From ben@petering.net

# "Normal" exception in try {} is caught
BEGIN {
  try {
  widdle;
  CATCH {
  say "caught​: $!";
  }
  }
}

# But "use" is not.
BEGIN {
  try {
  use NonExistent; # dies, not caught.
  CATCH {
  say "caught​: $!";
  }
  }
}

Output (with Rakudo pulled 2009-08-08 06​:40 UTC, how on earth do you
specify revisions with git?)

  caught​: Could not find non-existent sub widdle
  Can't find ./NonExistent in @​*INC
  in Main (src\gen_setting.pm​:445)

(I'm not sure if this is correct, but I'm struggling to imagine why
this would be useful behavior.)

From ben@petering.net

In the second example, I'm not sure if that exception is a Control
exception... even if so, the problem still exists, because substituting
CONTROL for CATCH in that example results in the same output.

ben@petering.net - Status changed from 'new' to 'open'

From @moritz

On Sat Aug 08 00​:00​:00 2009, missingthepoint wrote​:

# But "use" is not.
BEGIN {
try {
use NonExistent; # dies, not caught.
CATCH {
say "caught​: $!";
}
}
}

Output (with Rakudo pulled 2009-08-08 06​:40 UTC, how on earth do you
specify revisions with git?)

$ git-rev-parse HEAD
c13df6b482fefa9bd6f70572978ce0b32ede6b32

that's the revision (or simpler​: git-show | head )

 caught​: Could not find non\-existent sub widdle
 Can't find \./NonExistent in @​\*INC
 in Main \(src\\gen\_setting\.pm​:445\)

(I'm not sure if this is correct, but I'm struggling to imagine why
this would be useful behavior.)

'use' is executed at compile time, and try { } only catches run time
exceptions. You'd be very much surprised to get no parse errors from
within a try { } block, would you?

At the time of 'use' the try block isn't even parsed yet.

Perl 5 does the same, btw​:

$ perl -we 'eval { use Nonexisting }'
Can't locate Nonexisting.pm in @​INC (@​INC contains​:
/home/moritz/cpan/lib /etc/perl /usr/local/lib/perl/5.10.0
/usr/local/share/perl/5.10.0 /usr/lib/perl5 /usr/share/perl5
/usr/lib/perl/5.10 /usr/share/perl/5.10 /usr/local/lib/site_perl .) at
-e line 1.
BEGIN failed--compilation aborted at -e line 1.

So I reject this report as "not a bug".

Cheers,
Moritz

@moritz - Status changed from 'open' to 'rejected'