What will be optimized version of the given code?

I have a block of code where i am checking checkboxes, comparing the contents of the datatable in the given code:

 foreach (DataRow dr in dtResult.Rows)
 {
    for (var i = 0; i < chkboxListWorkTypes.Items.Count; i++)
    {
      if (chkboxListWorkTypes.Items[i].Value.Equals(dr["WorkTypeID"].ToString()))
      {
         chkboxListWorkTypes.It开发者_开发百科ems[i].Selected = true;
      }
    }
 }

Any labmda or linq expression would be great.

---

Real optimization will be a data binding.

---

This is what I do

 foreach (DataRow dr in dtResult.Rows)
 {
    string cWorkTypeID = dr["WorkTypeID"].ToString();
    for (var i = 0; i < chkboxListWorkTypes.Items.Count; i++)
    {
         if(chkboxListWorkTypes.Items[i].Value.Equals(cWorkTypeID))
           chkboxListWorkTypes.Items[i].Selected =true;
    }
 }

if the ID on the checkboxes is diferent (that probably is),

 foreach (DataRow dr in dtResult.Rows)
 {
    string cWorkTypeID = dr["WorkTypeID"].ToString();
    for (var i = 0; i < chkboxListWorkTypes.Items.Count; i++)
    {
         if(chkboxListWorkTypes.Items[i].Value.Equals(cWorkTypeID))
         {
            chkboxListWorkTypes.Items[i].Selected = true;
            break;
         }
    }
 }

---

Though it does practically the same thing behind the scenes as @Aristos answer, doing it like the following will cut your lines of code down a bit:

foreach (DataRow dr in dtResult.Rows)
{
    ListItem item = chkboxListWorkTypes.Items.FindByValue(dr["WorkTypeID"].ToString());
    if (item != null)
    {
         item.Selected = true;
    }
}

---

You're essentially performing a JOIN operation. There are 3 types of join algorithms in common use - nested loop, hash, and merge. Your code is using the nested loop algorithm - which seems appropriate for table sizes of 10 and 5, unless one or both are already sorted (in which case, a merge join may be more appropriate).

I can't imagine this being a bottleneck in any real-world application - but, we can possibly improve it a little bit with some assumptions.

1. Let's start by hoisting variables, to ensure they're not being recalculated (the JIT-optimizer will probably do this automatically, but it doesn't hurt):

   
    int chkBoxCount = chkboxListWorkTypes.Items.Count;
    foreach (DataRow dr in dtResult.Rows)
    {
       string rowValue = dr["WorkTypeID"].ToString();
       for (var i = 0; i < chkBoxCount; i++)
       {
         var chkBox = chkboxListWorkTypes.Items[i];
         if (chkBox.Value.Equals(rowValue))
         {
            chkBox.Selected = true;
         }
       }
    }
   

2. On the assumption that rowValue is a string, let's use a cast:

   
    int chkBoxCount = chkboxListWorkTypes.Items.Count;
    foreach (DataRow dr in dtResult.Rows)
    {
       string rowValue = (string)dr["WorkTypeID"];
       for (var i = 0; i < chkBoxCount; i++)
       {
         var chkBox = chkboxListWorkTypes.Items[i];
         if (chkBox.Value.Equals(rowValue))
         {
            chkBox.Selected = true;
         }
       }
    }
   

3. Since we're never unselecting a checkbox, and a bool compare should be faster on average than a string compare:

   
    int chkBoxCount = chkboxListWorkTypes.Items.Count;
    foreach (DataRow dr in dtResult.Rows)
    {
       string rowValue = (string)dr["WorkTypeID"];
       for (var i = 0; i < chkBoxCount; i++)
       {
         var chkBox = chkboxListWorkTypes.Items[i];
         if (!chkBox.Selected && chkBox.Value.Equals(rowValue)) 
         {
            chkBox.Selected = true;
         }
       }
    }
   

4. Assuming each checkbox value is unique, we can stop iterating when we've found a result:

   
    int chkBoxCount = chkboxListWorkTypes.Items.Count;
    foreach (DataRow dr in dtResult.Rows)
    {
       string rowValue = (string)dr["WorkTypeID"];
       for (var i = 0; i < chkBoxCount; i++)
       {
         var chkBox = chkboxListWorkTypes.Items[i];
         if (!chkBox.Selected && chkBox.Value.Equals(rowValue)) 
         {
            chkBox.Selected = true;
            break; // stop looking
         }
       }
    }
   

5. Since you have 10 rows, and 5 checkboxes - we should reverse the loops. Since we break on the inner loop when value is found, making the inner loop the larger of the 2 is more efficient:

   
    int chkBoxCount = chkboxListWorkTypes.Items.Count;
    for (var i = 0; i < chkBoxCount; i++)
    {
      var chkBox = chkboxListWorkTypes.Items[i];
      if (chkBox.Selected) continue;
      foreach (DataRow dr in dtResult.Rows)
      {
         string rowValue = (string)dr["WorkTypeID"];
         if (chkBox.Value.Equals(rowValue)) 
         {
            chkBox.Selected = true;
            break;
         }
      }
    }
   

6. Assuming you don't care about culture or case sensitive comparisons, we can use OrdinalIgnoreCase:

   
    int chkBoxCount = chkboxListWorkTypes.Items.Count;
    for (var i = 0; i < chkBoxCount; i++)
    {
      var chkBox = chkboxListWorkTypes.Items[i];
      if (chkBox.Selected) continue;
      foreach (DataRow dr in dtResult.Rows)
      {
         string rowValue = (string)dr["WorkTypeID"];
         if (string.Equals(chkBox.Value, rowValue, StringComparison.OrdinalIgnoreCase)) 
         {
            chkBox.Selected = true;
            break;
         }
      }
    }
   

That's about the best I can do. I'd be surprised if there was any measurable difference on such a small dataset. So, here's what you should really do:

 var workTypeIds = dtResult.Rows.Cast<DataRow>().Select(dr => (string)dr["WorkTypeId"]);
 foreach (var chk in chkBoxListWorkTypes) 
 {
     if (workTypeIds.Contains(chk.Value))
     {
        chk.Selected = true;
     }
 }    

Or, with an Each extension:

 var workTypeIds = dtResult.Rows.Cast<DataRow>().Select(dr => (string)dr["WorkTypeId"]);
 chkBoxListWorkTypes.Where(c => workTypeIds.Contains(c.Value)).Each(c => c.Selected = true);

which may be slightly slower, but reads a lot better IMO.