many of the ast regression tests assume the ast-open package I took a quick look at this today the ksh regression tests should be independent of the non-builtin ast commands, but its not worth the effort to non-ast-ize the rest of the tests because we don't want to be testing architecture specific native implementation niggles along with the ast unit under test
On Fri, 13 Aug 2010 03:27:57 +0200 Elias Pipping wrote: > I'm also seeing this: > shcomp-io.ksh[346]: read -n3 from fifo not working > Which occurs randomly, and this (which happened two out of two times): post something if you can get it to reproduce > locale.sh[263]: LANG change not seen by function -- expected > $'(libshell,3,46)\nAn error occurred.\n(libshell,3,46)', got > $'(debug,message,libshell,"An error occurred.")\nAn error > occurred.\n(debug,message,libshell,"An error occurred.")' we're not seeing this on our linux.i386-64 > Something's wrong with the nmake man page by the way: > pipp...@bogus ~ $ ls -l $(find build_no_ccflags/ -name nmake.1) > -rw-r--r-- 1 pipping pipping 43 2010-08-13 02:56 > build_no_ccflags/arch/linux.i386-64/man/man1/nmake.1 > -rw-r--r-- 1 pipping pipping 43 2010-08-13 02:56 > build_no_ccflags/arch/linux.i386-64/src/cmd/nmake/nmake.1 > pipp...@bogus ~ $ cat $(find build_no_ccflags/ -name nmake.1) > make: makerules.mo: cannot read base rules > make: makerules.mo: cannot read base rules nmake.1 is generated look back in the make.out logs and see if can spot any errors _______________________________________________ ast-users mailing list [email protected] https://mailman.research.att.com/mailman/listinfo/ast-users
