>Number:         12838
>Category:       bin
>Synopsis:       new expr(1) is totally broken
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    bin-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri May 04 16:30:00 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     Ross Harvey
>Release:        NetBSD-current
>Organization:
	
>Environment:
	
System: NetBSD sigmet 1.5U NetBSD 1.5U (skb) #28: Thu Apr 26 19:28:16 PDT 2001 ross@sigmet:/usr/ross/skb i386
Architecture: i386
Machine: i386
>Description:
	The new expr(1) implements priorities that are different from
	every other commonly used unix system. It broke a script used
	by my mailer.

	Software 
>How-To-Repeat:
	$ expr 2 \> 1 \* 17

	NetBSD returns 17, every one else returns 0.
>Fix:
	Don't replace software that has been tested for years with casually
	done rewrites. Fix the bugs. Change is the enemy of reliability.

	If a rewrite is necessary, have a senior developer do it. 

	If a rewrite is necessary, have it reviewed by several serious
	developers.

	Don't commit widely used tools in critical paths without basic
	testing. (Never mind, _serious_ testing.)

	Don't check in code just because it compiles and runs successfully
	once.

	If you rewrite or extensively change a core utility, include
	real regression tests. This change included a mere TWO calls
	to expr as regression tests, both of which tested only a
	completely obscure feature.

	Working on NetBSD is a _serious_ activity and must be done
	thoughtfully and with care. Find developers who understand this.
>Release-Note:
>Audit-Trail:
>Unformatted: