Re: [NEW] databases/postgresql-odbc

2007-11-20 Thread Julian Leyh
On 17:49 Tue 23 Oct , Paul Irofti wrote:
 I've attached some configs that can be used in order to lighten the test
 process, hope this will make submitting this port faster.

Works for me, thanks.

Julian



Re: [NEW] databases/postgresql-odbc

2007-10-23 Thread Paul Irofti
On Tue, Oct 09, 2007 at 05:05:24PM +0300, Paul Irofti wrote:
 On Wed, Oct 03, 2007 at 06:26:48PM +0300, Paul Irofti wrote:
  This is the postgresql odbc driver for iodbc.  Works fine for amd64 and
  i386. 
  
  pkg/DESCR:
  The official PostgreSQL ODBC Driver
  
  
  
  I've attempted to make this a multipackage for unicode but failed 
  miserably. 
  
  If anyone is willing to help me make it into a multipackage please 
  send a patch or guide me through emails, here is what I've put extra
  inside the Makefile for the MULTI_PACKAGES:
  
  MULTI_PACKAGES=-main   -unicode
  COMMENT-main=  The official PostgreSQL ODBC Driver
  COMMENT-unicode=   The official PostgreSQL ODBC Driver with unicode
  support
  PKGNAME-main=  postgresql-odbc-${VERSION}
  PKGNAME-unicode=   postgresql-odbc-unicode-${VERSION}
  CONFIGURE_ARGS+=   --with-iodbc --without-unixodbc
  --enable-pthreads
  .if !defined(SUBPACKAGE)
  CONFIGURE_ARGS+=   --disable-unicode
  .endif
  
  I can build both packages separately, but when I do a `make package' 
  there is no second configure+compile stage and the -unicode package 
  errors out.
 
 It seems that there's no need for two packages, the unicode enabled
 driver can handle plain ascii as well. Here is the new port, tested on
 i386 and amd64 with both unicode and ascii.

I've attached some configs that can be used in order to lighten the test
process, hope this will make submitting this port faster.

There are two database dumps (one using simple ASCII and the other UTF8)
and two config files .odbc*.ini that should be placed in $HOME.

Afterwards running iodbctest and selecting DSN=test or DSN=testw will
drop you off to an sql cli where further testing may be done.
--
-- PostgreSQL database dump
--

SET client_encoding = 'UTF8';
SET standard_conforming_strings = off;
SET check_function_bodies = false;
SET client_min_messages = warning;
SET escape_string_warning = off;

--
-- Name: SCHEMA public; Type: COMMENT; Schema: -; Owner: postgres
--

COMMENT ON SCHEMA public IS 'Standard public schema';


--
-- Name: public; Type: ACL; Schema: -; Owner: postgres
--

REVOKE ALL ON SCHEMA public FROM PUBLIC;
REVOKE ALL ON SCHEMA public FROM postgres;
GRANT ALL ON SCHEMA public TO postgres;
GRANT ALL ON SCHEMA public TO PUBLIC;


--
-- PostgreSQL database dump complete
--

[ODBC Data Sources]
test  = postgresql
testw = postgresql

[test]
Driver  = /usr/local/lib/psqlodbcw.so
Description = Test Database
database= test

[ODBC]
Trace = 0
TraceAutoStop = 0
TraceFile = sql.log

[testw]
Driver  = /usr/local/lib/psqlodbcw.so
Description = Test Unicode Database
database= uni_test
[ODBC Drivers]
postgresql = Installed

[postgresql]
Driver = /usr/local/lib/psqlodbcw.so
Setup  = /usr/local/lib/psqlodbcw.so

[ODBC Connection Pooling]
PerfMon= 0
Retry Wait = 
--
-- PostgreSQL database dump
--

SET client_encoding = 'SQL_ASCII';
SET standard_conforming_strings = off;
SET check_function_bodies = false;
SET client_min_messages = warning;
SET escape_string_warning = off;

--
-- Name: SCHEMA public; Type: COMMENT; Schema: -; Owner: postgres
--

COMMENT ON SCHEMA public IS 'Standard public schema';


SET search_path = public, pg_catalog;

SET default_tablespace = '';

SET default_with_oids = false;

--
-- Name: users; Type: TABLE; Schema: public; Owner: bulibuta; Tablespace: 
--

CREATE TABLE users (
id integer NOT NULL,
address character(120),
name character(50)
);


ALTER TABLE public.users OWNER TO bulibuta;

--
-- Name: users_id_seq; Type: SEQUENCE; Schema: public; Owner: bulibuta
--

CREATE SEQUENCE users_id_seq
INCREMENT BY 1
NO MAXVALUE
NO MINVALUE
CACHE 1;


ALTER TABLE public.users_id_seq OWNER TO bulibuta;

--
-- Name: users_id_seq; Type: SEQUENCE OWNED BY; Schema: public; Owner: bulibuta
--

ALTER SEQUENCE users_id_seq OWNED BY users.id;


--
-- Name: users_id_seq; Type: SEQUENCE SET; Schema: public; Owner: bulibuta
--

SELECT pg_catalog.setval('users_id_seq', 37, true);


--
-- Name: id; Type: DEFAULT; Schema: public; Owner: bulibuta
--

ALTER TABLE users ALTER COLUMN id SET DEFAULT nextval('users_id_seq'::regclass);


--
-- Data for Name: users; Type: TABLE DATA; Schema: public; Owner: bulibuta
--


--
-- Name: unique id; Type: CONSTRAINT; Schema: public; Owner: bulibuta; 
Tablespace: 
--

ALTER TABLE ONLY users
ADD CONSTRAINT unique id PRIMARY KEY (id);


--
-- Name: public; Type: ACL; Schema: -; Owner: postgres
--

REVOKE ALL ON SCHEMA public FROM PUBLIC;
REVOKE ALL ON SCHEMA public FROM postgres;
GRANT ALL ON SCHEMA public TO postgres;
GRANT ALL ON SCHEMA public TO PUBLIC;


--
-- PostgreSQL database dump complete
--



Re: [NEW] databases/postgresql-odbc

2007-10-09 Thread Paul Irofti
On Wed, Oct 03, 2007 at 06:26:48PM +0300, Paul Irofti wrote:
 This is the postgresql odbc driver for iodbc.  Works fine for amd64 and
 i386. 
 
 pkg/DESCR:
 The official PostgreSQL ODBC Driver
 
 
 
 I've attempted to make this a multipackage for unicode but failed 
 miserably. 
 
 If anyone is willing to help me make it into a multipackage please 
 send a patch or guide me through emails, here is what I've put extra
 inside the Makefile for the MULTI_PACKAGES:
 
 MULTI_PACKAGES=-main   -unicode
 COMMENT-main=  The official PostgreSQL ODBC Driver
 COMMENT-unicode=   The official PostgreSQL ODBC Driver with unicode
 support
 PKGNAME-main=  postgresql-odbc-${VERSION}
 PKGNAME-unicode=   postgresql-odbc-unicode-${VERSION}
 CONFIGURE_ARGS+=   --with-iodbc --without-unixodbc
 --enable-pthreads
 .if !defined(SUBPACKAGE)
 CONFIGURE_ARGS+=   --disable-unicode
 .endif
 
 I can build both packages separately, but when I do a `make package' 
 there is no second configure+compile stage and the -unicode package 
 errors out.

It seems that there's no need for two packages, the unicode enabled
driver can handle plain ascii as well. Here is the new port, tested on
i386 and amd64 with both unicode and ascii.


postgresql-odbc.tgz
Description: application/tar-gz


[NEW] databases/postgresql-odbc

2007-10-03 Thread Paul Irofti
This is the postgresql odbc driver for iodbc.  Works fine for amd64 and
i386. 

pkg/DESCR:
The official PostgreSQL ODBC Driver



I've attempted to make this a multipackage for unicode but failed 
miserably. 

If anyone is willing to help me make it into a multipackage please 
send a patch or guide me through emails, here is what I've put extra
inside the Makefile for the MULTI_PACKAGES:

MULTI_PACKAGES=-main   -unicode
COMMENT-main=  The official PostgreSQL ODBC Driver
COMMENT-unicode=   The official PostgreSQL ODBC Driver with unicode
support
PKGNAME-main=  postgresql-odbc-${VERSION}
PKGNAME-unicode=   postgresql-odbc-unicode-${VERSION}
CONFIGURE_ARGS+=   --with-iodbc --without-unixodbc
--enable-pthreads
.if !defined(SUBPACKAGE)
CONFIGURE_ARGS+=   --disable-unicode
.endif

I can build both packages separately, but when I do a `make package' 
there is no second configure+compile stage and the -unicode package 
errors out.



Re: [NEW] databases/postgresql-odbc

2007-10-03 Thread Paul Irofti
And the actual port (-:


postgresql-odbc.tgz
Description: application/tar-gz


Re: [NEW] databases/postgresql-odbc

2007-10-03 Thread Marc Balmer

Paul Irofti wrote:

This is the postgresql odbc driver for iodbc.  Works fine for amd64 and
i386. 



is this dependent on a specific postgresl version?


pkg/DESCR:
The official PostgreSQL ODBC Driver



I've attempted to make this a multipackage for unicode but failed 
miserably. 

If anyone is willing to help me make it into a multipackage please 
send a patch or guide me through emails, here is what I've put extra

inside the Makefile for the MULTI_PACKAGES:

MULTI_PACKAGES=-main   -unicode
COMMENT-main=  The official PostgreSQL ODBC Driver
COMMENT-unicode=   The official PostgreSQL ODBC Driver with unicode
support
PKGNAME-main=  postgresql-odbc-${VERSION}
PKGNAME-unicode=   postgresql-odbc-unicode-${VERSION}
CONFIGURE_ARGS+=   --with-iodbc --without-unixodbc
--enable-pthreads
.if !defined(SUBPACKAGE)
CONFIGURE_ARGS+=   --disable-unicode
.endif

I can build both packages separately, but when I do a `make package' 
there is no second configure+compile stage and the -unicode package 
errors out.






Re: [NEW] databases/postgresql-odbc

2007-10-03 Thread Marc Balmer

Paul Irofti wrote:

On Wed, Oct 03, 2007 at 10:32:31PM +0200, Marc Balmer wrote:

Paul Irofti wrote:

This is the postgresql odbc driver for iodbc.  Works fine for amd64 and
i386. 

is this dependent on a specific postgresl version?



Yes, the 8.2 series. There are other versions for the postgresql
equivalents.


I will most probably soon change the Postgresql port to support more than
one version (8.1.x, 8.2.x., 8.3.x), one of these versions would then be of
course the default version.

we should dicuss and talk about how we can handle this is dependent ports.



Re: [NEW] databases/postgresql-odbc

2007-10-03 Thread Paul Irofti
On Wed, Oct 03, 2007 at 10:55:58PM +0200, Marc Balmer wrote:
 Paul Irofti wrote:
 On Wed, Oct 03, 2007 at 10:32:31PM +0200, Marc Balmer wrote:
 Paul Irofti wrote:
 This is the postgresql odbc driver for iodbc.  Works fine for amd64 and
 i386. 
 is this dependent on a specific postgresl version?

 Yes, the 8.2 series. There are other versions for the postgresql
 equivalents.

 I will most probably soon change the Postgresql port to support more than
 one version (8.1.x, 8.2.x., 8.3.x), one of these versions would then be of
 course the default version.

 we should dicuss and talk about how we can handle this is dependent ports.

There is an equivalent driver for each postgres version, so there should
be no problem. I'll watch the changes and mails sent over to ports@ to
see when you submit the patch and I'll try to adapt mine accordingly.