Some Viz Calculations Being Dropped

[jdunkerley]

Reading a 1mm row Excel file, the visualization is not always rendered.
Interrupted exception also shows up (but thats another issue).

from Standard.Base import all
from Standard.Table import all
import Standard.Visualization

main =
    node1 = Data.read 'C:\\Users\\jdunk\\Downloads\\1m.xlsx' ..Sheet
    integer1 = node1.row_count
    table1 = node1.column_info

https://docs.google.com/spreadsheets/d/1hGx9W7qeIUvBvJMlR-a2fy6k_TtRFh_f/edit?usp=sharing&ouid=105802918667196204721&rtpof=true&sd=true

Logs:
enso-project-manager-2024-12-16.0.zip

[hubertp]

I can't seem to reproduce that problem. I would need to see it live, I'm afraid. Also, the exchanges between GUI and LS would be much appreciated.

[hubertp]

I think I got it now and also got a bit more logs.
I believe what happens is:

1. Computation gets interrupted and visualizations get the missing method on exception, as reported in InterruptException/ClosedByInterruptException are still showing up in long running computations #11896
2. That value becomes a bit sticky and somehow expression doesn't get updated
3. Eventually, visualization will get calculated (not really dropped despite the title of this ticket) but it may take a reaaaaaly long time until that happens due to numerous interrupts

[enso-bot]

Hubert Plociniczak reports a new STANDUP for the provided date (2024-12-17):

Progress: Investigating vis issues, can't seem to reproduce the main problem but created #11896 for a related issue. Continued doing small tweaks to native image - needs testing with electron. It should be finished by 2024-12-19.

Next Day: Next day I will be working on the #11896 task. Continue investigating the issue

[hubertp]

Just to prove my point, I can see execution of visualizations being done once

[org.apache.poi.openxml4j.opc.OPCPackage] The close() method is intended to SAVE a package. This package is open in READ ONLY mode, use the revert() method instead!

is being reported in the logs.

Note that:
a) in IDE it seems that table visualization of the read Excel file is already being shown for a while so it seems as if the computation hung
b) nothing is being reported in the logs (apart from usual ping/pong) for over 2 minutes and suddenly boom, The close() method.... method appears and execution of the visualization is finished.