Comments for Andrew Murphy's E-Memory http://andrew-murphy.co.uk Development “how-to”s and gotchas that I always forget. Mon, 06 Sep 2010 16:56:12 +0000 http://wordpress.org/?v=2.7.1 hourly 1 Comment on C# 3.0 Implicitly Typed Local Variables by Andrew Murphy http://andrew-murphy.co.uk/?p=95&cpage=1#comment-10 Andrew Murphy Wed, 11 Mar 2009 18:51:37 +0000 http://andrew-murphy.co.uk/?p=95#comment-10 Try it. It does! Or rather "int i = (i = 20)" does, which is what I meant to type :-). I didn't believe it though and it would never pass a code review but there you are, for some reason it does. The example is lifted directly from the Microsoft documentation <a href="http://msdn.microsoft.com/en-us/library/bb384061.aspx" rel="nofollow">here</a>, under remarks. Try it. It does! Or rather “int i = (i = 20)” does, which is what I meant to type :-). I didn’t believe it though and it would never pass a code review but there you are, for some reason it does.

The example is lifted directly from the Microsoft documentation here, under remarks.

]]> Comment on C# 3.0 Implicitly Typed Local Variables by Matt http://andrew-murphy.co.uk/?p=95&cpage=1#comment-9 Matt Sat, 07 Mar 2009 04:47:49 +0000 http://andrew-murphy.co.uk/?p=95#comment-9 I'd hazard a guess that even: int i = (i + 20); won't compile due to it being unassigned when you are trying to use it! Matt :) I’d hazard a guess that even:

int i = (i + 20);

won’t compile due to it being unassigned when you are trying to use it!

Matt :)

]]> Comment on C# 3.0 Enhancements (brief summary) by admin http://andrew-murphy.co.uk/?p=71&cpage=1#comment-3 admin Wed, 18 Feb 2009 12:03:08 +0000 http://andrew-murphy.co.uk/?p=71#comment-3 I agree with everything that you said (I've just not posted that far into the enhancements yet) although LINQ to SQL is being discouraged I believe due to the inevitable architectural silliness that it encourages. I'll get through C# 3.0 just in time for C# 4.0 hopefully. I agree with everything that you said (I’ve just not posted that far into the enhancements yet) although LINQ to SQL is being discouraged I believe due to the inevitable architectural silliness that it encourages.

I’ll get through C# 3.0 just in time for C# 4.0 hopefully.

]]> Comment on C# 3.0 Enhancements (brief summary) by Jeremy http://andrew-murphy.co.uk/?p=71&cpage=1#comment-2 Jeremy Wed, 18 Feb 2009 09:16:09 +0000 http://andrew-murphy.co.uk/?p=71#comment-2 I agree with you about implicit typing. I haven't seen much take-up of this, luckily. Good for quick prototyping - very bad for production quality. Extension methods can be very useful - I could add some convenience methods to the interface where they logically belonged without changing the interface itself. Also means that none of the classes that implemented the interface required any modification. Big win. LINQ against SQL Server seems to run into limitations very quickly or maybe I did not persevere for long enough. Maybe a little like C++ templates, really simple to use but as soon as you run into a problem, the effort of trying to fix it becomes huge. The LINQ extensions to IEnumerable are very useful - especially Count() - but I am a big fan of IEnumerable because of its immutability. Lamda expressions and Expression trees are world changing like you say but I suspect that they will never really be understandable for all us mere mortals. At least LINQ makes them simpler. I agree with you about implicit typing. I haven’t seen much take-up of this, luckily. Good for quick prototyping - very bad for production quality.

Extension methods can be very useful - I could add some convenience methods to the interface where they logically belonged without changing the interface itself. Also means that none of the classes that implemented the interface required any modification. Big win.

LINQ against SQL Server seems to run into limitations very quickly or maybe I did not persevere for long enough. Maybe a little like C++ templates, really simple to use but as soon as you run into a problem, the effort of trying to fix it becomes huge.

The LINQ extensions to IEnumerable are very useful - especially Count() - but I am a big fan of IEnumerable because of its immutability.

Lamda expressions and Expression trees are world changing like you say but I suspect that they will never really be understandable for all us mere mortals. At least LINQ makes them simpler.

]]>