Dear Maintainer,

there is one file on my computer that crashes localsearch-ext. That's
ok, it is unavoidable, especially with plugins to handle many file
types. The problem is that when there is such a crash, instead of
marking the file as problematic somehow, localsearch keeps retrying,
every second, all day long, without any limit. The output of journalctl
is flooded with

sept. 16 21:32:02 abc localsearch-3[5221]: Extractor subprocess died unexpectedly: Child process killed by signal 11
sept. 16 21:32:03 abc systemd-coredump[732120]: [🡕] Process 732112 (localsearch-ext) of user 59682 terminated abnormally without generating a coredump.
░░ Subject: Process 732112 (localsearch-ext) dumped core
░░ Defined-By: systemd
░░ Support: https://www.debian.org/support
░░ Documentation: man:core(5)
░░
░░ Process 732112 (localsearch-ext) crashed and dumped core.
░░
░░ This usually indicates a programming error in the crashing program and
░░ should be reported to its vendor as a bug.
sept. 16 21:32:03 abc localsearch-3[5221]: Extractor subprocess died unexpectedly: Child process killed by signal 11
sept. 16 21:32:03 abc systemd[4462]: drkonqi-coredump-launcher.socket: Too many incoming connections (16), dropping connection.
sept. 16 21:32:04 abc systemd-coredump[732137]: [🡕] Process 732130 (localsearch-ext) of user 59682 terminated abnormally without generating a coredump.

Or in dmesg (which seems to have a clever limiter on repetitive messages)
[52293.643987] traps: localsearch-ext[751238] general protection fault ip:7fd241187419 sp:7ffc702b4208 error:0 in libc.so.6[16a419,7fd241045000+168000]


I am going to disable localsearch, but I hope someone can configure it
to be less eager...

(the problematic file seems to be icu_conversion_data.c.gz.abw from
tensorflow 2.21, but that's not what this report is about)