Test Failures in Latest CVS

Test Failures in Latest CVS

[Vin Shelton]

Hello,

The 5/21 build passes all tests, but builds from the current CVS do not:

...

../../../src/zsh-2003-05-23/Test/B02typeset.ztst: starting.
ZTST_test:76: no such file or directory: 
../../../src/zsh-2003-05-23/Test/ztst.zsh:421: bad math expression: operand expected at `'
: all tests successful.
ZTST_cleanup:2: command not found: rm

...

../../../src/zsh-2003-05-23/Test/E01options.ztst: starting.
*** /tmp/zsh.ztst.out.15277	Fri May 23 20:25:18 2003
--- /tmp/zsh.ztst.tout.15277	Fri May 23 20:25:18 2003
***************
*** 1,12 ****
  inside1 scalar-local-export
! inside2 scalar-local-export
  inside3 scalar-export
  outside1 scalar
! outside2 scalar
  inside3 scalar-export
  inside1 scalar-export
! inside2 scalar-local-export
  inside3 scalar-export
  inside1 scalar-export
! outside2 scalar
  inside3 scalar-export
--- 1,12 ----
  inside1 scalar-local-export
! inside2 scalar-export
  inside3 scalar-export
  outside1 scalar
! inside2 scalar-export
  inside3 scalar-export
  inside1 scalar-export
! inside2 scalar-export
  inside3 scalar-export
  inside1 scalar-export
! inside2 scalar-export
  inside3 scalar-export
Test ../../../src/zsh-2003-05-23/Test/E01options.ztst failed: output differs from expected as shown above for:
  showit() { local v;
             for v in first second third; do
               eval print \$$v \$\{\(t\)$v\}
	     done;
           }      
  setit() { typeset -x first=inside1;
            typeset +g -x second=inside2;
            typeset -g -x third=inside3;
            showit;
          }
  first=outside1 second=outside2 third=outside3
  unsetopt globalexport
  setit
  showit
  setopt globalexport
  setit
  showit
  unfunction setit showit
Was testing: GLOBAL_EXPORT option
../../../src/zsh-2003-05-23/Test/E01options.ztst: test failed.

This is on a

Linux zion.rcn.com 2.4.21-0.13mdkcustom #4 SMP Sat Mar 8 23:08:38 EST 2003 i686 unknown unknown GNU/Linux

system, but I believe I saw the same failures on my Solaris systems at
work.

HTH,
  Vin

Re: Test Failures in Latest CVS

[Peter Stephenson]

Vin Shelton wrote:
> Hello,
> 
> The 5/21 build passes all tests, but builds from the current CVS do not:

This is because I changed OPT_ISSET() to return either 1 or 0, based on
noticing that some code expects that, but other bits of code were using
OPT_ISSET() to test for specific bits.

Index: Src/builtin.c
===================================================================
RCS file: /cvsroot/zsh/zsh/Src/builtin.c,v
retrieving revision 1.101
diff -u -r1.101 builtin.c
--- Src/builtin.c	15 May 2003 09:39:57 -0000	1.101
+++ Src/builtin.c	24 May 2003 22:39:26 -0000
@@ -2145,7 +2145,7 @@
     }
 
     if (!(OPT_ISSET(ops,'g') || OPT_ISSET(ops,'x') || OPT_ISSET(ops,'m')) || 
-	OPT_ISSET(ops,'g') == 2 || *name == 'l' ||
+	OPT_PLUS(ops,'g') || *name == 'l' ||
 	(!isset(GLOBALEXPORT) && !OPT_ISSET(ops,'g')))
 	on |= PM_LOCAL;
 
@@ -2263,7 +2263,7 @@
 		returnval = 1;
 		continue;
 	    }
-	    if (OPT_PLUS(ops,'m') == 2 && !asg->value) {
+	    if (OPT_PLUS(ops,'m') && !asg->value) {
 		scanmatchtable(paramtab, pprog, on|roff, 0,
 			       paramtab->printnode, printflags);
 		continue;
@@ -2351,7 +2351,7 @@
     int on = 0, off = 0, pflags = 0;
 
     /* Do we have any flags defined? */
-    if (OPT_ISSET(ops,'u') == 2)
+    if (OPT_PLUS(ops,'u'))
 	off |= PM_UNDEFINED;
     else if (OPT_MINUS(ops,'u') || OPT_ISSET(ops,'X'))
 	on |= PM_UNDEFINED;

-- 
Peter Stephenson <pws@pwstephenson.fsnet.co.uk>
Work: pws@csr.com
Web: http://www.pwstephenson.fsnet.co.uk

Re: Test Failures in Latest CVS

[Vin Shelton]

Peter Stephenson <pws@pwstephenson.fsnet.co.uk> writes:

> Vin Shelton wrote:
>> Hello,
>> 
>> The 5/21 build passes all tests, but builds from the current CVS do not:
>
> This is because I changed OPT_ISSET() to return either 1 or 0, based on
> noticing that some code expects that, but other bits of code were using
> OPT_ISSET() to test for specific bits.

Yes, that works.  Thanks, Peter!  I love the zsh testsuite!

  - Vin