FreeTDS Development Group

[Frediano Ziglio <freddy77 AT gmail.com>]

Hi,
   time for a new release!

Although the stable branch is kept updated with bugs and minor features, new 
big
features, build improvements and potential compatibility changes are not 
ported
to the stable branch.

The reasons for a new release are UTF-8 support for MSSQL 2019 and the removal
of the obsolete (and wrong) "8.0" version which won't be recognized anymore.

Here (available also in NEWS file) changes for next release:

Summary of Changes in release 1.3
--------------------------------------------
User visible (not in a particular order):
- Generic:
  - Support UTF-8 columns using MSSQL 2019;
  - Do not accept TDS protocol versions "4.6" (never really supported) and
    "8.0";
  - Minor portability issues;
  - Fix log elision for login;
  - Detect some possible minor memory failure in application;
  - Support long (more than 64k) SSPI packets (never encountered but you
    never know);
  - Fix unicode columns for ASA database;
  - Avoid using BCP with old protocols;
  - (*) Fix bulk copy using big endian machines;
  - (*) Fix Sybase uni(var)char and unsigned types for big endian machines;
  - (*) Do not send nullable data during bulk copy if type is not nullable;
- ODBC:
  - Added "Timeout" setting;
- Applications:
  - Improve defncopy utility:
    - Fix some declaration;
    - Fix Sybase support;
  - (*) Fix datacopy and freebcp logging;
- CT-Library:
  - Minor fix for variant type;
  - Better support for timeout setting;
  - (*) Support some missing types (like nullable unsigned integers) for
    Sybase;
- DB-library:
  - Unify date format (all systems can use the same syntax);
  - (*) Allows to pass 0 as type for bcp_bind;
  - (*) Fix DBSETLSERVERPRINCIPAL macro;
  - (*) Do not limit queries length for bcp using Sybase;
  - (*) Add KEEP_NULLS to BCP hints.

(*) Feature backported in stable 1.2 branch.

Implementation:
- Move replacement headers under freetds directory for coherence with other
  internal headers;
- Lots of style updates;
- Optimize UTF-8 encoding for ODBC, reuse common code.

Kind Regards,
  Frediano