DBD::Unify --  a Unify interface for Perl 5 using DBI.

This is the Database Driver for the Unify family of database products.
See http://www.unify.com/Products/DataServer/default.aspx for Unify
DataServer products.

See the ChangeLog file for a complete history

REQUIREMENTS

   It has been ported and tested on DataServer 6.3AB, 6.3BE, 8.1CC, 8.1CE,
   8.1DAB, 8.2AC, 8.2AD, 8.2B, 8.2BC, 8.3E/64, 8.3F/64, 8.3I/64, 8.3KAB/64,
   and 9.1B, but this does not mean it might not work on U2000 databases or
   other DataServer releases.

   I use DBI 1.637 and perl-5.8.7 through perl5.26.0, but might get it to
   work with DBI 0.93 or higher and perl 5.6.2 (the modern toolchain will
   require 5.8.1 as of Apr 2013). But then again, in that case you should
   consider an upgrade. As of DBD-Unify-0.40 I've made DBI-1.42 prerequisite.
   As of DBD-Unify-0.70, I have made perl-5.6 a prerequisite.

   The test case however uses scalarIO to catch errors, so if you use a
   perl that does not support scalarIO (e.g. perlIO enabled), you will
   have to disable the failing tests that use it yourself and hope for
   the best (or better, upgrade to a perl that supports scalarIO).

   If you plan to use DataServer 8.1 or higher database in Unicode encoding
   (UTF-8), use perl5.8.4 or up. Note that $LANG will need to match your
   database locale. See the Unicode site for the differences between the
   Unicode standards if you want to know what to expect
   (http://www.unicode.org/versions/enumeratedversions.html) and
   http://www.unicode.org/history/publicationdates.html for their age.

     Perl version
     Stable  Devel   Unicode
     ------  ------  -------
             5.27.2   10.0.0  UCD 10.0.0 released 20 Jun 2017
     5.26.0  5.25.3    9.0.0  UCD  9.0.0 released 21 Jun 2016
     5.24.0  5.23.0    8.0.0  UCD  8.0.0 released 17 Jun 2015
     5.22.0  5.21.1    7.0.0  UCD  7.0.0 released 16 Jun 2014
     5.20.0  5.19.5    6.3.0  UCD  6.3.0 released 30 Sep 2013
     5.18.0  5.17.1    6.2.0  UCD  6.2.0 released 26 Sep 2012
     5.16.0  5.15.8    6.1.0  UCD  6.1.0 released 31 Jan 2012
     5.14.0  5.13.7    6.0.0  UCD  6.0.0 released 11 Oct 2010
     5.12.0  5.11.3    5.2.0  UCD  5.2.0 released 01 Oct 2009
     5.10.1  5.10.1    5.1.0  UCD  5.1.0 released 04 Apr 2008
     5.10.0  5.9.5     5.0.0  UCD  5.0.0 released 14 Jul 2006
     5.8.7   5.8.7     4.1.0  UCD  4.1.0 released 31 Mar 2005
     5.8.5   5.8.5     4.0.1  UCD  4.0.1 released    Mar 2004
     5.8.1   5.8.1     4.0.0  UCD  4.0.0 released 24 Apr 2003
     5.8.0   5.8.0     3.2.0  UCD  3.2.0 released 02 Apr 2002
     5.7.x   5.7.x     3.1.1  UCD  3.1.1 released    Aug 2001
                       3.0.x  UCD  3.0.0 released    Sep 1999
                              UCD  1.0.0 draft was planned spring 1991 :)

   Use 'corelist -a Unicode' to see the complete list of what version of
   Unicode is shipped with what version of perl.

   One user has reported success with perl 5.005_03 on SunOS 5.7 with
   gcc-2.7.2.2, DBI-1.30, DBD-Unify-0.24 with DataServer-7.2.

   Todd Zervas reported working versions:
   - 0.65 working on 32bit 9.0G on Linux 2.4.21-47.EL
   - 0.76 working on 32bit 9.0G on Linux 2.6.18-92.1.22.el5PAE
   - 0.77 working on 32bit 9.0G on Linux 2.6.18-128.1.14.el5PAE
                   with GNU g++ 4.1.2, perl-5.8.8, and DBI-1.609

BUILDING, TESTING AND INSTALLING:

   This will only work if DBI is installed on a recent version of perl ;-)
   and you are working in a valid Unify environment. Makefile.PL will stop
   with a warning if you are not.

          # perl Makefile.PL

   This step will ask you if you need networked databases. Probably you do
   not, which is the default answer. Choosing 'Yes' will include two more
   libraries that are broken in 8.1CC for HP-UX. Choosing 'No' will prefer
   U2000ul.a and S2000ul.a over U2000u.a and S2000u.a resulting in smaller
   binaries with local-only support.

   If you are using gcc, it might be needed to explicitly tell Unify to
   use it, because ucc defaults to cc (always):

          # export UPPNAME=/usr/local/pa20_32/bin/gcc
          # export UCCNAME=/usr/local/pa20_32/bin/gcc
          # export ULDNAME=/usr/local/pa20_32/bin/gcc

   Another approach would be:

          # ln -s /usr/local/pa20_32/bin/gcc cc
          # export PATH=".:$PATH"

   Now you need to build the objects:

          # make

   And if that succeeds, run the tests:

          # make test

   Make test will use the current database and create a table "xx". If
   the test fails, table xx might still exist. If it already exists,
   it might get lost, so be aware!

   Make test will try to test all DBI features found in the DBI-1.19 docs
   even some that are not DBD dependent. Partly to show that it still works
   for DBD::Unify, and partly to find what features are still to be
   implemented. Features known to fail will issue an ok when ran with
   make test, but will show additional comments with a todo tag when run
   directly (e.g. 'perl t/11-dbi-dbh.t'). Note that the t/21-uni-regex test
   skips the test for 'SHLIKE' because this will cause core dumps. Advice:
   do *not* use SHLIKE, but use REGLIKE (which is quite easy to convert to
   from SHLIKE). All versions above A/SQL 8.2AC should have this problem
   solved. The test is enabled by default. If you have an older (crashing)
   version, disable this test by removing the hash in line

     #if ("This test is known to fail") { ok ("shlike will core", 1) } else

   If you happen to have a DataServer only product (no accell), make
   will probably complain about a missing ACCELL.a from Makefile.PL
   The workaround seems to be to remove the reference to ACCELL.a from
   Makefile.PL (reported by Ron Kuris)

          # make install

   Will install the DBD::Unify extension into the default perl path's. If
   you plan to run different perl versions and/or different Unify versions
   alongside eachother, use the installu.pl script instead, which installs
   DBD::Unify in the Unify tree [ .../lib/perl ], and extend the environment
   variable $PERL5LIB with $UNIFY/perl

         # make installu

   As with all modules: Patches are welcome.

SOURCE CODE

   Recent changes can be (re)viewed in the public GIT repository at
   https://github.com/perl5-dbi/DBD-Unify
   Feel free to clone your own copy:

    $ git clone https://github.com/perl5-dbi/DBD-Unify DBD-Unify

   or get it as a tgz:

    $ wget --output-document=DBD-Unify.tgz \
           https://github.com/perl5-dbi/DBD-Unify/archive/master.tar.gz

SHARED BUILDS on HP-UX

   As of somewhere in the DataServer 8.2 track, Unify has finally indulged
   to my request for shared library prepared objects, at least on HP-UX,
   where they only ship static libraries. Objects inside these libraries
   are now built with the +Z option, so it is easy to convert the static
   libraries (.a) to shared libraries (.sl). The Makefile build process
   will propose such transition, and - once succeeded - use it for the now
   shared object, which is *much* smaller than when built statically. This
   does not always work. If things fail, choose for the static route.

NAMING CONFLICTS

   In their enormous wisdom, Unify have chosen a function named "inflate"
   to be exported in their libraries. This will conflict with libz. If
   you use any module that uses a shared lib that requires libz, this
   will result in core dumps that are hard to trace. A possible solution
   is to rename the function *after* building. This is - of course - very
   very very dirty, but if it works, it will circumvent the problem:

    $ make
    $ perl -pi -e's/\0inflate\0/\0InFlate\0/g' \
         blib/arch/auto/DBD/Unify/Unify.so
    $ make test

RESTRICTIONS:

   This is a basic port for Unify. At the moment of writing, this DBD
   does not support all DBI elements, methods and attributes. I'll add
   these the moment I need them myself, or one of you will supply them
   as patches.

   DATE and TIME fields are not fully supported. The database interaction
   works, but there are some issues left to iron out in the user
   interface. We're working on it. Suggestions welcome.

   TEXT and BINARY fields are supported as of version 0.60, but probably
   not fully tested. Feedback and patches welcome.

   Other known restrictions:

   # Incomplete or out-of-date documentation, though I keep trying to
     keep in pace with what I change

   # Some database attributes not (yet) implemented. (AutoCommit et all)

   # Most statement attributes are now implemented. Those needed for
     $sth->fetchrow_hashref () now work, but STORE_attrib () doesn't.

     DBD::Unify tries to return the {TYPE} attribute in ANSI/ODBC DBI
     compliant coding, and has the original Unify field type available
     in the {uni_types} attribute. For AMOUNT and HUGE AMOUNT however,
     ANSI has no defined codes. These will be returned as SQL_FLOAT and
     SQL_DOUBLE, with the {SCALE} set to 2.

   And deficiencies/shortcommings that come with Unify:

   # Unify does not supply the "NULLABLE" indicator when describing
     fields, the returned value is as documented always 2.

   # Unify does not return the desired length of non-character type
     fields, but the /maximum/ PRECISION for those fields

   # Unify has no VARCHAR concept

   # Unify only distributes static libraries, so the module might
     become rather big. Libraries on AIX are always shared, and some
     of Unify's libraries can safely be converted to shared libs,
     but that is no guarantee to work.

   # At least until 8.2, Unify library pool does *NOT* allow more then
     one different database open at the same time from within the same
     process (even when the first database has correctly disconnected
     and closed) :-(

     This problem is known  with Unify and registered in their bug
     database as 'usb100963: disconnect'. Unify Sacramento has finally
     located the bug's source after yet again I reported this in
     bug report 22960. DBD-Unify 0.25 and up have a workaround for all
     versions up to and including 8.2AD. It's fixed in 8.2B.

TODO:

   See module documentation for specific TODO items

   # Remove as many restrictions as possible
   # Add more tests
   # Investigate if LIKE and SHLIKE operators in positional parameters
     indeed do not work (I've seen that, but found no time yet to look
     into this more). Currently it cores on SHLIKE. See t/21-uni-regex.t
   # Try finding ways to increase performance, squeezing every last
     nanosecond out of each handle
   # If you're into Unify internals, try finding out why describing
     positional parameters in where clauses (not in selected fields)
     causes a core dump on the field name. Look in dbdimp.ic in function
     dbd_prm_describe () for comment about OSF/1.

ACKNOWLEDGEMENTS

   Thanks to Tom Poage <[email protected]> for keeping
   the speed in the early development and help in adding the missing
   features in the earlier stages of this DBD interface.

   Thanks to Todd Zervas <[email protected]> for digging in the
   binary issues on DS 9.0.

COPYRIGHT AND LICENSE

   Copyright (C) 1999-2017  H.Merijn Brand <[email protected]>

   This library is free software; you can redistribute it and/or modify
   it under the same terms as Perl itself.

As always, have the appropriate amount of fun!

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.