mirror of
				git://git.code.sf.net/p/cdesktopenv/code
				synced 2025-03-09 15:50:02 +00:00 
			
		
		
		
	tests/sh_match.sh: redirect ulimit to 2>/dev/null
macOS 12.2.1 doesn't seem to like the -M, -v or -d ulimit options:
  sh_match.sh[502]: FAIL: test_xmlfragment1/0/testfile1.xml:
  Expected empty stderr, got $'test1_script.sh[2]: ulimit: 1048576:
  limit exceeded [Invalid argument]\ntest1_script.sh[3]: ulimit:
  1048576: limit exceeded [Invalid argument]\ntest1_script.sh[4]:
  ulimit: 1048576: limit exceeded [Invalid argument]'
The 'Invalid argument' addition is caused by errno==EINVAL and
suggests the OS either doesn't support setting this limit, or
support for it was somehow disabled.
src/cmd/ksh93/tests/sh_match.sh:
- Redirect standard error for ulimit commands to 2>/dev/null. If
  they fail it's pretty inconsequential and it's not related to
  actual ${.sh.match} testing at all.
Resolves: https://github.com/ksh93/ksh/issues/459
Thanks to @posguy99 for the report.
			
			
This commit is contained in:
		
							parent
							
								
									4ca578bde4
								
							
						
					
					
						commit
						fc5bd8e8c3
					
				
					 1 changed files with 6 additions and 4 deletions
				
			
		| 
						 | 
				
			
			@ -130,9 +130,11 @@ function test_xmlfragment1
 | 
			
		|||
	typeset -r testscript='test1_script.sh'
 | 
			
		||||
cat >"${testscript}" <<-TEST1SCRIPT
 | 
			
		||||
	# memory safeguards to prevent out-of-control memory consumption
 | 
			
		||||
	ulimit -M \$(( 1024 * 1024 ))
 | 
			
		||||
	ulimit -v \$(( 1024 * 1024 ))
 | 
			
		||||
	ulimit -d \$(( 1024 * 1024 ))
 | 
			
		||||
	{
 | 
			
		||||
		ulimit -M \$(( 1024 * 1024 ))
 | 
			
		||||
		ulimit -v \$(( 1024 * 1024 ))
 | 
			
		||||
		ulimit -d \$(( 1024 * 1024 ))
 | 
			
		||||
	} 2>/dev/null
 | 
			
		||||
 | 
			
		||||
	# input text
 | 
			
		||||
	xmltext="\$( < "\$1" )"
 | 
			
		||||
| 
						 | 
				
			
			@ -149,7 +151,7 @@ cat >"${testscript}" <<-TEST1SCRIPT
 | 
			
		|||
		nameref ar="\$1"
 | 
			
		||||
 | 
			
		||||
		# fixme:
 | 
			
		||||
		# - We want to enforce standard conformance - does ~(Exp) or ~(Ex-p) does that ?
 | 
			
		||||
		# - We want to enforce standard conformance - does ~(Exp) or ~(Ex-p) do that ?
 | 
			
		||||
		dummy="\${xmltext//~(Ex-p)(?:
 | 
			
		||||
			(<!--.*-->)+?|			# xml comments
 | 
			
		||||
			(<[:_[:alnum:]-]+
 | 
			
		||||
| 
						 | 
				
			
			
 | 
			
		|||
		Loading…
	
	Add table
		Add a link
		
	
		Reference in a new issue