new installation - subsetting behavr table failing

[estebanbeck]

Hi,

I installed a new computer with ubuntu 22.04 and install R and Rstudio. I am using Rstudio 2022.06.0 build 452

I can get the data and everything looks good. For instance, this kind of code runs ok:

ggetho(dt, aes(y=x_rel))+
  stat_pop_etho() +
  stat_ld_annotations()

but if I subset the database like this:

ggetho(dt[t>4700 & t<7100], aes(y=x_rel))+
  stat_pop_etho() +
  stat_ld_annotations()

I get this error:

Error in seq.default(from = first_d, to = right, by = period) :
  wrong sign in 'by' argument

The same happens if I subset by a metavariable employing xmv().

Any idea of which is the problem? I guess that I got new versions of everything and something is failing...

[qgeissmann]

Hello old friend, Thanks for reporting.

• can you try to break it down to smaller parts? E.g. what happens when you run just dt[t>4700 & t<7100], or just ggetho(dt[t>4700 & t<7100], aes(y=x_rel)),...
• Can you reproduce that with some toy data dt <- toy_dam_data()?
• Can you check you have data in dt[t>4700 & t<7100]
• If dt[t>4700 & t<7100] fail on its own, what happens with dt[between(t, 4700, 7100)]

[estebanbeck]

Hello!

I can do dt[t>4700 & t<7100] or even dt[xmv(fly)>10], that works. And I do have data in the time window (I can see it when I don't subset).

The problem comes with the line stat_ld_annotations(). This bit works when I don't have any subsetting, but as soon as I add the subset it fails. The same happens with dt[between(t, 4700, 7100)]. The error is the one I mentioned originally.

However, it looks like I can't reproduce the error with the toy data I got from zenodo...