I found a slightly different variant that works by changing the function parameters from SQLBoolean and SQLString to bool and string:

using System.Data.SqlTypes;

using System.Text.RegularExpressions;

using Microsoft.SqlServer.Server;

public partial class UserDefinedFunctions

{

public static readonly RegexOptions...

That was a thought however responsibilities and server authorities are split for DBA here and I've had problems taking that approach with other DBs, hence my search for a method...

Thanks. I'll give that a shot.

Thanks. You'd think people who publish an article titled "How to Truncate All Tables in a Database" would put that little caveat in there...

I got it working with Ragged Right. Since there are no files in the target directory I have to set the columns up individually in Advanced, but it works.

I...

I deleted the invalidly formatted flat file. I deleted the flat file connection. I added a new flat file connection defined as ragged right. I was unable to set the...

That's what I did. Turns out the error message was that the "0" in the expression and the field value were different data types. I changed the formula to use...

I recreated the report from scratch and now it works. I must have mucked up the table/grouping definitions somehow the first time through. RS isn't very forgiving....

This is the process for dynamic grouping that is documented in several places. I've never done it before but it is well documented.

Do you have another approach to dynamic grouping?

I had thought of something like that, or a script task in front of the enumerator to check for the existence of any files and fail if there are no...

I think I just answered my own question. While I was writing up the post something lurked in the back of my mind about temp tables vs. table variables being...

Problem is I have 75 columns and I'd have to reset each one. None of the target table columns are varchar(50) and if the metadata doesn't match between source and...

There's an option, I think it's new to 2008, to "guess" the datatypes by scanning "n" rows where "n" can be specified. The problem with that is I have numeric...

I was thinking there should be an option to map the input source schema to the output schema but there doesn't seem to be one.

I suppose there's a way to...

Nobody has a solution for this? Really??:crazy: