www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - Creator of ZeroMQ and AMQP comments on error handling

reply =?ISO-8859-1?Q?Jos=E9_Armando_Garc=EDa_Sancio?= <jsancio gmail.com> writes:
Hey everyone,

Apologies for sending this to the D mailing list but I just read a
very interesting article on error handling and was interested in your
opinion.

I just read an article by Martin S=FAstrik, one of the creators of AMQP
and ZeroMQ, on using C++ for writing infrastructure/system software.
At first glance most of his objects seem to apply to D. What do you
all think?

Why should I have written ZeroMQ in C, not C++ - http://www.250bpm.com/blog=
:4

Thanks,
-Jose
Jul 15 2012
parent reply "Tobias Pankrath" <tobias pankrath.net> writes:
 What do you all think?


All his arguments about C++ exceptions hold for plain return 
values, too.
Jul 15 2012
next sibling parent reply =?ISO-8859-1?Q?Jos=E9_Armando_Garc=EDa_Sancio?= <jsancio gmail.com> writes:
On Sun, Jul 15, 2012 at 3:00 PM, Tobias Pankrath <tobias pankrath.net> wrote:

 What do you all think?



 All his arguments about C++ exceptions hold for plain return values, too.


Yes but he would said that is not the point of his article. I think
his point is that if you decide not to use exception in C++, for the
reason he and others have mentioned, then C++'s classes/structs become
nothing more than C structs with an uglier/more verbose syntax.

I am wondering if you can extend his argument to D. Implementing RIIA
becomes more verbose. The scope keyword is useless. Thoughts?

Thanks,
-Jose
Jul 15 2012
next sibling parent "Paulo Pinto" <pjmlp progtools.org> writes:
On Sunday, 15 July 2012 at 22:27:02 UTC, José Armando García 
Sancio wrote:

 On Sun, Jul 15, 2012 at 3:00 PM, Tobias Pankrath 
 <tobias pankrath.net> wrote:

 What do you all think?



 All his arguments about C++ exceptions hold for plain return 
 values, too.


 Yes but he would said that is not the point of his article. I 
 think
 his point is that if you decide not to use exception in C++, 
 for the
 reason he and others have mentioned, then C++'s classes/structs 
 become
 nothing more than C structs with an uglier/more verbose syntax.

 I am wondering if you can extend his argument to D. 
 Implementing RIIA
 becomes more verbose. The scope keyword is useless. Thoughts?

 Thanks,
 -Jose


He is completely wrong.

C++ is a multiparadigm language, you can do much more than just 
plain OO. It is a safer language than C, while offering higher 
level abstractions to several programming paradigms. D as a 
better C++ follows these principles as well.

To this day I still don't understand why some developers prefer 
the pain of C  to safer and more productive languages for systems 
programming.


--
Paulo
Jul 16 2012
prev sibling parent "Tobias Pankrath" <tobias pankrath.net> writes:
On Sunday, 15 July 2012 at 22:27:02 UTC, José Armando García 
Sancio wrote:

 On Sun, Jul 15, 2012 at 3:00 PM, Tobias Pankrath 
 <tobias pankrath.net> wrote:

 What do you all think?



 All his arguments about C++ exceptions hold for plain return 
 values, too.


 Yes but he would said that is not the point of his article.


I read the article like this:

Exceptions are bad because they spawn "undefined behaviour", thus 
I don't want to use exceptions. However if I try that in a 
language that is designed to use exceptions, i'll run into 
problems.

And my thoughts about this are: The problems he seems to have 
with exceptions are more or less the same with error codes. So 
I'd say the premise of this article is wrong and the conclusion 
useless. But what if exception where really much worse than error 
codes and you want to avoid them in C++?

Yeah .. you get the same problems than in C, i.e. split 
initialization. So whatever happens you not worse of than in C.

And "what if developer puts code in here that my throw although 
he must not?" is  more or less equivalent then: "What if the 
developer ignores all error codes ..?"

You need to play by the rules and you can't force people to do 
this (Java proofed this). What you need to do is to make 
following rules easier than breaking.
Jul 16 2012
prev sibling parent reply Jonathan M Davis <jmdavisProg gmx.com> writes:
On Sunday, July 15, 2012 15:26:53 Jos=C3=A9 Armando Garc=C3=ADa Sancio =
wrote:

 On Sun, Jul 15, 2012 at 3:00 PM, Tobias Pankrath <tobias pankrath.net=
=20


wrote:

 What do you all think?


=20
 All his arguments about C++ exceptions hold for plain return values=




, too.

=20
 Yes but he would said that is not the point of his article. I think
 his point is that if you decide not to use exception in C++, for the
 reason he and others have mentioned, then C++'s classes/structs becom=


e

 nothing more than C structs with an uglier/more verbose syntax.
=20
 I am wondering if you can extend his argument to D. Implementing RIIA=



 becomes more verbose. The scope keyword is useless. Thoughts?


C++ and D are designed to use exceptions. It's virtually impossible to=20=

completely escape them. However, because D has nothrow, you can at leas=
t=20
guarantee that your code doesn't throw them (though you still have to w=
orry=20
about Errors)., even if it gets ugly in some cases like constructors, f=
orcing=20
you to do nonsense like two-part initialization, and C++ doesn't have t=
hat. D=20
also has stuff like scope and  std.exception.collectException which can=
 help=20
cleanup exception-handling code quite a bit. So, D's better off than C+=
+ is,=20
even if the situation is similar.

But I tend to disagree with anyone who argues that error codes are bett=
er.=20
Exceptions force you to actually deal with the errors that occur, where=
as=20
error codes can easily be skipped. At least some of that complication i=
s=20
intrinsic to the problem, and using error codes instead of exceptions j=
ust=20
means that you're probably not handling a lot of the errors correctly.

- Jonathan M Davis
Jul 15 2012
parent "David Nadlinger" <see klickverbot.at> writes:
On Sunday, 15 July 2012 at 23:17:15 UTC, Jonathan M Davis wrote:

 C++ and D are designed to use exceptions. It's virtually 
 impossible to
 completely escape them.


In many C++ projects, exceptions are disabled at compile time. 
There are only very few parts of STL which are off-limits then.

Sure, if you are using external code, it can still throw 
exceptions, but you have the same problem when using that piece 
of code from an exception-less language.

David

(this post is mainly an experiment if posting from Links works)
Jul 16 2012