python-sybase-misc

[python-sybase] ct_send error

[Grant M. <gmc...@in...>]

Hi,

I received the following error when trying to do an update:

ct_send(): user api layer: external error: This routine cannot be called
because another command structure has results pending.

This update occurs after a select using a Cursor object which I haven't yet
closed. If I close the select cursor object prior to trying to do the update
I no longer get this problem. I found reference to this problem in a Sybase
manual which mentioned that ct_results needed to return a value before the
results set was considered processed. Do I need to call .return_status on
the select cursor before I can use another cursor to do an update? Or do I
have to close the select cursor before doing any updates?

Regards,

Grant M.

P.S. both the select and the update were to the same table.

[python-sybase] Windows compile

[Glen J. <gl...@be...>]

I have been trying to get python-sybase to compile for Windows (XP),
but have not had any luck finding the right compiler. I have looked
for it at Microsoft, E-Bay, and several computer stores that recycle
old versions of software.

I tried using the Borland compiler. Converting the sybase libraries
seemed like an okay thing to do, but when the build script began
complaining about the installed Python libraries, I thought it was
time to quit. Converting those libraries to another format did not
seem like an okay thing to do.

Can anyone tell me how to deal with this? If not, I have to go back
to embedding isql calls inside python scripts.....

Thanks in advance for any suggestions.

Glen Jackson
============
Museum Informatics Project

Re: [python-sybase] space symbol at the end of string and binary params

[Dave C. <dj...@ob...>]

Andrey Plotnikov wrote:
> Hi All!
> 
> I use sybase-python and freetds with MSSQL.
> And I have the following problem: the space symbol is added to the 
> string and binary parameters.
> For example:
>  c.execute("insert into Tmp (f1) values (@s)",{"@s":"test"})
> As a result, field f1 value is "test "
> For string params its not vastly actual, but binary data can be passed 
> only through params and space symbol at the end is serious problem.
> 
> FreeDTS version is 0.62.3
> I tried sybase-python 0.36 and 0.37.
> 
> P.S. Sorry for my english...

It is probably worth raising this on one of the FreeTDS mailing lists.

- Dave

-- 
http://www.object-craft.com.au

[python-sybase] space symbol at the end of string and binary params

[Andrey P. <ap...@sb...>]

Hi All!

I use sybase-python and freetds with MSSQL.
And I have the following problem: the space symbol is added to the 
string and binary parameters.
For example:
  c.execute("insert into Tmp (f1) values (@s)",{"@s":"test"})
As a result, field f1 value is "test "
For string params its not vastly actual, but binary data can be passed 
only through params and space symbol at the end is serious problem.

FreeDTS version is 0.62.3
I tried sybase-python 0.36 and 0.37.

P.S. Sorry for my english...

-- 
Andrey Plotnikov

Re: [python-sybase] Executing SQL asynchronously

[Skip M. <sk...@po...>]

    >> I need to perform inserts and updates to a Sybase database
    >> asynchronously.  Is there already a mechanism in sybase-python to do
    >> this?  I see the Sybase module uses threads extensively.

    Dave> I suppose the best way to do this would be to perform the SQL
    Dave> operations in one or more separate Python threads.

That's pretty much what I came up with.  My initial tests at the command
prompt led me to believe that maybe I'd done something wrong, but I think I
was fine.  An AsyncConnection class has its own hidden cursor and a Queue.
The client gets a fake cursor that does nothing more than stuff the (sql,
params) tuple into the queue where it's picked up and executed by the real
cursor running in a daemon thread.

Skip

Re: [python-sybase] Executing SQL asynchronously

[Dave C. <dj...@ob...>]

Skip Montanaro wrote:
> I need to perform inserts and updates to a Sybase database asynchronously.
> Is there already a mechanism in sybase-python to do this?  I see the Sybase
> module uses threads extensively.

I suppose the best way to do this would be to perform the SQL operations 
in one or more separate Python threads.

- Dave

-- 
http://www.object-craft.com.au

[python-sybase] Executing SQL asynchronously

[Skip M. <sk...@po...>]

I need to perform inserts and updates to a Sybase database asynchronously.
Is there already a mechanism in sybase-python to do this?  I see the Sybase
module uses threads extensively.

Thx,

-- 
Skip Montanaro
sk...@po...

[python-sybase] python-sybase module trouble

[Alexey M. <sa...@ri...>]

Hello,

I'v try to install python-sybase module (verison 0.37). It completelly 
compiled and installed.
but i can't use it because of this:

morsov@router ~ $ python
Python 2.3.3 (#1, Feb  7 2005, 18:57:53)
[GCC 3.3.3 20040412 (ALT Linux, build 3.3.3-alt5)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
 >>> import Sybase
 >>> db = Sybase.connect("***","***","***")
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
   File "/usr/lib/python2.3/site-packages/Sybase.py", line 1005, in connect
     strip, auto_commit, delay_connect, locking)
   File "/usr/lib/python2.3/site-packages/Sybase.py", line 867, in __init__
     self.connect()
   File "/usr/lib/python2.3/site-packages/Sybase.py", line 886, in connect
     status = conn.ct_connect(self.dsn)
   File "/usr/lib/python2.3/site-packages/Sybase.py", line 200, in 
_servermsg_cb
     raise DatabaseError(_fmt_server(msg))
Sybase.DatabaseError: Msg 21, Level 10
TDS: unexpected token 95
 >>>

i'v compile it with Freetds.
environment vars

morsov@router ~ $ echo $SYBASE $LD_LIBRARY_PATH
/usr /usr/lib

with it i'v compiel this module.

tsql and sqsh connect to this database.

Can i fix this somehow ?

-- 
WBR,
Alexey Morsov

[python-sybase] Anyone have Python-Sybase compiled binaries for Windows 2000/XP?

[Bradley F. <br...@re...>]

I'm having a bear of a time getting the environment right for compiling =
the
Sybase module on Windows, following the instructions.  I've spent a =
great
deal of time already and still can't make python and bcc happy.

If someone has a set of binaries they could email to me, I'd greatly
appreciate it.  Using it locally from windows saves me the trouble of =
always
remote-debugging through Komodo.

Thanks

Bradley

RE: [python-sybase] Errors building on Debian?

[Jan J. <j2...@mu...>]

>On Linux you will be more sucessful it you install the Sybase module and 
>using it to talk to SQL Server.

Any links on how to accomplish that?

Re: [python-sybase] Errors building on Debian?

[Dave C. <dj...@ob...>]

Jan Johansson <j2...@mu...>Jan Johansson wrote:
> I am trying to install http://www.object-craft.com.au/projects/mssql/ on my
> Debian system, but when i try to build, i end up with the below. 
> 
> hooch:~/Download/Asterisk/MSSQL-0.09# python setup.py install

[snip lots of errors]

FreeTDS is a bit of a moving target, and the MSSQL module is not 
maintained as actively as the Sybase module.  MSSQL works best when 
compiled using the db library from MS on Windows :-).

On Linux you will be more sucessful it you install the Sybase module and 
using it to talk to SQL Server.

- Dave

-- 
http://www.object-craft.com.au

Re: [python-sybase] Large text input

[Dave C. <dj...@ob...>]

thi...@ya... wrote:
> Any help on trying to update a column in a table that
> is large ie 18k +. I tried to use the cursor and did
> an execute "update table set column=@data" but it
> raises an error saying something about the data is to
> large for a data field even though the field in the
> database is text and large enough to hold it. 
> 
> I am a newbie using this library and using sybase in
> general so hopefully this is easy one to answer.
> Thanks

 From memory there is a way to do this using the CT library, but it does 
not fit the way that the DB-API is supposed to work.

If there is enough interest I could go back and read the documentation, 
implement the missing CT functions, and then try to wrap the 
functionality in some higher level Python.

- Dave

-- 
http://www.object-craft.com.au

Re: [python-sybase] Building python-sybase with ubuntu hoary

[Marcos P. <ma...@bu...>]

Disregard this message. The whole linux setup was falling apart. I blame
disk corruption.

I have reinstalled and it works fine.

El mar, 05-04-2005 a las 17:31 +0200, Marcos Sánchez Provencio escribió:
> The module seems to build fine, but I am not able to import it. Has
> anybody been successful?
> 
> marcos@quina:~ $ python
> Python 2.4.1 (#2, Mar 30 2005, 21:51:10)
> [GCC 3.3.5 (Debian 1:3.3.5-8ubuntu2)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> import Sybase
> Traceback (most recent call last):
>   File "<stdin>", line 1, in ?
>   File "/usr/lib/python2.4/site-packages/Sybase.py", line 20, in ?
>     from sybasect import *
> ImportError: /lib/libc.so.6: undefined symbol: register_match6
> 
> -----====-----
> And this is the ldd output:
> 
> marcos@quina:~ $  ldd /usr/local/lib/python2.4/site-packages/sybasect.so
> ldd: /usr/local/lib/python2.4/site-packages/sybasect.so: No existe el
> fichero o el directorio
> marcos@quina:~ $  ldd /usr/lib/python2.4/site-packages/sybasect.so
>                 libct.so.1 => /usr/lib/libct.so.1 (0xb7faa000)
>         libpthread.so.0 => /lib/tls/i686/cmov/libpthread.so.0
> (0xb7f9a000)
>         /lib/libc.so.6 => /lib/libc.so.6 (0xb7f97000)
>         libnsl.so.1 => /lib/tls/i686/cmov/libnsl.so.1 (0xb7f83000)
>         libc.so.6 => /lib/tls/i686/cmov/libc.so.6 (0xb7e56000)
>         /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)
> -----====-----
> SYBASE is /usr
> freetds-dev deb package version is 0.61-6.1
> python2.4 deb package version is 2.4.1-0
> 
> 
> 
> _______________________________________________
> Python-sybase mailing list
> Pyt...@ww...
> https://www.object-craft.com.au/cgi-bin/mailman/listinfo/python-sybase

Re: [python-sybase] Large text input

[Skip M. <sk...@po...>]

    Jeff> Oh boy.  This one is *not* an easy one to solve.  The sybase
    Jeff> protocol doesn't handle blocks larger than 16k. (I hope I'm
    Jeff> actually wrong about this.)  In order to send larger blocks you
    Jeff> have to use the lower level libraries.  

Can you do something like this?

    chunk = data[:1000]
    del data[:1000]
    c.execute("update table set column=@chunk where ...",
              {"@chunk": chunk})
    while data:
        chunk = data[:1000]
        del data[:1000]
        c.execute("update table set column=concat(column,@chunk) where ...",
                  {"@chunk": chunk})

Kind clunky, but might do the trick in a pinch.  (I assume concat() is a
standard SQL function.  YMMV.)

Skip

Re: [python-sybase] Large text input

[Jeff Y. <jyo...@of...>]

Oh boy.  This one is *not* an easy one to solve.   The sybase protocol
doesn't handle blocks larger than 16k. (I hope I'm actually wrong about
this.)  In order to send larger blocks you have to use the lower level
libraries.  You have to look up the row that you want to insert the data
into, get a handle to the field you want, and then repeatedly send
sub-16k chunks of data.

It's not pretty.   I'd be glad to contribute the code to do it, but
unfortunately it is my employer's property.  (*ugh*)  I'd ask my boss
to release it, but he's on vacation at the moment.

All I can say is, sob now.   Sob very hard.

-jeff


On Apr 11, 2005, at 4:16 PM, thi...@ya... wrote:

> Any help on trying to update a column in a table that
> is large ie 18k +. I tried to use the cursor and did
> an execute "update table set column=@data" but it
> raises an error saying something about the data is to
> large for a data field even though the field in the
> database is text and large enough to hold it.
>
> I am a newbie using this library and using sybase in
> general so hopefully this is easy one to answer.
> Thanks

[python-sybase] Large text input

[<thi...@ya...>]

Any help on trying to update a column in a table that
is large ie 18k +. I tried to use the cursor and did
an execute "update table set column=@data" but it
raises an error saying something about the data is to
large for a data field even though the field in the
database is text and large enough to hold it. 

I am a newbie using this library and using sybase in
general so hopefully this is easy one to answer.
Thanks


		
__________________________________ 
Do you Yahoo!? 
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/

[python-sybase] Errors building on Debian?

[Jan J. <j2...@mu...>]

I am trying to install http://www.object-craft.com.au/projects/mssql/ on my
Debian system, but when i try to build, i end up with the below. 

hooch:~/Download/Asterisk/MSSQL-0.09# python setup.py install
running install
running build
running build_py
not copying MSSQL.py (output up-to-date)
running build_ext
building 'mssqldb' extension
creating build/temp.linux-i686-2.2
gcc -DNDEBUG -g -O3 -fno-strict-aliasing -Wall -Wstrict-prototypes -fPIC
-DHAVE_FREETDS -DHAVE_DBALTBIND -DHAVE_DBCLROPT -DHAVE_DBCURCMD
-DHAVE_DBCURROW -DHAVE_DBISOPT -DHAVE_DBNUMCOMPUTE -DHAVE_DBRETTYPE
-I/usr/include -I/usr/include/python2.2 -c mssqldb.c -o
build/temp.linux-i686-2.2/mssqldb.o
gcc -DNDEBUG -g -O3 -fno-strict-aliasing -Wall -Wstrict-prototypes -fPIC
-DHAVE_FREETDS -DHAVE_DBALTBIND -DHAVE_DBCLROPT -DHAVE_DBCURCMD
-DHAVE_DBCURROW -DHAVE_DBISOPT -DHAVE_DBNUMCOMPUTE -DHAVE_DBRETTYPE
-I/usr/include -I/usr/include/python2.2 -c loginrec.c -o
build/temp.linux-i686-2.2/loginrec.o
gcc -DNDEBUG -g -O3 -fno-strict-aliasing -Wall -Wstrict-prototypes -fPIC
-DHAVE_FREETDS -DHAVE_DBALTBIND -DHAVE_DBCLROPT -DHAVE_DBCURCMD
-DHAVE_DBCURROW -DHAVE_DBISOPT -DHAVE_DBNUMCOMPUTE -DHAVE_DBRETTYPE
-I/usr/include -I/usr/include/python2.2 -c dbproc.c -o
build/temp.linux-i686-2.2/dbproc.o
dbproc.c:199: warning: `wrap_DbProcIntFunc_bool' defined but not used
gcc -DNDEBUG -g -O3 -fno-strict-aliasing -Wall -Wstrict-prototypes -fPIC
-DHAVE_FREETDS -DHAVE_DBALTBIND -DHAVE_DBCLROPT -DHAVE_DBCURCMD
-DHAVE_DBCURROW -DHAVE_DBISOPT -DHAVE_DBNUMCOMPUTE -DHAVE_DBRETTYPE
-I/usr/include -I/usr/include/python2.2 -c databuf.c -o
build/temp.linux-i686-2.2/databuf.o
databuf.c: In function `get_buff_value':
databuf.c:86: error: `DBVARYCHAR' undeclared (first use in this function)
databuf.c:86: error: (Each undeclared identifier is reported only once
databuf.c:86: error: for each function it appears in.)
databuf.c:86: error: parse error before ')' token
databuf.c:87: error: parse error before ')' token
databuf.c:90: error: `DBBIT' undeclared (first use in this function)
databuf.c:90: error: parse error before ')' token
databuf.c: In function `DataBuf_set':
databuf.c:155: error: `DBVARYCHAR' undeclared (first use in this function)
databuf.c:155: error: parse error before ')' token
databuf.c:160: error: parse error before ')' token
databuf.c:161: error: parse error before ')' token
databuf.c:169: error: `DBBIT' undeclared (first use in this function)
databuf.c:169: error: parse error before ')' token
databuf.c: In function `DataBuf__init__':
databuf.c:270: error: `DBVARYCHAR' undeclared (first use in this function)
databuf.c:273: error: `DBBIT' undeclared (first use in this function)
error: command 'gcc' failed with exit status 1
hooch:~/Download/Asterisk/MSSQL-0.09#

_______________________________________________
Tutor maillist  -  Tu...@py...
http://mail.python.org/mailman/listinfo/tutor

[python-sybase] Sybase module 0.37 released

[Dave C. <dj...@ob...>]

WHAT IS IT:

The Sybase module provides a Python interface to the Sybase relational
database system.  It supports all of the Python Database API, version
2.0 with extensions.

NOTES:

The 0.37 release is identical to 0.37pre3 as no problems were reported
with the prerelease.

This release contains a number of small bugfixes and patches received
from users.

I have been unable to find the source of the memory leak reported here:
 
http://www.object-craft.com.au/pipermail/python-sybase/2004-December/000346.html

The test program I wrote follows:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
import sys
import Sybase

db = Sybase.connect(..., auto_commit=True)
db.execute('''
if exists (select name from sysobjects where name = "sp_test_leak")
begin
     drop procedure sp_test_leak
end
''')
db.execute('''
create procedure sp_test_leak
     @arg int
as
     select @arg
''')
for i in range(200):
     for j in range(1000):
         c = db.cursor()
         c.callproc('sp_test_leak', {'@arg': 12345 })
     sys.stdout.write('%3d\r' % i)
     sys.stdout.flush()
sys.stdout.write('\n')
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

If someone is able to modify this and come up with a leaking result I
am interested in working on the fix.

You can build for FreeTDS like this:

      python setup.py build_ext -D HAVE_FREETDS -U WANT_BULKCOPY
      python setup.py install


The module is available here:

 
http://www.object-craft.com.au/projects/sybase/download/sybase-0.37pre3.tar.gz

The module home page is here:

     http://www.object-craft.com.au/projects/sybase/

CHANGES SINCE 0.36:

* Fix FreeTDS compilation and rearrange header includes to remove warnings.

* Cursor state initialisation fix from Skip Montanaro.

* Callback declaration fix on Windows from Vadim Beloborodov.

* Cursor output parameters now work when parameters are passed as a
   sequence.

* Output parameters now work for FreeTDS 0.62.4.

     1> create procedure sp_test_output
     2>     @num int, @result int output
     3> as
     4>     select @result = @num
     5> go

     params = c.callproc('sp_test_output', {'@num': 12345,
                                            '@result': Sybase.OUTPUT(1)})
     print params['@result']

* The CS_STATUS_RESULT result set is now consumed internally in the
   Cursor and does not appear in the result sets consumed by the fetch
   and nextset methods.

   The return value from the stored procedure is available in the
   .return_status member of the Cursor.  It will only contain a
   meaningful value once all of the row result sets have been consumed.

   Note that this does not work with FreeTDS 0.62.4.  The return_status
   seems to always be 0.  Research shows that the problem is probably
   in the CT emulation layer as tsql displays the correct value, but
   sqsh displays 0.

* Output hook patch from Ty Sarna has been applied.

* Applied patch from Andre Sedinin to improve error handling.

* Improved detection of SYBASE_OCS.

-- 
http://www.object-craft.com.au

[python-sybase] Building python-sybase with ubuntu hoary

[Marcos P. <ma...@bu...>]

The module seems to build fine, but I am not able to import it. Has
anybody been successful?

marcos@quina:~ $ python
Python 2.4.1 (#2, Mar 30 2005, 21:51:10)
[GCC 3.3.5 (Debian 1:3.3.5-8ubuntu2)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import Sybase
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/lib/python2.4/site-packages/Sybase.py", line 20, in ?
    from sybasect import *
ImportError: /lib/libc.so.6: undefined symbol: register_match6

-----====-----
And this is the ldd output:

marcos@quina:~ $  ldd /usr/local/lib/python2.4/site-packages/sybasect.so
ldd: /usr/local/lib/python2.4/site-packages/sybasect.so: No existe el
fichero o el directorio
marcos@quina:~ $  ldd /usr/lib/python2.4/site-packages/sybasect.so
                libct.so.1 => /usr/lib/libct.so.1 (0xb7faa000)
        libpthread.so.0 => /lib/tls/i686/cmov/libpthread.so.0
(0xb7f9a000)
        /lib/libc.so.6 => /lib/libc.so.6 (0xb7f97000)
        libnsl.so.1 => /lib/tls/i686/cmov/libnsl.so.1 (0xb7f83000)
        libc.so.6 => /lib/tls/i686/cmov/libc.so.6 (0xb7e56000)
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000)
-----====-----
SYBASE is /usr
freetds-dev deb package version is 0.61-6.1
python2.4 deb package version is 2.4.1-0

[python-sybase] Re: [DB-SIG] Sybase module 0.37pre2 released

[Dave C. <dj...@ob...>]

Dave Cole wrote:
> The module is available here:
> 
> 
> http://www.object-craft.com.au/projects/sybase/download/sybase-0.37pre1.tar.gz 

Ooops.  Make that:

http://www.object-craft.com.au/projects/sybase/download/sybase-0.37pre2.tar.gz

- Dave

-- 
http://www.object-craft.com.au

[python-sybase] Sybase module 0.37pre2 released

[Dave C. <dj...@ob...>]

WHAT IS IT:

The Sybase module provides a Python interface to the Sybase relational
database system.  It supports all of the Python Database API, version
2.0 with extensions.

NOTES:

This release contains a number of small bugfixes and patches received
from users.

I have been unable to find the source of the memory leak reported here:
 
http://www.object-craft.com.au/pipermail/python-sybase/2004-December/000346.html

The test program I wrote follows:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
import sys
import Sybase

db = Sybase.connect(..., auto_commit=True)
db.execute('''
if exists (select name from sysobjects where name = "sp_test_leak")
begin
     drop procedure sp_test_leak
end
''')
db.execute('''
create procedure sp_test_leak
     @arg int
as
     select @arg
''')
for i in range(200):
     for j in range(1000):
         c = db.cursor()
         c.callproc('sp_test_leak', {'@arg': 12345 })
     sys.stdout.write('%3d\r' % i)
     sys.stdout.flush()
sys.stdout.write('\n')
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

If someone is able to modify this and come up with a leaking result I
am interested in working on the fix.

You can build for FreeTDS like this:

      python setup.py build_ext -D HAVE_FREETDS -U WANT_BULKCOPY
      python setup.py install


The module is available here:

 
http://www.object-craft.com.au/projects/sybase/download/sybase-0.37pre1.tar.gz

The module home page is here:

     http://www.object-craft.com.au/projects/sybase/

CHANGES SINCE 0.36:

* Cursor state initialisation fix from Skip Montanaro.

* Callback declaration fix on Windows from Vadim Beloborodov.

* Cursor output parameters now work when parameters are passed as a
   sequence.

* Output parameters now work for FreeTDS 0.62.4.

     1> create procedure sp_test_output
     2>     @num int, @result int output
     3> as
     4>     select @result = @num
     5> go

     params = c.callproc('sp_test_output', {'@num': 12345,
                                            '@result': Sybase.OUTPUT(1)})
     print params['@result']

* The CS_STATUS_RESULT result set is now consumed internally in the
   Cursor and does not appear in the result sets consumed by the fetch
   and nextset methods.

   The return value from the stored procedure is available in the
   .return_status member of the Cursor.  It will only contain a
   meaningful value once all of the row result sets have been consumed.

   Note that this does not work with FreeTDS 0.62.4.  The return_status
   seems to always be 0.  Research shows that the problem is probably
   in the CT emulation layer as tsql displays the correct value, but
   sqsh displays 0.

* Output hook patch from Ty Sarna has been applied.

* Applied patch from Andre Sedinin to improve error handling.

* Improved detection of SYBASE_OCS.

- Dave

-- 
http://www.object-craft.com.au

Re: [python-sybase] Sybase module 0.37pre1 released

[Dave C. <dj...@ob...>]

Skip Montanaro wrote:
>     Dave> This release contains a number of small bugfixes and patches
>     Dave> received from users.
> 
>     ...
> 
> I don't see this change:
> 
> *** /Users/skip/src/sybase-0.36/Sybase.py       Sun Apr 27 05:54:35 2003
> --- /Users/skip/tmp/Sybase.py   Sat Mar 19 16:46:17 2005
> ***************
> *** 436,441 ****
> --- 471,477 ----
>   
>       def start(self, arraysize):
>           self._arraysize = arraysize
> +         self._set_state(_LAZY_FETCHING)
>           status = self._cmd.ct_send()
>           if status != CS_SUCCEED:
>               self._raise_error(Error, 'ct_send')
> 
> You created that in response to some problems one of my users was having.

Ooops.  I will put together a pre2 release with that fix and the 
__stdcall patch from Vadim.

I think I should change my strategy for releasing.  Previously I had 
been posting small patches then waiting for feedback.  That seems not to 
work.  A better technique might be to release as soon as a patch is 
received and wait for feedback from that.

> I also don't see anything related to Python's builtin datetime module.  I
> can whip up a patch based upon my local changes if you like.  mx.DateTime is
> nice, but overkill for most things and represents yet another thing that
> needs to be installed.
> 
> Finally, way back when I first started using the Sybase module I added a
> converters arg to the Connection class.  I realize that's not necessarily
> everybody's idea of the best way to implement type converters, but it allows
> me to pretty transparently use whatever date/time module I have available.
> (Some users at work have old code that expects times to be floats in epoch
> seconds as returned by time.time().  This allows them to easily use that
> with minimal changes to their code.)

I like the idea of doing the conversion via a configuration set of 
converters.

Rather than have the set of converters replace the standard set, maybe 
the converters passed to the connection should be merged with the 
standard set for that connection.

- Dave

-- 
http://www.object-craft.com.au

Re: [python-sybase] Sybase module 0.37pre1 released

[Skip M. <sk...@po...>]

    Dave> This release contains a number of small bugfixes and patches
    Dave> received from users.

    ...

I don't see this change:

*** /Users/skip/src/sybase-0.36/Sybase.py       Sun Apr 27 05:54:35 2003
--- /Users/skip/tmp/Sybase.py   Sat Mar 19 16:46:17 2005
***************
*** 436,441 ****
--- 471,477 ----
  
      def start(self, arraysize):
          self._arraysize = arraysize
+         self._set_state(_LAZY_FETCHING)
          status = self._cmd.ct_send()
          if status != CS_SUCCEED:
              self._raise_error(Error, 'ct_send')

You created that in response to some problems one of my users was having.

I also don't see anything related to Python's builtin datetime module.  I
can whip up a patch based upon my local changes if you like.  mx.DateTime is
nice, but overkill for most things and represents yet another thing that
needs to be installed.

Finally, way back when I first started using the Sybase module I added a
converters arg to the Connection class.  I realize that's not necessarily
everybody's idea of the best way to implement type converters, but it allows
me to pretty transparently use whatever date/time module I have available.
(Some users at work have old code that expects times to be floats in epoch
seconds as returned by time.time().  This allows them to easily use that
with minimal changes to their code.)

-- 
Skip Montanaro
sk...@po...

[python-sybase] Sybase module 0.37pre1 released

[Dave C. <dj...@ob...>]

WHAT IS IT:

The Sybase module provides a Python interface to the Sybase relational
database system.  It supports all of the Python Database API, version
2.0 with extensions.

NOTES:

This release contains a number of small bugfixes and patches received
from users.

I have been unable to find the source of the memory leak reported here:
 
http://www.object-craft.com.au/pipermail/python-sybase/2004-December/000346.html

The test program I wrote follows:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
import sys
import Sybase

db = Sybase.connect(..., auto_commit=True)
db.execute('''
if exists (select name from sysobjects where name = "sp_test_leak")
begin
     drop procedure sp_test_leak
end
''')
db.execute('''
create procedure sp_test_leak
     @arg int
as
     select @arg
''')
for i in range(200):
     for j in range(1000):
         c = db.cursor()
         c.callproc('sp_test_leak', {'@arg': 12345 })
     sys.stdout.write('%3d\r' % i)
     sys.stdout.flush()
sys.stdout.write('\n')
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

If someone is able to modify this and come up with a leaking result I
am interested in working on the fix.

You can build for FreeTDS like this:

      python setup.py build_ext -D HAVE_FREETDS -U WANT_BULKCOPY
      python setup.py install


The module is available here:

 
http://www.object-craft.com.au/projects/sybase/download/sybase-0.37pre1.tar.gz

The module home page is here:

     http://www.object-craft.com.au/projects/sybase/

CHANGES SINCE 0.36:

* Cursor output parameters now work when parameters are passed as a
   sequence.

* Output parameters now work for FreeTDS 0.62.4.

     1> create procedure sp_test_output
     2>     @num int, @result int output
     3> as
     4>     select @result = @num
     5> go

     params = c.callproc('sp_test_output', {'@num': 12345,
                                            '@result': Sybase.OUTPUT(1)})
     print params['@result']

* The CS_STATUS_RESULT result set is now consumed internally in the
   Cursor and does not appear in the result sets consumed by the fetch
   and nextset methods.

   The return value from the stored procedure is available in the
   .return_status member of the Cursor.  It will only contain a
   meaningful value once all of the row result sets have been consumed.

   Note that this does not work with FreeTDS 0.62.4.  The return_status
   seems to always be 0.  Research shows that the problem is probably
   in the CT emulation layer as tsql displays the correct value, but
   sqsh displays 0.

* Output hook patch from Ty Sarna has been applied.

* Applied patch from Andre Sedinin to improve error handling.

* Improved detection of SYBASE_OCS.

- Dave

-- 
http://www.object-craft.com.au

[python-sybase] Python-sybase on MacOSX

[Glen J. <gl...@be...>]

Has anybody succeeded in compiling the python-sybase package for Mac 
OSX?  I am
unable to compile ANY Sybase module on my G4 (OS 10.3.8,  ASE 12.5.1, 
GCC 3.3).
I have tried python-sybase, sybperl, DBI-Sybase. All fail for the same 
reason, the linker
can't find the library routines.

If there is a success story, I would appreciate getting some details...

Glen Jackson
=========
gl...@be...