Import, stuck when reading some metadata.ods files

Depending on the metadata.ods file lines/columns the ReadODS.toTable() method takes a very long time to read the file.

Hypothesis

The odftoolkit (https://odftoolkit.org/) java library takes a lot of time to read virtual? lines or columns.

Solution 0

Add a console message at the beginning [1] and at the end [2] of reading ODS files to help user diagnoze abnormally long reading time of a metadata.ods file:

[1 Reading metadata.ods...] [2 Done.]

Solution

Fix the reading problem

(from redmine: issue id 3455, created on 2023/10/13 by Matthieu Decorde)

• Relations:
  • relates #3452 (closed)

added import type::bug + 1 deleted label

changed milestone to %TXM 0.8.4

added status::feedback label

Cells and Rows repetitions were not well managed by the lib

mentioned in issue #3452 (closed)

0.8.4 (2024-10-29 10h41) Ubuntu 20.04

OK ? : J'ai refait depuis le départ (re-création du corpus par import depuis les sources originales) la situation qui avait manifesté le bug, cf. #3452 (closed) Je ne rencontre plus le bug. Cela dit, comme le processus est long (import, export, chargement, annotation, sauvegarde de l'annotation), peut-être que la situation n'est pas exactement la même qu'il y a deux ans.

Pas sur que le fichier que tu as utilisé contienne toutes les lignes virtuelles qui provoquaient le bug.

Après il y a un test automatisé qui refait le test à chq build de TXM donc ca doit être ok ;)

closed

removed status::à tester label

reopened

When Excel read&write ODS files it creates the virtual lines with the maximum row and column capacity of librefoffice :

• all lines ends with 19k empty cells
• all sheets ends with 1M empty rows

changed milestone to %TXM 0.8.4.1

changed milestone to %TXM 0.8.4

marked this issue as related to #4195 (closed)

closed