From finjulhich at gmail.com  Mon Aug 25 09:17:57 2025
From: finjulhich at gmail.com (MM)
Date: Mon Aug 25 09:18:27 2025
Subject: [odb-users] internal compiler error: in cl_optimization_save,
	at options-save.cc:293
Message-ID: <CA+Jb2D1SPanR5WZEmLTTKXpYLFjMXdSgE=-RnHte4VXo01q69g@mail.gmail.com>

Trying to create a reproducible example here, but no success yet.

In the meantime, perhaps this is a known issue

odb -d pgsql --std c++17 --profile boost --generate-query -I...
-I/usr/include --options-file .../odb.options --changelog
.../odb/...._pgsql.xml .../....hpp

*** WARNING *** there are active plugins, do not report this as a bug
unless you can reproduce it without enabling any plugins.
Event                            | Plugins
PLUGIN_START_UNIT                | odb
PLUGIN_PRAGMAS                   | odb
PLUGIN_OVERRIDE_GATE             | odb
<stdin>: internal compiler error: in cl_optimization_save, at
options-save.cc:293

I believe I built odb with gcc 15.0, while now my system has 15.2.1

strings $(which odb) | grep -i gcc
libgcc_s.so.1
GCC_3.0
/usr/local/src/odb-gcc-15/odb-2.5.0/odb/pregenerated/odb/options.cxx
odb.environment = [strings] CPATH CPLUS_INCLUDE_PATH GCC_EXEC_PREFIX
COMPILER_PATH
GCC: (GNU) 15.1.1 20250521 (Red Hat 15.1.1-2)
GCC: (GNU) 15.0.1 20250418 (Red Hat 15.0.1-0)
RV:running gcc 15.0.1 20250418
BV:annobin gcc 15.0.1 20250418
.gcc_except_table

I would think minor version changes of gcc don't require a odb rebuild

Rds,
MM
From boris at codesynthesis.com  Tue Aug 26 03:14:23 2025
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Tue Aug 26 03:14:35 2025
Subject: [odb-users] internal compiler error: in cl_optimization_save, at
	options-save.cc:293
In-Reply-To: <CA+Jb2D1SPanR5WZEmLTTKXpYLFjMXdSgE=-RnHte4VXo01q69g@mail.gmail.com>
References: <CA+Jb2D1SPanR5WZEmLTTKXpYLFjMXdSgE=-RnHte4VXo01q69g@mail.gmail.com>
Message-ID: <boris.20250826091226@codesynthesis.com>

MM <finjulhich@gmail.com> writes:

> I believe I built odb with gcc 15.0, while now my system has 15.2.1
> 
> I would think minor version changes of gcc don't require a odb rebuild

GCC does not guarantee any stability for its plugins API (which is
essentially all of GCC). So I would suggest that you rebuild with
GCC 15.2.1 and see if that makes the error go away.

From roberto.minarelli.de at gmail.com  Tue Aug 26 05:56:33 2025
From: roberto.minarelli.de at gmail.com (roberto minarelli)
Date: Tue Aug 26 07:46:02 2025
Subject: [odb-users] Inserting CLOB greater having size greater then 12288
In-Reply-To: <boris.20240201111915@codesynthesis.com>
References: <CADPyhbAMzF8n2h1EYY1dKFxuUZUtO=4a3vx9BicLnWLztqVbhw@mail.gmail.com>
	<boris.20240112122542@codesynthesis.com>
	<CADPyhbDwxrw+Dd7ichPa2HZaMxjtvwJ=_bGepcianGdHy5SfdQ@mail.gmail.com>
	<boris.20240115122858@codesynthesis.com>
	<CADPyhbDrw6gc2WD=w8zUbzpNODNADNJqz+DnsUwrBCFpVpthbA@mail.gmail.com>
	<CADPyhbCzvhoxAXJQPQEAzeHA3O9D9iTMnUyZwvmufMQqn7rmwA@mail.gmail.com>
	<boris.20240117150128@codesynthesis.com>
	<CADPyhbC4g+EJYEVvrq0muJNr2wppJq+yrjT+nQSSZyGWTwLy=g@mail.gmail.com>
	<CADPyhbDsbf-cCMcH+3gcG_YpoUNkHxTSGaYD7s=+6ngUoP+9Vg@mail.gmail.com>
	<CADPyhbCvw7ciSQicGm3=jp2AiaAfZtQJRVsM8eSPm1+xmCk85A@mail.gmail.com>
	<boris.20240201111915@codesynthesis.com>
Message-ID: <CADPyhbD=ozTWqJp=+X3AZq+q0DqEN+yTaR3bOHy=XKa+VOQq2w@mail.gmail.com>

Hi again Boris,
I am stuck on this problem
I have a table that is like
Table_A
{
string UUID
string sLob CLOB
}

I start to insert a lob of 70k but it doesn't fit
So I reduced the length until It is successfully inserted
This happens with

OL 8.8
libodb-2.5.0-b.25.so
libodb-oracle-2.5.0-b.25.so
Oracle 19c

if I try with the same library and the same OS but Oracle 11g it fits

So what could be the cause?
I tried also to debug odb library but didn't understand what happens

the Oracle client is always the 12.2.0

Do you believe that installing the odb 2.5.0 will solve the situation?

Please let me know as soon as possible

Thank you in advance

Regards
Roberto Minarelli Della Valle

Il giorno gio 1 feb 2024 alle ore 10:26 Boris Kolpackov <
boris@codesynthesis.com> ha scritto:

> roberto minarelli <roberto.minarelli.de@gmail.com> writes:
>
> > have you an update for this issue?
>
> No, I haven't had a chance to look into this and unfortunately won't
> be able to for the foreseeable future.
>
>
> > I tried to debug the callback but I couldn't step into it
>
> Try to set the breakpoint inside the callback. This should allow
> you to step through it without having to go through OCI code (which
> I assume is what's causing the difficulty).
>
>
> > My guess is that the callback is called three times, each of which with a
> > chunk of 4Kb of maximum dimension when writing
> > And it is called maximum three times
> > first time with OCI_FIRST_PIECE
> > second time with OCI_NEXT_PIECE
> > and finally with OCI_LAST_PIECE
>
> What is the size of the data you are trying to insert and will it fit
> into the three 4Kb chunks?
>
> Do a little bit of digging and thinking on your own, I am willing to
> help guide you in the right direction but "it doesn't work, help"
> attitude is not going to get us far with this approach.
>
From boris at codesynthesis.com  Tue Aug 26 07:49:24 2025
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Tue Aug 26 07:49:37 2025
Subject: [odb-users] Inserting CLOB greater having size greater then 12288
In-Reply-To: <CADPyhbD=ozTWqJp=+X3AZq+q0DqEN+yTaR3bOHy=XKa+VOQq2w@mail.gmail.com>
References: <CADPyhbDwxrw+Dd7ichPa2HZaMxjtvwJ=_bGepcianGdHy5SfdQ@mail.gmail.com>
	<boris.20240115122858@codesynthesis.com>
	<CADPyhbDrw6gc2WD=w8zUbzpNODNADNJqz+DnsUwrBCFpVpthbA@mail.gmail.com>
	<CADPyhbCzvhoxAXJQPQEAzeHA3O9D9iTMnUyZwvmufMQqn7rmwA@mail.gmail.com>
	<boris.20240117150128@codesynthesis.com>
	<CADPyhbC4g+EJYEVvrq0muJNr2wppJq+yrjT+nQSSZyGWTwLy=g@mail.gmail.com>
	<CADPyhbDsbf-cCMcH+3gcG_YpoUNkHxTSGaYD7s=+6ngUoP+9Vg@mail.gmail.com>
	<CADPyhbCvw7ciSQicGm3=jp2AiaAfZtQJRVsM8eSPm1+xmCk85A@mail.gmail.com>
	<boris.20240201111915@codesynthesis.com>
	<CADPyhbD=ozTWqJp=+X3AZq+q0DqEN+yTaR3bOHy=XKa+VOQq2w@mail.gmail.com>
Message-ID: <boris.20250826134621@codesynthesis.com>

roberto minarelli <roberto.minarelli.de@gmail.com> writes:

> Do you believe that installing the odb 2.5.0 will solve the situation?

Yes, I remember us fixing something in this area for the 2.5.0 release.
So I think there is a good change upgrading to 2.5.0 will help.

From roberto.minarelli.de at gmail.com  Wed Aug 27 10:17:26 2025
From: roberto.minarelli.de at gmail.com (roberto minarelli)
Date: Thu Aug 28 04:00:46 2025
Subject: [odb-users] upgrading to odb 2.5.0
Message-ID: <CADPyhbDKdd8Kmo3UVJKgjDdGwM4Gq=s2A+_CUcNNNZS_2AMXUA@mail.gmail.com>

Hi Boris,
I am currently trying to upgrade to odb 2.5.0
But whenever I run the command
bpkg build odb@https://pkg.cppget.org/1/stable
there is this error
https://pkg.cppget.org/1/stable/packages.manifest:245971:8: error:
unexpected enable clause

I am building from source packages because there is not the odb-oracle
library in the precompiled ones, and furthermore I don't see a way to tell
the compiler I want odb to work with

I also tried the offline installation but didn't not work

Any idea of what can I do to overcome the problem?

Thank you in advance

Best Regards

Roberto Minarelli Della Valle
From boris at codesynthesis.com  Thu Aug 28 04:02:45 2025
From: boris at codesynthesis.com (Boris Kolpackov)
Date: Thu Aug 28 04:02:57 2025
Subject: [odb-users] upgrading to odb 2.5.0
In-Reply-To: <CADPyhbDKdd8Kmo3UVJKgjDdGwM4Gq=s2A+_CUcNNNZS_2AMXUA@mail.gmail.com>
References: <CADPyhbDKdd8Kmo3UVJKgjDdGwM4Gq=s2A+_CUcNNNZS_2AMXUA@mail.gmail.com>
Message-ID: <boris.20250828100128@codesynthesis.com>

roberto minarelli <roberto.minarelli.de@gmail.com> writes:

> But whenever I run the command
>
> bpkg build odb@https://pkg.cppget.org/1/stable
>
> there is this error
>
> https://pkg.cppget.org/1/stable/packages.manifest:245971:8: error: unexpected enable clause

Run `b --version`, if you are not using 0.17.0 or later, then you
will need to upgrade build2 first.

From roberto.minarelli.de at gmail.com  Thu Aug 28 16:34:19 2025
From: roberto.minarelli.de at gmail.com (roberto minarelli)
Date: Mon Sep  1 09:57:29 2025
Subject: [odb-users] upgrading to odb 2.5.0
In-Reply-To: <boris.20250828100128@codesynthesis.com>
References: <CADPyhbDKdd8Kmo3UVJKgjDdGwM4Gq=s2A+_CUcNNNZS_2AMXUA@mail.gmail.com>
	<boris.20250828100128@codesynthesis.com>
Message-ID: <CADPyhbBp_Ok+SK1W-R+gwtYUf+TVpiWXu+VduWnBCEeD7wX77Q@mail.gmail.com>

I upgraded build2 to 0.17.0
now I would like to install odb to a different dir

I tried this

mkdir odb-build
cd odb-build

bpkg create -d odb-gcc-12 cc     \
  config.cxx=g++                  \
  config.cc.coptions=-O3          \
  config.bin.rpath=$HOME/lib \
  config.install.root=$HOME/odb

cd odb-gcc-12

bpkg build --configure-only libstudxml-1.1.0.tar.gz
bpkg build --configure-only libcutl-1.11.0.tar.gz
bpkg build odb-2.5.0.tar.gz

and I did expected to found odb installed
but I didn't get nor the include files nor the odb in the directory I
specified

I tried more than one solution so what is missing?

Thank you
Regards

Roberto Minarelli Della Valle

Il giorno gio 28 ago 2025 alle ore 10:02 Boris Kolpackov <
boris@codesynthesis.com> ha scritto:

> roberto minarelli <roberto.minarelli.de@gmail.com> writes:
>
> > But whenever I run the command
> >
> > bpkg build odb@https://pkg.cppget.org/1/stable
> >
> > there is this error
> >
> > https://pkg.cppget.org/1/stable/packages.manifest:245971:8: error:
> unexpected enable clause
>
> Run `b --version`, if you are not using 0.17.0 or later, then you
> will need to upgrade build2 first.
>
From enmarantispam at gmail.com  Sun Aug 31 01:52:26 2025
From: enmarantispam at gmail.com (NIkolai Marchenko)
Date: Mon Sep  1 09:57:29 2025
Subject: [odb-users] any access to sequence without explicitly naming it?
Message-ID: <CA+CE65w5UghX-B=rEWD3C8_VkLpn2DESwO0LPsxr=TBOh+w-0Q@mail.gmail.com>

Note: the name of the initial message may be misleading and I may be in
need of an entirely different solution.

I have encountered a weird use case trying to use the new points_to
functionality that I am not sure how to deal with:

suppose an object contains another object. an inner objects points_to the
outer object with on delete cascade option and both are not persistent yet.
The objects both have auto id quint64 keys. Since they both don't have id
yet, I cannot assign an id to points_to and when I persist the _outer_
object it says that inner isn't persistent yet. but when I persist an
_inner_ object it also fails because foreign key constraint is not
satisfied. the schema is created with deferred constraint checking ... but
I they need to have matching ids beforehand.

The only way I see of breaking this loop is to access a sequence for the
outer object inisde the transaction and reserving the key to it that I can
feed into points_to... but as far as I can see odb doesn't have a native
function to access a sequence. This means I have to .execute("raw sql
here") and this is potentially breaking on refactors. Am I missing
something obvious on how to deal with those?
From enmarantispam at gmail.com  Sun Aug 31 03:01:56 2025
From: enmarantispam at gmail.com (NIkolai Marchenko)
Date: Mon Sep  1 09:57:29 2025
Subject: [odb-users] Re: any access to sequence without explicitly naming it?
In-Reply-To: <CA+CE65w5UghX-B=rEWD3C8_VkLpn2DESwO0LPsxr=TBOh+w-0Q@mail.gmail.com>
References: <CA+CE65w5UghX-B=rEWD3C8_VkLpn2DESwO0LPsxr=TBOh+w-0Q@mail.gmail.com>
Message-ID: <CA+CE65z0TUgHHRC0Q20Due7r4rR7ALSGqeR_YJ4tG=y3TbOoSA@mail.gmail.com>

It's further compounded by the fact that pre-persist callbacks are const
only. so to even query that sequence I have to write my own pre-persist
hook to call for these objects that lives completely outside odb
architecture.

On Sun, Aug 31, 2025 at 8:52?AM NIkolai Marchenko <enmarantispam@gmail.com>
wrote:

> Note: the name of the initial message may be misleading and I may be in
> need of an entirely different solution.
>
> I have encountered a weird use case trying to use the new points_to
> functionality that I am not sure how to deal with:
>
> suppose an object contains another object. an inner objects points_to the
> outer object with on delete cascade option and both are not persistent yet.
> The objects both have auto id quint64 keys. Since they both don't have id
> yet, I cannot assign an id to points_to and when I persist the _outer_
> object it says that inner isn't persistent yet. but when I persist an
> _inner_ object it also fails because foreign key constraint is not
> satisfied. the schema is created with deferred constraint checking ... but
> I they need to have matching ids beforehand.
>
> The only way I see of breaking this loop is to access a sequence for the
> outer object inisde the transaction and reserving the key to it that I can
> feed into points_to... but as far as I can see odb doesn't have a native
> function to access a sequence. This means I have to .execute("raw sql
> here") and this is potentially breaking on refactors. Am I missing
> something obvious on how to deal with those?
>