I've just uploaded DBD::ODBC 1.34_5 to the CPAN.
I know I often say this but my intention is to release this as 1.35 soon unless
any major issues are found. The total changes since the last full release are:
=head2 Changes in DBD::ODBC 1.34_5 February 17 2012
[BUG FIXES]
* The 40UnicodeRoundTrip tests counts could be 1 off in some cases.
* Fix for t/03batt.t which could fail a test if the data source had
no table - Kenichi Ishigaki
* If a driver misbehaves during global destruction e.g. SQLFreeStmt
fails but no error is available DBD::ODBC issues an error saying
an error occurred but no error diagnostics could be found. This is
pointless and irritating during global destruction. This stems
from a change in 1.28. Thanks to Peter Rabbitson for reporting
and suggested fix.
[CHANGE IN BEHAVIOUR]
* Prior to this release if you called selectall_* methods with a
non-select statement DBD::ODBC would raise an error saying "no
select statement currently executing". See RT 68720. After
discussions on dbi-dev the concensus seems to be that issuing a
warning in this case is better so that is what I've done. As a
result t/rt_68720.t has been removed and
t/85_selectall_non_select.t has been added.
[DOCUMENTATION]
* dbc_getdiagfield was incorrectly named odbc_getdiagrec in the pod
* add DBI version required for StrictlyTyped and DiscardString to
pod
* Added new FAQ on why a transaction may be committed when
AutoCommit is turned off.
[OTHER]
* Make examples\odbc_diag.pl more tolerant of drivers which do not
handle diagnostic calls properly.
* Make t/40UnicodeRoundTrip.t work with SQLite - Kenichi Ishigaki
* Make t/odbc_describe_parameter.t work with SQLite - Kenichi
Ishigaki
* Add 80_odbc_diags.t based on the same file in examples
=head2 Changes in DBD::ODBC 1.34_4 February 5 2012
[BUG FIXES]
* When odbc_getdiag* methods were added they installed themselves
into DBI but did not set IMP_KEEP_ERR so calling them cleared
DBI's errors.
=head2 Changes in DBD::ODBC 1.34_3 February 3 2012
[BUG FIXES]
* Linking against unixODBC was working by accident on most UNIX
machines and depended on the order of the files in /usr/lib (or
wherever) and what files there were (e.g. an archive or a shared
object). Same applied to iODBC but it was more broken especially
on machines where libiodbc.so.N.N existed but there was no
libiodbc.so which could lead to no adding the shared object at
all. I doubt anyone really noticed this but I did eventually on
Ubuntu where libiodbc.so.N.N existed but libiodbc.so did not.
[ENHANCEMENTS]
* Added experimental odbc_getdiagrec and odbc_getdiagrec methods,
examples/odbc_diag.pl and examples/params_in_error.pl.
[DOCUMENTATION]
* New FAQ entries.
=head2 Changes in DBD::ODBC 1.34_2 January 25 2012
[BUG FIXES]
* Fixed rt73734 - debian moved where unixODBC libs are stored.
* Fixed memory leak of the parameter status array introduced in
previous release when execute_for_fetch used. When the statement
handle is destroyed the parameter status array was not freed.
[ENHANCEMENTS]
* Added environment variable PERL_DBD_ODBC_PREFER_UNIXODBC as a
synonym for -x from Rafael Kitover (Caelum).
[DOCUMENTATION]
* Add a deviation from DBI spec for type_info_all.
[OTHER]
* Added example execute_for_fetch.pl
=head2 Changes in DBD::ODBC 1.34_1 December 11 2011
[ENHANCEMENTS]
* Added experimental execute_for_fetch support and associated
attributes odbc_batch_size and odbc_disable_array_operations.
Martin
--
Martin J. Evans
Easysoft Limited
http://www.easysoft.com
