Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺

@kibeha

Danish geek 🤓 Oracle SQL & PL/SQL Developer ♠ ACE Director 4️⃣2️⃣ #SYM42 👨‍🍳 Likes to cook 🚀 Reads sci-fi 🍺 Beer Enthusiast 🏢 Cegal Danmark A/S 🧑‍💻

238 Folgt584 Follower572 Posts

Bluesky

Ist das dein Profil?Übernimm es — dann gehört dir kibeha.dirty.social.Profil übernehmen

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· 13h

The COMMIT is actually not a COMMIT, it is just a label because of missing semicolon after END LOOP. And no, the compiler does not complain. It is documented that the compiler does NOT check whether the label used in END LOOP matches the actual label of the loop (in this case an unlabeled loop.)

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· 14h

#SpotTheBug 😁

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· 10d

Thank you @oracleace.bsky.social for being appreciative of us #OracleACE ♠️ folks for another year 🙏

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· 14d

Hmm... didn't know that queries were written in PL/SQL, but Oracle docs says so, so it must be true 😇

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· 28d

"Dataveloper"... I resemble that 😇 source: jorriss.com/about/

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· 29d

Obvious, right? 😇 (Source: freefall.purrsia.com)

Freefall comic. Transcript:

Sam: Telling the computer what I want to have done. It makes my job easy. I don't know how it works. I don't know what I'm actually doing.
Sam: I've always assumed the other guys know what they're doing. If they're using A.I. as well... this could be a problem.
Sam: It makes it too obvious that almost nobody knows what they're actually doing. That destroys one of the illusions societies are built on.

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Jun 24

Yup. You’re *not* helping yourself by letting an LLM write your abstract submission for a conference! Read the blog post - the advice at the end may be a way to use an LLM that can actually be useful to you 👍 #OracleACE #OracleUGs

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Jun 10

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Jun 3

Another fiscal year gone by - a new one begins. Happily I once again have qualified to be awarded Oracle ACE Director for the new year by #OracleACE program. It's great to be part of the community and meet a lot of great people - live and online. Thanks to @oracleace.bsky.social for having me 🙏

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Jun 1

Leaving Slovenia after our first visit, @dbandycapp.bsky.social and I. *Very* positive experience, both the country and the #MakeIT2026 conference. Hoping to be back one day 😁

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 29

Picture of @zorantica.bsky.social ‘s assistant that helped him in his “I know what You did last Summer… and Winter… and Autumn… and Spring” session on auditing at #MakeIT2026 🐈

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 27

Registered at #SIOUG #MakeIT2026 Conference early morning buzz is starting. We’re gonna have so much fun here, @dbandycapp.bsky.social and I 😁

DBAndyCapp and myself at the registration desk of MakeIT 2026 conference

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 26

My first visit to Slovenia, I'm excited to be a speaker at SIOUG MakeIT 2026! I'm on twice tomorrow: - Visualizing PL/SQL with PL/Scope, 26ai Property Graph and APEX Graph Plug-in - 3-hour Hands-On-Lab on analytic functions. Check it out: makeit.si #MakeIT2026 #OracleDB #OracleAPEX #OracleACE

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 18

Is @connormcd.bsky.social really comfortable with a vibe coded shirt? How about a vibe coded database design? We’re about to find out at #apexconn26

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 17

The @oracleace.bsky.social #OracleACE Adventure is starting at #apexconn26

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 17

Very organized in Germany. Even male discharge has to be sorted, so of course I used 🍺 😇

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· May 11

So who's taking over after us when history repeats? 😇

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Apr 30

#POUG2026, here I come \o/ So awesome that @pougorg.bsky.social accepted my session "Shape-Changing SQL" and put it on the agenda: poug.org/edycja/poug2... See you in Krakow in September. Be sure to get your tickets while they last: poug2026.konfeo.com/en/groups </happydance>

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Apr 30

💯

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Apr 22

💯 percent truth! You >must< learn #SQL

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Apr 13

Well, half fixed 😄

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Apr 10

Strange, a couple weeks ago a colleague tried in a 19.29 to use this backported 'system:' wallet path and got "ORA-29248: an unrecognized WRL was used to open a wallet". Have you gotten it to work in 19? @martindba.bsky.social : What could cause the ORA-29248?

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Apr 6

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Mar 25

Welcome our new overlords 🤔

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Mar 24

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Mar 6

These work fine too. I'd hazard the guess that they can be checked and evaluated more efficiently. Also it has the advantage of error messages specifying which assertion is violated.

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Mar 6

Actually, this one works in a single assertion, turning it into NOT EXISTS with AND instead of EXISTS with OR. But one assertion per parent table might be able to evaluate the checks faster and check only when needed. Whether that'd outweigh overhead of many assertions - who knows 😉

create assertion Notes_RefRecId_MultiFK check (
not exists (
select null
from Notes n
where not exists (
select null
from DebTab d
where n.RefFileId = 11
and d.RecId = n.RefRecId
) and not exists (
select null
from CreTab c
where n.RefFileId = 15
and c.RecId = n.RefRecId
) and not exists (
select null
from StoTab s
where n.RefFileId = 27
and s.RecId = n.RefRecId
)
)
);

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Mar 6

So I got my conditional/subset foreign key working. But multi-parent foreign key I need to rethink how to phrase. So far my attempts have raised: ORA-08663: Set operators are not supported. ORA-08697: SYS owned tables are not supported. ORA-08695: A subquery within an OR condition is not supported.

create assertion Notes_RefRecId_MultiFK check (
all (
select n.RefFileId, n.RefRecId
from Notes n
) nref
satisfy (
exists (
select null
from DebTab d
where nref.RefFileId = 11
and d.RecId = nref.RefRecId
UNION ALL
select null
from CreTab c
where nref.RefFileId = 15
and c.RecId = nref.RefRecId
UNION ALL
select null
from StoTab s
where nref.RefFileId = 27
and s.RecId = nref.RefRecId
)
)
);

create assertion Notes_RefRecId_MultiFK check (
all (
select n.RefFileId, n.RefRecId
from Notes n
) nref
satisfy (
exists (
select null
where exists (
select null
from DebTab d
where nref.RefFileId = 11
and d.RecId = nref.RefRecId
) OR exists (
select null
from CreTab c
where nref.RefFileId = 15
and c.RecId = nref.RefRecId
) OR exists (
select null
from StoTab s
where nref.RefFileId = 27
and s.RecId = nref.RefRecId
)
)
)
);

create assertion Notes_RefRecId_MultiFK check (
all (
select n.RefFileId, n.RefRecId
from Notes n
) nref
satisfy (
exists (
select null
from dict d
where d.FileId = nref.RefFileId
and d.ColId = 0
and (
(
nref.RefFileId = 11
and exists (
select null
from DebTab d
where d.RecId = nref.RefRecId
)
) or (
nref.RefFileId = 15
and exists (
select null
from CreTab c
where c.RecId = nref.RefRecId
)
) or (
nref.RefFileId = 27
and exists (
select null
from StoTab s
where s.RecId = nref.RefRecId
)
)
)
)
)
);

Kim Berg Hansen: SQL ♠ 🤓 👨‍🍳 🍺@kibeha· Mar 6

I tried and got error: ORA-08689: CREATE ASSERTION failed ORA-08673: Equijoin "D"."FILEID"="N"."REFFILEID" found does not meet the criteria to do a FAST validation.