switch statement

[James Lin]

If you look at the preliminary documentation released at the end of June in
the MSDN library, you'll notice a little blurb on the switch statement. The
document says that there are no "fall-throughs" between each case, but a
break is still required to exit the switch statement. See

http://msdn.microsoft.com/library/pr...spec_8_7_2.htm

But in the September issue of MSDN Magazine, there is an article on C# that
states that there ARE implied break statements between each case. See

http://msdn.microsoft.com/msdnmag/is...arp/csharp.asp

Does this indicate that the design of C#, VB.NET, and .NET in general is in
more flux than we realize? (Or that the author of the magazine article is
simply wrong? 8) )

Thanks,
- Jim
--
James Lin
jlin@ugcs.caltech.edu

http://www.ugcs.caltech.edu/~jlin/

[Robert Scoble]

Keep in mind that anything you read in a magazine is at least four weeks old
(and is probably older than that).

So, just by age alone I would go with the MSDN library's information.

Robert Scoble
http://conferences.devx.com

###


"James Lin" <jlin@ugcs.caltech.edu> wrote in message
news:397a5823$1@news.devx.com...
> If you look at the preliminary documentation released at the end of June
in
> the MSDN library, you'll notice a little blurb on the switch statement.
The
> document says that there are no "fall-throughs" between each case, but a
> break is still required to exit the switch statement. See
>
> http://msdn.microsoft.com/library/pr...spec_8_7_2.htm
>
> But in the September issue of MSDN Magazine, there is an article on C#
that
> states that there ARE implied break statements between each case. See
>
> http://msdn.microsoft.com/msdnmag/is...arp/csharp.asp
>
> Does this indicate that the design of C#, VB.NET, and .NET in general is
in
> more flux than we realize? (Or that the author of the magazine article is
> simply wrong? 8) )
>
> Thanks,
> - Jim
> --
> James Lin
> jlin@ugcs.caltech.edu
>
> http://www.ugcs.caltech.edu/~jlin/
>
>
>

[James Curran]

"James Lin" <jlin@ugcs.caltech.edu> wrote in message
news:397a5823$1@news.devx.com...
> Does this indicate that the design of C#, VB.NET, and .NET in general is
in
> more flux than we realize? (Or that the author of the magazine article is
> simply wrong? 8) )

The MSDN magazine technically isn't wrong, since it never said that
breaks weren't needed. My guess is the either the author glossed over that
point, in an attempt to hit as many points as possible in a brief article,
or wrote the article based on sketchier internal documentation, which itself
did not make the point as clearly as the later version.

--
Truth,
James Curran
http://www.NJTheater.com (Professional)
http://www.NovelTheory.com (Personal)
http://www.BrandsForLess.com (Day Job)

[James Lin]

"James Curran" <jamescurran@mvps.org> wrote in message
news:397dd23b$1@news.devx.com...
> The MSDN magazine technically isn't wrong, since it never said that
> breaks weren't needed.

Yes, it did:

* * *

Another programmer-friendly feature is the improvement over C++ in the way
switch statements work. In C++, you could write a switch statement that fell
through from case to case. For example, this code

switch (i)
{
case 1:
FunctionA();

case 2:
FunctionB();
Break;
}

would call both FunctionA and FunctionB if i was equal to 1. C# works like
Visual Basic, putting an implied break before each case statement.

* * *

> My guess is the either the author glossed over that point, in an attempt
> to hit as many points as possible in a brief article, or wrote the article
> based on sketchier internal documentation, which itself did not make the
> point as clearly as the later version.

That's what I'm thinking too. I understand that details are subject to
change at this early stage.

- Jim
--
James Lin
jlin@ugcs.caltech.edu

http://www.ugcs.caltech.edu/~jlin/

[point]

That is *not* programmer friendly,
that is programmer 'idiot proofing'.

Programmer friendly gives the programmer more control not
less.

besides, MS states that the C# reference is not final,
*not final* is usually implied with any MS product
so if they are explicit about *not final* then you can
bet it will change drastically.


James Lin wrote in message <397e9012$3@news.devx.com>...
>"James Curran" <jamescurran@mvps.org> wrote in message
>news:397dd23b$1@news.devx.com...
>> The MSDN magazine technically isn't wrong, since it never
said that
>> breaks weren't needed.
>
> Yes, it did:
> ***
> Another programmer-friendly feature is the improvement
over C++ in the way
> switch statements work. In C++, you could write a switch
statement that fell
> through from case to case. For example, this code
>
> switch (i)
> {
> case 1:
> FunctionA();
>
> case 2:
> FunctionB();
> Break;
> }
>
> would call both FunctionA and FunctionB if i was equal to
1. C# works like
> Visual Basic, putting an implied break before each case
statement.
>

[Eric Gunnerson]

C# does not look like VB. When you write some code as part of a case,
control cannot fall off the end of the case and into the next case; it must
go somewhere else, via a break, return, goto, or thrown exception.

In other words, it's an error to write something like:

string s = "hello";

switch (s)
{
case "hello":
case "goodbye":
// processing here;
// no break;

case "how are you":
// more processing here.
break;
}

There must be a break on the "no break" line. If you want to have
fall-through, you must make it explicit with a goto case.


"James Lin" <jlin@ugcs.caltech.edu> wrote in message
news:397e9012$3@news.devx.com...
> "James Curran" <jamescurran@mvps.org> wrote in message
> news:397dd23b$1@news.devx.com...
> > The MSDN magazine technically isn't wrong, since it never said that
> > breaks weren't needed.
>
> Yes, it did:
>
> * * *
>
> Another programmer-friendly feature is the improvement over C++ in the way
> switch statements work. In C++, you could write a switch statement that
fell
> through from case to case. For example, this code
>
> switch (i)
> {
> case 1:
> FunctionA();
>
> case 2:
> FunctionB();
> Break;
> }
>
> would call both FunctionA and FunctionB if i was equal to 1. C# works like
> Visual Basic, putting an implied break before each case statement.
>
> * * *
>
> > My guess is the either the author glossed over that point, in an attempt
> > to hit as many points as possible in a brief article, or wrote the
article
> > based on sketchier internal documentation, which itself did not make the
> > point as clearly as the later version.
>
> That's what I'm thinking too. I understand that details are subject to
> change at this early stage.
>
> - Jim
> --
> James Lin
> jlin@ugcs.caltech.edu
>
> http://www.ugcs.caltech.edu/~jlin/
>
>
>

[Paul Vick [MSFT]]

Actually, this is very much like VB, just that VB doesn't require the break.
There is no fallthrough in VB select case statements.

Paul

"Eric Gunnerson" <ericgu@nospam.microsoft.nospam.com> wrote in message
news:398f14b3$1@news.devx.com...
> C# does not look like VB. When you write some code as part of a case,
> control cannot fall off the end of the case and into the next case; it
must
> go somewhere else, via a break, return, goto, or thrown exception.
>
> In other words, it's an error to write something like:
>
> string s = "hello";
>
> switch (s)
> {
> case "hello":
> case "goodbye":
> // processing here;
> // no break;
>
> case "how are you":
> // more processing here.
> break;
> }
>
> There must be a break on the "no break" line. If you want to have
> fall-through, you must make it explicit with a goto case.
>
>
> "James Lin" <jlin@ugcs.caltech.edu> wrote in message
> news:397e9012$3@news.devx.com...
> > "James Curran" <jamescurran@mvps.org> wrote in message
> > news:397dd23b$1@news.devx.com...
> > > The MSDN magazine technically isn't wrong, since it never said that
> > > breaks weren't needed.
> >
> > Yes, it did:
> >
> > * * *
> >
> > Another programmer-friendly feature is the improvement over C++ in the
way
> > switch statements work. In C++, you could write a switch statement that
> fell
> > through from case to case. For example, this code
> >
> > switch (i)
> > {
> > case 1:
> > FunctionA();
> >
> > case 2:
> > FunctionB();
> > Break;
> > }
> >
> > would call both FunctionA and FunctionB if i was equal to 1. C# works
like
> > Visual Basic, putting an implied break before each case statement.
> >
> > * * *
> >
> > > My guess is the either the author glossed over that point, in an
attempt
> > > to hit as many points as possible in a brief article, or wrote the
> article
> > > based on sketchier internal documentation, which itself did not make
the
> > > point as clearly as the later version.
> >
> > That's what I'm thinking too. I understand that details are subject to
> > change at this early stage.
> >
> > - Jim
> > --
> > James Lin
> > jlin@ugcs.caltech.edu
> >
> > http://www.ugcs.caltech.edu/~jlin/
> >
> >
> >
>
>

[Matt Searle]

How about introducing a "CarryOnMate" command if you don't want it to
break - Or I remember a "Continue" command from somewhere.

Matt#
http://csharpindex.com

"Eric Gunnerson" <ericgu@nospam.microsoft.nospam.com> wrote in message
news:398f14b3$1@news.devx.com...
> C# does not look like VB. When you write some code as part of a case,
> control cannot fall off the end of the case and into the next case; it
must
> go somewhere else, via a break, return, goto, or thrown exception.
>
> In other words, it's an error to write something like:
>
> string s = "hello";
>
> switch (s)
> {
> case "hello":
> case "goodbye":
> // processing here;
> // no break;
>
> case "how are you":
> // more processing here.
> break;
> }
> <snip>

[Joe Mayo]

I know I'm going to catch it for even suggesting this, but there is a way to
make a C# switch act like a C/C++ switch:

switch (count) {
case 0:
// do zero stuff
goto one;
case 1:
one:
// do one stuff
goto two;
case 2:
two:
// do two stuff
break;
default:
// do default stuff
break;
}

--
Joe Mayo
C# Station - Information, Links, and Other
Resources for the C# Programming Language
http://www.csharp-station.com/


"Matt Searle" <matt@8001dotcom> wrote in message
news:3997fdf2@news.devx.com...
> How about introducing a "CarryOnMate" command if you don't want it to
> break - Or I remember a "Continue" command from somewhere.
>
> Matt#
> http://csharpindex.com
>
> "Eric Gunnerson" <ericgu@nospam.microsoft.nospam.com> wrote in message
> news:398f14b3$1@news.devx.com...
> > C# does not look like VB. When you write some code as part of a case,
> > control cannot fall off the end of the case and into the next case; it
> must
> > go somewhere else, via a break, return, goto, or thrown exception.
> >
> > In other words, it's an error to write something like:
> >
> > string s = "hello";
> >
> > switch (s)
> > {
> > case "hello":
> > case "goodbye":
> > // processing here;
> > // no break;
> >
> > case "how are you":
> > // more processing here.
> > break;
> > }
> > <snip>
>
>

[ralph]

Hey! I know you.

You are the guy that was holding the "NMAKE RULES!" sign when the IDE Convention
bus rolled by.

I love it. <smile>