Edgewaters

Windows 8 in a VM

2011-12-01T00:02:00.003-05:00

I finally got my VM NAT working thanks to the comments in an installation blog post. The key was changing the Adapter Type (Network - Advanced) to Intel PRO /1000 MT Desktop. As soon as I restarted the VM, it recognized the adapter.

Windows 8 on my Dell Duo has COMPLETELY REVOLUTIONIZED my tablet. With Windows 7, the Duo was better in "netbook mode" but now I rarely flip to the keyboard. The split on-screen keyboard is a huge improvement and the pinch-zoom on IE 10 is FANTASTIC.

Supported Extensions for CRM 2011

2011-09-07T11:20:00.001-04:00

I thought it was worth highlighting the official Do’s and Don’ts of CRM extension.

It seems obvious, but no modifying the physical database schema is permitted (except indexes):

Modifications to the physical schema of the database, other than adding or updating indexes. This includes any actions performed against the database without going through System Customization. Modifying tables, stored procedures, or views in the database is not supported. Adding tables, stored procedures, or views to the database is also not supported because of referential integrity or upgrade issues. Adding indexes is supported per the guidelines in the “Microsoft Dynamics CRM 2011 Implementation Guide”. This applies to all Microsoft Dynamics CRM databases and the Microsoft Dynamics CRM for Microsoft Office Outlook local database.

One infraction that I have seen cause disastrous results is saved for last:

Editing a solutions file to edit any solution components other than Ribbons, Forms, SiteMap or Saved Queries not supported. Defining new solution components by editing the solutions file is not supported. Editing Web Resource files exported with a solution is not supported. Editing the contents of a managed solution is not supported.

TFS Build Service Tips

2011-08-17T11:38:00.000-04:00

My organization is hosting TFS for multiple customers. To keep everyone isolated (and happy) we have multiple Team Project Collections, but the downside is we need to setup a separate Build Controller for each Site Collection (see MSDN article for more pros/cons). TFS 2010 changed to only allow a 1 to 1 pairing between a TFS Collection and a Build Controller.

Setting up a Build Controller is fairly straightforward. Here are some tips that I have found helpful:

1. Create an AD account for the service. It’s easier to grant permissions to output folders and network shares using named users/groups. Do this before running the Build Controller wizard – it will add the necessary groups and permissions for you.

2. Run the TFS Setup and only select “Build Controller” for the install.

3. Use the Build Controller wizard at the end of the install to connect to your Team Project Collection.

4. If Visual Studio is not installed on the Build Controller, copy the MS Build targets from a developer machine to:
C:\Program Files (x86)\MSBuild\Microsoft\VisualStudio\v10.0

5. Add a Build project to copy our compiled output to an integration server (see example file below)

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

</ItemGroup>

<Copy

SourceFiles="@(MySourceFiles)"

DestinationFolder="\\<Network Share>"

</Target>

</Project>

CRM ideas and products I discovered at WPC 2011

2011-07-20T09:45:00.001-04:00

CRM Solutions

There are some new change management options available in CRM 2011. With Managed Solutions, you can achieve a layered effect where one solution builds on the next and it is even possible to isolate and rollback individual solutions. This is particularly important for teams that have multiple simultaneous development projects proceeding at different rates/release dates. There is a good discussion of the implications of solutions as well as fundamentals of the entire CRM development spectrum in the new CRM Developer’s Training Kit. This is tricky stuff, so make sure you get it working in a “sandbox” before trying this in your main Dev branch.

CRM Marketplace

I just attended the Microsoft Worldwide Partner Conference where I had the chance to talk to a number of ISVs and Microsoft engineers. There are a lot of cool products emerging around the CRM platform. There are two products that I am particularly keen to try out: NeuDesic’s Pulse and ADX Studios portals.

Pulse is a way to bring social-like conversations into CRM and SharePoint and looks like it would have fantastic application wherever team collaboration is going on.

ADX Studios (makers of the LINQ based SDK that Microsoft ships with CRM) has a number of ready-to-run public facing portals that have built-in claims based authentication to re-use Facebook, Google, and Live accounts.

Plugins, Workflows and Dialogs

Finally, I just came across this helpful CRM team blog post that outlines the best time to use a plugin vs workflow.

Synchronizing CRM Records in Outlook using Local Data Groups

2011-07-06T15:52:00.000-04:00

My customer wanted to have all the Contacts owned by their Business Unit synchronized to Outlook and that is easily accomplished using the Local Data Groups feature!

Bing Map Browser for CRM 2011

2011-07-01T17:39:00.001-04:00

There is a very usable Silverlight Bing Map Browser on Codeplex. The documentation is a little light so here are a couple of tips on setup:

There are two different HTML resources

1. LocateAddress – this is used on a form like Contact to find the Lat/Lon of an address

2. BingMapBrowser – this is the control that maps multiple records based on the FetchXML you configure

Mapping a Contact

I followed the “configure entity form” instructions and added the web resource in a section on a new Contact form. One thing that was confusing to me, the control only works with Lat/Lon BUT it will do an address lookup and enter Lat/Lon on your Contact form if you press the “FillLatLong” button.

Mapping multiple records

I left the default BingMapConfig.xml settings and the browser mapped all my Contacts and Accounts. You can test it by clicking on the link in the URL property of the BingMapBrowser web resource. To add a link in the left nav, you need to export and modify your Sitemap (WARNING: always keep a backup copy). I added “Maps” to the Workplace bar:

</Titles>

</SubArea>

Cool OneNote feature

2011-06-29T14:37:00.001-04:00

I was already very impressed by OneNote, but I am blown-away after stumbling upon the built-in OCR capability. When I did a content search for “Attendee” it not only found the image, it even highlighted the word on the image!

Removing the Dell DataSafe Upgrade Discount pop-up

2011-06-27T16:30:00.003-04:00

I was baffled by a Dell product pop-up that did not appear to have a way to disable/ignore it...until I unlocked the taskbar and resized it. After that, the entire ad was displayed, including the option "Don't remind me of dicount offerings". Now if I can just get Dell to reimburse 30 minutes of life.

Mail Merge Templates No Entity error

2011-06-22T12:05:00.001-04:00

I removed an entity that we were no longer using, but the Mail Merge Templates were still there. CRM throws an error when there are templates that are missing their entity. I discovered this by running the CRM Diag Tool and the CRM Trace Log Viewer.

To fix this, I had to manually set the records to “deleted” in the CRM database (DeletionStateCode=1). Lesson learned: remove Mail Templates before deleting an entity!

Oracle regexp_replace function

2011-06-02T16:41:00.001-04:00

I needed to remove non-ASCII characters from an Oracle column and this regular expression did the trick nicely:

update TABLE set COLUMN = regexp_replace( COLUMN,'[^a-z,_, ,A-Z,0-9,@,&,$,:,.,%,'']','');

CRM 4.0 SRS Connector "Same Account" Setup Error

2011-05-20T15:13:00.001-04:00

After performing my steps for troubleshooting SSRS issues in CRM, I realized my CRM environment was having a double-hop error saying

Cannot create a connection to data source 'Portal'.

I was getting an error while trying to install the CRM SRS Connector:

A Microsoft Dynamics CRM Server component is using the same account as the instance of SQL Server Reporting Services

My SSRS Service was running as Network Service. I changed it to run as an AD user account that has database permissions and that took care of the setup error and SRS connector completed.

After changing the CRM SSRS Data Source I got an error saying:

Then I remembered my CRM was on Rollup 11, so I installed the SRS Rollup 11. After an IISRESET my reports finally came up.

Rendering HTML in SSRS 2008 Reports

2011-05-10T11:59:00.000-04:00

Some of the data in my report was stored with simple HTML formatting tags. It turns out that the expressions I had on my report was already a “placeholder” and SSRS Placeholders have the ability to set the Markup Type to “HTML – Interpret HTML tags as styles”.

Note: I had to click THREE times on the expression to drill past the Rectangle and Text Box containers to finally select the expression placeholder.

SOLVED: Connecting to Team Foundation Server 2010 (TFS 2010) from Visual Studio 2008 (VS2008)

2011-03-22T11:45:00.001-04:00

It took two different forum posts for me to get this working, so this deserves permanent blog recognition. The key was that Visual Studio 2008 doesn’t have a concept of TFS “Collections” so you have to help it by providing the URL to the exact collection you are connecting to. Not sure what the URL is? Open up the collection in VS 2010 and copy the URL property in the Collection Properties. Here are the steps from the forum post:

To connect to TFS 2010 using VS2008, you need to install the below components in mentioned order. The Installation order is really important to make it work.

- Install Visual Studio 2008
- Install Team Explorer 2008
- Install Visual Studio 2008 SP1 (This will upgrade both VS2008 and Team Explorer 2008 to SP1)
- Install VSTS 2008 Forward Compatibility Update

As TFS2010 introduces the notion of Team Project Collections. After the installation is completed successfully, you should provide the entire URL to connect TFS2010 server.

You can express the connection string as follows:

http://<serverName>:<port>/<vdir>/<collectionName> (e.g. http://TfsServer:8080/Tfs/ProjectCollectionName )

The <vdir> is an optional path for the TFS web sites specified by the administrator during setup. By default it is “tfs”.

Hiding Custom ISV.Config Buttons in CRM 4.0

2011-03-10T13:13:00.004-05:00

I needed to hide/show my ISV.config buttons (note the ToolTip is the key here):

 <Entity name="new_myentity">  
  <ToolBar>  
   <Button AccessKey="P" Icon="/_imgs/button_approve.ico"   
     ValidForCreate="1" ValidForUpdate="1"   
     JavaScript="ApproveApplication();">  
    <Titles>  
     <Title LCID="1033" Text="Approve" />  
    </Titles>  
    <ToolTips>  
     <ToolTip LCID="1033" Text="Approve"/>  
    </ToolTips>  
   </Button>  
  </ToolBar>  
 </Entity>

So I found some sources that led me to this function that I call in the Entity’s OnLoad event:

 // hide ISV buttons that don't apply

 function hideISVButton(buttonTitle) {
   var comps = document.getElementsByTagName('li');
   for (var i = 0; i < comps.length; i++) {
     if (comps[i].title == buttonTitle)
     {
       comps[i].style.display = "none";
       break;
     }
   }
 }

***Interesting to note about the ISV.Config JavaScript…it calls a function that is also defined in the Entity’s OnLoad…a much cleaner alternative to real JS in the ISV.Config file:

 ApproveApplication = function() {
 
  // Your function code here
 
 }

Fun with Enums: Calculate next Monday

2011-03-09T13:17:00.001-05:00

I needed to calculate a couple of interesting dates for the Notary system.

Next Monday:
DateTime.Now.AddDays(8 - (int)DateTime.Now.DayOfWeek)

The last day of the birth month, 4 years from now:
new DateTime(DateTime.Now.Year + 4, dobDt.Month, 1).AddMonths(1).AddDays(-1);

Scribe Console Access Denied Error

2011-03-01T13:59:00.001-05:00

Today I tried to create an Integration Process for a new DTS I had just finished, but every time the Console tried to access the file system I got an “Access Denied” error. Turned out that restarting the Scribe AdminServer restored whatever permissions the Console needed.

Date Validation is SSIS

2011-02-16T12:28:00.000-05:00

I needed to import a large amount of data from an Access database and the standard SQL Server Management Studio would fail when it ran into a record with a bad date. I ended up saving the import as an SSIS package (file system) and then editing the DTSX file. I was able to override bad dates to Null by adding a Derived Column with a date validation that replaced the existing column in the dataset.

Full SSIS validation expression:

((DT_Date)Date_to_validate < (DT_DATE)"1753-1-1") ? NULL(DT_DATE) : (DT_Date)Date_to_validate

CRM Outlook Client Error "input string was not in a correct format"

2011-02-10T15:21:00.001-05:00

I ran into an issue with the CRM Outlook client configuration wizard that turned out to be a quick registry edit:

Set sPositiveSign to “+” at

HKEY_CURRENT_USER.Control Panel.International

Make sure to update the Default and all profiles under HKEY_USERS if the person running RegEdit is not the actual user (e.g. a desktop support person)

Automate your CRM Plugin deployment using the PluginDeveloper

2011-02-02T10:08:00.003-05:00

Automating the Plugin deployment process has helped our team reduce errors and speed
up the development and debugging of our custom plugins and workflows. Using
a combination of the PluginDeveloper included in the SDK (<sdk install>\tools\plugindeveloper\),
a batch file, and
configuring an external tool in Visual Studio allows our CRM build manager to deploy
plugins to Dev, Test or Production directly from Visual Studio.

Our team uses separate register.xml files for Dev, Test and Prod that are
set in Visual Studio to Copy always in the Copy to Output Directory
in the Visual Studio’s File Properties window. The batch file copies
the project’s output directory to the PluginDeveloper’s directory and
prompts the user to select which environment to deploy to.

Build the PluginDeveloper Executable

1. Follow the Readme.doc in the project root.

2. Copy the RunPD.bat file to the PluginDeveloper.exe’s
location (we use a common file system path \CRM\tools\PluginDeveloper)

Configuring Visual Studio’s External Tools

1. Select the Tools – External Tools …

2. Click Add and enter the details for the batch file

a. For Arguments specify $(BinDir) for select Binary
Directory from the options

Create register.xml Files for Each CRM Environment

1. The batch file expects to find files named:

a. DEV register.xml

b. TEST register.xml

c. register.xml

Running the Tool from Visual Studio

1. Select Tools – Run Plugin Developer

2. At the menu, type 1,2 or 3 for the desired environment and
press Enter

3. Enter password when prompted

*Credit goes to DosTips.com for
the menu framework

Fool Proof Data Imports in CRM

2011-01-28T10:07:00.000-05:00

Sometimes it can be a challenge to get the data maps to work properly, but using the “Automatic” map is almost foolproof. Follow these basic steps:

1. Create an advanced find with all of the entity’s “custom columns” (ignore CRM columns like created by, status, etc)

2. Export the results to Excel and remove the hidden GUID column (Column A)

3. Using the exported data as a guide for picklist values, insert the new data

4. Run the Import wizard. You should see “Automatic” in the Data Map

CRM 4.0 Required TCP Ports

2011-01-20T12:49:00.001-05:00

I was trying to install CRM on a Server 2003 connecting to SQL 2008 and ran into this error:

The warning message:
|Warning| Check SqlServerAgentValidator : Warning: Setup was unable to verify that SQL Server Agent (SqlAgent$crm) was running.

The error message:
| Error| Check CrmSqlDomainValidator : Failure: The SQL Server ‘{0}’ is unavailable.

Windows Server 2003 requires TCP/UDP port 445 to be open between CRM and the Database. Once the firewall was opened, the installation went off without a hitch. In my experience with Server 2008, this port was not required. Microsoft has a comprehensive port list, but my list is below:

Ports I request for CRM on Server 2008 installs:

TCP 1433

TCP 135

Ports I request [now] for CRM on Server 2003 installs:

TCP 1433

TCP 135

TCP/UDP 445

Using CRM's TargetRetrieveDynamic to retrieve a DynamicEntity via Guid

2011-01-20T09:19:00.001-05:00

This is a snippet from the SDK example: <sdk loction>\sdk\server\reference\cs\crmservice\dynamicentityretrieve.cs

// Create the retrieve target.

TargetRetrieveDynamic targetRetrieve = new TargetRetrieveDynamic();

// Set the properties of the target.

targetRetrieve.EntityName = EntityName.contact.ToString();

targetRetrieve.EntityId = created.id;

// Create the request object.

RetrieveRequest retrieve = new RetrieveRequest();

// Set the properties of the request object.

retrieve.Target = targetRetrieve;

retrieve.ColumnSet = new AllColumns();

// Indicate that the BusinessEntity should be retrieved as a DynamicEntity.

retrieve.ReturnDynamicEntities = true;

// Execute the request.

RetrieveResponse retrieved = (RetrieveResponse)service.Execute(retrieve);

// Extract the DynamicEntity from the request.

DynamicEntity entity = (DynamicEntity)retrieved.BusinessEntity;

Troubleshooting CRM SSRS Errors

2011-01-15T09:24:00.001-05:00

I had a perfectly happy production CRM environment humming along nicely…until I tried to install a second CRM instance that used the same SSRS server. After the install (and rebooting the SSRS server), my production reports stopped working and the new CRM’s reports we’re not working either.

I went through an episode of this w/ MS Premier Support during the first CRM install and I remembered some of the troubleshooting steps we took back then.

The first thing I tried was to enable remote errors in SSRS. That only helped a little because CRM still displayed the friendly error page, “Report cannot be displayed”. To have CRM give me the detailed error message I had to set the DevErrors key to “true” in the CRM Web.Config and now I can see the error. After adding the SSRS service account to my new CRM’s AD Group PrivReportingGroup and restarting the SSRS service, my “new” CRM reports started working, but the old ones were still broken. Opening Regedit on the SSRS server to HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\MSCRM I can see my configdb key is pointing to the wrong server (this is what the CRM Data Connecor uses). When I changed that key back to my production database instance (similar to the procedure for moving a CRM database) and rebooting my SSRS server, the reports worked again.

Summary:

Step One – Enable Remote Errors on SSRS

Step Two – Change CRM Web.Config DevErrors key to “On”

Step Three – Check the Active Directory groups (Start -- Run -- DSA.MSC )

Step Four – Check Registry Keys

Microsoft support confirmed my predicament. To do what I want (two CRMs on one SSRS server) I need to install a second SSRS instance.

Setting the Primary Key of a Custom Entity in a Plugin

2011-01-07T17:17:00.001-05:00

I needed to set the primary key Guid on a custom entity in the Pre stage of a Plugin. It required adding a KeyProperty to my DynamicEntity.

Guid requestId = Guid.NewGuid();

Key newReqKey = new Key(requestId);

entity.Properties.Add(new KeyProperty("new_myentityid", newReqKey));

Troubleshooting CRM Entity Publish Errors

2011-01-07T12:42:00.001-05:00

Here are a few things to keep in mind when working w/ entities in CRM

1. When you export your entities, make sure you export all N:1 related entities. When you select an entity and click export, there is a warning dialogue that reminds you of this, but nobody reads it.

2. Make sure you have Published your entities after importing them. I had an entity that was “broken”, but I couldn’t delete it until I had published it. Running a query like this will help you determine if your entity has been published: SELECT * FROM OrganizationUIBase WHERE ObjectTypeCode = <offending entity code>.

3. Finally, if you cannot delete a published entity, Compare the fields in the Form with the entity attributes. Microsoft Support gave me a tool to run a FormXML comparison. Two times now there has been an attribute in the form was missing from the entity. Once you add the attribute back (type doesn’t matter), you can delete the entity and start fresh.

Two ways to build a CRM QueryExpression

2010-12-20T17:08:00.000-05:00

I always wondered what FetchXML was useful for and I’m starting to see its value. I ran into a wall with the new Linq based Advanced Developer Extensions and have reverted back to the “stone-age” web services. Having to build QueryExpressions by hand led me to the discovery that I could actually use FetchXML syntax to build my expressions by using a FetchXmlToQueryExpressionRequest Combining this with Stunnware’s FetchXML builder might be even faster once you get the hang of it. Certainly makes debugging a complex join query snappy.

Check out these two ways to get the exact same dataset:

private static void FetchAppsToProcess(Microsoft.Crm.SdkTypeProxy.CrmService crmWebSvc)

{

string fxml =

@"<fetch mapping='logical' count='50' version='1.0'>" +

" <entity name='new_transistionapplication'>" +

" <attribute name='new_transistionapplicationid' />" +

" <filter>" +

" <condition attribute='createdon' operator='lt' value='9/17/2010' />" +

" <condition attribute='new_lastname' operator='eq' value='Smith' />" +

" </filter>" +

" </entity>" +

"</fetch>";

//string xmlResult = crmWebSvc.Fetch(fxml);

FetchXmlToQueryExpressionRequest fxmlReq =

new FetchXmlToQueryExpressionRequest() { FetchXml = fxml };

FetchXmlToQueryExpressionResponse expression =

(FetchXmlToQueryExpressionResponse)crmWebSvc.Execute(fxmlReq);

var request = new RetrieveMultipleRequest { Query = expression.Query, ReturnDynamicEntities = true };

var response = (RetrieveMultipleResponse)crmWebSvc.Execute(request);

}

private static void GetAppsToProcess(Microsoft.Crm.SdkTypeProxy.CrmService crmWebSvc)

{

List<string> reqBoards = new List<string>();

var con = new ConditionExpression

{

AttributeName = "createdon",

Operator = ConditionOperator.LessThan,

Values = new object[] { DateTime.Parse("9/17/2010") }

};

var conName = new ConditionExpression

{

AttributeName = "new_lastname",

Operator = ConditionOperator.Equal,

Values = new[] { "Smith" }

};

var filter = new FilterExpression();

filter.AddCondition(con);

filter.AddCondition(conName);

filter.FilterOperator = LogicalOperator.And;

// The Entity returning the results

var expression = new QueryExpression

{

EntityName = "new_transistionapplication",

Criteria = filter

};

expression.ColumnSet = new ColumnSet(new string[] { "new_transistionapplicationid", "new_requestedboardappointments" });

var request = new RetrieveMultipleRequest { Query = expression, ReturnDynamicEntities = true };

var response = (RetrieveMultipleResponse)crmWebSvc.Execute(request);

}

Installing SSRS - SocketException: No such host is known

2010-12-17T12:26:00.001-05:00

I just about went crazy trying to do a SSRS install on a named instance. Every time I ran through the database configuration, I got this error.

System.Net.Sockets.SocketException: No such host is known

at System.Net.Dns.GetAddrInfo(String name)

at System.Net.Dns.InternalGetHostByName(String hostName, Boolean includeIPv6)

at System.Net.Dns.GetHostEntry(String hostNameOrAddress)

at ReportServicesConfigUI.RSDatabase.IsLocalDbServer(String dbServer)

at ReportServicesConfigUI.RSDatabase.GrantRSConnectionRights()

at ReportServicesConfigUI.RSDatabase.CreateNewDatabase()

Turns out, this issue has a simple, but hidden workaround. Specify the Server Name as outlined on this blog post:

It seems if you use a connection format like “ServerName\InstanceName,Port”, then SSRS Configuration manager is able to connect without requiring the SQL Browser service.

Without SQL Browser service running (either shutdown or blocked behind the firewall), SSRS Configuration Manager still does not work with “ServerName,Port\InstanceName” format or via SQL Aliases.

Daddy Bread

2010-12-14T16:18:00.001-05:00

After reading the article, Artisian Bread in Five Minutes a Day, I have been making about 1-2 loaves and a pizza every week. My kids have dubbed it, “Daddy bread” and it goes fast! As you might imagine we go through a decent amount of flour, so we buy 25lbs bags at Costco. This picture of six coffee cans is the latest supply of flour. I can almost smell the bread cooking!

How to Hide System Views in CRM 4.0

2010-12-13T16:14:00.002-05:00

This post could also be titled, how to pass configuration to CRM Plugins. I found a sample solution in the MSDN Code Gallery but the installation details were left to the imagination. Fortunately another person posted on how to hard-code a plugin with the Assembly Registration information needed. First you register the assembly, then add a step with this information:

Message: RetrieveMultiple
Primary Entity: savedquery
Stage of Execution: Post Stage
Execution Mode: Synchronous

Unsecure Configuration: <the XML from the sample config file>

There was a problem with the code from the gallery in that the constructor for plugins seems to have changed. Plugin constructors require two string parameters as detailed in this blog post. The first is the “unsecure configuration” and the second is the “secure configuration”.

The nice thing is that the views are hidden as soon as you “Update” the Step Registration. No need to restart CRM or anything fancy.

You can download the source of my plugin here (6 Kb zip) (you may have to update the references to the SDK dlls).

How to set a recipient in the CRM Send Email window via URL querystring

2010-12-02T08:40:00.001-05:00

I need to let the user edit the email before sending, so email templates seem to be the best choice. I love how CRM set the Regarding to my current entity. In addition I wanted to be able to set the To: field based on a contact record.

MSDN - URL Addressable Forms and Views

and here:

Working With All MS CRM DataTypes In Javascript

and ultimately, I figured out the querystring parameters that allow you set a PartyList attribute (To is a PartyList - thanks to Stunnware tool for that detail). So by adding these parameters to the original Send E-Mail link, the email opens pre-populated:

&partytype=2&partyname=John%20Doe&partyId={1FF3C7A9-D8BC-DF11-8337-005056BC7B56}

Microsoft CRM SDK Advanced Developer Extensions Linq Cache Issue

2010-11-30T13:22:00.000-05:00

There has been a significant improvement in the XRM Linq provider in the 4.0.13 SDK that allows you to specify how you want your DataContext to use caching. One thing I noticed was I had to change my default Service to use the disabled cached to get it to use the Microsoft.Xrm.Client.Services.OrganizationService which is the “un-cached” service.

There seems to be a noticeable speed improvement for the initial connection time and I don’t see any difference in query time using the un-cached service.

Sample web.config entries (2kb XML file)

Checking Java memory usage on a Linux server

2010-11-22T09:03:00.001-05:00

A customer reported slowness in a production system, so I ran the following commands to check the state of the memory usage (since recently we had out-of-memory errors):

ps -ef|grep java

jmap -heap <pid>

The output:

Attaching to process ID 1715, please wait...

Debugger attached successfully.

Server compiler detected.

JVM version is 14.0-b16

using thread-local object allocation.

Parallel GC with 2 thread(s)

Heap Configuration:

MinHeapFreeRatio = 40

MaxHeapFreeRatio = 70

MaxHeapSize = 2015363072 (1922.0MB)

NewSize = 2686976 (2.5625MB)

MaxNewSize = 17592186044415 MB

OldSize = 5439488 (5.1875MB)

NewRatio = 2

SurvivorRatio = 8

PermSize = 21757952 (20.75MB)

MaxPermSize = 134217728 (128.0MB)

Heap Usage:

PS Young Generation

Eden Space:

capacity = 368246784 (351.1875MB)

used = 259936376 (247.89464569091797MB)

free = 108310408 (103.29285430908203MB)

70.58754815900849% used

From Space:

capacity = 11534336 (11.0MB)

used = 11502000 (10.969161987304688MB)

free = 32336 (0.0308380126953125MB)

99.71965443004261% used

To Space:

capacity = 12451840 (11.875MB)

used = 0 (0.0MB)

free = 12451840 (11.875MB)

0.0% used

PS Old Generation

capacity = 122290176 (116.625MB)

used = 76169032 (72.64044952392578MB)

free = 46121144 (43.98455047607422MB)

62.285487265959944% used

PS Perm Generation

capacity = 88211456 (84.125MB)

used = 39361424 (37.53797912597656MB)

free = 48850032 (46.58702087402344MB)

44.62166909477155% used

Severn River Cruise on the Mobjack Bay

2010-11-15T08:20:00.000-05:00

The weather around Veterans Day 2010 was absolutely stunning; Crystal clear blue skies, 65F, and a fresh 15-20Kts out of the NE. I got permission to take the Alberg 22 docked near Yorktown, VA for a two-day solo cruise. With the wind creating Small Craft Advisory conditions, I was thinking of heading NW up the York see West Point from the water and then find a little creek to tuck into. As I headed out into the York, I was surprised to find ideal flat water conditions. I realized there was quite a bit of protection from the land, but it was so tempting to head east…so that’s what I did. As I got nearer to the Guinea flats, the wind picked up. As I cleared the land and the full fetch of the Mobjack was felt, the waves were a short steep 3-5’ seas. The Alberg could really have used a reef at this point, but easing the sheet just a little helped with the weather helm. The boat kept a stead 4.5-5 kts with a nice motion.

Being a mid-November Thursday, there weren’t a lot of boats on the wather, so I was relieved to see a 50’ M/V and a little later, I saw what looked like two boats sailing really close together. Later this turned out to be a gaff-rigged schooner that I followed into the Severn River. I headed up the North branch and noticed what looked like a large sailboat behind a spit of land. The tide was about 3’ above normal so it seemed like my 3’ draft should have no trouble getting back there. I was rewarded with a beautiful little anchorage that was surrounded by some gorgeous Matthew’s County properties.

I didn’t have too many chances to snap pictures while sailing (solo SCA kept my attention), but the anchorage afforded an amazing sunset:

http://picasaweb.google.com/snyd1437/ChesapeakBay#

IIS 404 error!!! It's now 2am...thank you very much Microsoft security team

2010-11-06T01:31:00.001-04:00

http://support.microsoft.com/kb/315122

By default, when IIS is installed on any version of the Windows Server 2003 family, IIS only serves static content (HTML).

Creating a WebReference to a specific CRM tenant

2010-10-28T07:30:00.001-04:00

Sometimes it’s the little things that can take forever to figure out. I inherited a web application that connected to a CRM tenant and after adding a few new attributes to CRM, I needed to update my WebReference. Unfortunately the last developer had been working in a VPC so my WebReference that I had set to the CRM server resulted in the proxy classes that were “generic” and did not contain my customizations. I remembered doing this before, but it had been awhile so I had to search for the answer (again):

<add key="crmservice.CrmServiceWsdl" value="http://<CRM Server>/mscrmservices/2007/crmservice.asmx?uniquename=<tenant>"/>

Setup a domain user account as the CrmAppPool identity

2010-10-05T16:06:00.000-04:00

It was difficult to find the documentation on exactly which groups / policies the CrmAppPool identity needed when changing from Network Service to a domain user account. Fortunately I found this Microsoft KB which gives you step-by-step instructions (note, I removed the CRM 3.0 portions) and it worked on the first try:

1. Include the domain account user in the following groups in Active Directory:

o The Domain Users Active Directory group

o The PrivUserGroup

o The SQLAccessGroup

To do this, follow these steps:

d. Log on to a server as a user who has the Domain Admin rights or the rights to update these groups.

e. Right-click the Domain Users group in Active Directory, and then click Properties.

f. In the Group name box, type the name of the user who is running the CRM Application Pool, and then click OK two times.

g. Repeat steps b and c for the PrivUserGroup group and for the SQLAccessGroup group.

If you have more than one Microsoft Dynamics CRM deployment installed, multiple groups exist in Active Directory. To determine the groups that you want to update, follow these steps.

For Microsoft Dynamics CRM 4.0

h. Run the following SQL statement against the MSCRM_config database:

select id, friendlyname from organization

i. Note the GUID. For example, the GUID may be C8AB1D52-9383-4164-B571-4C80D46674E3 Org Name.

j. Find the PrivUserGroup group and the SQLAccessGroup group in Active Directory. The group name contains the GUID that you noted in step b.

2. Include the domain account user in the following groups in the Microsoft Dynamics CRM server:

o The local IIS_WPG group

o The local CRM_WPG group

Note The domain account user must have the following local user rights:

o The Impersonate a client after authentication right

o The Log on as a service right

To do this, follow these steps:

d. In the Microsoft Dynamics CRM server, click Start, point to Administrative Tools, and then click Local Security Policy.

e. Expand Local Policies, and then click User Rights Assignment.

f. Right-click Impersonate a client after authentication, and then click Properties.

g. Click Add User or Group.

Note You may have to click Location to select the domain instead of the local computer.

h. In the Group name box, type the name of the user who is running the CRM Application Pool, and then click OK two times.

i. Repeat steps 2c through 2e for the Log on as a service right.

3. Restart Internet Information Services (IIS). To do this, click Start, click Run, type IISRESET, and then click OK.

Connect to CRM Web Service using impersonation

2010-08-17T11:18:00.003-04:00

My Dev and Test servers live in two different domains and I am developing an external “portal” for public users to interface with CRM using web services. I am developing locally, but my CRM server is on a different domain, so I need to use a different domain user to connect to the web service. I am using the new SDK and the connection string was easy to change from “Integrated” to “AD”:

<add name="Crm"

connectionString="Authentication Type=AD;Server=http://CRM/Org1;User ID=USA\user1;Password=pass1"

But I also need to use the MetadataService to lookup some picklist values and that requires connecting the “old school” way. We were already using the Crm connection string to configure the MetadataService, so I just needed to check the Auth type and grab the User ID and Password:


public MetadataService GetMetadataServiceForConnString()
{
    string serverPath = GetConfigAttribute("server=http://");
    string[] serverPathParts = serverPath.Split('/');
    string connServer = serverPathParts[0];
    string connOrgName = serverPathParts[1];
    string username = null;
    string password = null;
    string domain = null;
    if (GetConfigAttribute("authentication type=").ToLower() == "ad")
    {
        // need to use active directory credentials
        string[] usernameParts = GetConfigAttribute("user id=").Split('\\');
        domain = usernameParts[0];
        username = usernameParts[1];
        password = GetConfigAttribute("password=");

    }
    return GetMetadataService(connServer, connOrgName, username, password, domain);
}

private string GetConfigAttribute(string searchString)
{
    string connCrm = System.Configuration.ConfigurationManager.ConnectionStrings["Crm"].ConnectionString;
    int pathPadding = searchString.Length;
    int pathStart = connCrm.ToLower().IndexOf(searchString.ToLower()) + pathPadding;
    int pathLength = connCrm.IndexOf(";", pathStart);
    if (pathLength == -1)
        pathLength = connCrm.Length;
    pathLength -= pathStart;
    return connCrm.Substring(pathStart, pathLength);
}

public MetadataService GetMetadataService(string server, string organizationName, string username, string password, string domain)
{
    CrmAuthenticationToken token = new CrmAuthenticationToken();
    token.AuthenticationType = AuthenticationType.AD; //Active Directory
    token.OrganizationName = organizationName;
    MetadataService service = new MetadataService();
    service.Url = string.Format("http://{0}/mscrmservices/2007/MetadataService.asmx", server);
    if (username == null)
    {
        service.UseDefaultCredentials = true;
    }
    else
    {
        service.UseDefaultCredentials = false;
        System.Net.ICredentials WebServiceCredentials = 
            new System.Net.NetworkCredential(username,password,domain);
        service.Credentials = WebServiceCredentials;
    }
    service.PreAuthenticate = true;
    service.CrmAuthenticationTokenValue = token;
    return service;
}

Use xp_cmdshell and BCP to Export SQL Server data to CSV

2010-08-05T14:19:00.001-04:00

This article describes how to use xp_cmdshell and BCP to export data to CSV. One problem I ran into was the error “Unable to open BCP host data-file”. This turned out to be a file permission issue. I had to give NETWORK SERVICE (the local account that runs my SQL Server service) Full Access rights to the queryout directory. I also had to use the –S flag to specify which SQL Instance (my server has two) to connect to:

EXEC xp_cmdshell 'bcp "SELECT * FROM <my database>.dbo.account" queryout "D:\queryout\bcptest.txt" -T -S "<my server>\<my instance>" -c -t,'

Get a list of all the tables in a SQL 2005/2008 database

2010-08-05T14:02:00.001-04:00

I thought this SQL 2005/2008 tip was very cool. This gets a list of all the tables in my database:

SELECT [name] FROM <my database>.sys.tables;

Grant Permission for SQL User for Membership Provider using Database Role

2010-07-21T14:03:00.000-04:00

The SQL User connecting to your Membershp Provider database (default named aspnetdb) should have the following database role:

aspnet_Membership_FullAccess

Otherwise you will get the following error:

EXECUTE permission denied on object 'aspnet_CheckSchemaVersion', database 'aspnetdb', schema 'dbo'.

Connection String for Membership Provider using a Trusted Connection over TCP/IP on a specific Port

2010-07-21T13:55:00.001-04:00

<add name="LocalSqlServer"

connectionString="Data Source=192.0.0.1,2116;

Network Library=DBMSSOCN;

Initial Catalog=aspnetdb;

Integrated Security=SSPI;"/>

Using data binding with Eval to set TemplateField button Visible property

2010-07-20T10:29:00.003-04:00

I needed to conditionally hide a delete button in a GridView. This data binding expression did the trick:


<asp:ImageButton ID="ImageButton1" runat="server" CausesValidation="False"
    CommandName="Delete" ImageUrl="~/_Images/cross.png" Text="Delete"
    Visible='<%# ((string)Eval("statuscodeLabel") == "Draft") %>' />

I also ran across this little beauty that calls a DAL class to do a check against the database (the data is hopefully cached):


<asp:HyperLink ID="lnkCert" runat="server" 
    Visible='<%# SomeDAL.UserCanDoSomething(CType(Container.DataItem, System.Data.DataRowView).Item("Code").ToString()) %>'
    NavigateUrl='<%# Eval("ID", "~/forms/AgencyForm.aspx?pid={0}&mode=cert") %>'>
    <img src="../images/tick.png" alt="Certify Agency Data" /></asp:HyperLink>

Pride goeth before a fall

2010-07-17T23:18:00.000-04:00

I assure you...this did not end well.

Using CRM Advanced Developer Extensions to Add Attachment to Contact

2010-07-16T12:25:00.001-04:00

This code sample is straight from MSDN, but it is so often used, I wanted to make sure I kept it handy:

// Add a note with a document attachment to the contact's record.

var attachment = File.OpenRead("capture.png");

var data = new byte[attachment.Length];

attachment.Read(data, 0, (int)attachment.Length);

var note = new Xrm.annotation()

{

subject = "Note subject...",

notetext = "Note Details....",

Contact_Annotation_id = contact.contactid,

objecttypecode = "contact",

documentbody = Convert.ToBase64String(data),

filename = Path.GetFileName(attachment.Name),

mimetype = "image/png"

};

crm.AddToannotations(note);

crm.SaveChanges();

Retrieve CRM Entity's Picklist using MetadataService

2010-07-09T15:44:00.002-04:00

This method returns a NameValueCollection of picklist items. CrmHelper is similar to the SDK utility sample.


public NameValueCollection GetAgencyRelationshipTypePicklist()
{
    CrmHelper helper = new CrmHelper();

    NameValueCollection picklist = new NameValueCollection();

    RetrieveAttributeRequest attribReq = new RetrieveAttributeRequest();
    attribReq.EntityLogicalName = "new_agencycontact";
    attribReq.LogicalName = "new_agencycontacttype";
    attribReq.RetrieveAsIfPublished = true;

    MetadataService service = helper.GetMetadataService(); 

    RetrieveAttributeResponse attribResp = service.Execute(attribReq) as RetrieveAttributeResponse;

    PicklistAttributeMetadata listData = attribResp.AttributeMetadata as PicklistAttributeMetadata;

    foreach (Option item in listData.Options)
    {
        picklist.Add(item.Value.formattedvalue, item.Label.UserLocLabel.Label);
    }

    return picklist;
}

Use JavaScript to hide an ASP.Net Button

2010-07-09T11:45:00.001-04:00

To be 508 compliant, my page must allow users to select items from a drop-down list with JavaScript disabled, so I added a “select” button (called btnAgSelect) next to the list. When JS is enabled, I do not want the button to be visible, so I used this in my Page.Load() event:

ScriptManager.RegisterStartupScript(

this ,this.GetType(),

"", "document.getElementById('" +

btnAgSelect.ClientID + "').style.visibility='hidden';",

true);

Windows XP Standby Wakes Up

2010-07-03T07:44:00.001-04:00

I got a new work laptop that I typically left on Standby, but it kept waking up in the middle of the night and draining the battery. I assumed it was the NIC’s Wake-On-LAN setting so I disabled the following setting (note the OS warning that someone had obviously ignored) using Hardware Manager (System – Hardware tab):

Finding CRM 4.0 Entities with Attachments by Querying annotationBase Table

2010-06-11T14:59:00.001-04:00

I needed to find all of the inactive records for an entity that had documents attached:

SELECT COUNT(*)
FROM   new_myEntityBase t
       JOIN annotationBase ab
         ON t.myentityid = ab.objectid
WHERE  statecode = 1

Move Flex Alert.show on the screen

2010-05-11T16:38:00.001-04:00

I have a tall Flex application that uses Alert.show( ) to display error messages. This created problems when the error messages were centered on the screen, but not visible when the screen was at an extreme scroll position (top or bottom). To solve this, I capture the MouseEvent and use it to set the Alert’s “Y” position. Note this code did not work UNTIL I used the PopUpManager.centerPopUp prior to moving:

var errorMsg:Alert = Alert.show(errmsg,"Error", Alert.OK, this);

PopUpManager.centerPopUp(errorMsg);

var mousePt:Point = new Point(event.localX, event.localY);

mousePt = event.target.localToGlobal(mousePt);

errorMsg.move(errorMsg.x, mousePt.y - 50); // try to vertical center it at button

Compiling the CRM SDK CrmServiceUtility Class without Web References

2010-05-04T10:49:00.005-04:00

According to Shan’s Blog post Using CRM SDK Instead of Web References, it is often preferable to use the SDK classes instead of the proxy classes generated when you add web references. Using the CrmServiceUtility class found in the 4.0 SDK in the \server\reference\cs folder is a perfect case in point. The way the code is written, you can compile it without a web reference simply by adding references to Microsoft.Crm.Sdk.dll, Microsoft.Crm.SdkTypeProxy.dll, and System.Web.Services and two using statements:

using Microsoft.Crm.SdkTypeProxy;
using Microsoft.Crm.SdkTypeProxy.Metadata;

Google Maps and Flex are beautiful together

2010-03-11T21:05:00.003-05:00

All I can say is WOW! Working with Google Maps in ActionScript is SOOOO much easier than HTML/JavaScript. In 5 minutes I had a working map and in 20 minutes of Google examples I had this:

http://edgewaters.s3.amazonaws.com/maps/Places.html

I am sold.

EveryTrail.com GPS Visualizer

2010-02-28T21:51:00.002-05:00

I went for a mountain bike ride today (the trails are finally dry enough) and I was packing my Holux GPS to record the trek. I was running the new GPS Sport Tracker softare on my Windows Mobile 6.0 phone and the logs are in GPX file format. I found EveryTrail.com that will create a Flash animation of your GPX trip:

45 Minutes in Powhite

Map your trip with EveryTrail

Windows Mobile GPS Sport Tracker Software

2010-02-24T14:55:00.001-05:00

I just discovered an excellent free (donations accepted) GPS software for Windows Mobile phones called GPS Sport Tracker. It has amazing features for running, biking, hiking, boating and probably many other things. I can’t wait to get this in the boat and fine tune the sail adjustments. The features I particularly like are:

Speed in 1/10s of MPH
Distance travelled
Average/Max speed
Elevation change
Data logging to GPX file format
Maps of your track
Profile of your elevation

This is fantastic work!

New in Tomcat 6.0.19 - Realm lock-out feature

2010-02-23T10:52:00.000-05:00

NOW they tell me. Man it pays to RTFM...

http://www.jeremythomerson.com/blog/2008/11/apachecon-securing-apache-tomcat-for-your-environment/

New in 6.0.19 [the] LockOut realm - it wraps around standard realms and provides a lock-out mechanism for multiple failed attempts for the same user. With this, there will also be the ability to have multiple realms for authentication - if any match, you get access - so you could use, for example, a tomcat users file for admins and a JNDI realm for users.

Full details here:
http://tomcat.apache.org/tomcat-6.0-doc/config/realm.html

You just wrap this around your existing realm (JDBC in our case). I got this working and it will serve as a "backup" to my custom login attempts table. The LockOut realm will prevent a brute force attack that hits the JAAS servlet directly.

SQL that return 1 or more rows, but is never empty

2010-02-22T16:33:00.002-05:00

The alternate title for this post is "how I worked around the Spring JdbcTemplate queryForObject method throwing an EmptyResultDataAccessException when the query returned zero rows", but that didn't seemed a little verbose.

I added a SELECT 0 as a table source and joined to that to ensure there would always be one row:


SELECT Coalesce(i.UserID, 0) UserID, Coalesce(i.Password, '') Password, Count(a.UserID) Attempts
FROM (SELECT 0 AS id) mstr LEFT JOIN
(SELECT 0 AS id, UserID, UserName, Password FROM User WHERE UserName = 'Scooby') i ON mstr.id = i.id
LEFT JOIN UserAttempts a ON i.UserID = a.UserID AND LastAttempt > 1266602785
GROUP BY i.UserID,i.Password

Using Spring JdbcTemplate to queryForObject example

2010-02-22T10:59:00.001-05:00

I needed to make a quick query to check some data before inserting a new record. The queryForObject allows you to create an inner RowMapper class to handle the ResultSet…in this case, I needed to return two integers so I mapped the results to an integer array with two elements:


JdbcTemplate springTemplate = new JdbcTemplate(this.currDs);
int[] user = springTemplate.queryForObject(
      "select UserID, count(*) from t_actor where orgID = ?",
      new Object[]{new Long(1212)},
      new RowMapper<int[]>() {

          public int[] mapRow(ResultSet rs, int rowNum) throws SQLException {
              int[] user = new int[2];
              user[0] = rs.getInt(0);
              user[1] = rs.getInt(1);
              return user;
          }
      });

Adding new folders in Eclipse

2010-02-19T11:33:00.003-05:00

I discovered a great feature/shortcut for adding new folders in Eclipse. Java/Flex packages sometimes have very deep folder structures which are a bit of a pain to recreate by hand. The good news is Eclipse will create the folder structure in one fell swoop if you enter it. Here's what I do: I find the folder in Windows Explorer and copy the location (e.g. com\adobe\crypto ) and paste it into the new folder dialog and viola!

Eclipse WTP Axis2 Web Service Tutorial Hints

2010-02-10T14:55:00.001-05:00

I have attempted to follow this Axis2 Tutorial and it has been painful. I will attempt to ease other’s pain by providing the mistakes that I made to prevent others from making the same…

When I got this error:

ClassDefNotFoundException org.apache.http.HttpResponseFactory

It was because I was using Axis2 ver 1.5.1. When I switched to 1.4.1 it was gone.

The tutorial says that opening http://localhost:8080/MyProject should display the Axis2 start page, but I found another Axis2 tutorial that said there was a bug and you actually needed to use:

http://localhost:8080/MyProject/axis2-web/index.jsp

Flex mx:Label htmlText and Font color behaviors

2010-02-09T16:15:00.001-05:00

I just discovered a few qwerks with the Flex Label’s htmlText property that I thought were worth passing along.

First off, Labels are limited to one line, so the default behavior of the <mx:htmlText> <![CDATA[ will not display:

<mx:Label>

<mx:htmlText>

<![CDATA[

* indicates required field.

]]>

</mx:htmlText>

</mx:Label>

Once I consolidated the the CDATA lines into one it displayed:

<mx:Label>

<mx:htmlText>

<![CDATA[ * indicates required field.]]>

</mx:htmlText>

</mx:Label>

…but the color still didn’t work. Turns out you have to use Hex to specify font color:

<mx:Label>

<mx:htmlText>

<![CDATA[ * indicates required field.]]>

</mx:htmlText>

</mx:Label>

The Adobe LiveDocs explain all this in more detail.

Flex Builder stopped generating HTML wrapper after adding VSS

2010-02-04T11:15:00.001-05:00

After adding my Flex project to VSS, the Flex Builder silently stopped generating the HTML wrapper files. To fix the problem, I unchecked the Generate HTML wrapper file in the project properties. I was prompted to delete the html-template directory and I did remove it (including the VSS linkage). I then changed the project properties back and the HTML wrappers generated again. Note, I did not check the html-template directory back in to VSS.

Improved Flex Validation: Display Error Tooltip

2010-02-02T16:06:00.004-05:00

I combined two validation concepts to arrive at my Improved Flex Validation. The key component is Joel Hooks Form Validation for the Lazy Programmer. I incorporated the UIComponent.callLater() hack for displaying error ToolTips and viola!

Full Source

Here is what the callLater() looks like:


...
    // display errorTip
    if (showErrorsImmediately && 
        currentControlIsValid == false && 
        supressEvents == false)
    {
        // cast the focussed control to a UIComp to use callLater
        var ffc:UIComponent = focusedFormControl as UIComponent;
        ffc.callLater(showDeferred, [focusedFormControl]);
    }
    
    return currentControlIsValid;
}

private function showDeferred(target:DisplayObject):void {
      target.dispatchEvent(new MouseEvent(MouseEvent.MOUSE_OUT));
      target.dispatchEvent(new MouseEvent(MouseEvent.MOUSE_OVER));
}

Flex CompareValidator custom Validator

2010-02-02T15:18:00.003-05:00

I have been using Matt Holden's Password Strength Indicator component with great success, but I was asked to make the input errors more obvious. The Flex Validators are a nice start, but their red-outline + tooltip approach is not the most intuitive. So, in conjunction with my improved Flex validation, I created a custom Validator for the password component to do the confirmation password comparison. The CompareValidator is generic enough that it can be reused for other components like account numbers or email addresses.


package core.util
{
import mx.validators.ValidationResult;
import mx.validators.Validator;


public class CompareValidator extends Validator
{
   public var valueToCompare:Object;
   public var errorMessage:String = "Value does not match.";

   public function CompareValidator()
   {
       super();
   }

   override protected function doValidation(value:Object):Array {
       var results:Array = [];
       var srcVal:Object = this.getValueFromSource();
 
       if (srcVal != valueToCompare) {
           results.push(new ValidationResult(true, null, "Match",errorMessage));
       }
       return results;

   }
}
}

The MXML looks like this:


  <coreutil:CompareValidator
      id="comparePasswords"
      source="{password2}"
      property="text"
      valueToCompare="{password.text}"
      errorMessage="Passwords do not match."
  />

Overriding the Java toString method using For loop

2009-12-23T13:36:00.002-05:00

This solution uses a combination of reflection, constants and regular expressions to build a meaningful toString output:


private String TO_STRING_FIELDS = "email isPKI isTrustedRoot isITU_TX509 caName"

    @Override public String toString() {
        try {
            StringBuilder result = new StringBuilder();
            //String newLine = System.getProperty("line.separator");
            String[] fieldList = TO_STRING_FIELDS.split( "[\\s,]+");
            Field curField = null;
            for (String f : fieldList) {
                curField = this.getClass().getDeclaredField(f);
                result.append("  " + f);
                result.append(": ");
                //requires access to private field:
                result.append( curField.get(this) );
            }
        } catch (Exception e) {
            System.out.println("Error in toString for class: " + this.getClass().getName() + "\n" + e.getMessage());
        }

        return null;
    }

Refurbishing a Morgan 30

2009-12-23T08:46:00.000-05:00

Hull Preparation and Painting

After all the work I did restoring my Flying Scot this summer, I can imagine what a labor (of love) this Morgan 30 refurb was. There are a lot of great tips and experience to be gleaned from his notes.

Flex Relative paths for embedded images in buttons with @Embed

2009-12-22T10:53:00.004-05:00

Flash, Flex and ActionScript: Relative paths for embedded images

Setting the @Embed source with an absolute path (based on flex_src root) :

<mx:Button icon="@Embed(source='/assets/blue_delete.png')" />

as3corelib Contains Crypto Utils for FLEX MD5 Hash

2009-12-02T15:51:00.001-05:00

I needed to store MD5 hashed passwords in the database and was delighted to find the as3corelib already contains the necessary libraries. While there are many more features, I only needed the MD5 Hash function so I only copied in the com.adobe.crypto.MD5 and com.adobe.utils.* files. With the .as files added to my Flex project, hashing the password was a simple as calling:

MD5.hash( passwordComp.getPassword() );

I verified that it returns the same hash as the MySQL MD5 function

UPDATE user_table SET password = MD5(‘change_me’) WHERE username = ‘sandman’

Dynamically position a Flex popup relative to a button mouse click

2009-10-26T15:36:00.001-04:00

I was having trouble with positioning a wide popup window that was in a component nested several containers deep. The PopUpManager.centerPopUp method would center on the container and push the window off the right edge of the screen. To correct this, I moved the popup over 100 pixels using:

popup.x = popup.x – 100;

That worked, but the problem would come back if the window grew wider. Bush league. To make it a bit more reliable, I changed to use the mouseEvent localX/Y properties. Note you have to convert the “local” coordinates to “global”:

private function showPopup(event:MouseEvent):void

{

pop = EnotaryRegReturnPopup(PopUpManager.createPopUp(this,EnotaryRegReturnPopup,true));

pop.title = "Please select return reasons.";

pop.showCloseButton =true;

pop.enotaryRegReturnList = this.enotaryRegReturnList;

pop.buildUnselectedReturns();

// position the window next to the "add" button

var pt:Point = new Point(event.localX, event.localY);

pt = event.target.localToGlobal(pt);

pop.x = pt.x - pop.width;

pop.y = pt.y;

pop.addEventListener("close",removeMe);

pop["cancelButton"].addEventListener("click", removeMe);

pop["okButton"].addEventListener("click",processReturns);

}

Delete rows from DataTable using DataRow

2009-10-05T16:01:00.000-04:00

I was surprised by the delete functionality in DataTable. It was my understanding that when you call DataRow.Delete it marks the row for deletion, but waits for a DataTable.AcceptChanges call before actually removing them, but when I tried to delete inside a For Each dr as DataRow in dt.Rows I was getting an error “Collection was modified”. Thankfully I ran across a thread that suggested walking backwards through the collection to avoid those types of errors. I ended up with this to solve my problem:

Do While i >= 0

If dt.Rows(i)("KEEP_ME"))) = False Then dt.Rows(i).Delete()

i -= 1

Loop

dt.AcceptChanges()

DOS BAT File Copy Rename with String Manipulation

2009-09-28T14:40:00.001-04:00

I needed to rename a directory of files to truncate the file names to the first 5 characters (plus the file extension). I found one example that showed how two write a FOR loop that broke out the filename and extension and passed it to a second BAT file:

for /f %%a IN (’dir /b *.txt’) do call runner.bat %%~na %%~xa

And using a second example that covered DOS string manipulation, I put together this batch file that accepts two parameters and does a substring to get the first 5 characters of the first parameter:

set file=%1
set file=%file:~0,5%
ren %1%2 %file%%2

Creating an Asynchronous ASP.NET Page To Initiate Batch Processes

2009-09-17T15:27:00.001-04:00

I recently created a page that gave the user the option of initiating a long-running batch process. The batch required repeatedly calling external web services that would have taken far too long for a user to wait for the traditional page processing model. But, by using ASP.NET asynchronous execution to initiate the batch process on a separate thread, the page response is excellent. I started by working through this excellent tutorial on asynchronous execution, but quickly ran into problems when I began calling web services. Creating an asynchronous process involves writing a method that can execute without needing anything from the HttpContext (Application, Page, Session, Cache, Configuration, etc.) since this is not available when you use BeginInvoke to run your subroutine. This presented me with an issue since my project had a library that we used to setup web service instances (with WSE security), but that library relied on the HttpContext’s Application object. To work around this limitation, I created a private inner class that encapsulated the web service references.

Private Class WebSvcInstances

Public FSI As MyWseSVc1 = GetMyWseSVc1

Public PSI As MyWseSVc2 = GetMyWseSVc2

End Class

The inner class was passed into the batch method while the Application object was still available and providing the objects the batch would need later.

To start off, create the subroutine you want to initiate asynchronously:

Private Sub InitiateBatch(ByVal wsRefs As WebSvcInstances)

Create a Delegate in your code behind with the same signature as the subroutine:

Private Delegate Sub Batch_Delegate(ByVal wsRefs As WebSvcInstances)

Then you can initiate your subroutine in a page event like so:

Dim batchDelegate As InitiateBatch_Delegate

Dim wsRefs As New WebSvcInstances

batchDelegate = New InitiateBatch_Delegate(AddressOf InitiateBatch)

Dim asyncResult As IAsyncResult

asyncResult = batchDelegate.BeginInvoke(wsRefs, Nothing, Nothing)

You can read up on BeginInvoke to learn more about the callback features. I didn’t need to be notified when the execution finished, so I just set those to Nothing

Before and AFTER

2009-09-17T12:23:00.005-04:00

It hasn't taken quite as long as the lack of blog posts would have you believe, but the refurbishment of Flying Scot 338 is nearly complete. In addition to the fairing and painting, I've just finished replacing the gooseneck and standing rigging, installing a boom whoopie, turtle bags and guy hooks. I hope to get the transom port, bow bag, and stem band on before the Boathouse Regatta this weekend. I am SO glad I have a Flying Scot since the class builder has been SUPER helpful including instructions with all of the replacement parts.

Painting recap:

After talking to another club member who did the same paint job (but rolled everything) I should have been more aggressive with sanding the primer. I was concerned about sanding too much off and not hiding the yellow gel coat so I wet sanded with 220. My friend said he used 80-100 on the primer which I now wish I had done (with a long board)...but time was limited. I didn't think I'd have a chance to put a 2nd primer coat on.
I am very impressed with the shine and durability of Interlux Brightsides so far. I don't think I've scratched any paint off, though it's only been a couple months of light use.
During the refurb, I decided to tackle the bottom as well, painting it with VC Performance (a 2-part epoxy) and given that experience, I think I would be more willing to attempt a 2-part poly for the topsides next time. A warning about VC Performance though: it is extremely thin (runny), being designed to spray on. They say it is possible to brush/roll it on, but it would seem far easier to spray given my limited experience. My "spray rig" consisted of a $99 compressor and $15 gravity fed gun, both from Harbor Freight.

Centerboard:

As earlier pictures showed, my centerboard was in terrible shape. The leading edge was so badly damaged, there was actually a large accumulation of clay between the lead balast and the outer fiberglass causing a bulging in the airfoil shape. I ended up cutting out a curved section of fiberglass from the leading edge (both sides) and repairing it with about 4-5 layers of 1.5oz mat. I faired the repair and other leading edge defects with Interlux Watertite and finished with two coast of VC Performance (after removing LOTS of bottom paint) the board looks almost new.

Removing boat bottom paint

2009-05-14T18:02:00.004-04:00

I read somewhere that the best tool for removing bottom paint is your checkbook. After my experience, I would have to agree, but it can be done. Apparently the pros sandblast the stuff off. Not being a pro, I turned to Google and read about several different methods for DIY: scraping, stripping, and sanding being the top three. In the end, I believe the most efficient way is to scrape AND sand.

Important tip: have a file in your pocket and keep the scraper sharp. The copper bottom paint dulls the scraper quickly. To sharpen, you have to lay the file down and hold the scraper perpendicular to the file. Just a few strokes makes a huge difference.

Flying Scot #338 had several layers of bottom paint so scraping removed a bunch of material, but did not remove the last layer of paint. The picture above shows the results of scraping on the right and after sanding on the left. I sanded using a 1-2 punch of a 6" sander with 50 grit lightly and carefully run over the paint to remove 95% and then a 5" orbital sander with 80 grit to get the last bit off and smooth out the agressive grit. This process took about 1 hour for the first side, second side TBD...

Fairing and repainting a Flying Scot

2009-04-27T08:38:00.001-04:00

I will be blogging regularly about this year’s refurbishment of Flying Scot 338. Last weekend I removed all of the hardware including rub rails. This weekend I removed the centerboard (with the help of the club hoist). Launched the 338 onto dry land – a very unnaturally feeling. And, with the help of 3 club members, flipped the hull onto low sawhorses. Next up, cutting the bad balsa out of the aft deck and removing the bottom paint. I feel itchy already.

Take a look at that centerboard. I got a chance to give it a close inspection while we were…uh…sideways a few weeks ago. I think I see why light air has never been kind to us!

Configuring Oracle XE connection pool in Glassfish and Using JPA

2009-04-09T14:49:00.004-04:00

I followed these instructions for creating an Oracle JDBC datasource. I needed to download the Oracle JDBC driver (ojdbc14.jar) into my <Glassfish>/lib directory and restart the server. After that the “Ping” was successful.

In “step 2” I had to specify 4 properties: user, password, url, and xa-driver-does-not-support-non-tx-operations.

Next I needed to create a JDBC resource that my persistence.xml file would reference.

With those tasks complete, I modified an existing EJB 3 JPA project to utilize the new connection pool. All that was necessary was to change two lines in the persistence.xml file. First was to point to the new JNDI name:

<jta-data-source>jdbc/AMRPool</jta-data-source>

And second was to change the “toplink.platform.class.name” (now deprecated) to:

After redeploying the JAR the CUSTOMER table was added to my schema. I re-ran the test client and was delighted to find a record in the table. Bravo to the Java community!

Excellent EJB 3 Persistence Tutorial

2009-04-08T13:56:00.004-04:00

It had been a long time since I have visited the land of Java. When I was a regular traveler there, I remember feeling inundated by the rapidly progressing proliferation of APIs. JMS, JNDI, RMI, CMP, and on…and on…and ON. The technologies and vendors were spawning and/or evolving so fast and from so many directions it was difficult to stay current. From what I can see the frameworks have matured and are stabilizing. While there are still many innovations occuring, the rate of change does not appear to be the dizzying pace that I remembered.

To get up-to-speed on the current J2EE release, I dutifully installed the latest version of the open source Java EE 5 implementation…something called GlassFish. It has a decidedly un-Microsoft feel about it. I had flashbacks of all the previous Weblogic/Tomcat/Apache configuration nightmares and had to steel myself for navigation through the narrow passages in the reefs surrounding a modern Java app server. I have found…in general…that Java technologies are much less tolerant of the neophyte and the learning curve seems significantly steeper than the reciprocal Microsoft technology (“platform agnostic”…I know…I know).

I also downloaded the Java EE tutorial and attempted to follow it using Eclipse instead of the recommend NetBeans because I had previously used Eclipse and thought it was generally a great IDE. In doing so, I ran into trouble with creating the EJB example that was entirely related to my inexperience…and the 50 different project types Eclipse now ships with out of the box. I left the Sun script and started working through simple JSP examples just to get some tiny amount of momentum built. After following the online documentation, success was not far behind.

Then I ran across the EJB 3 persistence topic and decided that looked interesting, particularly hibernate.org. The top tutorial on the matter…as declared by Google, was from Web Age Solutions. It was short enough that it didn’t scare me off so I dove in with low expectations of a working example. I was pleasantly surprised by the author’s accuracy and brevity. The logical and successful progression was refreshing and rewarding. By the end of the tutorial, I had a basic understanding of the Java Persistence API (JPA) and a new level of respect for the current Java specification. Bravo!

http://www.webagesolutions.com/knowledgebase/javakb/jkb006/index.html

Data densification using Oracle Partitioned Outer Joins

2009-03-04T09:30:00.002-05:00

I ran across a feature of Oracle 10g called Partitioned Outer Joins while tracking down the answer to another problem. Rather than get side tracked, I made a note to check them out at a later time and I’m glad I did. I frequently get specifications for reports where the data has gaps, but the report should be continuous, regardless of the presence of the data. The typical example is when you want a customer sales report, but some months a customer has no orders. Oracle-Developer.Net has a great article explaining how to create just such a query (and proper use of WITH). The first helpful piece of SQL is a WITH clause to generate a “time dimension” of continuous months

WITH year_months
AS (SELECT To_char(Add_months(DATE '2004-01-01',ROWNUM - 1), 'YYYYMM') AS year_month
FROM dual
CONNECT BY ROWNUM < 12)

Adding the Partition clause turns out to be quite straightforward:

SELECT co.name,
ym.year_month,
Nvl(Sum(co.amt),0) AS total_amount
FROM year_months ym
LEFT OUTER JOIN customer_orders co
PARTITION BY (co.name)
ON (To_char(co.dt,'YYYYMM') = ym.year_month)
GROUP BY co.name,
ym.year_month
ORDER BY co.name,
ym.year_month;

Use Oracle Table Function to Encapsulate and Modularize your Data Layer

2009-02-25T14:47:00.003-05:00

I finally ran into a situation where I wanted the functionality of a Table Function so I was forced to learn the PL/SQL way to do this. In this case, the benefits of accepting parameters made the Table Function a clear winner. To start off you need a table:

CREATE TABLE cool_people (
cp_id NUMBER(8,0),
first_name VARCHAR2(30),
last_name VARCHAR2(30))

Next you need a new TYPE that describes a single “row” of your Table Function’s output. In my contrived example, I’m only returning a formatted name:

create or replace TYPE PEOPLE_OBJ AS OBJECT(

PERSON_NAME VARCHAR(100));

Now that we have created the new type of OBJECT, we can declare a TABLE of that new type:

create or replace TYPE PEOPLE_TABLE AS TABLE OF PEOPLE_OBJ;

Now we can create our function that accepts a parameter of CHAR:

CREATE OR REPLACE FUNCTION
     TFN_GET_COOL_PEOPLE (formatName CHAR)
     RETURN PEOPLE_TABLE PIPELINED
IS
     fName VARCHAR(30);
     lName VARCHAR(30);
     p1 PEOPLE_OBJ := PEOPLE_OBJ(NULL);
     CURSOR C1 IS
             SELECT FIRST_NAME,
                    LAST_NAME
             FROM   COOL_PEOPLE;

BEGIN
      OPEN C1;
      LOOP
              FETCH C1
              INTO  fName,
                    lName;
           
              EXIT
      WHEN C1%NOTFOUND;
              IF formatName           = 'Y' THEN
                      p1.PERSON_NAME := lName
                      || ', '
                      || fName;
              ELSE
                      p1.PERSON_NAME := fName
                      || ' '
                      || lName;
              END IF;
              PIPE ROW (p1);
      END LOOP;
      RETURN;
END TFN_GET_COOL_PEOPLE;

You can then use the function in a FROM clause:


SELECT * FROM TABLE(TFN_GET_COOL_PEOPLE('N'));

Private stored procedures in an Oracle package

2009-02-23T12:19:00.001-05:00

I never knew you could write a procedure in a package that isn’t declared in the package declaration. It’s kind of like a private function/subroutine since it can’t be called by anything outside of the package…which is how I discovered this “feature”. I was trying to call an “undeclared” procedure from a different package. I had already written another procedure that called it from its own package, so I was mighty confused about the error I was getting: “MY_PROC must be declared” but now it makes perfect sense.

One other discovery I made is that you can pass NULL as a parameter to another procedure instead of declaring a null valued variable.

Using GeoIP's GeoLite to find an IP address City using C# .Net

2009-01-06T10:58:00.000-05:00

There’s a terrific open source product offered by MaxMind called GeoLite City that is a binary (or CSV) database for doing IP location lookups. There are many flavors of the GeoIP API that provide optimized access to the .DAT file, including C# and VB.Net (Country only). The C# version is a collection of 5 classes and is very easy to use:

Dim ls As New LookupService("C:\GeoIP\GeoLiteCity.dat")
Dim IPLoc As Location = ls.getLocation(IPString)

Long2IP in VB.Net and Narrowing Conversion

2008-12-30T14:32:00.001-05:00

This is post #2 demonstrating how VB.Net makes things harder than C#. I am receiving IP data as a type Long number. It turns out that PHP has this helpful built-in function called long2ip that will do the translation back to the familiar “192.0.1.100” string for you.

I found a few examples of VB code that could do the same, but as best I (and the compiler) could tell, they relied on functions that were not carried into VB.Net. The problem is very simple though, or so I thought. All I need to do is narrow the Long into a Byte and then shift the bits a Byte at a time. In C# this is a piece of cake, but try as I might I could not convince the .Net CLR that I was intentionally overflowing the variable. Despite trying every combination of Option Strict Off/On with CType, CByte, DirectCast, the runtime dutifully threw an OverflowException. I was unable to find a way to do a narrowing conversion in VB.Net.

I gave up and created a separate C# class library for my helper class based on this code I found:

public static void ToBigEndian(long number, byte[] outputArray) {
int length = outputArray.Length;
outputArray[length - 1] = (byte)number;
for (int i = length - 2; i >= 0; i--)

outputArray[i] = (byte)(number >> (8 * (i + 1)));

}

With that, it was trivial to create my own version of Long2IP:

Public Function Long2IP(ByVal IP As Long) As String
Dim myIp(3) As Byte
IPHelp.ToBigEndian(IP, myIp)
Return myIp(0) & "." & myIp(1) & "." & myIp(2) & "." & myIp(3)
End Function

My first time (geo) caching

2008-12-20T16:25:00.005-05:00

There are some days that I say, "we just have to blog about this." Memories that I know we will want to revisit. Today was one of those days. Last night (after a enjoyable visit to the botanical gardens) we realized that we had absolutely nothing scheduled...a rare event for a Saturday. It was a good day for a family hike, but the neighborhood trails were undoubtedly muddy from all of the rain, so we decided to try the local park built on historic coal mines. There was also the promise of a long-awaited geocaching adventure.

I think the thrill of the hunt really livens up a walk in the woods for little people. In the future we will definitely incorporate that into our hikes! We struck out on the first cache (apparently it is often muggled), but after much searching we found our second hide. The hike on the old rail grade, through the tunnel, and around the lake was quite challenging for our little guys, but they did great!

Then to top it off, we enjoyed a filling lunch at a mama-favorite: Cheeburger Cheeburger. The kids had a great time playing peek-a-boo with the staff :)

Use ROWNUM to return multiple non-aggreate GROUP BY columns

2008-12-19T15:21:00.001-05:00

Somebody needs to explain something to me. Why in the world can’t we use some of the aggregate functions without having to GROUP BY every other column? I know things like SUM and AVG won’t work, but why can’t I do the following:

SELECT name, toy_category, Max(cool_factor) FROM toys GROUP BY toy_category

And have it just return the coolest toys? Now that is a very simplistic example, but I needed to do something very similar to this. The best way I found was to use Oracle’s ROWNUM column on a sorted resultset to allow me to join back to said resultset. This also seems like the perfect time to use a WITH Clause (my other post about using WITH Clause) since we need to hit that data several times and preserve the ROWNUM. The first thing we need to do is get the list of Toys into a named query ordered by the column we want to Max() – the order is very important:

WITH temp_all_toys AS (SELECT name, toy_category, cool_factor FROM toys ORDER BY cool_factor)

Now that we have all the toys in a discreet list we can look for the Max(ROWNUM) for each toy_category.

SELECT Max(rownum) as max_rownum FROM temp_all_toys GROUPY BY toy_category

The output of that query is pretty dull. Just a bunch of random looking numbers…but they are really the “primary key” for our temp table so if we join them back to the temp_all_toys table, we can get at any of the columns! Thanks for nothing GROUP BY. There is one last trick I found for joining back: for some reason my query wouldn’t work unless I joined to a second subquery to get at the ROWNUM:

WITH temp_all_toys AS (SELECT name, toy_category, cool_factor FROM toys ORDER BY cool_factor)

SELECT
all_recs.*
FROM
(SELECT rownum as rnum, name, toy_category FROM temp_all_toys GROUPY BY toy_category) all_recs
JOIN
(SELECT Max(rownum) as max_rownum FROM temp_all_toys GROUPY BY toy_category) max_recs
ON all_recs.rnum = max_recs.max_rownum

Create multiple Oracle temporary tables using WITH Clause

2008-12-19T14:05:00.003-05:00

According to the Oracle docs, the WITH clause behaves much like a temporary table and is a great way to clean up lengthy SQL:

The WITH clause (formally known as subquery_factoring_clause) enables you to reuse the same query block in a SELECT statement when it occurs more than once within a complex query. WITH is a part of the SQL-99 standard. This is particularly useful when a query has multiple references to the same query block and there are joins and aggregations. Using the WITH clause, Oracle retrieves the results of a query block and stores them in the user's temporary tablespace. Note that Oracle Database does not support recursive use of the WITH clause.

This is exactly what I was looking for to solve a difficult task involving ROWNUM (ROWNUM post). What’s more, you are not limited to only a single “named query”. You can define as many as you want by separating them with a comma:

WITH query_name AS (subquery)

     [, query_name AS (subquery) ]...

Calling a PHP SOAP 1.2 web service using Visual Studio and VB.Net

2008-12-12T11:49:00.001-05:00

Oh the fun of learning new technologies. The endless errors, outbursts of profanity, stumping dead-ends…and finally, the trill of success! Nothing beats it in my book. This story is about my latest “trophy” (a fishing metaphor), creating a PHP web service and then calling it from a .NET console application. Once you get everything working, it all seems so simple…but rest assured, creating a WSDL file by hand can be a tricky experience if you are new to one of the technologies in question. First things first, to follow this example you need to download the sample application that contains a couple of PHP files and console application. I tried to boil this down to the Hello World level so it would be easiest to understand. I assume you already have PHP setup and have enabled the SOAP extension. Here is the part that took me awhile to get working. I wanted a web service that took a string as a parameter and returned a string. Apparently Visual Studio requires this “complexType” description in the “types” section; otherwise you could just define your parameters in the “message” section and be done with it.

<s:element name="GetReportsListRequest">

<s:complexType>

<s:sequence>

<s:element minOccurs="1" maxOccurs="1" name="Category" type="s:string"/>

</s:sequence>

</s:complexType>

</s:element>

<s:element name="GetReportsListResponse">

<s:complexType>

<s:sequence>

<s:element minOccurs="1" maxOccurs="1" name="ReportsList" type="s:string"/>

</s:sequence>

</s:complexType>

</s:element>

When VS2005 reads this WSDL (lots of XML omitted in the above – see download), it generates the following stub:

Public Function GetReportsList(ByVal Category As String) As String

Now another tricky part was getting the PHP service to return the string in the correct object hierarchy. It took the following code to do so:

function GetReportsList($P) {

$reportList = "Report1|Report2" . $P->Category;

return array("ReportsList"=>$reportList);

}

For an example of how to return complex types, see the references section at the end of this article.

One other quirky thing was how to call the web service from a PHP client (I’ll throw that in for free). You have to wrap your parameter array in a blank array because of the WSDL GetReportsListRequest hierarchy:

$return = $client->__soapCall("GetReportsList",array(array( "Category" => "abc")));

References:

PHP SOAP Manual
User Comment in SoapServer.addFunction documentation
Trouble with Visual Studio and WSDL
WSDL Essentials

A .Net developer's guide to MySQL

2008-12-10T08:25:00.003-05:00

I was ~~given the opportunity~~ forced to get to know all things MySQL so these are the observations of a Microsoft/Oracle developer’s first foray into MySQL.

The MySQL Downloads page is the one-stop for the latest MySQL goodies. As of this writing the Community Server (free) was at release 5.1. Once you install the database, you quickly realize a need for some additional tools. The MySQL GUI Tools bundle includes the essential Administrator and QueryBrowser applications that will be familiar to any Microsoft developer. Now that you can create and maintain your new RDBMS, you need some drivers to connect the .NET code to. I tested both the ODBC and the Native (Connector/Net) drivers and from what I can see, I’d go with the Native drivers for development, but I’d also install the ODBC drivers for general Microsoft bliss (Visio, Excel, etc.).

Working with the Native driver is nearly identical to any other ADO.Net provider, save that everything is prefixed with “MySql” (e.g. MySqlConnection, MySqlCommand, etc.). After adding a project reference to MySql.Data, you need only import MySql.Data and MySql.Data.MySqlClient and the rest is completely straightforward:


Imports System.Web.Services
Imports System.Web.Services.Protocols
Imports System.ComponentModel
Imports MySql.Data
Imports MySql.Data.MySqlClient



<System.Web.Services.WebService(Namespace:="http://tempuri.org/")> _
<System.Web.Services.WebServiceBinding(ConformsTo:=WsiProfiles.BasicProfile1_1)> _
<ToolboxItem(False)> _
Public Class ReportService
    Inherits System.Web.Services.WebService

    Dim _connString As String = "server=localhost;database=csrm;user id=root;password=****;"

    <WebMethod()> _
    Public Function ReportList(ByVal group As String) As Report()
        Dim newList As New ArrayList
        Dim newItem As Report


        Dim conn As New MySqlConnection(_connString)
        Try
            conn.Open()

            Dim cmd As New MySqlCommand("SELECT * FROM report", conn)
            Dim reader As MySqlDataReader = cmd.ExecuteReader(CommandBehavior.SingleResult)

            While reader.Read
                newItem = New Report()
                newItem.Code = reader("Code")
                newItem.Name = reader("Name")
                newList.Add(newItem)
            End While

            reader.Close()

        Catch ex As Exception
            Throw New ApplicationException("Error occured: " & ex.Message, ex)
        Finally
            conn.Close()
        End Try

        Return newList.ToArray(GetType(Report))
    End Function

    <WebMethod()> _
    Public Function ReportOutput(ByVal reportCode As String) As DataTable
        Dim dt As New DataTable("Report")

        Dim conn As New MySqlConnection(_connString)
        Try
            conn.Open()

            Dim cmd As New MySqlCommand("ServiceReport", conn)
            cmd.CommandType = CommandType.StoredProcedure
            cmd.Parameters.Add(New MySqlParameter("IN_REPORT_CODE", reportCode))

            Dim da As New MySqlDataAdapter(cmd)
            da.Fill(dt)

        Catch ex As Exception
            Throw New ApplicationException("Error occured: " & ex.Message, ex)
        Finally
            conn.Close()
        End Try

        Return dt
    End Function

End Class


<Serializable()> _
Public Class Report

    Private _code As String
    Public Property Code() As String
        Get
            Return _code
        End Get
        Set(ByVal value As String)
            _code = value
        End Set
    End Property

    Private _name As String
    Public Property Name() As String
        Get
            Return _name
        End Get
        Set(ByVal value As String)
            _name = value
        End Set
    End Property

    Private _description As String
    Public Property Description() As String
        Get
            Return _description
        End Get
        Set(ByVal value As String)
            _description = value
        End Set
    End Property

End Class

The syntax for creating a stored procedure is slightly different than PL-SQL, but seems easy enough:


DELIMITER $$

DROP PROCEDURE IF EXISTS `csrm`.`ServiceReport`$$
CREATE DEFINER=`root`@`localhost` PROCEDURE  `csrm`.`ServiceReport`(
    IN IN_REPORT_CODE varchar(5)
)
BEGIN

SELECT * FROM emailAddr;

END $$

DELIMITER ;

Using Oracle SQL Developer Copy Wizard to create local XE schemas

2008-12-05T14:23:00.000-05:00

Teleworking is great. Actually, it’s a modern wonder. There are tons of benefits to the employee and employer. In addition to better QoL scores, in many cases teleworking boosts developer productivity thanks to fewer distractions. There are some disadvantages however. Working remotely places one at the whim of residential bandwidth. Once considered ginormous, 10Mbps connections are almost the norm in many suburban areas, however local conditions can vary and an outage can be devastating. Even when the bits are flowing, downloading large record sets can clog even the biggest pipe costing the developer time during repeated compile/test cycles. That brings us to development “unplugged”. If you are a Microsoft shop, running a local copy of SQL Server express is nothing new, but if your application is tied to an Oracle backend, replicating that was typically beyond the skill set/toolset of the average code jockey. Oracle 10g Express (XE) and SQL Developer has changed all that. Now the process is as simple as adding a new user (see screenshot) to your local database with the browser admin utility and using the dead-simple Copy Wizard in SQL Developer’s Tools menu. Unplugged is fast, but don’t forget to keep your integration server in-synch (daily at a minimum) so your work is not lost if the workstation croaks. Note, the tns-free connection string did make my forehead wrinkle:

Data Source=(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=localhost)(PORT=1521)))(CONNECT_DATA=(SID=XE)(SERVER=DEDICATED)));user id=<user>;password=<password>;

Tip: keep multiple connection strings in your web.config to be able to switch between the local and integration server; whichever key is last, “wins”.

Making the AJAX Control Toolkit 508 Compliant - TabContainer

2008-12-02T15:03:00.001-05:00

The new ASP.NET AJAX Control Toolkit has many useful new controls, but being constrained by Section 508 can make life difficult for those of us wanting to make use of asynchronous browser techniques. Enter the TabContainer control. The TabContainer couldn’t be easier to use. Right out of the box you get a beautiful tab layout using familiar markup:

<ajaxToolkit:TabContainer runat="server" ID="Tabs" Height="138px">

<ajaxToolkit:TabPanel runat="server" ID="Panel3" HeaderText="Email">

Email:

<asp:TextBox ID="emailText" runat="server" />

<asp:Button ID="Button1" runat="server" Text="Save" OnClick="SaveProfile" />

Hit Save to cause a full postback.

</ContentTemplate>

</ajaxToolkit:TabPanel>

This results in a nice looking tabbed control:

Now the bad news: if you turn off JavaScript to test 508 accessibility, nothing is displayed because the tabs are rendered as a DIV with style set to visibility:hidden;

<div id="ctl00_ContentPlaceHolder1_Tabs" class="ajax__tab_xp" style="width:402px;visibility:hidden;">

I couldn’t find an easy way to get at the DIV output other than to “overwrite” the style with an Attribute added on PreRender:

Private Sub Tabs_PreRender(ByVal sender As Object, ByVal e As System.EventArgs) Handles Tabs.PreRender

Tabs.Attributes.Add("style", "visibility:visible")

Tabs.ActiveTab.Attributes.Add("style", "visibility:visible")

End Sub

Now that will get your DIV contents visible again, but the tabs no longer work (more JavaScript), so I changed the TabContainer to use a HeaderTemplate and added NoScript tags with buttons:

Signature and Bio

<asp:Button ID="btnTab0" runat="server" CommandArgument="0"

Text="Signature and Bio Tab" OnCommand="Tab_Command" /></noscript>
</HeaderTemplate>

Add a simple OnCommand handler and you are all set. Note, I prefer to set the handler in this way (declaratively) since I tend to copy/paste my html code and I ~~might~~ always forget to hook up the handler in the CodeBehind:

Protected Sub Tab_Command(ByVal sender As Object, ByVal e As System.Web.UI.WebControls.CommandEventArgs)

Tabs.ActiveTabIndex = Integer.Parse(e.CommandArgument)

End Sub

Without any CSS or HTML trickery this is what you wind up with:

It’s not the most beautiful looking page, but then again, screen readers are by definition not concerned with “look and feel”.