Perform a search within the Client scopes to bring up the Client Results page. Click on Site Actions and Edit Page. Edit the Search Core Results web part by clicking the Modify Shared Web Part from the Edit (dropdown) menu:

On the left hand side pane, under the Results Query Options, click inside the Selected Columns to make the blue ellipsis button appear. Click on the blue ellipsis button:

At the bottom of the text, before the </Columns> tag, add a column entry for each managed property that was created for the BDC content source:

Click OK.

Next click on the XSL Editor under the Data View Properties:

The XSL entry will appear in the text dialog.  It may be easier to copy and paste the code out into a different code editor and then paste back into the dialog for the remaining XSLT steps.

XSLT: Overriding the results URL

The tasks outlined in this section are for the scenario where we want to show a custom URL in the results versus the one returned from the normal search results. While this may not apply to everyone, the steps show how to add a new variable based on a custom search column and use it within the overall XSLT.

Locate the “This template is called for each result” section of the code.

Add a new variable name entry for the ClientURL under the two predefined variable entries (all variables should be in lowercase regardless of how the managed property was defined):

Next, find and change all “$url” entries to “$clienturl”.

Find the “xsl:value-of select=”url”" entry within the hithighlightedproperties condition code (when statement). Change the “url” to “title”.

Locate the span class for srch-URL (span class=”srch-URL”). Within this section, locate the when condition for the hit highlighted properties as shown:

Replace the “hithighlightedproperties/HHUrl” with “clienturl”

Several lines down, replace the “url” in  “xsl:value-of select=”url”" with clienturl (xsl:value-of select=”clienturl”)

Click Save, Click OK on the web part tool pane and publish the page. The client search results should now show the custom client URL that is being returned from the BDC.

XSLT: Creating the Additional Reference Links within the Results

If you perform a Google search and you hit a website’s core address (such as Microsoft.com), there are additional reference links that appear within the search results:

You can add these to the MOSS Search results as well. The reference links can be displayed using standard HTML. This HTML should be placed after the “p class=”srch-Metadata” section (therefore after the p end-tag) :

The “ENTER THE HTML HERE” message will not appear – this was entered for demonstration purposes.

The following is sample code within a <div> block for rendering the additional link references. The appropriate URLs should be retrieved from the BDC and entered appropriately (the columns used here are generalurl, contacturl, documenturl, and referenceurl):

Blank rows or cells are placed for spacing purposes. Additional URL properties will require additional variable entries as entered above (e.g. xsl:variable name=”generalurl” select=”generalurl”)

Here is a sample output:

Conclusion

We’ve done many things with the MOSS Search capabilities but obviously there are many more features and functionality. If you would like to see other search topics covered please send them along to mann@rdacorp.com.

Custom Search Page Creation

The out-of-the-box Search Center actually contains both a Search Page and a Search Results page for All Sites and People. Therefore, to be consistent, we should also create a Search Page and Search Results Page for our new Clients scope.

First we will create the Search Page. Go to the Search Center site and select Create Page from the Site Actions menu:

On the Create Page screen enter Clients for the Title of the page and the URL name. Select the Search Page layout and then click on Create:

The new page will appear in Design mode. Edit the Search Box web part and enter “ClientResults.aspx” in the Target search results page URL setting (under Miscellaneous):

Click OK and Publish the page.

Creating the Search Results Page

The creation of the search results page is the same process. Go to the Search Center site and select Create Page from the Site Actions menu:

On the Create Page screen enter Client Results as the Title of the page and ClientResults for URL name. Select the Search Results Page layout and then click on Create:

The new page will appear in Design mode. Edit the Search Box web part and enter “ClientResults.aspx” in the Target search results page URL setting (under Miscellaneous):

Click OK and Publish the page.

Creating New Tabs

Go back to the main Search Center again and Edit the Page. Click on Add Tab near the top:

This will bring up a screen to add a new tab (for Search Pages). Enter Clients for the tab name and Clients.aspx as the page (remember this is for the search pages, the results are next):

Click Ok and publish the page.

Perform a search under the All Sites scope such that the results page appears. Edit this page and click on the Add New Tab towards the left of the screen:

This will bring up a screen to add a new tab again but this time it will be for the Search Results Pages. Enter Clients for the tab name and ClientResults.aspx as the page:

So now the Search Center should be loaded up with a Client Search Page, Client Search Results Page, and the appropriate tabs.

Updating the Client Scope

Now that we have the client results page created, we need to go back and update the scope such that any search within the client scope uses the proper results page. Therefore, back in Central Administration under the Search Settings, click on the View Scopes within the Scopes section:

On the next screen, select the Clients scope to bring up the scope properties and then click on the Change Scope Settings link:

This will bring up the Edit Scope screen. Simply change the Target Results Page setting to the “Specify a different page for searching this scope” and enter “ClientResults.aspx” within the Target results page box:

Click OK. The scope will need to be updated but that happens automatically.

What’s Next?

We are coming down the home-stretch. Everything is now set-up for the customization of the search results. The final article will go into details in how we can use the managed properties and the results XSL to generate customized search results within our clients scope.

Now that the BDC source has been created and crawled, managed properties must be mapped to the crawled properties such that those columns may be used within the search results (XSLT). The steps in this article need to be repeated for each data element that is to be used within the search results. For scenario purposes, a Client URL is being provided by the BDC and this is the link we want to show in the search results (versus the standard BDC properties page).

From the Configure Search Settings page and select the “Metadata property mappings” link:

On the Metadata Property Mappings page click the New Managed Property button:

On the New Managed Property page, enter a name for the property, an applicable description, and select the type of information (most values will probably be the default of Text). After this has been entered, click on the Add Mapping button:

On the Crawled Property Mapping dialog, select the Business Data category. The list of available properties should appear. Find the appropriate property and select it. (If there are too many properties, enter a part of the name and use the Find button to narrow down the list).

Click OK.  Back on the New Managed Property screen, select the “Allow this property to be used in scopes” checkbox and then click OK.

Repeat for the remaining BDC data elements.

A full crawl will be needed on the BDC content source once the managed properties have been added.

What’s Next?

This was a short and sweet discussion but very necessary to continue. The next effort is to create the custom results page for use within the client search scope and therefore the next article will show the steps to successfully accomplish this task.

The out-of-the-box search features of MOSS 2007 include two scopes, All Sites and People; but what about customers or clients? Many organizations are in business because of their customer base; so when adding new scopes and search content, the customers are the next best bet (pun definitely intended).

Therefore this article will demonstrate the process of creating a Client scope for use in search from a BDC source. It assumes that a Client BDC definition file has been created and imported into the MOSS implementation. The customer/client is simply a business scenario and the steps outlined can be used for any BDC application.

Creating a Content Source for the Client BDC

The first step is to add a content source to the search settings for the Client BDC. From Central Administration, select the SSP under the Shared Administrative Services menu. Select the Search Settings link:

On the Configure Search Settings screen, select the “Content sources and crawl schedules” link:

On the Manage Content Sources screen, select the New Content Source button:

The Add Content Source screen will appear. Enter a name for the content source and select Business Data as the type. The Applications section will refresh. Select the “Crawl selected applications” and make sure the instance checkbox is checked:

Enter the appropriate Full and Incremental Crawl schedules as applicable.

At the bottom of the screen, select the “Start full crawl…” checkbox and then click OK:

The Manage Content Sources screen should appear with the new content source entry:

Before proceeding, make sure the full crawl is completed and the status is Idle.

Adding the Client Scope

Return to the Configure Search Settings page and select the “View Scopes” link under the Scopes section:

On the View Scopes page, click on the New Scope button:

On the New Scopes page, enter a title for the scope. For now, make sure the “Use the default Search Results Page” is selected. We will generate a new search results page in future articles.

Click OK, the screen should return the View Scopes page. The new scope should be listed but with an Empty status. Click on the Add Rules link under this status:

On the Add Scope Rule page, select Content Source. This will make the Content Source section appear. Select the BDC content source created for the search previously. Leave the Behavior to the default “Include” and click OK:

Return to the Configure Search Settings screen and select the “Start update now” under the Scopes section:

The scope should begin the update process. Refreshing the screen will show the % complete on the Update status line. We are done with the SSP settings at this point.

Adding the Scope to the Site Collection

Open a new browser and navigate to the main MOSS web site. Select Site Settings from the Site Actions menus (if there is a extended menu, select “Modify All Site Settings”).

Under the Site Collection Administration section click on the Search Scopes link:

On the View Scopes page, select the Search Dropdown link:

On the Edit Scope Display Group screen, check the checkbox next to Clients scope. Click OK.

Back on the View Scopes page, select the Advanced Search link:

On the Edit Scope Display Group screen, check the checkbox next to Clients scope. Click OK.

Navigate once again to the main MOSS site. The Clients scope should now appear in the search drop-down box:

Searching within this scope should yield results from the BDC.

What’s Next?

Now that the content source and scope has been created, in order to customize the search results, managed properties need to be mapped to the crawled content. The next article will explain the steps to create the managed properties.

Search Service Settings

As explained in the introduction, the server that runs the search service needs to have the Office SharePoint Server Search service started and configured. This is facilitated from Central Administration (CA). From within CA, under Operations, click on Services on server:

This will bring up the Services on Server page. At the bottom of the Services on Server page, the services that are Stopped and Started are shown:

The Office SharePoint Server Search service needs to be started if it is stopped. This can be facilitated by clicking the Start link under the Action column:

Clicking on Start (or on the service name if it is already started) will bring up the configuration page for the search service:

For the scenarios presented in this series, the server will be used for both Indexing and Querying so both checkboxes at the top of the configuration screen will be checked. The contact email address should contain a support address that can be used if there are search-related issues (this is typically used when searching outside of the domain so people can yell at you for crawling their servers and slowing them down). The farm service account should be entered accordingly in the next section of the configuration.

The farm search service account should be a separate domain account that has (or will have) read access to all content that will be crawled by the search service; anything that the service account cannot access will not be indexed and therefore will not appear in any search results.

The index performance determines essentially determines how hard the indexer will work. The crawl process does put strain on the SQL Server that stores the content. The description on the screen somewhat explains this, however, the selection here may depend on the full and incremental crawl schedules (the schedules will be shown in the next section). The default setting is partly reduced so that setting is fine for this series. Once again, the setting here could affect overall performance; this is regardless of which web-front-end it uses to crawl.

Which brings us to the next setting in which all web-front-ends can be used to crawl or a dedicated server may be selected. The selection here depends on how many web-front-ends are in the farm and/or which servers are in a load-balance cluster along with, once again, the crawl schedules.

Search Settings

Once the service has been started on one of the servers in the farm, the next step is determining what will be crawled and how often. Of course, there are out-of-the-box default settings which will be shown here. The remainder of this article series will build upon the default configuration and expand the overall search experience within MOSS.

The search settings are governed by the Shared Services Administration within the SSP that was setup during the MOSS installation process. Clicking on the SSP name will bring up the overall settings.

Clicking on Search settings will show the main search settings screen in which most of the modifications within this series will occur.

Content Sources

The content sources can be seen by clicking on the “Content source and crawl schedules” link from the Search Settings screen:

The screen that follows is the Manage Content Sources page. The out-of-the-box implementation automatically has the “Local Office SharePoint Service sites” content source which defaults to the top-level site collection.

In this series we will add a new content source such that business data can be included within the search environment.

Scopes

Towards the bottom of the Search Settings page lives the Scopes section. Scopes are “views” into the crawled content that can be used to define the boundaries of a user search query (hence the name “scope”).  Clicking on the “View scopes” link will show the View Scopes page.

We will create a new scope for our business data named Clients within this series.

Metadata Properties

The Metadata property mappings link will bring up the Metadata Properties page. From that page, new managed properties can be generated. This will enable us to essential map column names to crawled content such that the search results may be customized.

In this series we will create new managed properties and map them to our BDC crawled content. We will then use those “columns” within the rendering of the search results.

What’s Next?

In the next article, we will add a new content source and scope to the search settings. The content source will be based on a BDC application (i.e. Business Data). THE TRUE CUSTOMIZATION STARTS NEXT SEGMENT – STAY TUNED!

High Level Search Concepts

Essentially, there are two main components of search, the indexing and the querying. The indexing is the process in which all documents and content are reviewed and many key items about the item (meta data) are stored. The querying is when a user wishes to search for an item. The user enters keywords or phrases and the query engine looks at the indexed content and returns items that match.

The indexing process is called “crawling” and the indexed content is called “crawled content”. The querying and the query results (search results) is the end-user experience of the MOSS Search capabilities. Usually in a MOSS farm, there is one server dedicated to perform the indexing and the querying, however, these functions may be split out onto two separate servers. For this series, the assumption is the same server provides both services to the farm.

Office SharePoint Server Search Service

The server that runs the search service needs to have the Office SharePoint Server Search service started and configured. This is facilitated within Operations of the Central Administration (CA) and is managed by the Services on server settings.

The settings for the service will be investigated in the next article.

Search Settings

Once the service has been started on one of the servers in the farm, the next step is determining what will be crawled and how often. Of course, there are out-of-the-box default settings which will be shown here. The remainder of this article series will build upon the default configuration and expand the overall search experience within MOSS.

The search settings are governed by the Shared Services Administration within the SSP that was setup during the MOSS installation process. Clicking on the SSP name will bring up the overall settings.

Clicking on Search settings will show the main search settings screen in which most of the modifications within this series will occur.

This is where the content sources can be defined. From MSDN, a content source is a collection of start addresses representing content that should be crawled by the search index component.

Search Results

The results are based on the “scope” that is used for searching. There are two main out-of-the-box scopes; All Sites and People. A scope carves out a piece of the crawled content and determines what should be searched. It can be thought of a defined view of the crawled content. During this series a new scope will be created for a customized search experience.

The results appear on page within the MOSS system (within the Search Center by default) using a pre-defined web-part page which contains several out-of-the-box search web-parts. The main web part is the Search Core Results and its configuration determines the appearance and layout of the results.

This series will also dive into the customization of the Search Core Results web-part to include additional search fields based on the business data content source.

What’s Next?

The next article will dig a little deeper into the settings and configurations described above. The entire series will build out a scenario of adding content from an external data source (using the Business Data Catalog) and then enabling MOSS to return customized results.

Glossary of Terms

Based on the above discussions, the key terms are defined as follows:

Business Data Catalog (BDC) – an Enterprise feature in MOSS that allows for the retrieval of external data from other systems or databases.

Content Source – a collection of start addresses representing content that should be crawled by the search index component (Source: MSDN)

Crawling – the process at which the search engine retrieves information from a content source and produces appropriate indexes and properties based on the content that is found.

Full Crawl – the process of crawling all of the content available within a content source; the reindexing of previous crawled content may occur

Incremental Crawl – the process of crawling only new content from the previous full or incremental crawl.

Managed Properties / Meta Data Properties – Managed properties are the set of properties that are part of the search user experience, so to include a crawled property value in search functionality, it must be mapped to a managed property in the Document property mappings. (Source: MSDN)

Scope  - search scopes are a collection of items grouped together based on a common element among the items within that scope, which help users broaden or narrow the scope of their searches. (Source: MSDN)

References

Enterprise Search Architecture – http://msdn.microsoft.com/en-us/library/ms570748.aspx