# TAP ADQL of MPC designations looses string case?

**URL:** <https://www.rubin.community/t/tap-adql-of-mpc-designations-looses-string-case/12934>\
**Category:** Data Products\
**Tags:** solarsystem, dp2\
**Created:** [November 10, 2026, 2:16pm UTC](https://www.rubin.community/t/tap-adql-of-mpc-designations-looses-string-case/12934 "2026-11-10T14:16:38Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jjkavelaars](https://sea2.discourse-cdn.com/flex002/user_avatar/www.rubin.community/jjkavelaars/32/2301_2.png) [@jjkavelaars](https://www.rubin.community/u/jjkavelaars)\
**Post date:** [November 10, 2026, 2:16pm UTC](https://www.rubin.community/t/tap-adql-of-mpc-designations-looses-string-case/12934/1 "2026-11-10T14:16:38Z")

</div>

Hi,

Perhaps already known. When I run queries on the mpc\_orbits table in the dp2 schema, I get mixed-case results when using a string-based query.

select \* from dp2.mpc\_orbits AS mpc\_orbits WHERE mpc\_orbits.packed\_primary\_provisional\_designation = ‘K15KH4Q’

Returns rows matching K15Kh4Q and K15KH4Q [which are two different objects].

Is that expected for an ‘=’ query ?

JJ

---

<div class="post-metadata">

**Author:** ![ctslater](https://sea2.discourse-cdn.com/flex002/user_avatar/www.rubin.community/ctslater/32/28_2.png) [@ctslater](https://www.rubin.community/u/ctslater)\
**Post date:** [November 13, 2026, 12:51am UTC](https://www.rubin.community/t/tap-adql-of-mpc-designations-looses-string-case/12934/2 "2026-11-13T00:51:38Z")

</div>

Hi JJ,

That’s definitely a surprise to me, but evidently mariadb (which underlies qserv) defaults to a case-insensitive collation for string comparison. The qserv team is testing a configuration change to the table that will make string comparison case-sensitive, and I expect we will be able to roll that out in the next ~week.

Thanks for the report!

-Colin
