Microsoft Office SharePoint Server (MOSS) 2007 As an Application Development Platform JANUARY 30, 2007 BY VISHWAS LEE
VISHWAS LELE
Vishwas Lele is Chief Technology officer (.NET Technologies) at Applied Information Sciences, Inc, where he has worked for the last fourteen years. In his current role, he is responsible for assisting organizations in envisioning, designing, and implementing enterprise solutions that are based on the .NET technologies. Vishwas also serves as the Microsoft Regional Director for the Washington DC area. As a Microsoft endorsed expert, he is regularly consulted by clients for his insight and informed perspective on implementing .NET based solutions. A regular industry speaker and author, he has presented at a number industry conferences as well as community user groups.
2 | Microsoft Office SharePoint Server (MOSS) 2007
TABLE OF CONTENTS OBJECTIVE ..................................................................................................................................................................... 4 WHY MOSS AS AN APPLICATION DEVELOPMENT PLATFORM? .................................................................................... 4 APPLICATION LAYERS ................................................................................................................................................... 6 Presentation Layers .................................................................................................................................................. 6 Site Definition ...................................................................................................................................................... 6 End User Customization ....................................................................................................................................... 8 Data Access Layer ..................................................................................................................................................... 9 List and Content Types ......................................................................................................................................... 9 Business Date Catalog ........................................................................................................................................ 13 Shared Services Layer ............................................................................................................................................. 15 Shared Service Provider (SSP) Concept .............................................................................................................. 15 OTHER PLATFORM CHARACTERISTICS ........................................................................................................................ 14 Extensibility ............................................................................................................................................................ 17 Provider Model .................................................................................................................................................. 17 Master Pages ...................................................................................................................................................... 17 ASP.NET Forms ................................................................................................................................................... 18 Custom Virtual Path Provider ............................................................................................................................ 18 Workflow Integration ......................................................................................................................................... 18 Business Intelligence (BI) Integration ................................................................................................................. 18 Toolsets .............................................................................................................................................................. 19 Development...................................................................................................................................................... 19 Deployment ....................................................................................................................................................... 19 Non‐Functional Attributes ...................................................................................................................................... 21 Scalability and Reliability .................................................................................................................................... 21 Localization ........................................................................................................................................................ 23 Consistent Object Model ................................................................................................................................... 23 LIMITATIONS ............................................................................................................................................................... 24 CONCLUSION .............................................................................................................................................................. 24 REFERENCES ................................................................................................................................................................ 25 WHO WE ARE .............................................................................................................................................................. 14
3 | Microsoft Office SharePoint Server (MOSS) 2007
OBJECTIVE
The purpose of this white paper is to present Microsoft Office SharePoint Server 2007 (MOSS) as a development platform for building rich collaborative web applications. First we will discuss the primary reasons and motivations for utilizing MOSS as an application development platform. We will then evaluate the specific features that enable companies and organizations to use MOSS as an application development platform. Finally, we will review and examine some of the fundamental requirements ‐ such as reliability and scalability ‐ which an application development platform must meet. Please note that core concepts described in this white paper are either MOSS features or are features that are inherited from underlying technologies on which MOSS is based, including Windows SharePoint Services (WSS) and ASP.NET 2.0. For a detailed feature breakdown, along with the different SKUs, please refer to reference {1 }. Please also note that in this white paper we will use the term SharePoint to collectively refer to these features. The primary audience for this paper includes architects, IT managers, and consultants involved in building rich collaborative web applications. The secondary audience is technical decision makers who want to make the business case for portals and collaborative web application investments.
WHY MOSS AS AN APPLICATION DEVELOPMENT PLATFORM? Regulatory compliance requires companies and organizations to securely manage records for long periods of time in a safe and unalterable state. The appropriate archival of documents is just one component of a larger records management process that also includes the collection, management, and expiration of corporate records (information important for the history, knowledge, or legal defense of a company or organization) in a consistent and uniform manner in accordance with a company’s corporate governance policies. ASP.NET is a rich platform for building web applications. The latest version (2.0) provides a number of enhancements including: Master Pages that allows for visual inheritance,Webparts that enable end user customizable controls, and Provider Model that allows for integrating custom data stores (as opposed to using AD). Additionally, Microsoft has introduced platforms/products such WSS and MOSS that build on ASP.NET technology to provide higher‐level building blocks such as the document library and lists, end‐user‐defined forms, search, personalization and workflow. The following diagram (Figure 1) illustrates how these technologies stack up. At the bottom of the stack is the Operating System Services layer that includes the ASP.NET 2.0 .NET framework. Above it is the Core Workspace Services layer provided by Windows SharePoint Services. This layer provides services such as content storage, security, administration and navigation. At the top,of the stack lies the Portal layer that provides services such as search, content management and business data catalog, etc. There are two primary motivations for choosing SharePoint as an application development platform: 1. The need to provision more than one website based on a logical grouping ‐ such as department, region or country ‐ rather than have one website that serves all users. For example, a company needs to develop a web application for its partners that allow them access to pertinent sales information. It would be perfectly reasonable to start out with an ASP.NET application, however as the application usage grows, partners would like to customize the site based on their own
4 | Microsoft Office SharePoint Server (MOSS) 2007
unique needs. The partners may want the sales information to surface differently (i.e. grouped by regions vs. grouped by cities), or they may want to co‐locate additional applications on the same page (i.e. a tax calculator). Rather than building all of this personalization in code, it is easier to provision a site for each partner that is based on a single common site blueprint. Each partner can then customize their site based on their specific needs via a single code base. 2. The need to manage un‐structured content (i.e. documents, web casts, etc.) along with the structured data stored in a SQL database. Most modern websites need to manage ever increasing digital content. A distinction between the structured content and unstructured content is that the former deals with data that can be viewed and managed using set‐based groupings (database views), whereas, the latter deals with data that is managed as explicit standalone entities along with the associated metadata. Not only is it important to manage the integrity and security of standalone entities (such as documents); it is also important to manage the relationships among them (for instance, the association relationship between a document and structured application data). Additionally, provisions need to be made to facilitate business process workflow (routing, approval, regulatory compliance, etc.), and document and folder‐level access control and search. Again, rather than building these capabilities into an application, it is far easier to rely on Content Management Services provided by SharePoint.
Figure 1: SharePoint Technology Stack
5 | Microsoft Office SharePoint Server (MOSS) 2007
APPLICATION LAYERS Like any development platform, SharePoint provides platform elements for building a layered application. Let us look at the different layers of a SharePoint based application: PRESENTATION LAYER
Site Definition Any web application is ultimately a collection of static or dynamically generated web pages. For manageability and other reasons, the web pages are typically grouped into logical organizational units such as sites or sub‐ sites. WSS has two core entities, SPWeb and SPSite, which support such an organization. A SPWeb entity is a collection of web pages and can be thought of as a sub‐site. A SPSite is a collection of SPWebs. A sample hierarchy is depicted in Figure 2 below. In this diagram, http://contoso is the root site (SPSite). Underneath the root site there is a collection of applications – Report Center, Blog, etc. – that are nested inside the root site. These are, of course, logical abstractions on top of a physical IIS publishing directory. Multiple SPSites can be hosted inside one IIS publishing directory. A multiple IIS server farm setup, represented by SPFarm and SPWebApplication, is also supported. We will discuss server farms later in this paper.
Figure 2: Site Hierarchy
6 | Microsoft Office SharePoint Server (MOSS) 2007
Together SPWeb and SPSite allow the creation of custom web application topologies by defining what is known as a site definition. A site definition is a blueprint of a web application. For example, a site definition would include a number of SPWebs and how they are nested. It will also include the structure of individual pages – such as navigation, contents and custom code ‐ that are part of a SPWeb instance. Once the template or site definition has been defined and registered, administrators can use them to provision sites. The following diagram (Figure 3) illustrates the concept of site provisioning.
Figure 3: Site Provisioning
The important thing to note in the above diagram is that all the provisioned sites of a certain type are based on one common definition, requiring the development team to manage only one template across all sites. In the event that the site needed to be modified after it has been provisioned, Features allows change to an existing functionality associated with a provisioned site. Examples of modifications include adding a new control on the page; updating the workflow logic; and adding a new action menu. Like SiteDefinition, Features is a collection of XML files that, once registered, becomes available to site administrators. Site administrators can use screens like the one shown in the Figure 4 below to enable or disable registered Features. To make modular provisioning easier, Features can be applied at different scopes of the SPWebApplication, SPSite and SPWeb. For example, a Feature applied at a SPSite scope is automatically available to all nested SPWebs. Feature referencing, wherein a SPWeb instance references a feature installed at the SPSite scope, is also supported. Feature referencing makes it possible to change the feature installed at a higher scope and have that change apply to all nested scopes where the feature is referenced. For advanced scenarios, there is support for setting up dependencies amongst individual features.
7 | Microsoft Office SharePoint Server (MOSS) 2007
Figure 4: Feature Activation
End User Customization In the previous section we saw how the site can be provisioned. Individual pages inside a provisioned site can also be customized using WebParts. WebParts are end user customizable reusable units of UI that implement a well‐known interface. A canonical example is a stock ticker WebPart. An end user can customize the stock ticker WebPart by including stocks of interest. End users with the appropriate permissions can personalize a page by adding or removing WebParts as well by reorganizing the layout of a web page by moving WebParts around. The biggest benefit of WebParts is the ability to leverage all the WebParts that ship with SharePoint [2]. Additionally, there is a thriving market for vendors that sell WebPart libraries. It is easy to develop custom WebParts by deriving from the ASP.NET WebPart class. The SharePoint infrastructure takes care of infrastructure for personalization. This includes the ability to: Allow users to add/remove Webparts and personalize them. Serialize all the Webparts, along with their state, and save them to the content data store. For detailed information on building Webparts please refer to Figure 8.
8 | Microsoft Office SharePoint Server (MOSS) 2007
DATA ACCESS LAYER
List and Content Types One of the most important benefits of using SharePoint as an application platform is the ability to use its storage containers for storing application data, especially in the form of a list ‐ which is the fundamental data structure in SharePoint. There are two kinds storage containers inside SharePoint: 1) Administration Store ‐ This store contains data related to site administration, containing information such as server farm and node setup, etc. ; and 2) Content Store ‐ This store contains data related to site hierarchy as well as the content associated with the site, including user data , layout, menu items, etc. The content database is where most application data will tend to reside. The use of SharePoint storage containers is centered on the notion of list. A list is a collection of items or rows and can be used to store application data. List and List Items are enabled via two core classes SPList and SPListItem. The SPList class represents a generic list of items or rows. While the SPListItem class, as the name suggests, represents a generic item that can be stored in SPList class instance. For example, a helpdesk application can store open request tickets in a list (let’s call it Open Tickets List). Each ticket would represent an individual row instance in the list and can itself be made up of multiple fields. Each field would represent some information about the ticket (the requestor’s name, date created, severity, etc.).SharePoint provides the common column types including (string, datetime, integer, etc.) that can be used to capture the data associated with the ticket. Once the Open Tickets List has been created, not only is the application storage (and the code to access it ) taken care of, the user interface to add, update and delete request tickets is also in place without the need to write any custom code. Of course this is a contrived example. Most applications will need to define fields that represent complex data structures. For example, Open Ticket may need to store the barcode of the equipment along with a picture as one field. To allow applications to define custom fields, SharePoint supports the notion of extensible fields. Extensible fields can be defined using a custom class that inherits from SPField class (or one of the other built‐in classes such as SPFieldChoice, SPFieldUrl, etc.). The inherited field class can define rendering, data validation, etc. Custom fields are useful but are inextricably tied to a list. In other words, there is no way to separate the data schema associated with a custom field from the list it belongs to. Fortunately, there is a notion of content types. Content types encapsulate a data schema in a reusable manner. Continuing with the previous example, a data schema that includes the barcode as well as a description can be defined using a content type. Once the content type is defined, it can be associated with lists to achieve the same functionality as provided by custom fields. Note that it is also possible to associate more than one content type with a list. Any change made to the content type will impact all lists that have associations to it. Content types also support the notion of inheritance. This means it is possible to build a hierarchy of content types as shown in Figure 5. In this figure Document is the base content type. Content types Dublin Core, Form and Custom1 derive from Document. So for example, Dublin Core will inherit the schema characteristics of Document. Any changes to the Document content type will automatically be propagated to Dublin Core and as a result be reflected in all lists that have associations to Dublin Core.
9 | Microsoft Office SharePoint Server (MOSS) 2007
Figure 5: Content Type & Inheritance
In addition to allowing application data to be captured inside the SharePoint data store, lists provide the following additional cross‐cutting features that can be used by applications to implement custom functionality: Alerts allow email notification to be sent out to all subscribers when a change is made to a List or a List item. For instance, this occurs when items are added, deleted or modified. It is possible to consolidate the email notifications based on the frequency (daily, weekly, etc). In addition to email notification, RSS‐based subscription is also available. Versioning on List, as well as on list items, is supported. This means it is possible to have multiple versions of the same documents. In addition, users can obtain exclusive right to modify a document using the check‐in and checkout functionality. Figure 6 shows an example of how versioning can be applied to a document in a list. Version 3.0 of a document is the public version. A new version (3.1) created from version 3.0 is only available to a group (Authors). An individual user can checkout the document exclusively, modify it and check it in to create version 3.2. At a later point in time, a new public version of document (4.0) is published. Auditing allows applications to track changes made to the List. This includes changes such as security changes, check‐in/check‐out operation, updates made to associated content types, etc. Events allow custom event handlers to be fired when a List is changed. For example, Add Item event is fired when a new item is added to the list. Multiple event handlers for one event are supported. Both synchronous as well asynchronous event handlers are supported. Recycle Bin allows deleted list items to be restored. Recycle bin is implemented as a two‐stage process. The first stage allows user restoration of deleted documents. The second stage is a system‐level document restoration.
10 | Microsoft Office SharePoint Server (MOSS) 2007
Security follows a cascading model wherein permissions flow down from a SPSite to a SPWeb to a SPList. Security granularity is an item. Search across one single site or across multiple sites is supported. Fields in the list can be marked as indexed to improve the query performance. Let us consider an example that will help us understand the applicability of lists: Imagine that we are required to build a web page that displays a list of webcast recordings. Users with administration permissions are allowed to upload new webcast recordings, and update (and delete) existing items in the list. Other users can only view existing items in the list. Further, we also have a UI requirement to customize the list rendering such that in addition to the name of the webcast and an icon (that allows users to initiate streaming), the list also allows users to download the associated presentation slide deck as well as view the description of the webcast. One approach for implementing the above requirement would be to define a webcast content type that has fields that correspond to the columns described above. We can then associate the webcast content type with a SharePoint list. Using CAML (Collaborative Application Markup Language) we can customize the UI of the list to meet our requirements. Note that we could have just created a list using site columns instead of defining a content type. The benefit of using a content type is that it is a reusable type that can be associated to other list instances. We can also create new content types that derive from webcast content type. SharePoint uses ASP.NET forms to allow users to insert and update list items. All data associated with the list is automatically stored in the content database. Figure 7 depicts a custom webcast list.
Figure 6: List Versioning
11 | Microsoft Office SharePoint Server (MOSS) 2007
Figure 7: Custom Webcast List
While it may not seem difficult to add a list control on an ASP.NET page and hook‐up some ADO.NET code to persist the data in the database, it should be noted that we have not written a single line of code thus far. We have relied on SharePoint list handling and content database to implement the list. But imagine if we are required to extend the above functionality. For example, it is required that each item of the list be secured individually. Users also want to subscribe to any changes to the list (new recordings) made to the list either via RSS or email. From a QA standpoint, a content management process needs to be enforced when a new webcast recording is uploaded requiring versioning, check‐in/check‐out, and approval workflow. Content management requirement invariably necessitates the ability to maintain an audit trail of changes as well as the ability to undelete an item that was inadvertently deleted. Last but not the least, a search function on the site should include the information about the webcasts. Now with the need for additional features, the custom ASP.NET solution is not easy. Fortunately, all of the above functions are provided by SharePoint List by default. We can even extend the SharePoint List behavior using event handlers. All of the information stored inside a list is accessible, not only via the SharePoint UI, also via the Object Model (Class Library as well as Web Service based OM). This means that processes outside the host process can access the list information – a key to building transparent applications that are reusable.
12 | Microsoft Office SharePoint Server (MOSS) 2007
Business Data Catalog In the previous section we discussed how application data can be stored inside SharePoint storage. In many cases however, the application data needs to reside outside of the SharePoint storage. Consider for example a line‐of‐business application that needs to be integrated into SharePoint. Typically, the line‐of‐business application will need to rely on a dedicated relational store that can meet its specific storage management (physical database schema), performance and query optimization, as well as custom reporting needs. For such applications the Business Data Catalog (BDC) service can be used to integrate data stored in external storage with SharePoint. By integrating external application data using BDC, it becomes possible to leverage built‐in SharePoint functions including the ability to add external data to a list, search, security, versioning, etc. Figure 8 illustrates the architecture of BDC.
Figure 8: Business Data Catalog Architecture
Central to the BDC architecture is the notion of a metadata that defines how the external data source can be integrated into SharePoint. Metadata contains information on how the external data can be accessed. BDC can access the data directly via a database instance or by invoking a web service‐based interface provided by the external system. In BDC terminology, each external application is a System. Discrete objects (roughly nouns) such as customer that are part of external applications are defined as Entities. The operations that can be invoked on Entities are represented as Methods. Methods can be of different types including a Finder method (that return instances of an Entity) and SpecificFinder (that return a specific instance of an Entity). The relationship between Entities, such as master detail, can be modeled as Associations. Figure 9 illustrates the syntax for setting up the BDC metadata.
13 | Microsoft Office SharePoint Server (MOSS) 2007
Once the metadata is completely defined it can be uploaded into SharePoint. At this point the external data can be integrated into SharePoint. For example, it is possible to create a list that uses a BDC Entity Customer as a custom column. It is interesting to note that all of the list semantics, including security, text search, etc., apply to a BDC‐enabled list. Under the covers, SharePoint is making calls to the external system to obtain the relevant data about the Entity. Also note that data is not cached inside a SharePoint list. Up to now we have discussed read‐only access to external data. It is also possible to enable write back to the external system as well. BDC allows Entities to be tagged with Actions. Actions are link to custom form (InfoPath, Web, etc.) that can be used submit information back to the external system.
Figure 9: BDC Metadata Example
Once the metadata is completely defined it can be uploaded into SharePoint. At this point the external data can be integrated into SharePoint. For example, it is possible to create a list that uses a BDC Entity Customer as a custom column. It is interesting to note that all of the list semantics, including security, text search, etc., apply to a BDC‐enabled list. Under the covers, SharePoint is making calls to the external system to obtain the relevant data about the Entity. Also note that data is not cached inside a SharePoint list. Up to now we have discussed read‐only access to external data. It is also possible to enable write back to the external system as well. BDC allows Entities to be tagged with Actions. Actions are link to custom form (InfoPath, Web, etc.) that can be used submit information back to the external system.
14 | Microsoft Office SharePoint Server (MOSS) 2007
SHARED SERVICES LAYER
SharePoint provides a number of building‐block services that are available to SharePoint applications. This includes services such as the Search Service, Single Sign‐on Service, Forms Services and Excel Services. These services can provide higher levels of functionality. For example, Forms Service allows applications to add interactive forms more easily than developing a custom ASPX‐based form. Before we look at examples of services offered by SharePoint, it is important to understand the concept of shared services. Shared services can be setup to run outside the process context of the SharePoint application (in a separate IIS virtual directory). Running the service in a separate process has two benefits: 1) Multiple applications across different topologies can share one instance of the service, thus reducing the overall load, and 2) Shared services can be administered separately from the SharePoint central administration. This means shared service administration can be delegated to users without the need to assign them administration rights on the sites. Figure 10 illustrates the architecture of shared services. There are one or more shared service providers configured. Each provider can in turn enable one or more services that are shared across SharePoint applications. Please note, however, that each application can only access one shared provider. One team site could be accessing a shared provider while another team site could be accessing another provider. A number of factors go into deciding how Shared Service Providers are setup, including the user load, process execution times, and topology of sites. For more information please refer to the capacity planning guide on TechNet [9]. The Shared Service Provider model is extensible and can be used to create a custom shared service by deriving from the SPService class. Shared Service Provider (SSP) concept
Figure 10: Shared Service Provider
Here is a brief description of the services provided: Excel Services Excel Services enable two primary functions: 1) It allows a server‐side rendering of an Excel workbook inside the browser, and 2) It exposes the calculations within a workbook as a web service endpoint. Each of these functions can help minimize the need for custom code by enabling business users to contribute some portion of the application.
15 | Microsoft Office SharePoint Server (MOSS) 2007
Forms Services Forms Services, as discussed earlier in this section, is designed to make it easier to add interactive forms to a SharePoint‐based application. Figure 11 illustrates the Forms Services architecture. An InfoPath designer can be used to design the forms. The completed InfoPath form can then be exposed to the clients by placing them on the SharePoint sever. Based on whether the client is a rich InfoPath client or a browser‐based client, automatically adjusts the rendering. For a browser client, a HTML equivalent of the form is rendered. When the end user submits the completed browser‐based form, XMLHTTP‐based Ajax calls are used to send the information back to the SharePoint server. SharePoint server in turn posts the information to the data source (such as web service end‐point) that the form was originally bound to. The rich InfoPath clients can post information back to the data source directly. The primary benefits of using Forms Services are: the built‐in forms designer, the data submitted via the form is strongly typed XML, and the ability to render forms inside a browser (as a standalone HTML page or as part of an ASP.NET page using the InfoPath user control). Single Sign‐on Service The Single Sign‐on (SSO) service provided by SharePoint allows applications to cache credentials to external systems. Search Service As the name suggests, Search Service enables searches across the contents of the site as well as external data sources. SharePoint‐based applications can incorporate search capability by either leveraging the built‐ in search pages or by programmatically invoking the search web service methods. Using XSLT it is also possible to transform the results returned by the search service.
Figure 11: Forms Services Architecture
16 | Microsoft Office SharePoint Server (MOSS) 2007
OTHER PLATFORM CHARACTERISTICS For a technology to be seen as an application platform, it must meet certain characteristics ‐ Scalability, Reliability and Extensibility [3]. In the following section, we make a case for SharePoint as an application development platform by highlighting and overview of SharePoint’s key features. EXTENSIBILITY
According to Wikipedia [4] “extensibility means that the system is so architected that the design includes hooks and mechanisms for expanding with new capabilities.” One of the major design objectives for SharePoint was to achieve a tight integration with ASP.NET – the core .NET web application development technology. Tight integration with ASP.NET has two important benefits for SharePoint developers: The platform is more approachable as it is based on familiar ASP.NET constructs. For example, SharePoint routing is implemented using ASP.NET Pipeline, a concept very familiar to the ASP.NET developers, and It provides the ability to leverage ASP.NET mechanisms for extending the out‐of‐the‐box capabilities. For example it is possible to use ASP.NET based forms to edit/insert items in the list. In this section we will look at some of the ASP.NET integration aspects in greater detail. Provider Model An ASP.Net Provider [5] is a software module that provides a uniform interface between a service and the data source. Consider the ASP.NET membership service that provides the functions such as login, password management, etc. Rather than directly accessing the membership data source, the membership service interacts with it via a provider. Being squarely based on ASP.NET, SharePoint sites can leverage the membership provider to store login credentials in a custom store in lieu of the default Active Directory‐based membership store. Another example where SharePoint can leverage the Provider model is the role provider. Role provider is used to store authorization information about the users. Please note, however, that SharePoint keeps a copy of membership and role information in its content database. This is how SharePoint maintains permissions (full control, read‐only etc.) and roles (administrator, contributor, etc.) that apply to a SharePoint objects (such as a web site or a list). SharePoint stores this additional information in its content database. Master Pages ASP.NET Master Pages allow web page layouts to be consistent across applications. A single master page can be used to define the look and feel and the standard behavior for a group of pages. The individual pages can then define the content that needs to be displayed. At runtime, ASP.NET merges the master page layout with the individual content pages. SharePoint utilizes ASP.NET master pages to allow a consistent look and feel. Figure 12 provides an example of a SharePoint team site master page. Master pages are a design time concept inside ASP.NET. This means the master page layout of a content page is determined at design time. With SharePoint, it is possible to dynamically pick an alternate master page from a list of master pages stored in the list called Master Page Gallery. SharePoint Master Pages can be edited using a WYSWYG editor such SharePoint Designer.
17 | Microsoft Office SharePoint Server (MOSS) 2007
Figure 12: SharePoint Team Site Master Page
ASP.NET Forms Earlier in the paper, we talked about how content types can be associated with a list. SharePoint provides a default form template for a list that allows users to either edit existing items in the list or to create new ones. Developers can customize the default template by providing an ASP.NET‐based custom form template. As you can imagine, various ASP.NET constructs, such as controls and validators, can be used to build these custom forms. Custom Virtual Path Provider Similar to the provider concept mentioned earlier, ASP.NET 2.0 supports the notion of a virtual path provider. Virtual path provider allows files with ASP.Net extensions (such as .ASPX, .ASMX, etc.) to be loaded from custom store (instead of the defaulting to the file system). SharePoint utilizes this extensibility option to build a SQL‐ based virtual provider. SQL virtual provider allows ASP .Net files to be loaded from the SharePoint content database. This is another example of how a key SharePoint implementation mechanism is based on a service provided by the ASP.NET layer. Workflow Integration Earlier in the paper, we mentioned the need to apply business process (i.e. approval or disposition) to unstructured content (i.e. documents). SharePoint integration with Windows Workflow Foundation (WF) allows this capability. WF is a component of .NET 3.0 that provides a programming model for development and execution of workflow‐based applications. SharePoint utilizes WF to allow workflow to be associated with list
18 | Microsoft Office SharePoint Server (MOSS) 2007
items or documents. For example, it is possible to kickoff an approval workflow when a new document is added to a document library. The approval process can be based on organizational needs (i.e. a single approver vs. multiple approvers). Based on the complexity of the workflow to be implemented, it is possible to choose between VS.NET‐based designer (custom code, multi‐site deployment, etc.), or SharePoint Designer (non‐ custom code, limited to a specific list or document library). SharePoint acts as a host for WF instances. When users need to interact with the Workflow instance (i.e., a workflow initiation screen that allows the user to pick approvers for the approval process), it is possible to do so using ASP.NET forms as well as Forms Services‐based InfoPath forms. Business Intelligence (BI) Integration An important portal requirement is the ability to surface Business Intelligence data. SharePoint allows BI data to be aggregated from different sources such as KPIs (Key Performance Indicators) defined inside SQL Server Analysis Services, reports defined inside SQL Server Reporting Services, and Excel Services based workbooks. The easiest way of achieving this integration is via the out‐of‐the‐box Webparts (a number of third party Webparts provided by BI vendors are also available). These Webparts allow SharePoint pages to consume these sources in real‐time. If the integration provided by Webparts is not adequate, it is possible to create custom consumers using a SharePoint‐defined interface. Additionally, filter Webparts can be used to personalize the information presented to the user. For example, it is possible to connect to an Analysis Services instance and filter the data based on a dimension such as a region or date. SharePoint also provides a BI site template (called the Report Center) to make it easy to setup a BI Dashboard. Toolsets No application development platform is complete without adequate tooling. In this section we will discuss the tools you will use for SharePoint development. Unfortunately, this is an area that needs improvement. For example, many of the common configuration tasks – like installing a custom workflow into SharePoint ‐ require manually creating feature XML, etc. Hopefully future versions of VS.NET will improve the tooling support. Development Visual Studio 2005 is the primary tool for development. As part of the November CTP, Microsoft has released Visual Studio extensions [11] for developing custom SharePoint applications. These extensions include Visual Studio project templates for Web Parts, site definitions, and list definitions; and a stand‐alone utility program, the SharePoint Solution Generator. Regular VS.NET development techniques of attaching to the w3c process or using debug breaks can be used for debugging Webparts and other custom SharePoint assemblies. In this sense, SharePoint development is similar to regular ASP.NET development. Deployment The following options are available for deploying SharePoint solutions: Stsadm Stsadm is a command‐line tool that allows SharePoint objects such as features and sites to be installed and activated. Please refer to the stsadm.exe documentation [6]. All the functionality available via stsadm tool is also available via the SharePoint object model.
19 | Microsoft Office SharePoint Server (MOSS) 2007
Figure 13: Solution based deployment
20 | Microsoft Office SharePoint Server (MOSS) 2007
Solution Packages Using the command line tool such as stsadm is a convenient way to deploy features individually during development; however, it does not lend itself well when deploying multi‐featured solutions to a number of different servers (for testing, staging or production). Solutions can be handy in such situations. Developers can package all the artifacts (site definitions, assemblies) into a solution package. Figure 13 depicts the solution‐based deployment. This figure provides an example of a solution that consists of feature manifests and template files, as well as assemblies. It is possible to define code access security policies that are associated to a solution. This is a way for developers to assert the code access security permissions needed for the solution to run successfully. Once the solution package is registered with the SharePoint configuration databases, administrators can activate them on the servers that make up the server farm. Windows Installer (MSI) In many cases it makes sense to use a hybrid approach that combines multiple solution packages with MSI. This approach allows for combining the power of solution packages with the flexibility of invoking custom actions, as well as ability to take the advantage MSI functions such as rollback.
NON-FUNCTIONAL ATTRIBUTES
Scalability and Reliability According to “Characteristics of scalability and their impact on performance” [7], scalability is a desirable property of a system, a network or a process, which indicates its ability to either handle growing amounts of work in a graceful manner, or to be readily enlarged. To achieve the desired scalability, SharePoint enables a
21 | Microsoft Office SharePoint Server (MOSS) 2007
variety of topology options that allow for increasing its throughput by adding hardware. SharePoint consists of the Web Server Tier, Application Tier, and Database Tier, as depicted in Figure 14 and discussed below. •
Web Server Tier. This tier comprises of one or more stateless web server nodes. These nodes can be load‐balanced using software (Network Load Balancing) or hardware (switch box) schemes. Based on the scalability requirements, additional web server nodes can be added.
•
Application Tier. This tier is a collection of application services such as Excel Services, Search Service, and Project Server. Many of these services (such as Excel Services and Query) can be installed on multiple nodes to improve the throughput, as the web server tier automatically load balances the requests it forwards to the application tier. The other benefit of adding more then one node to the application is to build redundancy into the system. Note that certain application services (such as Index) do not support redundancy.
•
Database Tier. This tier is where the configuration database and all the content databases reside. Based on the scalability needs, content can be broken up into multiple instances to distribute load on the database tier. Clustering or mirroring options are available for supporting redundancy.
Figure 14: SharePoint Tiers
22 | Microsoft Office SharePoint Server (MOSS) 2007
Localization It is quite likely that the SharePoint applications you develop will need to be localized. SharePoint supports the following two mechanisms to achieve this localization: 1) Feature Localization – any XML file in a feature or site definition can be tokenized to allow feature localization, and 2) Language Packs – each solution can define sets of XML resource files resource binaries, etc. SharePoint will then load the appropriate language pack based on the locale of the incoming request. Consistent Object Model Almost all the services enabled via the SharePoint UI are also exposed via the SharePoint object model. Furthermore, the object model is available via the class library for in‐process access, as well as a web service interface for remote access. Figure 15 illustrates a snapshot of web services that relate to the SharePoint store. Services are available that allow for accessing and manipulating data residing inside SharePoint objects such as Sites, Lists, etc.
Figure 15: SharePoint Store Web Services
23 | Microsoft Office SharePoint Server (MOSS) 2007
LIMITATIONS It is important to note that the SharePoint development platform is still evolving, and as a result, there are always limitations. For example, the end user reporting on list data is not easy unless the list data is transferred to another reporting‐enabled data source. Similarly, cross‐site searching across large lists can be expensive because of limited indexing options (for instance, only one column can be indexed). Another area of limitation related to the lists is the transactional update: there is no way to bracket multiple list operations inside an ACID transaction. To address many of these limitations with really large lists, you may want to consider placing the data in an external store and integrat ing it with SharePoint using a service such as the BDC. Finally, a server OS (Windows Server 2003) is required for SharePoint development. This means that developer workstations need to be have a server OS (or use some virtualization techniques). CONCLUSION According to a Microsoft press release [10], over 75 million licenses of SharePoint were sold until May 2006. With the momentum of broad adoption behind it, SharePoint has transitioned from a portal product into a platform for building collaborative web applications. Developers can reap productivity gains by leveraging building blocks such as built‐in list functions, forms and personalization services, and at the same time, have the flexibility to drop down into the ASP.NET layer as needed. Many of the collaborative functions needed by modern websites such as unstructured content management, workflow, and compliance are core platform services. Applications built on this platform will be in position to leverage future enhancements to the platform including an expected closer integration with BizTalk Server and Windows Communication Foundation.
24 | Microsoft Office SharePoint Server (MOSS) 2007
REFERENCES
[1] SKU Comparison Matrix http://download.microsoft.com/download/1/d/c/1dc632e8‐71e1‐466f‐8a2f‐ c940f1438e0a/SharePointProductsComparison.xls [2] To MOSS or not to MOSS, Web Parts may be the answer http://www.wssdemo.com/blog/Lists/Posts/Post.aspx?ID=190 [3]. What You Need To Know About Using Office as a Development Platform http://msdn.microsoft.com/msdnmag/issues/06/08/BusinessApps/ [4] Extensibility http://en.wikipedia.org/wiki/Extensibility [5] Provider Toolkit http://msdn2.microsoft.com/en‐us/asp.net/aa336558.aspx http://download.microsoft.com/download/2/a/e/2aeabd28‐3171‐4b95‐9363‐ 22150625a6a5/aspnet%20provider%20model.pdf [6] Command‐Line Operations http://technet2.microsoft.com/windowsserver/WSS/en/library/f9f9a3eb‐ce46‐4dbb‐a15c‐ 9fad9eb32ec71033.mspx?mfr=true [7] ScalabilityAndré B. Bondi, ‘Characteristics of scalability and their impact on performance’, Proceedings of the 2nd international workshop on Software and performance, Ottawa, Ontario, Canada, 2000, ISBN 1‐58113‐195‐X, pages 195 ‐ 203 [8] A Developer’s Introduction to Web Parts http://msdn2.microsoft.com/en‐us/library/ms916848.aspx [9] Planning and architecture for Office SharePoint Server 2007 http://technet2.microsoft.com/Office/en‐us/library/0a7b2b45‐f633‐46d2‐a4fd‐78691d4b8f631033.mspx [10]SharePoint Server Conference May 2006 http://www.microsoft.com/presspass/press/2006/may06/05‐15SPConference06PR.mspx [11] Windows SharePoint Services 3.0 Tools: Visual Studio 2005 Extensions http://www.microsoft.com/downloads/details.aspx?familyid=19F21E5E‐B715‐4F0C‐B959‐ 8C6DCBDC1057&displaylang=en
25 | Microsoft Office SharePoint Server (MOSS) 2007
WHO WE ARE
For 25 years, Applied Information Sciences (AIS) has provided software and systems engineering services to government agencies and businesses around the world. Our analysis‐driven approach to business problems combined with our commitment to deadlines and budgets results in successful projects and long‐term relationships with our customers, employees and partners. For 13 years AIS has been a Microsoft Managed Gold Partner, the highest partner relationship offered by Microsoft. Additionally, several key AIS employees have been selected to be members of Microsoft Partner Advisory Councils. Joining a select few from around the world, these council members have the opportunity to review and provide feedback for Microsoft technologies and products. This unique partner status results in special product trainings and technical readiness programs years ahead of their market release, enabling AIS to help our clients prepare with the long‐term planning of their technology evolution. AIS, Inc. is headquartered in Reston, Virginia with additional offices in Columbia, Maryland and Austin, Texas. Contact Us
Corporate Headquarters 11400 Commerce Park Drive Suite 600 Reston, VA 20191 Phone: 703.860.7800 Fax: 703.860.7820 Central Regional Office 7718 Wood Hollow Drive Suite 150 Austin, TX 78731 Phone: 512.651.5220 Fax: 512.651.5241 Contact Information by Department Sales Phone: 703.860.7815
WWW.APPLIEDIS.COM
26 | Microsoft Office SharePoint Server (MOSS) 2007
800‐AIS‐4553