On time series, when the first table is not valid, the linelist construction crashes.

[yves-amevoin]

No description provided.

[yves-amevoin]

A new section should be also a valid table. Meaning a new section could be everywhere. The simplest way to deal with the issue is to work on it at import level either:

• By removing non valid tables from the time series analysis
• Or By adding way to identify valid new sections on time series analysis.

This is a thornier question and need more development.