Re: [open-cobol-list] cobol binaries

[Sergey Kashyrin]

Helder,

For me it sounds like a problem with a flex/bison.
1. Make sure you are using flex 2.5.4 and bison 2.1 (or at least 1.875 but as I remember I've got a problem with 1.875 on some 
platform)
2. Make sure that all your *.l and *.y files DO NOT have CR+LF at the end of 
the lines but only LF (on UNIX).

If after that some test will fail, please send me the subdirectory with all 
files of the test (for example tests/run.dir/001)
It also make a sense to get the latest tarball from Roger and do a "clean" 
install (http://www.sim-basis.de/open-cobol-0.33.tar.gz)

Regards,
Sergey

----- Original Message ----- 
From: "Helder Velez" <address@hidden>
To: <address@hidden>
Cc: <address@hidden>; <address@hidden>
Sent: Tuesday, May 30, 2006 9:58 PM
Subject: cobol binaries


> Thanks for all your excellent work.
> (and all the other guys of the opencobol team especially Keisuke Nishida and
> Roger While)
>
> under Windows 32-bit your precompiled binaries have done ok  (under VS2003
> and after poking some values on the environment variables PATH,INCLUDE...) .
> Program mt.exe (about manifest file) is not found but it seemed not
> necessary.
>
> Now my problem is :
> I have 64bit solaris 5.11 snv36 (sun nevada build 36)  in a x86_64 (amd64)
> architecture
> I didnt have done successfull tests using 0.33 sources
> almost all tests in make check have failed .
> config.gess returns  i386-pc-solaris2.11
> ( I didnt find where the opencobol figures out that it is a 64bit OS )
>
> arch gives i86pc ; uname -a returns: SunOS newpc 5.11 snv_36 i86pc i386
> i86pc;
> /usr/bin/isalist returns: amd64 pentium_pro+mmx pentium_pro pentium+mmx
> pentium i486 i386 i86
>
> I've tried the gcc3 compiler envirnoment  and also the solaris compiler
> (sunstudio 11)
>
> (I'have included somewere.h  two #define   PACKAGE_VERSION  e PATCH_LEVEL )
>
> make[4]: Entering directory
> `/export/home/geral/open-cobol/open-cobol-0.33/tests'
> ## ---------------------------------------- ##
> ## OpenCOBOL 0.33 test suite: Syntax Tests. ##
> ## ---------------------------------------- ##
>  1: COPY: file not found                         ok
>  2: COPY: replacement order                      FAILED (copy.at:61)
>  3: COPY: separators                             FAILED (copy.at:89)
>  4: COPY: partial replacement                    FAILED (copy.at:117)
>  5: COPY: recursive replacement                  FAILED (copy.at:146)
>  6: Undefined data name                          ok
>  7: Undefined group name                         ok
>  8: Undefined data name in group                 ok
>  9: Reference not a group name                   ok
> 10: Incomplete 01 definition                     ok
> 11: Same labels in different sections            ok
> 12: Redefinition of 01 items                     ok
> 13: Redefinition of 01 and 02 items              ok
> 14: Redefinition of 02 items                     ok
> 15: Redefinition of 77 items                     ok
> 16: Redefinition of 01 and 77 items              ok
> 17: Redefinition of 88 items                     ok
> 18: Ambiguous reference to 02 items              ok
> 19: Ambiguous reference to 02 and 03 items       ok
> 20: Ambigous reference with qualification        ok
>
> ..... ok until end of systax
> .... but from now on almost all tests have failed
>
> Please send `syntax.log' and all information you think might help:
>
>   To: <address@hidden>
>   Subject: [OpenCOBOL 0.33] syntax: 2 3 4 5 failed
>
>
> FAIL: ./syntax
> ## ------------------------------------- ##
> ## OpenCOBOL 0.33 test suite: Run Tests. ##
> ## ------------------------------------- ##
>  1: DISPLAY literals                             FAILED (fundamental.at:38)
>  2: DISPLAY literals, Decimal Point is COMMA     FAILED (fundamental.at:68)
>  3: Hexadecimal literal                          FAILED
> (fundamental.at:102)
>  4: DISPLAY data items with VALUE clause         FAILED
> (fundamental.at:135)
>  5: DISPLAY data items with MOVE statement       FAILED
> (fundamental.at:173)
>  6: non-numeric subscript                        ok
>  7: The range of subscripts                      ok
>  8: Subscript out of bounds (1)                  FAILED (subscripts.at:100)
>  9: Subscript out of bounds (2)                  FAILED (subscripts.at:123)
> 10: Value of DEPENDING ON N out of bounds (lower)FAILED (subscripts.at:146)
> 11: Value of DEPENDING ON N out of bounds (upper)FAILED (subscripts.at:169)
> 12: Subscript bounds with ODO (lower)            ok
> 13: Subscript bounds with ODO (upper)            ok
> 14: Subscript bounds with ODO                    FAILED (subscripts.at:236)
> 15: Subscript by arithmetic expression           FAILED (subscripts.at:261)
> 16: Separate sign positions                      FAILED (ref-mod.at:49)
>
> _________________________________________________________________
> MSN Messenger: converse com os seus amigos online.
> http://messenger.msn.com.br