more information from catch(...)

[Khuong Dinh Pham]

How can i get more information about the exception when using catch(...)?
Thx in advance

[Ian]

Khuong Dinh Pham wrote:

How can i get more information about the exception when using catch(...)?

You can't.

You have to try a catch specific typed first, like

catch( const std::exception& e )
{
}
catch( ... )
{
}

Ian

[Stefan Näwe]

Khuong Dinh Pham wrote:

How can i get more information about the exception when using catch(...)?
Thx in advance

No.

--
Stefan Naewe
naewe.s_AT_atla s_DOT_de

[Alf P. Steinbach]

* Khuong Dinh Pham:

How can i get more information about the exception when using catch(...)?

You can rethrow the exception (using "throw;") and catch it again with more
specific catch clauses.

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Owen Jacobson]

On Wed, 12 Oct 2005 08:33:57 +0000, Alf P. Steinbach wrote:

* Khuong Dinh Pham:

How can i get more information about the exception when using catch(...)?

You can rethrow the exception (using "throw;") and catch it again with more
specific catch clauses.

Of course, if you're going to do that you might as well catch the specific
exceptions in the first place... Unless I'm missing something important,
which is entirely possible.

-O

[Alf P. Steinbach]

* Owen Jacobson:

On Wed, 12 Oct 2005 08:33:57 +0000, Alf P. Steinbach wrote:

* Khuong Dinh Pham:

How can i get more information about the exception when using catch(...)?

You can rethrow the exception (using "throw;") and catch it again with more
specific catch clauses.

Of course, if you're going to do that you might as well catch the specific
exceptions in the first place... Unless I'm missing something important,
which is entirely possible.

Yes, the possibility of centralizing the exception type determination and
relevant info extraction in a function, thus reducing code redundancy.

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Nick Keighley]

Alf P. Steinbach wrote:

* Owen Jacobson:

On Wed, 12 Oct 2005 08:33:57 +0000, Alf P. Steinbach wrote:

* Khuong Dinh Pham: How can i get more information about the exception when using catch(...)?

You can rethrow the exception (using "throw;") and catch it again with more
specific catch clauses.

Of course, if you're going to do that you might as well catch the specific
exceptions in the first place... Unless I'm missing something important,
which is entirely possible.

Yes, the possibility of centralizing the exception type determination and
relevant info extraction in a function, thus reducing code redundancy.

hmm. that sounds like something I want to do. Could you expand on this
a bit?
I'd like to put a bit long list of catches in a function but quite see
how to
do that.
--
Nick Keighley

[Rolf Magnus]

Alf P. Steinbach wrote:

* Owen Jacobson:

On Wed, 12 Oct 2005 08:33:57 +0000, Alf P. Steinbach wrote:

> * Khuong Dinh Pham:
>> How can i get more information about the exception when using
>> catch(...)?
>
> You can rethrow the exception (using "throw;") and catch it again with
> more specific catch clauses.

Of course, if you're going to do that you might as well catch the
specific
exceptions in the first place... Unless I'm missing something important,
which is entirely possible.

Yes, the possibility of centralizing the exception type determination and
relevant info extraction in a function, thus reducing code redundancy.

Why would I need to catch the exception in the first place if the only thing
to do with it is to throw it again?

[Alf P. Steinbach]

* Nick Keighley:

Alf P. Steinbach wrote:

* Owen Jacobson:

On Wed, 12 Oct 2005 08:33:57 +0000, Alf P. Steinbach wrote:
> * Khuong Dinh Pham:> How can i get more information about the exception when using catch(...)?
>
> You can rethrow the exception (using "throw;") and catch it again with more
> specific catch clauses.

Of course, if you're going to do that you might as well catch the specific
exceptions in the first place... Unless I'm missing something important,
which is entirely possible.

Yes, the possibility of centralizing the exception type determination and
relevant info extraction in a function, thus reducing code redundancy.

hmm. that sounds like something I want to do. Could you expand on this
a bit?
I'd like to put a bit long list of catches in a function but quite see
how to
do that.

Trust me, you don't _want_ to do that. But if you're using a bunch of
libraries that can thow various kinds of non-standard exceptions it can be a
more clean approach than the alternatives. It goes like this, off the cuff:

std::string exceptionText()
{
try
{
throw;
}
catch( unsigned x )
{
return "tulso tones exception #" + stringFrom( x );
}
catch( CExpectNot* pOuch )
{
std::string const result =
"Mucho Fishy Casserole exception: " + pOuch->Massage();
pOuch->Destroy();
return result;
}
catch( std::exception const& x )
{
return x.what();
}
catch( ... )
{
return "Unknown exception";
}
}

void foo()
{
try
{
lib1Func(); lib2Func(); lib3Func();
}
catch( ... )
{
throw std::runtime_er ror( exceptionText() );
}
}

and then bar(), similar, and so on.

This technique won't work with version 6.0 and earlier of MSVC, because of a
compiler bug.

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?

[Alf P. Steinbach]

* Rolf Magnus:

Alf P. Steinbach wrote:

* Owen Jacobson:

On Wed, 12 Oct 2005 08:33:57 +0000, Alf P. Steinbach wrote:

> * Khuong Dinh Pham:
>> How can i get more information about the exception when using
>> catch(...)?
>
> You can rethrow the exception (using "throw;") and catch it again with
> more specific catch clauses.

Of course, if you're going to do that you might as well catch the
specific
exceptions in the first place... Unless I'm missing something important,
which is entirely possible.

Yes, the possibility of centralizing the exception type determination and
relevant info extraction in a function, thus reducing code redundancy.

Why would I need to catch the exception in the first place if the only thing
to do with it is to throw it again?

See my reply to Nick Keighley.

--
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on usenet and in e-mail?