Developing on SharePoint 2010 (Day 2)

a) SharePoint Server Architecture

b) SharePoint Object Hierarchy

So as you can see the SharePoint Object model is relatively straightforward. Here we have a very simple slip of code which start from reaching all services of the Farm and navigate through to the item and file level objects.

   1:  foreach (SPService svc in SPFarm.Local.Services)

   2:  {

   3:      if (svc is SPWebService)

   4:      {

   5:          foreach (SPWebApplication webApp in ((SPWebService)svc).WebApplications)

   6:          {

   7:              foreach (SPSite site in webApp.Sites)

   8:              {

   9:                  foreach (SPWeb web in site.AllWebs)

  10:                  {

  11:                      foreach (SPList list in web.Lists)

  12:                      {

  13:                          foreach (SPListItem i in list.Items)

  14:                          {

  15:                              // do something

  16:                          }

  17:                      }

  18:   

  19:                      foreach (SPFile f in web.Files)

  20:                      {

  21:                          // do something

  22:                      }

  23:                  }

  24:              }

  25:          }

  26:      }

  27:  }

Oracle Procedures (version 11.1)

General Related Data Dictionary Objects error$ source$   DBA ALL USER dba_arguments all_arguments user_arguments dba_errors all_errors user_errors dba_object_size all_object_size user_object_size dba_procedures all_procedures user_procedures dba_source all_source user_source System Privileges Related To Procedures create procedure alter any procedure   create any procedure   debug any procedure   drop any procedure   execute any procedure Object Privileges GRANT execute ON <procedure_name>; Privileges to tables and views granted through roles may not be valid within a procedure. See the section on AUTHID. GRANT execute ON testproc TO uwclass;   Stored Procedure No Parameters CREATE OR REPLACE PROCEDURE <procedure_name> IS BEGIN   <code here> END <procedure_name>; / CREATE OR REPLACE PROCEDURE no_param IS BEGIN   dbms_output.put_line('No Params'); END no_param; / set serveroutput on exec no_param; Single IN Parameter CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> IN <data_type>) IS BEGIN   <code here> END <procedure_name>; / CREATE OR REPLACE PROCEDURE in_param (mesg VARCHAR2) IS BEGIN   dbms_output.put_line(mesg); END in_param; / set serveroutput on exec in_param('Single IN Parameter'); OUT Parameter CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> OUT <data_type>) IS BEGIN   <code here> END <procedure_name>; / CREATE OR REPLACE PROCEDURE out_param(mesg OUT VARCHAR2) IS BEGIN   mesg := 'Single OUT Parameter'; END out_param; / set serveroutput on DECLARE s VARCHAR2(50); BEGIN   out_param(s);   dbms_output.put_line(s); END; / CREATE OR REPLACE PROCEDURE out_param (mesg OUT VARCHAR2) IS BEGIN   mesg := 'Single OUT Parameter'; END out_param; / SQL> var x VARCHAR2(30) SQL> exec out_param(:x) SQL> print x IN OUT Parameter CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> IN OUT <data_type>) IS BEGIN   <code here> END <procedure_name>; / CREATE OR REPLACE PROCEDURE inout_param (mesg IN OUT VARCHAR2) IS BEGIN   mesg := mesg || ' an IN OUT Parameter'; END inout_param; / set serveroutput on DECLARE s VARCHAR2(50) := 'This procedure uses'; BEGIN   inout_param(s);   dbms_output.put_line(s); END; / Multiple Parameters CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> IN <data_type>, <parameter_name> OUT <data_type>, <parameter_name> IN OUT <data_type>) IS BEGIN   <code here> END <procedure_name>; / CREATE OR REPLACE PROCEDURE many_params ( mesg1 IN     VARCHAR2, mesg2 OUT   VARCHAR2, mesg3 IN OUT VARCHAR2) IS BEGIN   mesg2 := mesg1 || 'Parameter As The OUT';   mesg3 := mesg3 || 'Returned'; END many_params; / set serveroutput on DECLARE iparm  VARCHAR2(50) := 'This is the IN '; oparm  VARCHAR2(50); ioparm VARCHAR2(50) := 'And This is the IN OUT '; BEGIN   many_params(iparm, oparm, ioparm);   dbms_output.put_line(oparm || ' ' || ioparm); END; /   Parameter DEFAULT Procedure Without Default CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> <data_type> <default_value>) IS BEGIN   <code> END; / CREATE OR REPLACE PROCEDURE no_default(num_rows PLS_INTEGER) IS BEGIN   FOR r IN (SELECT object_name FROM all_objects             WHERE rownum < num_rows+1)   LOOP     dbms_output.put_line(r.object_name);   END LOOP; END no_default; / set serveroutput on exec no_default exec no_default(5); Procedure With Default CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> <data_type> DEFAULT <default_value>) IS BEGIN   <code> END; / CREATE OR REPLACE PROCEDURE with_default ( num_rows PLS_INTEGER DEFAULT 20) IS BEGIN   FOR r IN (SELECT object_name FROM all_objects             WHERE rownum < num_rows+1)   LOOP     dbms_output.put_line(r.object_name);   END LOOP; END with_default; / set serveroutput on exec with_default exec with_default(5);   Parameter Notations Positional Notation exec <procedure_name> (<parameter>,<parameter>); CREATE OR REPLACE PROCEDURE positional ( min_nr PLS_INTEGER DEFAULT 100, max_nr PLS_INTEGER DEFAULT 1000) IS BEGIN   FOR r IN (SELECT table_name FROM user_tables             WHERE num_rows BETWEEN min_nr AND max_nr)   LOOP     dbms_output.put_line(r.table_name);   END LOOP; END positional; / set serveroutput on exec positional; exec positional(1); exec positional(1000, 500000); Named Notation exec <procedure_name> (<parameter_name> => <parameter>); exec positional; exec positional(min_nr => 1); exec positional(max_nr => 500); exec positional(max_nr => 5000); exec positional(max_nr => 10000); exec positional(max_nr => 50000); exec positional(max_nr => 999999); Mixed Notation exec <procedure_name> (<parameter>, <parameter_name> => <parameter>); exec positional(10, max_nr => 1000); exec dbms_stats(USER, cascade=>TRUE);   Parameter NOCOPY Note: NOCOPY is a hint ... not a directive ... and may be ignored. NOCOPY Performance Demo CREATE OR REPLACE PROCEDURE <procedure_name> ( <parameter_name> NOCOPY <data_type> <default_value>) IS BEGIN   <code> END; / -- Note: This first demo shows an example of NOCOPY not working conn / as sysdba GRANT execute ON dbms_crypto TO uwclass; conn uwclass/uwclass CREATE OR REPLACE PROCEDURE default_out (retval OUT VARCHAR2) AS BEGIN   retval := dbms_crypto.randombytes(32); END default_out; / CREATE OR REPLACE PROCEDURE nocopy_out (retval OUT NOCOPY VARCHAR2) AS BEGIN   retval := dbms_crypto.randombytes(32); END nocopy_out; / CREATE OR REPLACE PROCEDURE nocopy_test (reps IN PLS_INTEGER) AS bt1 TIMESTAMP(9); et1 TIMESTAMP(9); bt2 TIMESTAMP(9); et2 TIMESTAMP(9); outval VARCHAR2(64); BEGIN   bt1 := SYSTIMESTAMP;   FOR i IN 1..reps LOOP     default_out(outval);   END LOOP;   et1 := SYSTIMESTAMP;   bt2 := SYSTIMESTAMP;   FOR i IN 1..reps LOOP     nocopy_out(outval);   END LOOP;   et2 := SYSTIMESTAMP;   dbms_output.put_line('Default: ' || TO_CHAR(et1-bt1));   dbms_output.put_line('No Copy: ' || TO_CHAR(et2-bt2)); END nocopy_test; / -- Note: Tried this with REF CURSORS returning 40K rows with, again,  -- no measurable difference. But now watch what happens when used in a  -- demo received from Tom Kyte on 2/9/2007. CREATE OR REPLACE PROCEDURE p3(x IN OUT dbms_sql.varchar2s) AS BEGIN   FOR i IN 1 .. 2000000 LOOP     x(i) := RPAD('*', 255, '*');   END LOOP;   RAISE PROGRAM_ERROR; END; / CREATE OR REPLACE PROCEDURE p4(x IN OUT NOCOPY dbms_sql.varchar2s) AS BEGIN   FOR i IN 1 .. 2000000 LOOP     x(i) := RPAD( '*', 255, '*' );   END LOOP;   RAISE PROGRAM_ERROR; END; / set serveroutput on set timing on DECLARE l_x dbms_sql.varchar2s; BEGIN   FOR i IN 1 .. 1000000 LOOP     l_x(i) := RPAD('*', 250, '*');   END LOOP;   p3(l_x); EXCEPTION   WHEN OTHERS THEN     dbms_output.put_line('count = ' || l_x.count); END; / DECLARE   l_x dbms_sql.varchar2s; BEGIN   FOR i IN 1 .. 1000000 LOOP     l_x(i) := RPAD( '*', 250, '*' );   END LOOP;   p4(l_x); EXCEPTION   WHEN OTHERS THEN     dbms_output.put_line('count = ' || l_x.count); END; / Note: If a subprogram exists with an unhandled exception, the value assigned to its OUT and IN OUT formal parameters are not copied into the corresponding actual parameters, the changes appear to roll back. However, when you specify NOCOPY, assignments to the formal parameters immediately affect the actual parameters. So, if the subprogram exits with an unhandled exception, the (possibly unfinished) changes are not "rolled back." Error Handling With NOCOPY CREATE OR REPLACE PROCEDURE raise_error ( p_Raise BOOLEAN, p_ParameterA OUT NOCOPY NUMBER) AS BEGIN   p_ParameterA := 7;   IF p_Raise THEN     RAISE DUP_VAL_ON_INDEX;   ELSE     RETURN;   END IF; EXCEPTION   WHEN OTHERS THEN     RETURN; END raise_error; / set serveroutput on DECLARE p_B BOOLEAN := TRUE; n NUMBER; BEGIN   raise_error(p_B, n);   dbms_output.put_line(n); END; / CREATE OR REPLACE PROCEDURE raise_error ( p_Raise      BOOLEAN, p_ParameterA IN OUT NOCOPY NUMBER) AS BEGIN   IF p_Raise THEN     RAISE DUP_VAL_ON_INDEX;   ELSE     p_ParameterA := 999;   END IF; EXCEPTION   WHEN OTHERS THEN     NULL; END raise_error; / DECLARE p_B BOOLEAN := FALSE; n NUMBER := 100; BEGIN   raise_error(p_B, n);   dbms_output.put_line(n); END; / DECLARE   p_B BOOLEAN := TRUE;   n NUMBER := 100; BEGIN   raise_error(p_B, n);   dbms_output.put_line(n); END; /   AUTHID Note: For DEFINER RIGHTS objects execute privileges must be granted directly to the user; not to a role. With CURRENT USER rights can be granted to a role. AUTHID DEFINER (default) stored procedures are executed in environment equivalent to the one you get after SET ROLE NONE. In other words, roles are disabled for PL/SQL and any privileges granted via roles do not apply unless you created the procedure with AUTHID CURRENT_USER, in which case role privileges do apply (but executing such procedures is a bit more expensive because Oracle has to evaluate the privileges on every call). AUTHID Demo AUTHID DEFINER (the default) run the procedure with the rights of the procedure's owner. AUTHID CURRENT_USER run the procedure with the rights of the executing schema. conn / as sysdba CREATE USER abc IDENTIFIED BY abc DEFAULT TABLESPACE uwdata TEMPORARY TABLESPACE temp QUOTA 10M ON uwdata; GRANT create session, create table, create procedure to abc; conn abc/abc CREATE TABLE t1 ( mycol VARCHAR2(20)); CREATE TABLE t2 ( yourcol NUMBER(10,2)); CREATE TABLE t3 ( ourcol DATE); conn uwclass/uwclass CREATE OR REPLACE PROCEDURE definer_test AUTHID DEFINER IS BEGIN   FOR rec IN (SELECT table_name FROM user_tables)   LOOP     dbms_output.put_line(rec.table_name);   END LOOP; END definer_test; / CREATE OR REPLACE PROCEDURE cu_test AUTHID CURRENT_USER IS BEGIN   FOR rec IN (SELECT table_name FROM user_tables)   LOOP     dbms_output.put_line(rec.table_name);   END LOOP; END cu_test; / set serveroutput on exec definer_test; exec cu_test; GRANT execute on definer_test TO abc; GRANT execute on cu_test TO abc; conn abc/abc set serveroutput on exec uwclass.definer_test; exec uwclass.cu_test; --====================================== conn abc/abc CREATE OR REPLACE PROCEDURE definer_test AUTHID DEFINER IS BEGIN   FOR rec IN (SELECT srvr_id FROM servers)   LOOP     dbms_output.put_line(rec.srvr_id);   END LOOP; END definer_test; / CREATE OR REPLACE PROCEDURE cu_test AUTHID CURRENT_USER IS BEGIN   FOR rec IN (SELECT srvr_id FROM servers)   LOOP     dbms_output.put_line(rec.srvr_id);   END LOOP; END cu_test; / set serveroutput on exec definer_test; exec cu_test; GRANT execute on definer_test TO uwclass; GRANT execute on cu_test TO uwclass; conn uwclass/uwclass set serveroutput on exec abc.definer_test; exec abc.cu_test;   Procedure Demos Calculate Business Days CREATE TABLE daterange ( beg_date DATE, end_date DATE, biz_days NUMBER(5)); INSERT INTO daterange VALUES (SYSDATE-10, SYSDATE+10, NULL); INSERT INTO daterange VALUES (SYSDATE-17, SYSDATE+10, NULL); INSERT INTO daterange VALUES (SYSDATE-22, SYSDATE+12, NULL); CREATE OR REPLACE PROCEDURE bizdays AUTHID CURRENT_USER IS CURSOR bd_cur IS SELECT beg_date, end_date, 0 FROM daterange; TYPE bDate_tab IS TABLE OF DATE INDEX BY BINARY_INTEGER; bdate bDate_tab; TYPE eDate_tab IS TABLE OF DATE INDEX BY BINARY_INTEGER; edate eDate_tab; TYPE bDay_tab IS TABLE OF NUMBER INDEX BY BINARY_INTEGER; bDays bDay_tab; NumDays PLS_INTEGER; TestDate DATE; BEGIN   OPEN bd_cur;   LOOP     FETCH bd_cur BULK COLLECT INTO bDate, eDate, bDays LIMIT 100;     EXIT WHEN bd_cur%NOTFOUND;     FOR i IN bdate.FIRST .. bdate.LAST LOOP       TestDate := bdate(i);       NumDays := 0;       FOR j IN 1 .. (edate(i) - bdate(i) + 1) LOOP         IF TO_CHAR(TestDate, 'D') BETWEEN '2' AND '6' THEN           NumDays := NumDays+1;         END IF;         TestDate := TestDate + 1;       END LOOP;       bDays(i) := NumDays;       UPDATE daterange       SET biz_days = bDays(i)       WHERE beg_date = bDate(i)       AND end_date = eDate(i);     END LOOP;   END LOOP;   COMMIT;   CLOSE bd_cur; END bizdays; /

Temporary Tables vs. Table Variables

	Temp Table	Table Variable
record transaction logs?	Yes	No
inside transaction scope?	Yes	No
cross execution batch?	Global temp table	No way
support index?	Yes	No
where to store?	Tempdb	Memory

Developing on SharePoint 2010 (Day 1)

This article is talking about developing web parts for SharePoint 2010 using Visual Studio 2010. If you are looking for articles for older version of SharePoint or Visual Studio I recommend this URL: http://www.aspfree.com/c/a/Windows-Scripting/Beginning-SharePoint-Web-Part-Development/

First of all, there are two types of web parts VS 2010 supports: Standard Web Part (which in VS 2010 Item Template list named Web Part) and Visual Web Part.

A standard web part is just a class inherited from abstract class WebPart. Below is the Inheritance Hierarchy of WebPart class.

   1:  WebControl

   2:  |---Panel

   3:      |---Part

   4:          |---WebPart

To add contents into standard web part we need override the CreateChildControls method and inside put code logic to create sub controls manually and add them into the Controls collection. See the code below:

   1:  protected override void CreateChildControls()

   2:  {

   3:      LiteralControl msg = new LiteralControl("test");

   4:      this.Controls.Add(msg);

   5:  }

The visual web part, in the other hand, give you a visual design UI to drag existed web control in. But if you check the source code, a visual web part is just a standard web part plus a Web.UI.UserControl. And in the overrided CreateChildControls method used a trick like this:

this.Controls.Add(Page.LoadControl(_controlPath));

One more thing for visual web part: the value _controlPath is the physical URL of the user control and the value of it is maintained by VS 2010.

How to convert UTF-8 to urlencoded string

As you know, System.Web.HttpUtility.UrlEncode can easily convert UTF-8 string into the format of %xx%xx so it can be used in a URL string. Now imagine you are not working on a ASP.NET project and you don’t want to add reference of System.Web.dll in your project, then how can you do the same thing? Below is one solution I found:

   1: private static string GetEncodedString(string username)

   2: {

   3:     StringBuilder sb = new StringBuilder();

   4:     byte[] btyes = System.Text.Encoding.Default.GetBytes(username);

   5:     for (int i = 0; i < btyes.Length; i++)

   6:     {

   7:         sb.Append(@"%" + Convert.ToString(btyes[i], 16));

   8:     }

   9:  

  10:     return sb.ToString();

  11: }

Now if you pass in chinese string 你好 it will turn out %c4%e3%ba%c3, which is exactly what I want. The reason I create this method is because I am trying to use HttpWebRequest component to directly post data to a web page and I need a way to submit UTF-8 string.

BTW, the only difference between my method and UrlEncode is my method will convert all characters in the passed in string to %xx format even it is ASCII but UrlEncode won’t do that.

Star Schema, Snowflake Schema and others

The star/snowflake schema is a way to implement multi-dimensional database (MDDB) functionality using a mainstream relational database.

Star schema resembles a star, one or more fact tables are surrounded by single level dimension tables (one completely de-normalized table per relationship). Dimension tables aren't normalized - that means even if you have repeating fields such as name or category no extra table is added to remove the redundancy.

Snowflake schema resembles a snowflake because comparing with Star Schema the dimension tables are further normalized or have parent tables.

Either way the fact tables are usually designed in third normal form because all data depends on either one dimension or all of them, not on combinations of a few dimensions.

Star Schema vs. Snowflake Schema

The reason for using a star schema is its simplicity from the users' point of view: queries are never complex because the only joins and conditions involve a fact table and a single level of dimension tables, without the indirect dependencies to other tables that are possible in a better normalized snowflake schema.

Snowflake schema generates more joins than a star schema during, which translates into longer queries. Therefore it is normally recommended to choose the star schema design over the snowflake schema for optimal performance.

Snowflake schema does have an advantage of providing more flexibility.

Disadvantage of Star Schema

A well designed database has to be in 3NF whereas the star schema's dimensions are de-normalized. Normally a huge number of redundant data existed in dimension tables.

For improving query performance many pre-processes (sum, grouping, resort etc) have been done on the dimension’s purpose and saved in the dimension table. When business logic changed to modify the related pre-processes are always hard.

When create new dimension table and build relationship with fact table, it can cause geometrically fact table size increasing.

How to Index Star/Snowflake Schema Data
Declare a primary key in each dimension table.
For star dimensions, declare foreign key (FK) relationships between each dimension table and the corresponding fact table. Additionally, for snowflake dimensions, declare FK relationships between each secondary dimension table and the primary dimension table that it augments.
Make sure that there is a primary key associated with the fact table(s), even if you have to use an IDENTITY field.
Declare indexes on each of the primary keys in the dimension tables and in the fact table(s).
Declare indexes on each of the foreign keys in the fact table(s).

WTF: UAC and rsAccessDenied error in Windows 2008 R2 (x64)

Here is today’s WTF, imaging you spend whole day installed SSRS and finally start creating your first report service, but when you are trying to deploy it you keep receiving error message regarding denied accessing, it is really confused me.

This is not the worst part yet, if you google the error type [rsAccessDenied] you can find the solution is accessing http://localhost/reports/ and assign proper user roles to the report service login user. But when login http://localhost/reports/ in Windows 2008 R2, you can see only a blank home page without any settings you can work with. This is because UAC blocked the dynamic content in this page. Even I added the URL to trusted site list, you can still only see a blank page, which really piss me off.

To overcome this issue, you have to turn off UAC and restart the computer, and then access http://localhost/reports/ you can see all the functionality you need.

Followed are some tips installing SSRS 2008 x64 on Windows 2008 R2 x64

First thing first I strongly suggest installing VS2008, SQL SERVER 2008 in the default path, otherwise you will suffer installation failure soon or later, especially for the Business Intelligence Development Studio.

Another thing you need know is Report Service is no longer relying on IIS. Even you stopped IIS service you can still access Report Service from URL http://localhost/Reports and http://localhost/ReportService.

Assuming you have already installed SSRS 2008 with SQLSERVER 2008, next step is configure database connection. To do this go to Microsoft SQL Server 2008/Configuration Tools/Reporting Service Configuration Manager, then click database page and connect the database server you want. The reporting service manager will create two databases ReportServer and ReportServerTempDB for itself.

Another thing you may need to know is the SSRS report web service asmx URLs are:
http://localhost/ReportServer/ReportService2005.asmx (to manage the service)
http://localhost/ReportServer/ReportExecution2005.asmx (to create reports)
Even you installed MSSQL2008 the services are still named 2005, isn't it kind of WTF. :)

SILVER - An easy way to remember asp.net lifecycle

S – Start
I – Initialize
L – Load
V – Validate
E – Event Handling
R – Render

Haha，really easy to remember.

TransactionScopeOption.Suppress

As you may know, Suppress scope will suppress the current ambient transaction, and if there is no ambient transaction present, it will not create any transaction at all.

But there is one thing you may not know, which is, if there is another TransactionScope further down the stack it will react as if there was no ambient transaction (and create one if necessary).

So if you have 3 scopes as follows: .Required -> .Suppress -> .Required: Then 2 transactions will be created. The innermost TransactionScope with .Required will see that there is no ambient transaction and will creates one for itself.

One thing to remember, to keep the code consistence, always call scope.Complete() at the end of your suppress TransactionScope even you know logically there will be no transaction created. There will be no any overhead caused by doing that.

101 LINQ Samples

No extra words, see the link below from Microsoft which covered all the knowledge you need to create any complicated Linq to SQL query.

http://msdn.microsoft.com/en-us/vcsharp/aa336746.aspx

Diagrams of determining the dimensions of elements

offsetWidth/offsetClient and clientWidth/clientHeight from Mozilla



offsetWidth/offsetClient and clientWidth/clientHeight from Microsoft

ViewData vs. TempData

ViewData and TempData are both used to pass data from Controller into View in ASP.NET MVC model. This article is trying to point out their's difference.

ViewData is only available in current request but TempData, which stored in user session, whose life will be around current and the right next requests. In the other word, if you are trying to share data between two continued request or during a redirection then TempData will be your choice.

BTW, to ensure you get the data by either ViewData or TempData you can use the code
<%=ViewData["name"] ?? TempData["name"]%> 

List<T> working with DataGridView

As you all know you can bind a List<T> into a DataGridView in .net 3.5. The DataGridView even automatically allows you to edit the bound List<T> data. But there are still some tips when working with DataGridView. This article is a quick guide to let you know how to make this kind of generic binding working properly even when you programmatically changing the source data.

Case 1: when you edited the data in a bound List<T>, simply call DataGridView.Refresh() to allow you changes be shown. Or you can set the DataSource of the DataGridView to null and reset the DataSource to the List<T>.

Case 2: when you added or removed records from the List<T> then you cannot refresh the DataGridView to see your changes, in this case you have to re-bind the DataSource (by re-bind I mean set the DataSource to null and then set it back to the List<T> otherwise the re-bind will not happen since the DataGridView will think the datasource didn't change)

Case 3: When you directly bind a List<T> to a DataGridView you will find it doesn't allow you adding or deleting record from the UI even you set the AllowUserToAddRows and AllowUserToDeleteRows. To allow you do these things in DataGridView UI we are going to use another Type BindingSource which implemented related interface to allow DataGridView add/edit/delete data from its data source. The sample code:

List<T> data = new List<T>();
BindingSource binding = new BindingSource();
inding.DataSource = data;
DataGridView.DataSource = binding;

Go back to Case 1 and 2, when you are using BindingSource instead of directly use List<T> now you have another way to refresh your DataGridView, BindingSource.ResetBindings(bool) which forces the DataGridView to reread all the items from the List<T> and refresh their displayed values.

How to add a Form into Tab control

Sometimes for reusing your old code you may want to add a form instead of a user control into a tab. To do so you need modify some properties of your form otherwise your form may not be showing properly.

The properties are:
FormBorderStyle = None; //remove the form border and title area, make it like a flat piece
TopLevel = false; //TopLevel control cannot be embeded
WindowState = Normal; //If you are going to set the Dock value, WindowState cannot be Maximized

The last thing is do not forget call form.Show(), otherwise the form is not going to render itself.

How to set K2 login model for Joomla 1.5

If you are not a patient man just like me then this article gonna help you.

Step1, download K2 full package and install it
Step2, enable K2 login model
Step3, assign existed user to K2 user group
Step4, to let the user publish new items, you have to change the related K2 user group to allow item editing, add items, editing level (All or Own), and publish items

You can ignore all the other settings and log in as the user, and then you can see a link Add New Item below your login user name now.

In case you want to know more about implement K2: A quick step-by-step guide on installing and setting up K2.