++ed by:

1 PAUSE user
1 non-PAUSE user.

H.Merijn Brand


DBD::Unify - DBI driver for Unify database systems


 # Examples marked NYT are Not Yet Tested, they might work
 #  all others have been tested.
 # man DBI for explanation of each method (there's more than listed here)

 $dbh = DBI->connect ("DBI:Unify:[\$dbname]", "", $schema, {
                         AutoCommit    => 0,
                         ChopBlanks    => 1,
                         uni_unicode   => 0,
                         uni_verbose   => 0,
                         uni_scanlevel => 2,
 $dbh = DBI->connect_cached (...);                   # NYT
 $dbh->do ($statement);
 $dbh->do ($statement, \%attr);
 $dbh->do ($statement, \%attr, @bind);

 $all = $dbh->selectall_arrayref ($statement);
 @row = $dbh->selectrow_array ($statement);
 $col = $dbh->selectcol_arrayref ($statement);

 $sth = $dbh->prepare ($statement);
 $sth = $dbh->prepare ($statement, \%attr);
 $sth = $dbh->prepare_cached ($statement);           # NYT
 @row = $sth->fetchrow_array;
 $row = $sth->fetchrow_arrayref;
 $row = $sth->fetchrow_hashref;
 $all = $sth->fetchall_arrayref;

 # Statement has placeholders like where field = ?
 $sth = $dbh->prepare ($statement);
 $sth->bind_param ($p_num, $bind_value);             # NYT
 $sth->bind_param ($p_num, $bind_value, $bind_type); # NYT
 $sth->bind_param ($p_num, $bind_value, \%attr);     # NYT
 $sth->bind_col ($col_num, \$col_variable);          # NYT
 $sth->bind_columns (@list_of_refs_to_vars_to_bind);
 $sth->execute (@bind_values);

 $cnt = $sth->rows;

 $sql = $dbh->quote ($string);

 $err = $dbh->err;
 $err = $sth->err;
 $str = $dbh->errstr;
 $str = $sth->errstr;
 $stt = $dbh->state;
 $stt = $sth->state;

 For large DB fetches the combination $sth->bind_columns ()
 with $sth->fetchrow_arrayref is the fastest (DBI


DBD::Unify is an extension to Perl which allows access to Unify databases. It is built on top of the standard DBI extension and implements the methods that DBI requires.

This document describes the differences between the "generic" DBD and DBD::Unify.


returned types

The DBI docs state that:

   Most data is returned to the perl script as strings (null values
   are returned as undef).  This allows arbitrary precision numeric
   data to be handled without loss of accuracy.  Be aware that perl
   may  not preserve the same accuracy when the string is used as a

Integers are returned as integer values (perl's IV's).

(Huge) amounts, floats, reals and doubles are returned as strings for which numeric context (perl's NV's) has been invoked already, so adding zero to force convert to numeric context is not needed.

Chars are returned as strings (perl's PV's).

Chars, Dates, Huge Dates and Times are returned as strings (perl's PV's). Unify represents midnight with 00:00, not 24:00.

    connect ("DBI:Unify:dbname[;options]" [, user [, auth [, attr]]]);

Options to the connection are passed in the data-source argument. This argument should contain the database name possibly followed by a semicolon and the database options which are ignored.

Since Unify database authorization is done using grant's using the user name, the user argument may be empty or undef. The auth field will be used as a default schema. If the auth field is empty or undefined connect will check for the environment variable $USCHEMA to use as a default schema. If neither exists, you will end up in your default schema, or if none is assigned, in the schema PUBLIC.

At the moment none of the attributes documented in DBI's "ATTRIBUTES COMMON TO ALL HANDLES" are implemented specifically for the Unify DBD driver, but they might have been inherited from DBI. The ChopBlanks attribute is implemented, but defaults to 1 for DBD::Unify. The Unify driver supports "uni_scanlevel" to set the transaction scan level to a value between 1 and 16 and "uni_verbose" to set DBD specific debugging, allowing to show only massages from DBD-Unify without using the default DBI->trace () call.

The connect call will result in statements like:

    SET TRANSACTION SCAN LEVEL 7;  -- if attr has { uni_scanlevel => 7 }

local database

    connect ("/data/db/unify/v63AB", "", "SYS")

It is recommended that the connect call ends with the attributes { AutoCommit = 0 }>, although it is not implemented (yet).

If you don't want to check for errors after every call use { AutoCommit = 0, RaiseError => 1 }> instead. This will die with an error message if any DBI call fails.


By default, this driver is completely Unicode unaware: what you put into the database will be returned to you without the encoding applied.

To enable automatic decoding of UTF-8 when fetching from the database, set the uni_unicode attribute to a true value for the database handle (statement handles will inherit) or to the statement handle.

  $dbh->{uni_unicode} = 1;

When CHAR or TEXT fields are retrieved and the content fetched is valid UTF-8, the value will be marked as such.


Though both the syntax and the module support connecting to different databases, even at the same time, the Unify libraries seem to quit connecting to a new database, even if the old one is closed following every rule of precaution.

To be safe in closing a handle of all sorts, undef it after it is done with, it will than be destroyed. (As of 0.12 this is tried internally for handles that proved to be finished)


 my $dbh = DBI->connect (...);
 my $sth = $dbh->prepare (...);
 $sth->finish;     undef $sth;
 $dbh->disconnect; undef $dbh;

or implicit:

 {   my $dbh = DBI->connect (...);
     {   my $sth = $dbh->prepare (...);
         while (my @data = $sth->fetchrow_array) {
         }  # $sth implicitly destroyed by end-of-scope
     }  # $dbh implicitly destroyed by end-of-scope
 $dbh->do ($statement)

This is implemented as a call to 'EXECUTE IMMEDIATE' with all the limitations that this implies.

commit and rollback invalidates open cursors

DBD::Unify does warn when a commit or rollback is issued on a $dbh with open cursors.

Possibly a commit/rollback/disconnect should also undef the $sth's. (This should probably be done in the DBI-layer as other drivers will have the same problems).

After a commit or rollback the cursors are all ->finish'ed, i.e. they are closed and the DBI/DBD will warn if an attempt is made to fetch from them.

A future version of DBD::Unify might re-prepare the statement.

Stuff implemented in perl


Just here for DBI. No use in telling the end-user what to do with it :)


There is no way for Unify to tell what data sources might be available. There is no central files (like /etc/oratab for Oracle) that lists all available sources, so this method will always return an empty list.


As DBI's quote_identifier () gladly accepts the empty string as a valid identifier, I have to override this method to translate empty strings to undef, so the method behaves properly. Unify does not allow to select NULL as a constant as in:

    select NULL, foo from bar;
prepare ($statement [, \%attr])

The only attribute currently supported is the dbd_verbose (or its alias uni_verbose) level. See "trace" below.

table_info ($;$$$$)
columne_info ($$$$)
foreign_key_info ($$$$$$;$)
primary_key ($$$)
uni_clear_cache ()

Note that these five get their info by accessing the SYS schema which is relatively extremely slow. e.g. Getting all the primary keys might well run into seconds, rather than milliseconds.

This is work-in-progress, and we hope to find faster ways to get to this information. Also note that in order to keep it fast across multiple calls, some information is cached, so when you alter the data dictionary after a call to one of these, that cached information is not updated.

For column_info (), the returned DATA_TYPE is deduced from the TYPE_NAME returned from SYS.ACCESSIBLE_COLUMNS. The type is in the ODBC range and the original Unify type and type_name are returned in the additional fields uni_type and uni_type_name. Somehow selecting from that table does not return valid statement handles for types currency and huge integer.

  Create as           sth attributes       uni_type/uni_type_name
  ------------------- -------------------  -------------------------
  amount              FLOAT             6   -4 AMOUNT (9, 2)
  amount (5, 2)       FLOAT             6   -4 AMOUNT (5, 2)
  huge amount         REAL              7   -6 HUGE AMOUNT (15, 2)
  huge amount (5, 2)  REAL              7   -6 HUGE AMOUNT (5, 2)
  huge amount (15, 2) REAL              7   -6 HUGE AMOUNT (15, 2)
  byte                BINARY           -2  -12 BYTE (1)
  byte (512)          BINARY           -2  -12 BYTE (512)
  char                CHAR              1    1 CHAR (1)
  char (12)           CHAR              1    1 CHAR (12)
  currency            DECIMAL           3    - ?
  currency (9)        DECIMAL           3    - ?
  currency (7,2)      DECIMAL           3    - ?
  date                DATE              9   -3 DATE
  huge date           TIMESTAMP        11  -11 HUGE DATE
  decimal             NUMERIC           2    2 NUMERIC (9)
  decimal (2)         NUMERIC           2    2 NUMERIC (2)
  decimal (8)         NUMERIC           2    2 NUMERIC (8)
  double precision    DOUBLE PRECISION  8    8 DOUBLE PRECISION (64)
  float               DOUBLE PRECISION  8    6 FLOAT (64)
  huge integer        HUGE INTEGER     -5    - ?
  integer             NUMERIC           2    2 NUMERIC (9)
  numeric             NUMERIC           2    2 NUMERIC (9)
  numeric (2)         SMALLINT          5    2 NUMERIC (2)
  numeric (6)         NUMERIC           2    2 NUMERIC (6)
  real                REAL              7    7 REAL (32)
  smallint            SMALLINT          5    2 NUMERIC (4)
  text                TEXT             -1   -9 TEXT
  time                TIME             10   -7 TIME

Currently the driver tries to cache information about the schema as it is required. When there are fields added, removed, or altered, references are added or removed or primary keys or unique hashes are added or removed it is wise to call $dbh->uni_clear_cache to ensure that the info on next inquiries will be up to date.


Stuff implemented in C (XS)


The DBI->trace (level) call will promote the level to DBD::Unify, showing both the DBI layer debugging messages as well as the DBD::Unify specific driver-side debug messages.

It is however also possible to trace only the DBD-Unify without the DBI->trace () call by using the uni_verbose attribute on connect () or by setting it later to the database handle, the default level is set from the environment variable $DBD_TRACE if defined:

  $dbh = DBI->connect ("DBI::Unify", "", "", { uni_verbose => 3 });
  $dbh->{uni_verbose} = 3;

As DBD::Oracle also supports this scheme since version 1.22, dbd_verbose is a portable alias for uni_verbose, which is also supported in DBD::Oracle.

DBD::Unify now also allows an even finer grained debugging, by allowing dbd_verbose on statement handles too. The default dbd_verbose for statement handles is the global dbd_verbose at creation time of the statement handle.

The environment variable DBD_VERBOSE is used if defined and overrules $DBD_TRACE.

  $dbh->{dbd_verbose} = 4;
  $sth = $dbh->prepare ("select * from foo");  # sth's dbd_verbose = 4
  $dbh->{dbd_verbose} = 3;                     # sth's dbd_verbose = 4
  $sth->{dbd_verbose} = 5;                     # now 5

Currently, the following levels are defined:

1 & 2

No DBD messages implemented at level 1 and 2, as they are reserved for DBI

  DBD::Unify::dbd_db_STORE (ScanLevel = 7)
  DBD::Unify::st_prepare u_sql_00_000000 ("select * from foo")
  DBD::Unify::st_prepare u_sql_00_000000 (<= 4, => 0)
  DBD::Unify::st_execute u_sql_00_000000
  DBD::Unify::st_destroy 'select * from parm'
  DBD::Unify::st_free u_sql_00_000000
  DBD::Unify::st 0x7F7F25CC 0x0000 0x0000 0x00000000 0x00000000 0x00000000
  DBD::Unify::st destroyed

Level 3 plus errors and additional return codes and field types and values:

  DBD::Unify::st_prepare u_sql_00_000000 ("select c_bar from foo where c_foo = 1")
      After allocate, sqlcode = 0
      After prepare,  sqlcode = 0
      After allocate, sqlcode = 0
      After describe, sqlcode = 0
      After count,    sqlcode = 0, count = 1
  DBD::Unify::fld_describe o_sql_00_000000 (1 fields)
      After get,      sqlcode = 0
  DBD::Unify::st_prepare u_sql_00_000000 (<= 1, => 0)
  DBD::Unify::st_execute u_sql_00_000000
      After open,     sqlcode = 0 (=> 0)
  DBD::Unify::st_fetch u_sql_00_000000
      Fetched         sqlcode = 0, fields = 1
      After get,      sqlcode = 0
       Field   1: c_bar: NUMERIC  4: (6030) 6030 ==
       Fetch done
  DBD::Unify::st_finish u_sql_00_000000
      After close,    sqlcode = 0
  DBD::Unify::st_destroy 'select c_bar from foo where c_foo = 1'
  DBD::Unify::st_free u_sql_00_000000
      After deallocO, sqlcode = 0
      After deallocU, sqlcode = 0

Level 4 plus some content info:

  DBD::Unify::st_fetch u_sql_00_000000
      Fetched         sqlcode = 0, fields = 1
      After get,      sqlcode = 0
       Field   1: [05 00 04 00 00] c_bar: NUMERIC  4: (6030) 6030 ==
       Fetch done

Level 5 plus internal coding for exchanges and low(er) level return codes:

  DBD::Unify::fld_describe o_sql_00_000000 (1 fields)
      After get,      sqlcode = 0
       Field   1: [05 00 04 00 FFFFFFFF] c_bar
  DBD::Unify::st_prepare u_sql_00_000000 (<= 1, => 0)

Level 6 plus destroy/cleanup states:

  DBD::Unify::st_free u_sql_00_000000
   destroy allocc destroy alloco    After deallocO, sqlcode = 0
   destroy alloci destroy allocp    After deallocU, sqlcode = 0
   destroy stat destroy growup destroy impset

No messages (yet) set to level 8 and up.

int dbd_bind_ph (SV *sth, imp_sth_t *imp_sth, SV *param, SV *value, IV sql_type, SV *attribs, int is_inout, IV maxlen)
SV *dbd_db_FETCH_attrib (SV *dbh, imp_dbh_t *imp_dbh, SV *keysv)
int dbd_db_STORE_attrib (SV *dbh, imp_dbh_t *imp_dbh, SV *keysv, SV *valuesv)
int dbd_db_commit (SV *dbh, imp_dbh_t *imp_dbh)
void dbd_db_destroy (SV *dbh, imp_dbh_t *imp_dbh)
int dbd_db_disconnect (SV *dbh, imp_dbh_t *imp_dbh)
int dbd_db_do (SV *dbh, char *statement)
int dbd_db_login (SV *dbh, imp_dbh_t *imp_dbh, char *dbname, char *user, char *pwd)
int dbd_db_rollback (SV *dbh, imp_dbh_t *imp_dbh)
int dbd_discon_all (SV *drh, imp_drh_t *imp_drh)
int dbd_fld_describe (SV *dbh, imp_sth_t *imp_sth, int num_fields)
void dbd_init (dbistate_t *dbistate)
int dbd_prm_describe (SV *dbh, imp_sth_t *imp_sth, int num_params)
SV *dbd_st_FETCH_attrib (SV *sth, imp_sth_t *imp_sth, SV *keysv)
int dbd_st_STORE_attrib (SV *sth, imp_sth_t *imp_sth, SV *keysv, SV *valuesv)
int dbd_st_blob_read (SV *sth, imp_sth_t *imp_sth, int field, long offset, long len, SV *destrv, long destoffset)
void dbd_st_destroy (SV *sth, imp_sth_t *imp_sth)
int dbd_st_execute (SV *sth, imp_sth_t *imp_sth)
AV *dbd_st_fetch (SV *sth, imp_sth_t *imp_sth)
int dbd_st_finish (SV *sth, imp_sth_t *imp_sth)
int dbd_st_prepare (SV *sth, imp_sth_t *imp_sth, char *statement, SV *attribs)
int dbd_st_rows (SV *sth, imp_sth_t *imp_sth)


As this module is probably far from complete, so will the TODO list most likely will be far from complete. More generic (test) items are mentioned in the README in the module distribution.

Handle attributes

Check if all documented handle (database- and statement-) attributes are supported and work as expected.

  local $dbh->{RaiseError}       = 0;
  local $sth->{FetchHashKeyName} = "NAME";
Statement attributes

Allow setting and getting statement attributes. A specific example might be

  $sth->{PrintError}       = 0;
  $sth->{FetchHashKeyName} = "NAME_uc";
3-argument bind_param ()

Investigate and implement 3-argument versions of $sth->bind_param ()

looks_as_number ()

Investigate if looks_as_number () should be used in st_bind (). Comments are in where it should.

Multiple open databases

Try finding a way to open several different Unify databases at the same time for parallel (or at least sequential) processing.


The DBI documentation in DBI, a lot of web pages, some very good, the Perl 5 DBI Home page (http://dbi.perl.org/), other DBD modules' documentation (DBD-Oracle is probably the most complete), the comp.lang.perl.modules newsgroup and the dbi-users mailing list (mailto:dbi-users-help@perl.org)


DBI/DBD was developed by Tim Bunce, who also developed the DBD::Oracle.

H.Merijn Brand developed the DBD::Unify extension.

Todd Zervas has given a lot of feedback and patches.


Copyright (C) 1999-2017 H.Merijn Brand

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