From: Eric Hodel <drbrain@...7.net>
Date: 2006-05-19T06:44:19+09:00
Subject: Re: Rescue clauses on do/end blocks?

--Apple-Mail-9-30243354
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
	charset=US-ASCII;
	delsp=yes;
	format=flowed

On May 18, 2006, at 11:22 AM, Jim Weirich wrote:

> Eric Hodel wrote:
>> Setting up and tearing down an exception handler for every block
>> invocation is going to be expensive.
>
> Couldn't the setup/teardown be done only if a rescue/ensure clause was
> actually present?  It also seems to me that this would mean the the
> setup/teardown would have to be in the block itself, not in the  
> calling
> method.

I forgot that the parser will just take care of it.  The patch was  
easy, applies to a recentish copy of the 1.8 branch:


--Apple-Mail-9-30243354
Content-Transfer-Encoding: 7bit
Content-Type: application/octet-stream;
	x-unix-mode=0644;
	name="do-rescue.patch"
Content-Disposition: attachment;
	filename=do-rescue.patch

Index: parse.y
===================================================================
RCS file: /src/ruby/parse.y,v
retrieving revision 1.307.2.34
diff -u -r1.307.2.34 parse.y
--- parse.y	19 Apr 2006 04:57:55 -0000	1.307.2.34
+++ parse.y	18 May 2006 21:36:19 -0000
@@ -1782,7 +1782,7 @@
 			$<num>1 = ruby_sourceline;
 		    }
 		  opt_block_var {$<vars>$ = ruby_dyna_vars;}
-		  compstmt
+		  bodystmt
 		  kEND
 		    {
 			$$ = NEW_ITER($3, 0, dyna_init($5, $<vars>4));
@@ -1845,7 +1845,7 @@
 			$<num>1 = ruby_sourceline;
 		    }
 		  opt_block_var {$<vars>$ = ruby_dyna_vars;}
-		  compstmt '}'
+		  bodystmt '}'
 		    {
 			$$ = NEW_ITER($3, 0, dyna_init($5, $<vars>4));
 			nd_set_line($$, $<num>1);
@@ -1857,7 +1857,7 @@
 			$<num>1 = ruby_sourceline;
 		    }
 		  opt_block_var {$<vars>$ = ruby_dyna_vars;}
-		  compstmt kEND
+		  bodystmt kEND
 		    {
 			$$ = NEW_ITER($3, 0, dyna_init($5, $<vars>4));
 			nd_set_line($$, $<num>1);

--Apple-Mail-9-30243354
Content-Transfer-Encoding: 7bit
Content-Type: text/plain;
	charset=US-ASCII;
	delsp=yes;
	format=flowed


(I'm still not sold on it, it was just really easy.)

> Thus making it hard to rescue exceptions from the calling method  
> (as some have suggested).

I don't believe it is possible to rescue exceptions from the calling  
method.  In fact, I believe this desire comes from a lack of  
understanding of how blocks/procs work.

-- 
Eric Hodel - drbrain@segment7.net - http://blog.segment7.net
This implementation is HODEL-HASH-9600 compliant

http://trackmap.robotcoop.com



--Apple-Mail-9-30243354--