ODBC Boolean handling

Aidan Mountford wrote:

I had a few issues with boolean handling in ODBC driver.

1) When a row is retrieved, and then a SQL_FETCH_FIRST is issued, the check
in convert.c
does not consider the fact that the value in the field has been altered to
be a '1' if the
backend handed it a 't'. The net result being that the first row on any
subsequent queries
has all it's boolean set to 0.

You seem to be right.
I would commit it as soon as I am able to have a cvs access(
I hope it would be before 7.2 release).

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

Hmm I don't understand this.
Please explain more in detail.

regards,
Hiroshi Inoue

I have reviewed the current CVS code and it looks quite different from
the patch you submitted. Any chance of you downloading the snapshot and
sending a context diff, rather than an ordinary one. I could apply that
easily.

I had a few issues with boolean handling in ODBC driver.

1) When a row is retrieved, and then a SQL_FETCH_FIRST is issued, the check
in convert.c
does not consider the fact that the value in the field has been altered to
be a '1' if the
backend handed it a 't'. The net result being that the first row on any
subsequent queries
has all it's boolean set to 0.

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

The following patch fixes both of these issues..

Cheers for building a fantastic product

Aidan

cvs diff convert.c (in directory C:\CVS-POSTGRES\pgsql\src\interfaces\odbc\)
Index: convert.c
===================================================================
RCS file: /home/projects/pgsql/cvsroot/pgsql/src/interfaces/odbc/convert.c,v
retrieving revision 1.44
diff -r1.44 convert.c
295c295,307
< 				if (s[0] == 'T' || s[0] == 't')
---

/* Aidan Mountford (aidan@oz.to) 1/08/2001:

if (s[0] == 'T' || s[0] == 't')

<<< This wont work...

When MoveFirst is called twice on

one set of tuples,

this will have the effect of setting

s[0] to 1 on the

first pass, and s[0] on the second.

This is bad ;)

*/

if (s[0] == 'T' || s[0] == 't' || s[0] ==

'1')
415c428
< len = 1;
---

418c431,449
< strcpy(rgbValueBindRow, value);
---

/* Aidan Mountford (aidan@oz.to)

1/08/2001:

When returning a CHAR

datatype, return -1

instead of +1.

That way if someone casts it

back to a boolean

again - it will work.
*/
if ( *(char *) value = '1')
{
strcpy(rgbValueBindRow, "-1");
len = 2;
}
else
{
strcpy(rgbValueBindRow, "0");
len = 1;
}

1160a1192,1193

1161a1195

.....................................................
. Cisco Certified Network Associate
. Microsoft Certified Professional
. 3Com 3Wizard
.
. Technical Director
. MindVision Interactive Pty. Ltd.
. http://www.mindvision.com.au
. PH: +61-8-8212-9544
. FAX: +61-8-8212-9644
. 48 Light Square
. Adelaide SA
. AUSTRALIA 5000
.....................................................

"Remember, There is always scope for things to be worse"

---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to majordomo@postgresql.org)

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

Here is the submitted patch so everyone can see it.

As Requested ...

-----Original Message-----
From: Bruce Momjian [mailto:pgman@candle.pha.pa.us]
Sent: Thursday, 2 August 2001 12:30 PM
To: Aidan Mountford
Cc: 'pgsql-patches@postgresql.org'
Subject: Re: [PATCHES] ODBC Boolean handling

I have reviewed the current CVS code and it looks quite different from
the patch you submitted. Any chance of you downloading the snapshot and
sending a context diff, rather than an ordinary one. I could apply that
easily.

I had a few issues with boolean handling in ODBC driver.

1) When a row is retrieved, and then a SQL_FETCH_FIRST is issued, the

check

in convert.c
does not consider the fact that the value in the field has been altered to
be a '1' if the
backend handed it a 't'. The net result being that the first row on any
subsequent queries
has all it's boolean set to 0.

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

The following patch fixes both of these issues..

Cheers for building a fantastic product

Aidan

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

You seem to be right.
I would commit it as soon as I am able to have a cvs access(
I hope it would be before 7.2 release).

Hiroshi, I will take care of applying the patch. Are you having CVS
troubles we can help with?

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

Bruce Momjian wrote:

Here is the submitted patch so everyone can see it.

I agree to the first change but am suspicious about the
second change.

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

Though Aidan is right, is -1 guaranteed to be true in
any application conversely ? It seems also different
from the ODBC spec. I don't have the solution now.

regards,
Hiroshi Inoue

Bruce Momjian wrote:

You seem to be right.
I would commit it as soon as I am able to have a cvs access(
I hope it would be before 7.2 release).

Hiroshi, I will take care of applying the patch. Are you having CVS
troubles we can help with?

My entry in hub.org seems to have gone about 25 days ago
by some accident. I got my entry in another host instead
but I don't know how to do with it.

regards,
Hiroshi Inoue

I wrote:

Bruce Momjian wrote:

Here is the submitted patch so everyone can see it.

I agree to the first change but am suspicious about the
second change.

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

Though Aidan is right, is -1 guaranteed to be true in
any application conversely ? It seems also different
from the ODBC spec. I don't have the solution now.

Please never apply the second part.
I would propose another idea instead soon(hopefully).

regards,
Hiroshi Inoue

-- 425,452 ----
break;

case PG_TYPE_BOOL:
!
if (cbValueMax > len)

<<<<<<< len isn't set at this point. >>>>>>>

{
! /* Aidan Mountford
(aidan@oz.to) 1/08/2001:
!
! When returning a CHAR
datatype, return -1
! instead of +1.
!
! That way if someone
casts it back to a boolean
! again - it will work.
! */
! if ( *(char *) value = '1')

<<<<<<<< '=' must be '=='. >>>>>>>>

Bruce Momjian wrote:

You seem to be right.
I would commit it as soon as I am able to have a cvs access(
I hope it would be before 7.2 release).

Hiroshi, I will take care of applying the patch. Are you having CVS
troubles we can help with?

My entry in hub.org seems to have gone about 25 days ago
by some accident. I got my entry in another host instead
but I don't know how to do with it.

I use postgresql.org for CVS. Not sure if that is the same as hub.org.
nslookup shows them with different IP addresses.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

Though Aidan is right, is -1 guaranteed to be true in
any application conversely ? It seems also different
from the ODBC spec. I don't have the solution now.

Please never apply the second part.
I would propose another idea instead soon(hopefully).

OK, I will not apply the patch and wait to hear from you.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

I am going to apply the first part only. Hiroshi has ideas on the
second part.

As Requested ...

-----Original Message-----
From: Bruce Momjian [mailto:pgman@candle.pha.pa.us]
Sent: Thursday, 2 August 2001 12:30 PM
To: Aidan Mountford
Cc: 'pgsql-patches@postgresql.org'
Subject: Re: [PATCHES] ODBC Boolean handling

I have reviewed the current CVS code and it looks quite different from
the patch you submitted. Any chance of you downloading the snapshot and
sending a context diff, rather than an ordinary one. I could apply that
easily.

I had a few issues with boolean handling in ODBC driver.

1) When a row is retrieved, and then a SQL_FETCH_FIRST is issued, the

check

in convert.c
does not consider the fact that the value in the field has been altered to
be a '1' if the
backend handed it a 't'. The net result being that the first row on any
subsequent queries
has all it's boolean set to 0.

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

The following patch fixes both of these issues..

Cheers for building a fantastic product

Aidan

cvs diff convert.c (in directory

C:\CVS-POSTGRES\pgsql\src\interfaces\odbc\)

Index: convert.c
===================================================================
RCS file:

/home/projects/pgsql/cvsroot/pgsql/src/interfaces/odbc/convert.c,v

retrieving revision 1.44
diff -r1.44 convert.c
295c295,307
< if (s[0] == 'T' || s[0] == 't')
---

/* Aidan Mountford (aidan@oz.to) 1/08/2001:

if (s[0] == 'T' || s[0] == 't')

<<< This wont work...

When MoveFirst is called twice on

one set of tuples,

this will have the effect of setting

s[0] to 1 on the

first pass, and s[0] on the second.

This is bad ;)

*/

if (s[0] == 'T' || s[0] == 't' || s[0] ==

'1')
415c428
< len = 1;
---

418c431,449
< strcpy(rgbValueBindRow, value);
---

/* Aidan Mountford (aidan@oz.to)

1/08/2001:

When returning a CHAR

datatype, return -1

instead of +1.

That way if someone casts it

back to a boolean

again - it will work.
*/
if ( *(char *) value = '1')
{
strcpy(rgbValueBindRow, "-1");
len = 2;
}
else
{
strcpy(rgbValueBindRow, "0");
len = 1;
}

1160a1192,1193

1161a1195

.....................................................
. Cisco Certified Network Associate
. Microsoft Certified Professional
. 3Com 3Wizard
.
. Technical Director
. MindVision Interactive Pty. Ltd.
. http://www.mindvision.com.au
. PH: +61-8-8212-9544
. FAX: +61-8-8212-9644
. 48 Light Square
. Adelaide SA
. AUSTRALIA 5000
.....................................................

"Remember, There is always scope for things to be worse"

---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to majordomo@postgresql.org)

-- 
Bruce Momjian                        |  http://candle.pha.pa.us
pgman@candle.pha.pa.us               |  (610) 853-3000
+  If your life is a hard drive,     |  830 Blythe Avenue
+  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

---------------------------(end of broadcast)---------------------------
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql.org so that your
message can get through to the mailing list cleanly

[ Attachment, skipping... ]

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

Here is the patchd I applied. It fixes item 1, but not item 2. Hiroshi
wants to work on that one.

As Requested ...

-----Original Message-----
From: Bruce Momjian [mailto:pgman@candle.pha.pa.us]
Sent: Thursday, 2 August 2001 12:30 PM
To: Aidan Mountford
Cc: 'pgsql-patches@postgresql.org'
Subject: Re: [PATCHES] ODBC Boolean handling

I have reviewed the current CVS code and it looks quite different from
the patch you submitted. Any chance of you downloading the snapshot and
sending a context diff, rather than an ordinary one. I could apply that
easily.

I had a few issues with boolean handling in ODBC driver.

1) When a row is retrieved, and then a SQL_FETCH_FIRST is issued, the

check

in convert.c
does not consider the fact that the value in the field has been altered to
be a '1' if the
backend handed it a 't'. The net result being that the first row on any
subsequent queries
has all it's boolean set to 0.

2) I had issues with some utilities that, when casting from string to
boolean, and having
the global 'bools as char' set, require that -1 = True, and 0 = false.

The following patch fixes both of these issues..

Cheers for building a fantastic product

Aidan

cvs diff convert.c (in directory

C:\CVS-POSTGRES\pgsql\src\interfaces\odbc\)

Index: convert.c
===================================================================
RCS file:

/home/projects/pgsql/cvsroot/pgsql/src/interfaces/odbc/convert.c,v

retrieving revision 1.44
diff -r1.44 convert.c
295c295,307
< if (s[0] == 'T' || s[0] == 't')
---

/* Aidan Mountford (aidan@oz.to) 1/08/2001:

if (s[0] == 'T' || s[0] == 't')

<<< This wont work...

When MoveFirst is called twice on

one set of tuples,

this will have the effect of setting

s[0] to 1 on the

first pass, and s[0] on the second.

This is bad ;)

*/

if (s[0] == 'T' || s[0] == 't' || s[0] ==

'1')
415c428
< len = 1;
---

418c431,449
< strcpy(rgbValueBindRow, value);
---

/* Aidan Mountford (aidan@oz.to)

1/08/2001:

When returning a CHAR

datatype, return -1

instead of +1.

That way if someone casts it

back to a boolean

again - it will work.
*/
if ( *(char *) value = '1')
{
strcpy(rgbValueBindRow, "-1");
len = 2;
}
else
{
strcpy(rgbValueBindRow, "0");
len = 1;
}

1160a1192,1193

1161a1195

.....................................................
. Cisco Certified Network Associate
. Microsoft Certified Professional
. 3Com 3Wizard
.
. Technical Director
. MindVision Interactive Pty. Ltd.
. http://www.mindvision.com.au
. PH: +61-8-8212-9544
. FAX: +61-8-8212-9644
. 48 Light Square
. Adelaide SA
. AUSTRALIA 5000
.....................................................

"Remember, There is always scope for things to be worse"

---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to majordomo@postgresql.org)

-- 
Bruce Momjian                        |  http://candle.pha.pa.us
pgman@candle.pha.pa.us               |  (610) 853-3000
+  If your life is a hard drive,     |  830 Blythe Avenue
+  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

---------------------------(end of broadcast)---------------------------
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql.org so that your
message can get through to the mailing list cleanly

[ Attachment, skipping... ]

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026