed props, not by reading the docs ;)]]></description>
		<content:encoded><![CDATA[<p>I agree that it&#8217;s not an easy find. But if you read the remarks on TechNet it says: This property must also be set to [true] to include all values from a multi-valued crawled property. If set to [false] it would only include the first value from a multi-valued crawled property.</p>
<p>I learned it myself by looking at the config for location and people managed props, not by reading the docs ;)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steen Westergaard</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-43</link>
		<dc:creator><![CDATA[Steen Westergaard]]></dc:creator>
		<pubDate>Thu, 03 Mar 2011 10:09:02 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-43</guid>
		<description><![CDATA[Thanks Mikael, with MergeCrawledProperties = true it works fine. It is a little bit confusing that MergeCrawledProperties  is called &quot;Include values from all crawled properties mapped&quot; in the GUI when you only want to map to one crawled property.]]></description>
		<content:encoded><![CDATA[<p>Thanks Mikael, with MergeCrawledProperties = true it works fine. It is a little bit confusing that MergeCrawledProperties  is called &#8220;Include values from all crawled properties mapped&#8221; in the GUI when you only want to map to one crawled property.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Trond Øivind Eriksen</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-42</link>
		<dc:creator><![CDATA[Trond Øivind Eriksen]]></dc:creator>
		<pubDate>Thu, 03 Mar 2011 08:44:55 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-42</guid>
		<description><![CDATA[Thanks for the comment, Steen. Please let me know if Mikael&#039;s approach doesn&#039;t work. I might just have another trick up my sleeve.]]></description>
		<content:encoded><![CDATA[<p>Thanks for the comment, Steen. Please let me know if Mikael&#8217;s approach doesn&#8217;t work. I might just have another trick up my sleeve.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mikael Svenson</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-41</link>
		<dc:creator><![CDATA[Mikael Svenson]]></dc:creator>
		<pubDate>Thu, 03 Mar 2011 08:21:58 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-41</guid>
		<description><![CDATA[Steen, make sure the managed property where you map your wholeword crawled properties have MergeCrawledProperties = true. If not you will only get one value.]]></description>
		<content:encoded><![CDATA[<p>Steen, make sure the managed property where you map your wholeword crawled properties have MergeCrawledProperties = true. If not you will only get one value.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Steen Westergaard</title>
		<link>http://blog.comperiosearch.com/blog/2011/03/02/property-extraction-fs4sp/#comment-40</link>
		<dc:creator><![CDATA[Steen Westergaard]]></dc:creator>
		<pubDate>Wed, 02 Mar 2011 21:49:33 +0000</pubDate>
		<guid isPermaLink="false">http://nuggets.comperiosearch.com/?p=271#comment-40</guid>
		<description><![CDATA[Great article – In addition to the out of the box extractors I have used the custom extractor Wholewords1 where you can build your own dictionary from scratch. Compared to the out of the box extractors I found a least one big disadvantage using a custom extractor. When you for example have two dictionary entries (key, value) Wine,Wine and Chardonnay,Chardonay and you use both  keys in a document, you only get a refiner values for one of the values which in my opinion makes it useless for searching. I know that dictionary keys are case sensitive and ensured to have an exactly match in my document. 

I have tried to add some new keys to the Person names extractor and it works fine with multiple key values in a document. 

Do you have any experience with custom extractors and do you know how to get around the problem described above?]]></description>
		<content:encoded><![CDATA[<p>Great article – In addition to the out of the box extractors I have used the custom extractor Wholewords1 where you can build your own dictionary from scratch. Compared to the out of the box extractors I found a least one big disadvantage using a custom extractor. When you for example have two dictionary entries (key, value) Wine,Wine and Chardonnay,Chardonay and you use both  keys in a document, you only get a refiner values for one of the values which in my opinion makes it useless for searching. I know that dictionary keys are case sensitive and ensured to have an exactly match in my document. </p>
<p>I have tried to add some new keys to the Person names extractor and it works fine with multiple key values in a document. </p>
<p>Do you have any experience with custom extractors and do you know how to get around the problem described above?</p>
]]></content:encoded>
	</item>
</channel>
</rss>
