Focus bug in sequences

[Akron]

It seems that sometimes focus queries are not crrectly embedded in sequences. I was sure we already had and fixed that issue, but sometimes it reappears. E.g. in the NKJP1M-SGJP instance, the following queries are not identical, although they should be:

contains(<base/s=s>, (focus({[nkjp/l="da(wa)?ć"][]{,5} do} [nkjp/p=subst]) zrozumienia))
contains(<base/s=s>, {[nkjp/l="da(wa)?ć"][]{,5} do} [nkjp/p=subst & orth=zrozumienia])

[margaretha]

Query Number	Query	Number of Results
1	contains(<base/s=s>, (focus({[nkjp/l="da(wa)?ć"][]{,5} do} [nkjp/p=subst]) zrozumienia))	6
2	contains(<base/s=s>, {[nkjp/l="da(wa)?ć"][]{,5} do} [nkjp/p=subst & orth=zrozumienia])	8

The 2 missing hits are in the following docs:

• NKJP/NKJP/200-4-000000301
• NKJP/NKJP/130-5-000001880

When adding the docs as a virtual corpus to Query 1, the 2 hits are found. So it seems that there is a problem with skipping documents.

[margaretha]

The NKJP instance uses Krill version 0.59.5. It seems that the bug has been fixed in a newer version of Krill, probably in f420cb341ead7e83005f53fbf81fb196e0f30aae.

[Akron]

You are right. Now that Java 11 is available everywhere, I was able to update the instance! Thank you!