By default, tablesorter text parser is doing a case insensitive compare. So the result of what you see, can be explained by the initial position of the characters in the table.

@Iodel: This issue has been assigned to you a year ago.
Could you please share a status update? Are you still working (or still plan to work) on this issue? Is there anything that others could help with? If you do not plan to work on this issue anymore, please remove yourself as assignee (via Add Action... → Assign / Claim in the dropdown menu) so others could work on it. Thanks a lot!

We use and have set Unicode collation for Categories (uca-cs for Czech Wikipedia). This is already configured for majority of Wikipedias. Why don't we use the same here?

In T47070#517583, @TheDJ wrote:

By default, tablesorter text parser is doing a case insensitive compare. So the result of what you see, can be explained by the initial position of the characters in the table.

Yes, but it produces not a good result, for example this is table-sorted:
á
A
a
b
B
C
c
Č
č
ď
d
Ď
It is a complete mess

Yes i know.

Right I remember now. This is of course because otherwise a, sorts after capital Z, which is even less desirable. Therefor, this is dependent on localeCompare, like T32674: Use sort collation config in JavaScript (jquery.tablesorter)

I see. But I doubt there is not any external library, that could handle this

Change 517266 had a related patch set uploaded (by TheDJ; owner: TheDJ):
[mediawiki/core@master] Tablesorter: Use localeCompare

https://gerrit.wikimedia.org/r/517266

Change 517266 merged by jenkins-bot:
[mediawiki/core@master] Tablesorter: Use localeCompare

https://gerrit.wikimedia.org/r/517266

In this line :

if ( upper === match && !lower === match ) {

I don't understand the !lower === match part. Because of the logical not !lower is forcibly a boolean, while match is a string. So the conditional is always false, isn't it?