That's what the using statement is for. The using clause automatically disposes of any IDisposable objects, including MemoryStream, thus it will call the appropriate close method when the scope leaves the using clause. Tech, life, family, faith: Give me a visit. I'm currently blogging about: Cops & Robbers Judah Himango

Here's a chapter from a design book on the subject Exceptions[^]. Also, Effective Java and More Effective C++ have good material. I have't seen Effective C#, but I suspect it has a similar take on them and they will all support your argument. I'm not sure the performance angle is the best one to pursue though. I think it is better to point out that it is simply misuse of the feature. The purpose of exceptions is to deal with exceptional conditions. Matt Gerrans

Dr Herbie wrote: Solution - in case anyone was interested ... Cool! And thanks for posting the solution! I've been wondering about what the answer might be, but certainly didn't have enough knowledge to contribute anything. Marc MyXaml Advanced Unit Testing YAPO

Static methods are like normal methods with respect to thread safety. You do need to protect them with some kind of synchronization mechanism.. By the way, are you sure you'd want to the use the MethodImpl attribute for synchronization? AFAIK, all methods with that attribute share a single lock, so if one thread is executing one method, no other thread will be able to execute *any* method with that attribute. Regards Senthil _____________________________ My Blog | My Articles | WinMacro

Thanks for the replies, it has put my mind at rest :) Dr Herbie Remember, half the people out there have below average IQs.

ASP.NET doesn't support Modal dialogs because that would be dependent on the browser supporting them. Also, web applications are stateless, which means that on each page request the server sees it as if it is the first time it has ever happened. ASP.NET provides some smokescreen illusions to make it appear as if it is a continuous application to make development easier, however this abstraction doesn't work out in many places the developer has to work around these limitations. "If a man empties his purse into his head, no man can take it away from him, for an investment in knowledge pays the best interest." -- Joseph E. O'Donnell Can't manage to P/Invoke that Win32 API in .NET? Why not do interop the wiki way!

:laugh: I gotta remember that line! Jeremy Kimball Catapultam habeo. Nisi pecuniam omnem mihi dabis, ad caput tuum saxum immane mittam. (I have a catapult. Give me all the money, or I will fling an enormous rock at your head)

Daniel Turini wrote: Did I mention before that I hate code generation tools? so you'd better never use microsoft products... TOXCCT >>> GEII power

If the class "owns" the objects (e.g., private fields) then, yes, you should dispose them. If the class is merely a container for the objects (e.g., a collection like ArrayList), then no, the class user is responsible for that. Due to technical difficulties my previous signature, "I see dumb people" will be off until further notice. Too many people were thinking I was talking about them... :sigh:

Arjan Einbu wrote: GUIDs are unique. Will they do? My intent for this design is to remove any explicit token passing. I already have a session object which has to be passed accross to the remote object to identify the user. I wanted to avoid having to explicitly pass any extra variables, so I was hoping that the thread/context/appdomain might contain something I could use to uniquely identify the client. E.g. Instead of: public void SAO:PerformAction(token ClientIdentifier, int parameter) { CheckSecurity(ClientIdentifier); DoStuff(parameter); } I could do: public void SAO:PerformAction(int parameter) { CheckSecurity( GetUniqueTokenFromCallersContext() ); DoStuff(parameter); } I'm trying to make the remoting interfaces as seamless as possible and I personally hate having to have an identifier token passed to every SAO method. X| Maybe I'm hoping for too much. :sigh: Dr Herbie. Remember, half the people out there have below average IQs.

OK, almost there. I continued delving around in MSDN and found the article : DataSets, Web Services, DiffGrams, Arrays, and Interoperability[^] So I now have a web method that returns an array of classes. Hooray.:) But is it still possible to return a typed collection (as opposed to an array)? Dr Herbie. Remember, half the people out there have below average IQs.

Using scripts from SQLServer is all very well, but we have to support ALL databases; SQLServer, Oracle, FoxPro, DBase, MySQL, MS Access, etc., (including any databases not yet written) CREATE TABLE syntax is similar between different DB vendors, but not identicle. Therefore we need a vendor-agnostic method. ADOX supplied this, as the driver created the actual SQL statement to create the table from a structure you created in code. I was hoping ADO.NET would have something similar, but it doesn't look promising.:( One option would be to get SQL Server to dump out table structures as XML, then use different XSLT files to transform the XML into database specific script files, but I was hoping to avoid any of that. Ho-hum, better dig that 'Learn XSLT in one easy lesson' book out.:sigh: Dr Herbie. Remember, half the people out there have below average IQs.

I'm not sure you can with the generic OLE provider. I do know you can with the native Oracle Provider for .NEt. Works alot like the native SQL provider. Mark Conger Sonork:100.28396