diff mbox series

[meta-oe,1/2] psqlodbc: upgrade 16.00.0000 -> 17.00.0006

Message ID 20251023144351.264756-1-skandigraun@gmail.com
State Accepted
Headers show
Series [meta-oe,1/2] psqlodbc: upgrade 16.00.0000 -> 17.00.0006 | expand

Commit Message

Gyorgy Sarvari Oct. 23, 2025, 2:43 p.m. UTC
Drop patch that was merged into this release.

Changelog: https://odbc.postgresql.org/docs/release.html

Signed-off-by: Gyorgy Sarvari <skandigraun@gmail.com>
---
 ...sure-to-call-AC_CHECK_HEADER_STDBOOL.patch | 45 -------------------
 ...c_16.00.0000.bb => psqlodbc_17.00.0006.bb} |  4 +-
 2 files changed, 2 insertions(+), 47 deletions(-)
 delete mode 100644 meta-oe/recipes-dbs/psqlodbc/files/0001-configure-make-sure-to-call-AC_CHECK_HEADER_STDBOOL.patch
 rename meta-oe/recipes-dbs/psqlodbc/{psqlodbc_16.00.0000.bb => psqlodbc_17.00.0006.bb} (92%)

Comments

Gyorgy Sarvari Oct. 23, 2025, 2:45 p.m. UTC | #1
After this patch there are still 2 more failing tests (they were present
in the previous version also), and I'm kinda clueless at this moment...

On 10/23/25 16:43, Gyorgy Sarvari via lists.openembedded.org wrote:
> The descrec test fails currently, due to the different characterwidth
> postgres uses depending on the current locale.
>
> If the locale is C or posix, then each character is 1 byte. Otherwise,
> if it's some UTF8 locale, then it is 4 bytes.
>
> The test case expects 4 byte width by default, making it fail. This backported patch
> allows for both 4 and 1 byte characters to be tested successfully.
>
> Signed-off-by: Gyorgy Sarvari <skandigraun@gmail.com>
> ---
>  ...xpected-output-file-for-descrec-test.patch | 72 +++++++++++++++++++
>  .../psqlodbc/psqlodbc_17.00.0006.bb           |  1 +
>  2 files changed, 73 insertions(+)
>  create mode 100644 meta-oe/recipes-dbs/psqlodbc/files/add-expected-output-file-for-descrec-test.patch
>
> diff --git a/meta-oe/recipes-dbs/psqlodbc/files/add-expected-output-file-for-descrec-test.patch b/meta-oe/recipes-dbs/psqlodbc/files/add-expected-output-file-for-descrec-test.patch
> new file mode 100644
> index 0000000000..af25b207c8
> --- /dev/null
> +++ b/meta-oe/recipes-dbs/psqlodbc/files/add-expected-output-file-for-descrec-test.patch
> @@ -0,0 +1,72 @@
> +From 453a8bdcaafde9fcd108a195ffea3ccb5c4d0a21 Mon Sep 17 00:00:00 2001
> +From: Christoph Berg <myon@debian.org>
> +Date: Mon, 16 Jun 2025 20:25:26 +0200
> +Subject: [PATCH] Add expected output file for descrec test in C locale (#131)
> +
> +When running with LC_ALL=C, SQL_DESC_OCTET_LENGTH matches the character
> +length. Add "10" as accepted value.
> +
> +The net diff to the existing descrec.out file is this:
> +
> + test/expected/descrec.out	2025-06-13 19:02:36.460302799 +0200
> + test/expected/descrec_2.out	2025-06-16 14:34:44.270543393 +0200
> + @@ -19,7 +19,7 @@
> +  -- Column 3 --
> +  SQL_DESC_NAME: col3
> +  SQL_DESC_TYPE: 12
> + -SQL_DESC_OCTET_LENGTH: 40
> + +SQL_DESC_OCTET_LENGTH: 10
> +  SQL_DESC_PRECISION: 0
> +  SQL_DESC_SCALE: 0
> +  SQL_DESC_NULLABLE: 0
> +
> +Fixes #51.
> +
> +Upstream-Status: Backport [https://github.com/postgresql-interfaces/psqlodbc/commit/453a8bdcaafde9fcd108a195ffea3ccb5c4d0a21]
> +Signed-off-by: Gyorgy Sarvari <skandigraun@gmail.com>
> +
> +---
> + test/expected/descrec_2.out | 34 ++++++++++++++++++++++++++++++++++
> + 1 file changed, 34 insertions(+)
> + create mode 100644 test/expected/descrec_2.out
> +
> +diff --git a/test/expected/descrec_2.out b/test/expected/descrec_2.out
> +new file mode 100644
> +index 0000000..6f61c49
> +--- /dev/null
> ++++ b/test/expected/descrec_2.out
> +@@ -0,0 +1,34 @@
> ++connected
> ++
> ++-- Column 1 --
> ++SQL_DESC_NAME: col1
> ++SQL_DESC_TYPE: 4
> ++SQL_DESC_OCTET_LENGTH: 0
> ++SQL_DESC_PRECISION: 0
> ++SQL_DESC_SCALE: 0
> ++SQL_DESC_NULLABLE: 0
> ++
> ++-- Column 2 --
> ++SQL_DESC_NAME: col2
> ++SQL_DESC_TYPE: 2
> ++SQL_DESC_OCTET_LENGTH: 6
> ++SQL_DESC_PRECISION: 4
> ++SQL_DESC_SCALE: 2
> ++SQL_DESC_NULLABLE: 1
> ++
> ++-- Column 3 --
> ++SQL_DESC_NAME: col3
> ++SQL_DESC_TYPE: 12
> ++SQL_DESC_OCTET_LENGTH: 10
> ++SQL_DESC_PRECISION: 0
> ++SQL_DESC_SCALE: 0
> ++SQL_DESC_NULLABLE: 0
> ++
> ++-- Column 4 --
> ++SQL_DESC_NAME: col4
> ++SQL_DESC_TYPE: -5
> ++SQL_DESC_OCTET_LENGTH: 8
> ++SQL_DESC_PRECISION: 0
> ++SQL_DESC_SCALE: 0
> ++SQL_DESC_NULLABLE: 0
> ++disconnecting
> diff --git a/meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb b/meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb
> index cead6789f4..2fbd8b78e4 100644
> --- a/meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb
> +++ b/meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb
> @@ -24,6 +24,7 @@ SRC_URI = "git://github.com/postgresql-interfaces/psqlodbc.git;protocol=https;br
>      file://psqlodbc-remove-some-checks-for-cross-compiling.patch \
>      file://psqlodbc-donot-use-the-hardcode-libdir.patch \
>      file://psqlodbc-fix-for-ptest-support.patch \
> +    file://add-expected-output-file-for-descrec-test.patch \
>      file://run-ptest \
>  "
>  
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#120935): https://lists.openembedded.org/g/openembedded-devel/message/120935
> Mute This Topic: https://lists.openembedded.org/mt/115911779/6084445
> Group Owner: openembedded-devel+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [skandigraun@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>
Ross Burton Oct. 31, 2025, 1:34 p.m. UTC | #2
On 23 Oct 2025, at 15:43, Gyorgy Sarvari via lists.openembedded.org <skandigraun=gmail.com@lists.openembedded.org> wrote:
> 
> -SRC_URI = "http://ftp.postgresql.org/pub/odbc/versions.old/src/${BPN}-${PV}.tar.gz \
> +PV_WITH_USCORE = "${@d.getVar('PV').replace('.', '_')}"
> +SRC_URI = "git://github.com/postgresql-interfaces/psqlodbc.git;protocol=https;branch=main;tag=REL-${PV_WITH_USCORE} \

Note that this recipe doesn’t set SRCREV, so this is now doing a remote tag lookup on every parse.

Ross
Yoann Congal Oct. 31, 2025, 1:45 p.m. UTC | #3
Le ven. 31 oct. 2025 à 14:35, Ross Burton via lists.openembedded.org
<ross.burton=arm.com@lists.openembedded.org> a écrit :

> On 23 Oct 2025, at 15:43, Gyorgy Sarvari via lists.openembedded.org
> <skandigraun=gmail.com@lists.openembedded.org> wrote:
> >
> > -SRC_URI = "
> http://ftp.postgresql.org/pub/odbc/versions.old/src/${BPN}-${PV}.tar.gz \
> > +PV_WITH_USCORE = "${@d.getVar('PV').replace('.', '_')}"
> > +SRC_URI = "git://
> github.com/postgresql-interfaces/psqlodbc.git;protocol=https;branch=main;tag=REL-${PV_WITH_USCORE}
> <http://github.com/postgresql-interfaces/psqlodbc.git;protocol=https;branch=main;tag=REL-$%7BPV_WITH_USCORE%7D>
> \
>
> Note that this recipe doesn’t set SRCREV, so this is now doing a remote
> tag lookup on every parse.
>

Nice catch, this is most likely what did break the meta-oe-mirror test :
https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors

Maybe we need to add a recipe qa warning about this...


> Ross
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#121232):
> https://lists.openembedded.org/g/openembedded-devel/message/121232
> Mute This Topic: https://lists.openembedded.org/mt/115911777/4316185
> Group Owner: openembedded-devel+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
> yoann.congal@smile.fr]
> -=-=-=-=-=-=-=-=-=-=-=-
>
>
Ross Burton Oct. 31, 2025, 2:19 p.m. UTC | #4
On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
> Nice catch, this is most likely what did break the meta-oe-mirror test : https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors

Yes, it would be.

> Maybe we need to add a recipe qa warning about this…

Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake —parse-only” stage to our CI, because one of our recipes was using a named tag too.

Ross
Yoann Congal Oct. 31, 2025, 3:06 p.m. UTC | #5
Le ven. 31 oct. 2025 à 15:20, Ross Burton <Ross.Burton@arm.com> a écrit :

> On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
> > Nice catch, this is most likely what did break the meta-oe-mirror test :
> https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors
>
> Yes, it would be.
>
> > Maybe we need to add a recipe qa warning about this…
>
> Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake
> —parse-only” stage to our CI, because one of our recipes was using a named
> tag too.
>

For reference, I created
https://bugzilla.yoctoproject.org/show_bug.cgi?id=16051

>
>
>
>
> Ross
Jose Quaresma Oct. 31, 2025, 3:22 p.m. UTC | #6
Ross Burton via lists.openembedded.org <ross.burton=
arm.com@lists.openembedded.org> escreveu (sexta, 31/10/2025 à(s) 14:20):

> On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
> > Nice catch, this is most likely what did break the meta-oe-mirror test :
> https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors
>
> Yes, it would be.
>
> > Maybe we need to add a recipe qa warning about this…
>
> Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake
> —parse-only” stage to our CI, because one of our recipes was using a named
> tag too.
>

Would it make sense to implement something like this on
the yocto-check-layer?

Jose


> Ross
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#121235):
> https://lists.openembedded.org/g/openembedded-devel/message/121235
> Mute This Topic: https://lists.openembedded.org/mt/115911777/5052612
> Group Owner: openembedded-devel+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
> quaresma.jose@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>
>
Gyorgy Sarvari Oct. 31, 2025, 3:40 p.m. UTC | #7
On 10/31/25 14:34, Ross Burton wrote:
> On 23 Oct 2025, at 15:43, Gyorgy Sarvari via lists.openembedded.org <skandigraun=gmail.com@lists.openembedded.org> wrote:
>> -SRC_URI = "http://ftp.postgresql.org/pub/odbc/versions.old/src/${BPN}-${PV}.tar.gz \
>> +PV_WITH_USCORE = "${@d.getVar('PV').replace('.', '_')}"
>> +SRC_URI = "git://github.com/postgresql-interfaces/psqlodbc.git;protocol=https;branch=main;tag=REL-${PV_WITH_USCORE} \
> Note that this recipe doesn’t set SRCREV, so this is now doing a remote tag lookup on every parse.
>

I'd like to say that it was a test to see if anyone's paying
attention... but in reality I just totally missed it. Thanks for the fix.
Yoann Congal Oct. 31, 2025, 3:49 p.m. UTC | #8
Le ven. 31 oct. 2025 à 16:22, Jose Quaresma <quaresma.jose@gmail.com> a
écrit :

>
>
> Ross Burton via lists.openembedded.org <ross.burton=
> arm.com@lists.openembedded.org> escreveu (sexta, 31/10/2025 à(s) 14:20):
>
>> On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
>> > Nice catch, this is most likely what did break the meta-oe-mirror test
>> :
>> https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors
>>
>> Yes, it would be.
>>
>> > Maybe we need to add a recipe qa warning about this…
>>
>> Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake
>> —parse-only” stage to our CI, because one of our recipes was using a named
>> tag too.
>>
>
> Would it make sense to implement something like this on
> the yocto-check-layer?
>

Sounds like a good idea to me. I'll add it to the ticket.

If I recall correctly, check-layer is a superset of qa checks so adding it
there should not be too hard.



> Jose
>
>
>> Ross
>> -=-=-=-=-=-=-=-=-=-=-=-
>> Links: You receive all messages sent to this group.
>> View/Reply Online (#121235):
>> https://lists.openembedded.org/g/openembedded-devel/message/121235
>> Mute This Topic: https://lists.openembedded.org/mt/115911777/5052612
>> Group Owner: openembedded-devel+owner@lists.openembedded.org
>> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
>> quaresma.jose@gmail.com]
>> -=-=-=-=-=-=-=-=-=-=-=-
>>
>>
>
> --
> Best regards,
>
> José Quaresma
>
Jose Quaresma Oct. 31, 2025, 4:02 p.m. UTC | #9
Yoann Congal <yoann.congal@smile.fr> escreveu (sexta, 31/10/2025 à(s)
15:49):

>
>
>
> Le ven. 31 oct. 2025 à 16:22, Jose Quaresma <quaresma.jose@gmail.com> a
> écrit :
>
>>
>>
>> Ross Burton via lists.openembedded.org <ross.burton=
>> arm.com@lists.openembedded.org> escreveu (sexta, 31/10/2025 à(s) 14:20):
>>
>>> On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
>>> > Nice catch, this is most likely what did break the meta-oe-mirror test
>>> :
>>> https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors
>>>
>>> Yes, it would be.
>>>
>>> > Maybe we need to add a recipe qa warning about this…
>>>
>>> Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake
>>> —parse-only” stage to our CI, because one of our recipes was using a named
>>> tag too.
>>>
>>
>> Would it make sense to implement something like this on
>> the yocto-check-layer?
>>
>
> Sounds like a good idea to me. I'll add it to the ticket.
>
> If I recall correctly, check-layer is a superset of qa checks so adding it
> there should not be too hard.
>

There are already functions check_command on the
scripts/lib/checklayer/__init__.py, so I think it's simple to implement.
I can take a look at this next week.

Jose


>
>
>
>> Jose
>>
>>
>>> Ross
>>> -=-=-=-=-=-=-=-=-=-=-=-
>>> Links: You receive all messages sent to this group.
>>> View/Reply Online (#121235):
>>> https://lists.openembedded.org/g/openembedded-devel/message/121235
>>> Mute This Topic: https://lists.openembedded.org/mt/115911777/5052612
>>> Group Owner: openembedded-devel+owner@lists.openembedded.org
>>> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
>>> quaresma.jose@gmail.com]
>>> -=-=-=-=-=-=-=-=-=-=-=-
>>>
>>>
>>
>> --
>> Best regards,
>>
>> José Quaresma
>>
>
Yoann Congal Oct. 31, 2025, 6:19 p.m. UTC | #10
Le ven. 31 oct. 2025 à 17:02, Jose Quaresma via lists.openembedded.org
<quaresma.jose=gmail.com@lists.openembedded.org> a écrit :

>
>
> Yoann Congal <yoann.congal@smile.fr> escreveu (sexta, 31/10/2025 à(s)
> 15:49):
>
>>
>>
>>
>> Le ven. 31 oct. 2025 à 16:22, Jose Quaresma <quaresma.jose@gmail.com> a
>> écrit :
>>
>>>
>>>
>>> Ross Burton via lists.openembedded.org <ross.burton=
>>> arm.com@lists.openembedded.org> escreveu (sexta, 31/10/2025 à(s) 14:20):
>>>
>>>> On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
>>>> > Nice catch, this is most likely what did break the meta-oe-mirror
>>>> test :
>>>> https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors
>>>>
>>>> Yes, it would be.
>>>>
>>>> > Maybe we need to add a recipe qa warning about this…
>>>>
>>>> Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake
>>>> —parse-only” stage to our CI, because one of our recipes was using a named
>>>> tag too.
>>>>
>>>
>>> Would it make sense to implement something like this on
>>> the yocto-check-layer?
>>>
>>
>> Sounds like a good idea to me. I'll add it to the ticket.
>>
>> If I recall correctly, check-layer is a superset of qa checks so adding
>> it there should not be too hard.
>>
>
> There are already functions check_command on the
> scripts/lib/checklayer/__init__.py, so I think it's simple to implement.
> I can take a look at this next week.
>

I was more thinking about adding a QA check outside of checklayer then
adding it to CHECKLAYER_REQUIRED_TESTS. This way users can still benefit
from the QA check without running checklayer (I would bet most users don't).
You can take https://bugzilla.yoctoproject.org/show_bug.cgi?id=16051 from
me if you plan to work on this.
Thanks!


> Jose
>
>
>>
>>
>>
>>> Jose
>>>
>>>
>>>> Ross
>>>>
>>>>
>>>>
>>>
>>> --
>>> Best regards,
>>>
>>> José Quaresma
>>>
>>
>
> --
> Best regards,
>
> José Quaresma
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#121241):
> https://lists.openembedded.org/g/openembedded-devel/message/121241
> Mute This Topic: https://lists.openembedded.org/mt/115911777/4316185
> Group Owner: openembedded-devel+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
> yoann.congal@smile.fr]
> -=-=-=-=-=-=-=-=-=-=-=-
>
>
Jose Quaresma Nov. 3, 2025, 11:21 a.m. UTC | #11
Yoann Congal <yoann.congal@smile.fr> escreveu (sexta, 31/10/2025 à(s)
18:19):

>
>
> Le ven. 31 oct. 2025 à 17:02, Jose Quaresma via lists.openembedded.org
> <quaresma.jose=gmail.com@lists.openembedded.org> a écrit :
>
>>
>>
>> Yoann Congal <yoann.congal@smile.fr> escreveu (sexta, 31/10/2025 à(s)
>> 15:49):
>>
>>>
>>>
>>>
>>> Le ven. 31 oct. 2025 à 16:22, Jose Quaresma <quaresma.jose@gmail.com> a
>>> écrit :
>>>
>>>>
>>>>
>>>> Ross Burton via lists.openembedded.org <ross.burton=
>>>> arm.com@lists.openembedded.org> escreveu (sexta, 31/10/2025 à(s)
>>>> 14:20):
>>>>
>>>>> On 31 Oct 2025, at 13:45, Yoann Congal <yoann.congal@smile.fr> wrote:
>>>>> > Nice catch, this is most likely what did break the meta-oe-mirror
>>>>> test :
>>>>> https://autobuilder.yoctoproject.org/valkyrie/#/builders/82/builds/494/steps/17/logs/errors
>>>>>
>>>>> Yes, it would be.
>>>>>
>>>>> > Maybe we need to add a recipe qa warning about this…
>>>>>
>>>>> Agreed. I discovered this when adding a “BB_NO_NETWORK=1 bitbake
>>>>> —parse-only” stage to our CI, because one of our recipes was using a named
>>>>> tag too.
>>>>>
>>>>
>>>> Would it make sense to implement something like this on
>>>> the yocto-check-layer?
>>>>
>>>
>>> Sounds like a good idea to me. I'll add it to the ticket.
>>>
>>> If I recall correctly, check-layer is a superset of qa checks so adding
>>> it there should not be too hard.
>>>
>>
>> There are already functions check_command on the
>> scripts/lib/checklayer/__init__.py, so I think it's simple to implement.
>> I can take a look at this next week.
>>
>
> I was more thinking about adding a QA check outside of checklayer then
> adding it to CHECKLAYER_REQUIRED_TESTS. This way users can still benefit
> from the QA check without running checklayer (I would bet most users don't).
> You can take https://bugzilla.yoctoproject.org/show_bug.cgi?id=16051 from
> me if you plan to work on this.
> Thanks!
>

Your approach is more appropriate than what I had in mind, so I'll leave
that for you to do.


>
>
>> Jose
>>
>>
>>>
>>>
>>>
>>>> Jose
>>>>
>>>>
>>>>> Ross
>>>>>
>>>>>
>>>>>
>>>>
>>>> --
>>>> Best regards,
>>>>
>>>> José Quaresma
>>>>
>>>
>>
>> --
>> Best regards,
>>
>> José Quaresma
>>
>> -=-=-=-=-=-=-=-=-=-=-=-
>> Links: You receive all messages sent to this group.
>> View/Reply Online (#121241):
>> https://lists.openembedded.org/g/openembedded-devel/message/121241
>> Mute This Topic: https://lists.openembedded.org/mt/115911777/4316185
>> Group Owner: openembedded-devel+owner@lists.openembedded.org
>> Unsubscribe: https://lists.openembedded.org/g/openembedded-devel/unsub [
>> yoann.congal@smile.fr]
>> -=-=-=-=-=-=-=-=-=-=-=-
>>
>>
>
> --
> Yoann Congal
> Smile ECS
>
diff mbox series

Patch

diff --git a/meta-oe/recipes-dbs/psqlodbc/files/0001-configure-make-sure-to-call-AC_CHECK_HEADER_STDBOOL.patch b/meta-oe/recipes-dbs/psqlodbc/files/0001-configure-make-sure-to-call-AC_CHECK_HEADER_STDBOOL.patch
deleted file mode 100644
index 9cc5ad9b86..0000000000
--- a/meta-oe/recipes-dbs/psqlodbc/files/0001-configure-make-sure-to-call-AC_CHECK_HEADER_STDBOOL.patch
+++ /dev/null
@@ -1,45 +0,0 @@ 
-From 5bfa9f1cd0523a0cad70cacf5ecf8c1aeb06c4d0 Mon Sep 17 00:00:00 2001
-From: Martin Jansa <martin.jansa@gmail.com>
-Date: Tue, 22 Apr 2025 17:47:02 +0000
-Subject: [PATCH] configure: make sure to call AC_CHECK_HEADER_STDBOOL
-
-before using ac_cv_sizeof_bool
-
-It was called after this conditional, so PG_USE_STDBOOL wasn't set
-even when it should be as shown in at the end of config.log:
-
-  ac_cv_header_stdbool_h=yes
-  ac_cv_sizeof_bool=1
-  ac_cv_type__Bool=yes
-  #define SIZEOF_BOOL 1
-  #define HAVE__BOOL 1
-  #define HAVE_STDBOOL_H 1
-
-* fixes:
-  https://github.com/postgresql-interfaces/psqlodbc/issues/110
-  https://github.com/postgresql-interfaces/psqlodbc/issues/94
-
-Upstream-Status: Submitted [https://github.com/postgresql-interfaces/psqlodbc/pull/112]
-
-Signed-off-by: Martin Jansa <martin.jansa@gmail.com>
----
- configure.ac | 4 +++-
- 1 file changed, 3 insertions(+), 1 deletion(-)
-
-diff --git a/configure.ac b/configure.ac
-index f35c825..ea363ca 100644
---- a/configure.ac
-+++ b/configure.ac
-@@ -28,9 +28,11 @@ AC_CHECK_SIZEOF([bool], [],
- #include <stdbool.h>
- #endif])
- 
-+AC_CHECK_HEADER_STDBOOL()
-+
- dnl We use <stdbool.h> if we have it and it declares type bool as having
- dnl size 1.  Otherwise, c.h will fall back to declaring bool as unsigned char.
--if test "$ac_cv_header_stdbool_h" = yes -a "$ac_cv_sizeof_bool" = 1; then
-+if test "$ac_cv_header_stdbool_h" = yes && test "$ac_cv_sizeof_bool" = 1; then
-   AC_DEFINE([PG_USE_STDBOOL], 1,
-             [Define to 1 to use <stdbool.h> to define type bool.])
- fi
diff --git a/meta-oe/recipes-dbs/psqlodbc/psqlodbc_16.00.0000.bb b/meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb
similarity index 92%
rename from meta-oe/recipes-dbs/psqlodbc/psqlodbc_16.00.0000.bb
rename to meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb
index dfd7f6a1e6..cead6789f4 100644
--- a/meta-oe/recipes-dbs/psqlodbc/psqlodbc_16.00.0000.bb
+++ b/meta-oe/recipes-dbs/psqlodbc/psqlodbc_17.00.0006.bb
@@ -19,11 +19,11 @@  HOMEPAGE = "https://odbc.postgresql.org/"
 LICENSE = "LGPL-2.0-only"
 LIC_FILES_CHKSUM = "file://license.txt;md5=6db3822fc7512e83087ba798da013692"
 
-SRC_URI = "http://ftp.postgresql.org/pub/odbc/versions.old/src/${BPN}-${PV}.tar.gz \
+PV_WITH_USCORE = "${@d.getVar('PV').replace('.', '_')}"
+SRC_URI = "git://github.com/postgresql-interfaces/psqlodbc.git;protocol=https;branch=main;tag=REL-${PV_WITH_USCORE} \
     file://psqlodbc-remove-some-checks-for-cross-compiling.patch \
     file://psqlodbc-donot-use-the-hardcode-libdir.patch \
     file://psqlodbc-fix-for-ptest-support.patch \
-    file://0001-configure-make-sure-to-call-AC_CHECK_HEADER_STDBOOL.patch \
     file://run-ptest \
 "