http://blogs.msdn.com/curth/archive/2008/07/31/this-is-dynamic.aspxRegarding our need to check-and-rethrow exceptions in IronPython , int19h asks "Shouldn't exception filters do the trick without the need to rethrow explicitly?" That's a good question. In fact, I had the exact same reaction when I first started lookingen-USCommunityServer 2.1 SP1 (Build: 61025.2)http://blogs.msdn.com/curth/archive/2008/07/31/this-is-dynamic.aspx#8797775Fri, 01 Aug 2008 09:40:26 GMT91d46819-8472-40ad-a661-2c78acb4018c:8797775int19h<p>Personally, I am of the opinion that anyone using this edge case in production code deserves to be drawn and quartered publicly... but yeah, I can understand the need to stay true to the spec.</p> http://blogs.msdn.com/curth/archive/2008/07/31/this-is-dynamic.aspx#8797792Fri, 01 Aug 2008 09:42:59 GMT91d46819-8472-40ad-a661-2c78acb4018c:8797792int19h<p>Even so, wouldn't it be possible to optimize the case where the value used in &quot;except&quot; cannot be changed later on (e.g., it's a name of a .NET class) to a filter block?</p> http://blogs.msdn.com/curth/archive/2008/07/31/this-is-dynamic.aspx#8800029Fri, 01 Aug 2008 17:29:30 GMT91d46819-8472-40ad-a661-2c78acb4018c:8800029curth<p>We don't do any identification of names to see if they match a .NET class name. &nbsp;The top of the file could say &quot;from System import ApplicationException&quot; or it could say &quot;class ApplicationException(object): pass&quot;. &nbsp;Or it could pick between the two based on some criterion. &nbsp;Either way, we won't know what the value of the symbol is until we actually get to execute the statement &quot;except ApplicationException&quot;.</p>