Comments on: My C# Wish List http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/ Thoughts while working and playing as a Software Developer Mon, 10 Jun 2013 15:59:00 +0000 hourly 1 http://wordpress.org/?v=3.4.2 By: Michael Robin http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-392 Michael Robin Tue, 12 Jun 2012 21:01:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-392 A reasonable typle syntax and tuple-unpacking/destructuring/assignment is useful and a good core language feature - and agree about something like a?.b?.c?.c -- this is a pain every programmer deals with every day. Other customizations like your syntax-embedding/string interpolation could be handled by making C# a better langage for meta-programming, so that your own DSL can be implemented within C# without writing seperate native code extensions, etc.   A reasonable typle syntax and tuple-unpacking/destructuring/assignment is useful and a good core language feature – and agree about something like a?.b?.c?.c — this is a pain every programmer deals with every day.
Other customizations like your syntax-embedding/string interpolation could be handled by making C# a better langage for meta-programming, so that your own DSL can be implemented within C# without writing seperate native code extensions, etc.  

]]> By: Paul Smith http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-381 Paul Smith Wed, 06 Jun 2012 20:52:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-381 The string tokenizing feature you describe would definitely be extremely useful. Implementing it in/on the string class itself could be very problematic though. I think the idea is a great candidate for a 'TokenString' type struct with overloaded implicit conversion operators to/from String that behave the way you describe.  The string tokenizing feature you describe would definitely be extremely useful. Implementing it in/on the string class itself could be very problematic though. I think the idea is a great candidate for a ‘TokenString’ type struct with overloaded implicit conversion operators to/from String that behave the way you describe. 

]]> By: Anil http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-377 Anil Wed, 06 Jun 2012 06:12:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-377  Example provided in "String Interpolation" section is present in PHP. But don't know why microsoft has not yet introduced such kind of feature in .Net  Example provided in “String Interpolation” section is present in PHP. But don’t know why microsoft has not yet introduced such kind of feature in .Net

]]> By: Chris Missal http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-366 Chris Missal Wed, 06 Jun 2012 02:01:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-366 Definitely. Or use something other than double quotes too. Powershell allows for single and double quotes, but interpolation only occurs when using the double quotes. Definitely. Or use something other than double quotes too. Powershell allows for single and double quotes, but interpolation only occurs when using the double quotes.

]]> By: Sam http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-365 Sam Wed, 06 Jun 2012 01:56:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-365 You could use some other operator in front of the string. Like @ changes the behaviour of a literal perhaps another operator could enable string interpolation. No breaking change that way You could use some other operator in front of the string. Like @ changes the behaviour of a literal perhaps another operator could enable string interpolation. No breaking change that way

]]> By: Rudi http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-364 Rudi Tue, 05 Jun 2012 07:30:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-364 Do all of these new features sound like Roslyn 'extensions' of some sort? I mean, once roslyn is available in the language (C#5?) could these be build as 'nuget-able' extensions in your project? Personally I'd rather prefer to keep C# lighter and have a 'pick-and-choose' way to include  language features.  e.g. I would not favour the 'unless' statement. Just refactor the if (!condition) to an if (ReverseCondition()) Do all of these new features sound like Roslyn ‘extensions’ of some sort?

I mean, once roslyn is available in the language (C#5?) could these be build as ‘nuget-able’ extensions in your project?

Personally I’d rather prefer to keep C# lighter and have a ‘pick-and-choose’ way to include  language features. 

e.g. I would not favour the ‘unless’ statement. Just refactor the if (!condition) to an if (ReverseCondition())

]]> By: Chris Missal http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-363 Chris Missal Mon, 04 Jun 2012 20:59:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-363 In all honesty, working a bit in PS is where some of these came from. In all honesty, working a bit in PS is where some of these came from.

]]> By: Steven Murawski http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-362 Steven Murawski Mon, 04 Jun 2012 20:55:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-362 Two of the three are native capabilities in PowerShell (string interpolation and anonymous tuples) and that always trips me up moving between PowerShell and C# Two of the three are native capabilities in PowerShell (string interpolation and anonymous tuples) and that always trips me up moving between PowerShell and C#

]]> By: Chris Bristol http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-361 Chris Bristol Mon, 04 Jun 2012 19:15:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-361 The compiler is going to need to do some sort of transform on the string anyway, so it can easily (and arguably should) check to be sure that the expression inside the $(...) block is valid. I think the biggest resistance for such a feature is that it is a breaking change; any program which contains a string to be printed of the form $(...) would be broken. The compiler is going to need to do some sort of transform on the string anyway, so it can easily (and arguably should) check to be sure that the expression inside the $(…) block is valid. I think the biggest resistance for such a feature is that it is a breaking change; any program which contains a string to be printed of the form $(…) would be broken.

]]> By: Michael McGuire http://lostechies.com/chrismissal/2012/05/31/my-c-wish-list/#comment-360 Michael McGuire Sun, 03 Jun 2012 21:54:00 +0000 http://lostechies.com/chrismissal/?p=134#comment-360 I Instapaper'ed this blog entry to post this very feature request later, but you beat me to it. I find the null coalescing operating to be basically useless, as I typically want to read some property off an object reference but have to check the object reference first. I Instapaper’ed this blog entry to post this very feature request later, but you beat me to it. I find the null coalescing operating to be basically useless, as I typically want to read some property off an object reference but have to check the object reference first.

]]>