issue_comments: 687648489

This data as json

html_url issue_url id node_id user created_at updated_at author_association body reactions performed_via_github_app issue
https://github.com/pydata/xarray/issues/4401#issuecomment-687648489 https://api.github.com/repos/pydata/xarray/issues/4401 687648489 MDEyOklzc3VlQ29tbWVudDY4NzY0ODQ4OQ== 6628425 2020-09-05T18:56:32Z 2020-09-05T18:56:32Z MEMBER

I looked into the nc_time_axis issue a bit more. It appears it is a label formatting problem. The ticks actually make sense, but are labeled in a confusing way. If I force the date format pattern to be "%Y-%m-%d" in nc_time_axis -- note there's no way to do this without modifying the source code -- I get a plot that looks like this:

We can see the ticks are really 0000-12-01, 0001-10-01, 0002-08-01, and 0003-06-01. The main problem, I think, is that nc_time_axis is choosing a monthly tick resolution, but using a format that would only make sense if the ticks had an annual resolution. So it should be picking a label format of "%Y-%m" instead of "%Y". In general it looks like there should be more work in nc_time_axis to make the logic that chooses the tick resolution and label format consistent.

I opened an issue in nc_time_axis to track this: https://github.com/SciTools/nc-time-axis/issues/48.

 
{
    "total_count": 2,
    "+1": 2,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   690624634