MIME-Version: 1.0
References: <CACxu=vJaKFNsYxooSnW1wEgsAO5u_v1XYBacfVJ14wgJV_PYeg@mail.gmail.com>
 <1342498.1729444411@sss.pgh.pa.us> <CACxu=vLXvpzN4X3k+9jsMt6ujuOvFVUSkA80t_cROSsF4y2jQQ@mail.gmail.com>
 <1445998.1729482404@sss.pgh.pa.us> <CACxu=vKEF8Qa-OaADFxf0uMg-xw6gH_CNCWd2s+xaqh-gY4=xg@mail.gmail.com>
 <2062830.1729625620@sss.pgh.pa.us> <2265411.1729699470@sss.pgh.pa.us>
 <CACxu=v++HNmss59yGUDkRny7g=M8tZ2YXF07AUXqKVGqcSfxGQ@mail.gmail.com>
 <2354718.1729737539@sss.pgh.pa.us> <2581216.1729794746@sss.pgh.pa.us>
 <CACxu=v+dn37zr8gx5xNP-EZY3OLtGLTHrbx_ZkCQc40HpyMLKA@mail.gmail.com>
 <1948345.1730500073@sss.pgh.pa.us> <CACxu=vL7i_U_iSNpREe8eCAMEyKHuPpk9THRpBhk+ar0U1EdOw@mail.gmail.com>
 <CACxu=vKLc6f5N8_DR58LKkE1eohWSxTvThTeGsLm7p7QH1aFBA@mail.gmail.com>
 <CACxu=vJf2S=ysun_h=zmYNu6oUM47+egbpX5mMC0X9BJK=EQwQ@mail.gmail.com> <CACxu=vK+S6BXN8ZYyBvqQBWrcwHXqtue1-ZuKO3+XtHGBYcDUQ@mail.gmail.com>
In-Reply-To: <CACxu=vK+S6BXN8ZYyBvqQBWrcwHXqtue1-ZuKO3+XtHGBYcDUQ@mail.gmail.com>
From: Pavel Stehule <pavel.stehule@gmail.com>
Date: Tue, 19 Nov 2024 19:44:49 +0100
Message-ID: <CAFj8pRCd6xcH-AYEyHFdGdU89O9JjZ-v-pyQnOwd9zNJkCEdhQ@mail.gmail.com>
Subject: Re: Using Expanded Objects other than Arrays from plpgsql
To: Michel Pelletier <pelletier.michel@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@lists.postgresql.org
Content-Type: multipart/alternative; boundary="000000000000b867b20627486fcf"
Archived-At: <https://www.postgresql.org/message-id/CAFj8pRCd6xcH-AYEyHFdGdU89O9JjZ-v-pyQnOwd9zNJkCEdhQ%40mail.gmail.com>
Precedence: bulk

--000000000000b867b20627486fcf
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

=C3=BAt 19. 11. 2024 v 18:51 odes=C3=ADlatel Michel Pelletier <
pelletier.michel@gmail.com> napsal:

>
>> So going back on the assumption I'm somehow not returning the right
>> pointer, but digging into the array code I'm pretty sure I'm following t=
he
>> same pattern and tracing my code path is calling EOHPGetRWDatum when I
>> return the object, I can't get it to stop on DeleteExpandedObject, so I
>> don't think plpgsql is deleting it, I'm going to keep investigating, sor=
ry
>> for the noise.
>>
>
> A couple years ago I tried to compress what I learned about expanded
> objects into a dummy extension that just provides the necessary
> boilerplate.  It wasn't great but a start:
>
> https://github.com/michelp/pgexpanded
>
> Pavel Stehule indicated this might be a good example to put into contrib:
>
>
> https://www.postgresql.org/message-id/CAFj8pRDnAnsm8pMosEys-TDRZpnCx1Kaxe=
PjV8HT6A1JPQtsCw@mail.gmail.com
>
> I've updated the repo to include points from our current discussion wrt
> multiple expansions in a plpgsql function, and added some test functions.
> The new pgexpanded type just keeps track of the number of times it's been
> expanded starting from an initialized value.  While it only stores a
> flattened integer, it follows the typical pattern of pallocing the value
> and pfreeing it with a memory context callback, so it should be covering
> most of the boilerplate needed to start a new expanded type.
>
> This would also be a good place to showcase and test the support function
> feature you've described to me.
>
> It occurs to me that including this example in contrib would be an easier
> way for us to collaborate on my existing issue instead of punting back an=
d
> forth on the list and would benefit all future expanded object developers=
.
> Do you think that's a good idea?  If this were in contrib you could acces=
s
> the code I'm working with directly and I can just follow along in my
> project.
>

another position can be src/test/modules - I think so your example is
"similar" to plsample

Regards

Pavel


>
> -Michel
>

--000000000000b867b20627486fcf
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><br></div><br><div class=3D"gmail_quote">=
<div dir=3D"ltr" class=3D"gmail_attr">=C3=BAt 19. 11. 2024 v=C2=A018:51 ode=
s=C3=ADlatel Michel Pelletier &lt;<a href=3D"mailto:pelletier.michel@gmail.=
com">pelletier.michel@gmail.com</a>&gt; napsal:<br></div><blockquote class=
=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg=
b(204,204,204);padding-left:1ex"><div dir=3D"ltr"><div class=3D"gmail_quote=
"><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;borde=
r-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr"><div d=
ir=3D"ltr"><div><br></div><div>So going back on the assumption I&#39;m some=
how not returning the right pointer, but digging into the array code I&#39;=
m pretty sure I&#39;m following the same pattern and tracing my code path i=
s calling EOHPGetRWDatum when I return the object, I can&#39;t get it to st=
op on DeleteExpandedObject, so I don&#39;t think plpgsql is deleting it, I&=
#39;m going to keep investigating, sorry for the noise.</div></div></div></=
blockquote><div><br></div>A couple years ago I tried to compress what I lea=
rned about expanded objects into a dummy extension that just provides the n=
ecessary boilerplate.=C2=A0 It wasn&#39;t great but a start:<div><br></div>=
<div><a href=3D"https://github.com/michelp/pgexpanded" target=3D"_blank">ht=
tps://github.com/michelp/pgexpanded</a></div><div><br></div><div>Pavel Steh=
ule indicated this might be a good example to put into contrib:</div><div><=
br></div><div><a href=3D"https://www.postgresql.org/message-id/CAFj8pRDnAns=
m8pMosEys-TDRZpnCx1KaxePjV8HT6A1JPQtsCw@mail.gmail.com" target=3D"_blank">h=
ttps://www.postgresql.org/message-id/CAFj8pRDnAnsm8pMosEys-TDRZpnCx1KaxePjV=
8HT6A1JPQtsCw@mail.gmail.com</a></div><br><div>I&#39;ve updated the repo to=
 include points from our current discussion wrt multiple expansions in a pl=
pgsql function, and added some test functions.=C2=A0 The new pgexpanded typ=
e just keeps track of the number of times it&#39;s been expanded starting f=
rom an initialized value.=C2=A0 While it only stores a flattened integer, i=
t follows the typical pattern of pallocing the value and pfreeing it with a=
 memory context callback, so it should be covering most of the boilerplate =
needed to start a new expanded type.</div><div><br></div><div>This would al=
so be a good place to showcase and test the support function feature you=
9;ve described to me.</div><div><br></div><div>It occurs to me that includi=
ng this example in contrib would be an easier way for us to collaborate on =
my existing issue instead of punting back and forth on the list and would b=
enefit all future expanded object developers.=C2=A0 Do you think that&#39;s=
 a good idea?=C2=A0 If this were in contrib you could access the code I&#39=
;m working with directly and I can just follow along in my project.</div></=
div></div></blockquote><div><br></div><div>another position can be src/test=
/modules - I think so your example is &quot;similar&quot; to plsample</div>=
<div><br></div><div>Regards</div><div><br></div><div>Pavel</div><div>=C2=A0=
<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8=
ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr=
"><div class=3D"gmail_quote"><div><br></div><div>-Michel</div></div></div>
</blockquote></div></div>

--000000000000b867b20627486fcf--