https://bugs.documentfoundation.org/show_bug.cgi?id=163872

--- Comment #5 from Justin L <[email protected]> ---
(In reply to raal from comment #0)
> 1.open file from bug 163865
That would be
ca7e9fa4-7667-4af3-a64d-8d1db535dd62-Ukraine_Support_Tracker_Release_19.xlsx
(attachment 197568)

This has 82 sheets, mostly with a few charts. The biggest sheet (Bilateral
Assistance, MAIN DATA) has approximately 4000 rows of data.

I confirmed that setting the global config to NOT recalculate row height on
file-open lets this document load. So that at least is a workaround available
with 24.8. See bug 124098.

For me, it failed to load the worksheet - I stopped waiting after 30 minutes. I
expect it has triggered either an error or a terribly unoptimized path. I see
at least one path through MaybeInterpret() which loops through 1 million rows
in ScIfJump.

It seems to be entering some kind of loop on nTab 45, column 2 in the first 54
rows. That is tab A.19 - which in LO 7.5 takes a minute or two to switch to. At
some point it runs HandleRefArrayForParallelism which returns bAllowThreading
and all the time is spend in here as it methodically chews through each row in
tab 75 (Refugees recorded) column J one by one until 1647.

So we must have terrible performance evaluating 
=INDEX($E:$E,MATCH($J$10,IF($B:$B=$B1648,$I:$I),1))

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to