[Tcsh] How to debug bug 74

[Christos Zoulas]

Hi, is there a test that fails right now? I don't see it. Or do you have a reproducer
for the bad cleanup?

Thanks,

christos

> On Apr 8, 2019, at 5:39 AM, Dr. Werner Fink <werner at suse.de> wrote:
> 
> On Mon, Apr 08, 2019 at 02:04:37PM +0200, Dr. Werner Fink wrote:
> 
>> --- tcsh-6.20.00/sh.c	2017-07-19 12:29:15.512471750 +0200
>> +++ tcsh-6.20.00/sh.c	2019-04-08 13:35:32.462771640 +0200
>> @@ -2188,14 +2188,15 @@ process(int catch)
>> 	    stderror(ERR_OLD);
>> 	}
>> 
>> -	postcmd();
>> 	/*
>> 	 * Execute the parse tree From: Michael Schroeder
>> 	 * <mlschroe at immd4.informatik.uni-erlangen.de> was execute(t, tpgrp);
>> 	 */
>> 	execute(t, (tpgrp > 0 ? tpgrp : -1), NULL, NULL, TRUE);
>> 	freesyn(t);
>> +	setcopy(STR_, InputBuf, VAR_READWRITE | VAR_NOGLOB);
>> 
>> +	postcmd();	/* The potential aliasrun() might destroy the parse tree */
>> 	/*
>> 	 * Made it!
>> 	 */
>> @@ -2204,7 +2205,6 @@ process(int catch)
>> 	    (void) check_window_size(0);	/* for window systems */
>> 	}
>> #endif /* SIG_WINDOW */
>> -	setcopy(STR_, InputBuf, VAR_READWRITE | VAR_NOGLOB);
>>     cmd_done:
>> 	if (cleanup_reset())
>> 	    cleanup_until(&paraml);
> 
> Hmm ... know the test case aliases.at at line 32 fails due
> reverse order of the output.  Any way to get the parse tree
> of the aliasrun() and process() not to interfere?
> 
> Werner
> 
> -- 
>  "Having a smoking section in a restaurant is like having
>          a peeing section in a swimming pool." -- Edward Burr
> -- 
> Tcsh mailing list
> Tcsh at astron.com
> https://mailman.astron.com/mailman/listinfo/tcsh