Archive for the ‘c#’ Category.

Role Based Access Control in ASP.Net MVC

Currently I am looking at access control systems, and how best to integrate them with ASP.Net MVC framework. While this framework already provides support for role based access control (RBAC), using the membership classes. I need to implement this on a legacy database, and some how integrate the old system with asp.net forms authentication. This post is about how I realised this, and acts a potential solution. If you can think of a better way, of find any devastating flaws, let me know. ;-)

The scenario is simple, we have four roles defined for the system. They are Students, Graduates, Staff and Administrators. Some staff can be graduates, (or even Students). Administrators are, of course staff! So how you model this? We already know of one bitwise trick from Michal’s post, so let us see how we can use bitwise operations to make this a reality!

First let us revise the results of the bitwise AND operations. You can check Wikipedia for full details.

1 & 0 = 0
0 & 1 = 0
0 & 0 = 0
1 & 1 = 1

Converting these back to decimal 1001 is 9 and 0101 is 5. So 9 & 5 = 8. If we convert each of these bits to represent a role in our system, we can come up with a table like this.

Bit 1 0 (false) Student
Bit 2 0 (false) Graduate
Bit 3 0 (false) Staff
Bit 4 1 (true) Admin

So a user of the system with a role number of 8 is an Admin, but in our case, an Admin is also a member of staff, and in fact, a member of staff could also be a student or a graduate. This is where using bitwise operations can really help model such a situation. To get it working, a staff member who is a student will have bits 1 and 3 set to true, while a graduate who is also a staff member will have bits 2 and 3 set to true. We can represent these roles in decimal as User(Staff & Graduate) = 6, while User (Staff & Student) = 5. Get the picture?

Let’s look at a simple real world example. First we have a User class, with a Role property of the type int. The reason we use an integer, is that is can be easily stored in the database.

    1     public class User {

    2 

    3         public string Name { get; set; }

    4         public int Role { get; set; }

    5         public bool IsInRole(Role role) {

    6             //todo

    7             return false;

    8         }

    9     }

We also need to create an enumeration, with a Flags attribute. The flags attribute tells the compiler that this enumeration can be treated as a bit field. We then define a value for each role. The reason for using exponents of 2 should become clearer later.

    1     [Flags]

    2     public enum Role {

    3         Student = 1,    // 0001

    4         Employer = 2,   // 0010

    5         Staff = 4,      // 0100

    6         Admin = 8       // 1000

    7     }

The menu of our website needs to be generated depending on the user role. The menu selection code below should generate the correct menu depending on the user role.

    1     <div class="LeftMenu"> 

    2 

    3         <% if (user.IsInRole(Role.Student)) %>

    4             <% Html.RenderPartial("StudentMenu"); %>

    5 

    6         <% if (user.IsInRole(Role.Graduate)) %>

    7             <% Html.RenderPartial("GraduateMenu"); %>

    8 

    9         <% if (user.IsInRole(Role.Staff)) %>

   10             <% Html.RenderPartial("StaffMenu"); %>

   11 

   12         <% if (user.IsInRole(Role.Admin)) %>

   13             <% Html.RenderPartial("AdminMenu"); %>

   14 

   15     </div>

Ok, so let see where the magic happens! If we AND (&) the user assigned role, with the role required, and we compare this result to the role required, we can determine if a user is in the role. Summarised, the end result of the AND operation needs to equal that of the role required. In user class we have the method:

    1         public bool IsInRole(Role role) {

    2             Role userRole = (Role)this.Role;

    3             return ((userRole & role) == role);

    4         }

Looking at some binary examples, we can see how it works. In the first example, an admin user wants accesses a graduate item.

Role Required Staff(4) 0 1 0 0
User Role Admin (8) 1 0 0 0
Result of & Access Denied (0) 0 0 0 0


It is clear that we have a problem here, because we said that admin could be both staff, and staff may also be graduates. What we need to do is add up the roles, so that this user will access both admin and staff content. Assigning the user the role of Admin and Staff is easy. All we do is:

    1             User user = new User();

    2             user.Role = (int)Role.Staff;

    3             user.Role += (int) Role.Admin;

And the resulting table is:

Role Required Staff(4) 0 1 0 0
User Role Admin + Staff (12) 1 1 0 0
Result of & Access Granted (4) 0 1 0 0

Now we can easily draw our menu depending on the roles assigned to a user. Adding or removing roles for a user is also easy, just add it or subtract it. I wrote a little project to go with this so you can test it our your self. Thanks to Michi for introducing this, and Dan for helping work it out!

Download the Roles sample project You’ll need to use nUnit to test it.

FluentNhibernate and Stored Procedures

I am evaluating FluentNHibernate (FNH), to see if it is suitable for a project I am working on. Disappointingly, FNH does not support Store procedures of the box. Of course, FNH is under the BSD licence, so I am sure those who are confident enough can implement this for the rest of us! This post will show how I got FNH to work with stored procedures, and can hopefully be followed as a working example.

FNH extends NHibernate, and automagically generates XML mapping files for your objects. Unfortunately, to get stored procedures to work, you need to take a step backwards, and create good old fashioned hbm.xml files, doing the mappings manually.

Firstly , let us look at the results of the stored procedure that we want to map.

ID enDescription cyDescription IsActive
1 Swansea Abertawe True
2 Cardiff Caerdydd True
3 Newport Cas Newydd False

The class that will use this data is called lookup.

The code for this class is:

    1 namespace Entities {

    2     public class Lookup  {

    3         public virtual int Id { get; set; }

    4         public virtual string EnDescription { get; set; }

    5         public virtual string CyDescription { get; set; }

    6         public virtual bool IsActive { get; set; }

    7     }

    8 }

 
This object will be used to populate a simple drop down list, so that a user can select their county.

When I started using FluentHNibernate, I wanted to totally avoid using XML mappings, so I skipped chapters 3 and 6 of Hibernate in Action. My first mistake! So for those attempting this, it may be worth your while understanding Hibernate mappings before you proceed. (You may also ask why I have the Java Book and my code is in C#, that is because I am quite used to working in different programming languages, so those who prefer examples in .Net examples check NHibernate in Action.)

Let’s move on to creating the mapping file.

IMPORTANT: When you add the mapping file to your project, make sure you set the Build Action to Embedded Resource!

I have created a Lookup.hbm.xml file, and the source is below:

    1 <?xml version="1.0" encoding="utf-8" ?>

    2 <hibernate-mapping xmlns="urn:nhibernate-mapping-2.2"

    3                    namespace="Entities">

    4     <class name="Lookup" table="dbo.sp_GetLookups" >

    5         <id name="Id" column="Id">

    6             <generator class="native" />

    7         </id>

    8         <property name="EnDescription" column="enDescription" />

    9         <property name="CyDescription" column="cyDescription" />

   10         <property name="IsActive" column="IsActive" />

   11         <loader query-ref="dbo.sp_GetLookups"/>

   12     </class>

   13 

   14     <sql-query name="dbo.sp_GetLookups" >

   15         <return alias="dbo.sp_GetLookups" class="Lookup">

   16                 <return-property name="Id" column="Id"/>

   17                 <return-property name="EnDescription" column="enDescription"/>

   18                 <return-property name="CyDescription" column="cyDescription"/>

   19                 <return-property name="IsActive" column="IsActive"/>

   20         </return>

   21         exec dbo.sp_GetLookups

   22     </sql-query>

   23 </hibernate-mapping>

 

To put it quite simply, lines 5 to 13 map my Lookup class to the columns in the stored procedure, while lines 16 to 20 map the results from the stored procedure my lookup class. Line 22 names the stored procedure. I am not sure if this is the best way to achieve the mappings, so any feedback would be appreciated.

Once your object is nicely mapped, you then need to update your fluent configuration. All you need to do is tell FNH to load hbmMappings from the current assembly. See the snippet below:

   1    .Mappings(m => {

   2            m.HbmMappings.AddFromAssembly(Assembly.GetExecutingAssembly());

   3            m.FluentMappings.AddFromAssembly(Assembly.GetExecutingAssembly());

   4     })

 

To retrieve the list of lookups, I do the following, which populates my results variable with a list of all my lookups.

    1    var sessionfactory = CreateSessionFactory();

    2    var session = sessionfactory.OpenSession();

    3    var results = session.GetNamedQuery("dbo.sp_GetLookups").List();

 

And that is it, the results variable now contains the list of lookups that I can use to populate my list control.

Top three C# questions on stackoverflow you should read

Here is a list of questions which I (!!) consider as TOP 3 C# (Visual Studio) questions (although the answers are probably the most interesting) on stackoverflow. Yes, I do love STO :)

stackoverflowtop5csharp

  1. Hidden Features of C#
  2. Visual Studio Optimizations
  3. Which C#/.Net blogs do you read?

If you need more see the most voted C# questions. Would be nice to see some more lists like that (Julien, what about Flex?, Fab PHP?).

Nhibernate and the Null object pattern

Is this dialog (exception) familiar to you?

null-object-pattern

Okay, maybe you haven’t seen this screen in particular but apart from the troubleshooting tips you should have seen it.
Continue reading ‘Nhibernate and the Null object pattern’ »

Create a multi languaged domain model with NHibernate and C#

Recently I’ve been developing an enterprise asp.net application with NHibernate which was required to fully support multi language (localization) on UI side as well as on data side. Whereas the former is easily implemented with .net resources the latter is not that straightforward as it seems. That articles talks about it and the solution I have chosen as I have never done it with NHibernate before. Continue reading ‘Create a multi languaged domain model with NHibernate and C#’ »

C# tips, tricks and things you should know

.net components During my previous holidays (i have to admit 1.5 months surfing in portugal) I had some literature with me which also included the book Programming .net components released with O’Reilly. It talks about the aspects of components in general and provides good examples how to achieve those in .net. Although these pages are a very good read (rather for experienced devs) for ppl interested in component development, I found it more interesting because of its general C# content. There are concepts & approaches, small tips & tricks and intersting things about C# which can be used not only in component based architectures and are therefore good to know anyway.

As I always make notes while reading “geek” books I thought: “so why not summarize them all up in a nice list and blog’em”. Here we go… Continue reading ‘C# tips, tricks and things you should know’ »

Singelton application with C#

A singelton application is an application which only allows running one instance of itself per machine. e.g. Microsoft Outlook is a singelton application. Everytime the user starts the application it will check if an instance is already running. If its not running it will launch it otherwise it will focus to the current one. Here is the source of how to create a singelton application with C# and .net … Continue reading ‘Singelton application with C#’ »

TimeRange object for C#

In one of my recent .net projects (C#) I had the need to implement an object which represents a time range which is defined by a start time and an end time. In addition it should detect collision with other time ranges and check if given TimeSpans are within the time range itself. Last but not least it should support parsing time ranges from strings and format the range in a nice way. For this I’ve written a small class which could be useful for some of you out there. Continue reading ‘TimeRange object for C#’ »