Comments on: A Rant of Sorts on Multiple .Net Web References http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/ Sat, 07 Nov 2009 20:20:37 +0000 http://wordpress.org/?v=wordpress-mu-1.0 by: Ryan Cromwell http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-384 Fri, 05 Jan 2007 16:30:21 +0000 http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-384 Brett, I have to disagree. Matt's suggestion is exactly the solution. As Scott says initially, the easy things are easy. Matt is showing the easy solution to a harder problem. Is it the easiest, no, but it's definitely easier than manually aggregating WSDL's and mucking with namespaces. Honestly, a simple batch file with 1 line per refernce would do everything "Web References" are doing for the simplest problem/solution. Brett,
I have to disagree. Matt’s suggestion is exactly the solution. As Scott says initially, the easy things are easy. Matt is showing the easy solution to a harder problem. Is it the easiest, no, but it’s definitely easier than manually aggregating WSDL’s and mucking with namespaces.

Honestly, a simple batch file with 1 line per refernce would do everything “Web References” are doing for the simplest problem/solution.

]]> by: Scott Balmos http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-383 Fri, 05 Jan 2007 16:21:52 +0000 http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-383 Indeed, I have to agree with Brett that this is a workaround, not a solution. A solution would be a fix within Visual Studio itself.  Indeed, I have to agree with Brett that this is a workaround, not a solution. A solution would be a fix within Visual Studio itself. 

]]> by: Brett Bim http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-382 Fri, 05 Jan 2007 15:15:43 +0000 http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-382 Regarding Matt's suggestion, I'm not so sure that's a resolution as much as it is a workaround. The supposed beauty of web references is that when (not if!) the reference changes, you can update it via Visual Studio and be on your way. We are currently experiencing the same pain from common types across services and are looking into making our common types more intelligent. They will be able, through reflection, to change themselves into the correct type for the correct service. Luckily, our common types are mostly dumb data structures so this is relatively straightforward. Maybe you could do something similar. Regarding Matt’s suggestion, I’m not so sure that’s a resolution as much as it is a workaround. The supposed beauty of web references is that when (not if!) the reference changes, you can update it via Visual Studio and be on your way.

We are currently experiencing the same pain from common types across services and are looking into making our common types more intelligent. They will be able, through reflection, to change themselves into the correct type for the correct service. Luckily, our common types are mostly dumb data structures so this is relatively straightforward. Maybe you could do something similar.

]]> by: Jimmy http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-381 Fri, 05 Jan 2007 15:06:34 +0000 http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-381 Scott, It's particularly ironic that you authored this post yesterday. As I leave this comment, my team is in the process of refactoring our common types so that they contain the ability to convert themselves from proxy generated common types to server side common types and vice versa. As of now, our common types are really just data structures. We are thinking of using reflection to negotiate the namespace on the proxy type. The structure of the type is known, so reflection is not necessary to populate the rest of the values. I'll let you know how this turns out. Scott,
It’s particularly ironic that you authored this post yesterday. As I leave this comment, my team is in the process of refactoring our common types so that they contain the ability to convert themselves from proxy generated common types to server side common types and vice versa. As of now, our common types are really just data structures. We are thinking of using reflection to negotiate the namespace on the proxy type. The structure of the type is known, so reflection is not necessary to populate the rest of the values. I’ll let you know how this turns out.

]]> by: Matt Dunn http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-380 Fri, 05 Jan 2007 03:02:39 +0000 http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-380 Hi Scott, This issue is resolved in .NET 2.0 with the introduction of the /sharetypes option for wsdl.exe (admittedly requires using the command line feature directly, not the Add Web Reference dialog in Visual Studio) heres an example of this feature: http://www.theserverside.net/tt/blogs/showblog.tss?id=WSStrikesBackP6 Cheers, Matt Hi Scott,

This issue is resolved in .NET 2.0 with the introduction of the /sharetypes option for wsdl.exe (admittedly requires using the command line feature directly, not the Add Web Reference dialog in Visual Studio) heres an example of this feature: http://www.theserverside.net/tt/blogs/showblog.tss?id=WSStrikesBackP6

Cheers,
Matt

]]> by: dovholuk http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-379 Thu, 04 Jan 2007 21:52:25 +0000 http://tssblog.blogs.techtarget.com/2007/01/04/a-rant-of-sorts-on-multiple-net-web-references/#comment-379 future blog suggestion: i'd personally like to see a blog entry around "polymorphism" and interoperability... are choice groups ok? substitution groups? abstract types? i personally abhor the first two methods preferring abstract types but i don't know how the "abstract" keyword fits in with all the "other" programming languages (ie: not c based) what are the interop blog thoughts on thats subject? future blog suggestion:
i’d personally like to see a blog entry around “polymorphism” and interoperability… are choice groups ok? substitution groups? abstract types?

i personally abhor the first two methods preferring abstract types but i don’t know how the “abstract” keyword fits in with all the “other” programming languages (ie: not c based)

what are the interop blog thoughts on thats subject?

]]>