Saturday, June 4, 2011

Tweener lists

 <style type="text/css">
        li label     
            background-color:White ;    
  li label:hover {    font-weightbold;    color#FFFFFF;    background-color#999999;    border-styleoutset;}
<input id="PersonIdTextBox" value="123" type="hidden" />
<input id="addSomething" title="add" type="button" value="add" />
<div >
  <ul class="divisions">
   <li data-personid="10" data-somethingid="20">
    <label class="removeSomething">remove</label>
    Something 1
      <li data-personid="200" data-somethingid=200">
    <label class="removeSomething">remove</label>
    Something 2
     <script type="text/javascript">
         $("#addSomething").click(function () {
             var personId = $("#PersonIdTextBox").val();
             var somethingName = "Something 3";
             var somethingId = "432";
             var removeLabel = '<label class="removeSomething">remove</label>';
             var listItem = '<li data-personid="' + personId + '" data-somethingid="' + somethingId + '">' + removeLabel + somethingName + '</li>';
         $(".removeSomething").live('click'function () {
             var personIdSelector = $(this).parent().data("personId");
             var somethingIdSelector = $(this).parent().data("somethingId");
             alert("here" + personIdSelector + " " + somethingIdSelector);

Monday, May 30, 2011

Scenario management for .ashx handlers.

    /// <summary>
    /// Scenario has three parts
    /// 1. A test that returns true or false
    /// 2. A method to run if the test is true.
    /// 3. A response object that has some details such as error messages.
    /// Benefits, your main entry point to the application is clean.
    /// The scenarios should be of a relatively similar type, this is not a service locator.
    /// For example the file up loader has three different save types, so there are three different scenarios.
    /// </summary>
    public class ScenarioHttpContext
        public Func<HttpContextbool> Test { getset; }
        public Func<HttpContextScenarioResponse> DoWork { getset; }
        public ScenarioHttpContext(Func<HttpContextbool> test, Func<HttpContextScenarioResponse> doWork)
            Test = test;
            DoWork = doWork;
        public static List<ScenarioResponse> RunActions(List<ScenarioHttpContext> scenarios, HttpContext context,bool runAll = false)
            Contract.Requires<ArgumentNullException>(context != null);
            Contract.Requires<ArgumentNullException>( scenarios != null);
            Contract.Requires<ArgumentOutOfRangeException>(scenarios != null || scenarios.Count > 0, "Cannot be null or have zero list items.");
            Trace.WriteLine("Starting ->" + MethodBase.GetCurrentMethod().Name); 
            var resp = new List<ScenarioResponse>();
            //if the test returns true then run the method
            foreach (var scenario in scenarios.Where(scenario => scenario.Test(context)))
                Trace.WriteLine(scenario + "Was called.");
                var a = scenario.DoWork(context);
                a.Name = scenario.ToString();
                if (!runAll) { break; }
            Trace.WriteLine(string.Format("A total of {0} scenarios evaluated true and ran.", resp.Count.ToString()));
            Trace.WriteLine("Ending ->" + MethodBase.GetCurrentMethod().Name);
            return resp;
    public class ScenarioResponse
        public string Name { getset; }
        public bool Success { getset; }
        public string ErrorMsg { getset; }
    public static class ScenarioHttpContextExtensions
       /// <summary>
       /// </summary>
       /// <param name="scenario">Unit of work</param>
       /// <param name="test">If statement to determine if this scenario should be run.</param>
       /// <param name="workToDo">The work to do if the scenario test is true</param>
        public static void AddNew(this List<ScenarioHttpContext> scenario ,Func<HttpContextbool> test, Func<HttpContext,ScenarioResponse > workToDo)
            Contract.Requires<ArgumentNullException>(scenario != null);
            scenario.Add(new ScenarioHttpContext(test, workToDo));

Saturday, May 21, 2011

jquery Autocomplete with previous dropdown

The html is simple we have a  dropdown box, ddlist and a textbox called textbox.
Please double check the names, I changed them in notepad.

This solution also fixes the issue where the dropdown list item is cashing the selected value from page load.

The first function erases the value of the textbox when the dropdown list changes.

$("#ddlist").change(function () {
    alert($('#ddlist option:selected').text());
This function kicks off the autocomplete process when the user clicks in the box.
If the value of the ddl is not good, then throw up a hint. 
$("#textbox").live("focus"function () {
    var marketVal = $('#ddlist option:selected').text();
    if (marketVal != '') {
        var sourceUrl = 'mypage.ashx?action=dosomething&ddlistName=' + $('#ddlist option:selected').text();
            source: sourceUrl,
            minLength: 2
            //select: function (event, ui) { $("#SubMarket").val(; }
    } else {
        alert("Please choose a Market item first");

Tuesday, May 3, 2011


Such a simple, but commonly over looked function.  When writing inline sql, (yes, it still happens), try to use string.format(...) vs. "text" + "text" + someVar.

I wish I had time to post samples of both, so you can see the readablity difference.

Wednesday, March 16, 2011

After listening to a podcast on Simple.Data I decided to check it out for a small project I was working on.
I like it and it works well for small sites that need an active record patten and more.

My first piece of advice is to use NuGet to add the libraries to your solution.  I don't think the download on the site gives you the latest version. Second,  understanding the dynamic type in C# 4.0 is helpful.

See Mark's wiki to see how works.

I had a design issue and here is how I worked around it.  I had properties in my class that were not defined in my database.  There may be an override to stop the errors.  My solution was to create a private class that was the same name as my table that only had the properties from the database.  Then I created a very simple mapping method. I could have used something dynamic, or named arguments, but I figured with a private class in my repository would give me flexibility to do things like datetime corrections and sure enough that happened.

Why go through all that?  Why not use NHibernate and be done. My application is going to be plugged into a CMS system and I do not know if the NHibernate session would play nicely with hosting package and I don't have time to find out.