[Summary view] [Print] [Text view]

   1  This document is written in pod format hence there are punctuation
   2  characters in odd places.  Do not worry, you've apparently got the
   3  ASCII->EBCDIC translation worked out correctly.  You can read more
   4  about pod in pod/perlpod.pod or the short summary in the INSTALL file.
   5  
   6  =head1 NAME
   7  
   8  README.BS2000 - building and installing Perl for BS2000.
   9  
  10  =head1 SYNOPSIS
  11  
  12  This document will help you Configure, build, test and install Perl
  13  on BS2000 in the POSIX subsystem.
  14  
  15  =head1 DESCRIPTION
  16  
  17  This is a ported perl for the POSIX subsystem in BS2000 VERSION OSD
  18  V3.1A or later.  It may work on other versions, but we started porting
  19  and testing it with 3.1A and are currently using Version V4.0A.
  20  
  21  You may need the following GNU programs in order to install perl:
  22  
  23  =head2 gzip on BS2000
  24  
  25  We used version 1.2.4, which could be installed out of the box with
  26  one failure during 'make check'.
  27  
  28  =head2 bison on BS2000
  29  
  30  The yacc coming with BS2000 POSIX didn't work for us.  So we had to
  31  use bison.  We had to make a few changes to perl in order to use the
  32  pure (reentrant) parser of bison.  We used version 1.25, but we had to
  33  add a few changes due to EBCDIC.  See below for more details
  34  concerning yacc.
  35  
  36  =head2 Unpacking Perl Distribution on BS2000
  37  
  38  To extract an ASCII tar archive on BS2000 POSIX you need an ASCII
  39  filesystem (we used the mountpoint /usr/local/ascii for this).  Now
  40  you extract the archive in the ASCII filesystem without
  41  I/O-conversion:
  42  
  43  cd /usr/local/ascii
  44  export IO_CONVERSION=NO
  45  gunzip < /usr/local/src/perl.tar.gz | pax -r
  46  
  47  You may ignore the error message for the first element of the archive
  48  (this doesn't look like a tar archive / skipping to next file...),
  49  it's only the directory which will be created automatically anyway.
  50  
  51  After extracting the archive you copy the whole directory tree to your
  52  EBCDIC filesystem.  B<This time you use I/O-conversion>:
  53  
  54  cd /usr/local/src
  55  IO_CONVERSION=YES
  56  cp -r /usr/local/ascii/perl5.005_02 ./
  57  
  58  =head2 Compiling Perl on BS2000
  59  
  60  There is a "hints" file for BS2000 called hints.posix-bc (because
  61  posix-bc is the OS name given by `uname`) that specifies the correct
  62  values for most things.  The major problem is (of course) the EBCDIC
  63  character set.  We have german EBCDIC version.
  64  
  65  Because of our problems with the native yacc we used GNU bison to
  66  generate a pure (=reentrant) parser for perly.y.  So our yacc is
  67  really the following script:
  68  
  69  -----8<-----/usr/local/bin/yacc-----8<-----
  70  #! /usr/bin/sh
  71  
  72  # Bison as a reentrant yacc:
  73  
  74  # save parameters:
  75  params=""
  76  while [[ $# -gt 1 ]]; do
  77      params="$params $1"
  78      shift
  79  done
  80  
  81  # add flag %pure_parser:
  82  
  83  tmpfile=/tmp/bison.$$.y
  84  echo %pure_parser > $tmpfile
  85  cat $1 >> $tmpfile
  86  
  87  # call bison:
  88  
  89  echo "/usr/local/bin/bison --yacc $params $1\t\t\t(Pure Parser)"
  90  /usr/local/bin/bison --yacc $params $tmpfile
  91  
  92  # cleanup:
  93  
  94  rm -f $tmpfile
  95  -----8<----------8<-----
  96  
  97  We still use the normal yacc for a2p.y though!!!  We made a softlink
  98  called byacc to distinguish between the two versions:
  99  
 100  ln -s /usr/bin/yacc /usr/local/bin/byacc
 101  
 102  We build perl using GNU make.  We tried the native make once and it
 103  worked too.
 104  
 105  =head2 Testing Perl on BS2000
 106  
 107  We still got a few errors during C<make test>.  Some of them are the
 108  result of using bison.  Bison prints I<parser error> instead of I<syntax
 109  error>, so we may ignore them.  The following list shows
 110  our errors, your results may differ:
 111  
 112  op/numconvert.......FAILED tests 1409-1440
 113  op/regexp...........FAILED tests 483, 496
 114  op/regexp_noamp.....FAILED tests 483, 496
 115  pragma/overload.....FAILED tests 152-153, 170-171
 116  pragma/warnings.....FAILED tests 14, 82, 129, 155, 192, 205, 207
 117  lib/bigfloat........FAILED tests 351-352, 355
 118  lib/bigfltpm........FAILED tests 354-355, 358
 119  lib/complex.........FAILED tests 267, 487
 120  lib/dumper..........FAILED tests 43, 45
 121  Failed 11/231 test scripts, 95.24% okay. 57/10595 subtests failed, 99.46% okay.
 122  
 123  =head2 Installing Perl on BS2000
 124  
 125  We have no nroff on BS2000 POSIX (yet), so we ignored any errors while
 126  installing the documentation.
 127  
 128  
 129  =head2 Using Perl in the Posix-Shell of BS2000
 130  
 131  BS2000 POSIX doesn't support the shebang notation
 132  (C<#!/usr/local/bin/perl>), so you have to use the following lines
 133  instead:
 134  
 135  : # use perl
 136      eval 'exec /usr/local/bin/perl -S $0 $1+"$@"}'
 137          if $running_under_some_shell;
 138  
 139  =head2 Using Perl in "native" BS2000
 140  
 141  We don't have much experience with this yet, but try the following:
 142  
 143  Copy your Perl executable to a BS2000 LLM using bs2cp:
 144  
 145  C<bs2cp /usr/local/bin/perl 'bs2:perl(perl,l)'>
 146  
 147  Now you can start it with the following (SDF) command:
 148  
 149  C</START-PROG FROM-FILE=*MODULE(PERL,PERL),PROG-MODE=*ANY,RUN-MODE=*ADV>
 150  
 151  First you get the BS2000 commandline prompt ('*').  Here you may enter
 152  your parameters, e.g. C<-e 'print "Hello World!\\n";'> (note the
 153  double backslash!) or C<-w> and the name of your Perl script.
 154  Filenames starting with C</> are searched in the Posix filesystem,
 155  others are searched in the BS2000 filesystem.  You may even use
 156  wildcards if you put a C<%> in front of your filename (e.g. C<-w
 157  checkfiles.pl %*.c>).  Read your C/C++ manual for additional
 158  possibilities of the commandline prompt (look for
 159  PARAMETER-PROMPTING).
 160  
 161  =head2 Floating point anomalies on BS2000
 162  
 163  There appears to be a bug in the floating point implementation on BS2000 POSIX
 164  systems such that calling int() on the product of a number and a small
 165  magnitude number is not the same as calling int() on the quotient of
 166  that number and a large magnitude number.  For example, in the following
 167  Perl code:
 168  
 169      my $x = 100000.0;
 170      my $y = int($x * 1e-5) * 1e5; # '0'
 171      my $z = int($x / 1e+5) * 1e5;  # '100000'
 172      print "\$y is $y and \$z is $z\n"; # $y is 0 and $z is 100000
 173  
 174  Although one would expect the quantities $y and $z to be the same and equal
 175  to 100000 they will differ and instead will be 0 and 100000 respectively.
 176  
 177  =head2 Using PerlIO and different encodings on ASCII and EBCDIC partitions
 178  
 179  Since version 5.8 Perl uses the new PerlIO on BS2000.  This enables
 180  you using different encodings per IO channel.  For example you may use
 181  
 182      use Encode;
 183      open($f, ">:encoding(ascii)", "test.ascii");
 184      print $f "Hello World!\n";
 185      open($f, ">:encoding(posix-bc)", "test.ebcdic");
 186      print $f "Hello World!\n";
 187      open($f, ">:encoding(latin1)", "test.latin1");
 188      print $f "Hello World!\n";
 189      open($f, ">:encoding(utf8)", "test.utf8");
 190      print $f "Hello World!\n";
 191  
 192  to get two files containing "Hello World!\n" in ASCII, EBCDIC, ISO
 193  Latin-1 (in this example identical to ASCII) respective UTF-EBCDIC (in
 194  this example identical to normal EBCDIC).  See the documentation of
 195  Encode::PerlIO for details.
 196  
 197  As the PerlIO layer uses raw IO internally, all this totally ignores
 198  the type of your filesystem (ASCII or EBCDIC) and the IO_CONVERSION
 199  environment variable.  If you want to get the old behavior, that the
 200  BS2000 IO functions determine conversion depending on the filesystem
 201  PerlIO still is your friend.  You use IO_CONVERSION as usual and tell
 202  Perl, that it should use the native IO layer:
 203  
 204      export IO_CONVERSION=YES
 205      export PERLIO=stdio
 206  
 207  Now your IO would be ASCII on ASCII partitions and EBCDIC on EBCDIC
 208  partitions.  See the documentation of PerlIO (without C<Encode::>!)
 209  for further posibilities.
 210  
 211  =head1 AUTHORS
 212  
 213  Thomas Dorner
 214  
 215  =head1 SEE ALSO
 216  
 217  L<INSTALL>, L<perlport>.
 218  
 219  =head2 Mailing list
 220  
 221  If you are interested in the VM/ESA, z/OS (formerly known as OS/390)
 222  and POSIX-BC (BS2000) ports of Perl then see the perl-mvs mailing list.
 223  To subscribe, send an empty message to perl-mvs-subscribe@perl.org.
 224  
 225  See also:
 226  
 227      http://lists.perl.org/showlist.cgi?name=perl-mvs
 228  
 229  There are web archives of the mailing list at:
 230  
 231      http://www.xray.mpe.mpg.de/mailing-lists/perl-mvs/
 232      http://archive.develooper.com/perl-mvs@perl.org/
 233  
 234  =head1 HISTORY
 235  
 236  This document was originally written by Thomas Dorner for the 5.005
 237  release of Perl.
 238  
 239  This document was podified for the 5.6 release of perl 11 July 2000.
 240  
 241  =cut