issue_comments

Table actions

3 rows where author_association = "MEMBER" and issue = 293345254 sorted by updated_at descending

✎ View and edit SQL

This data as json, CSV (advanced)

Suggested facets: created_at (date), updated_at (date)

user 2

issue 1

  • roll doesn't handle periodic boundary conditions well · 3 ✖

author_association 1

  • MEMBER · 3 ✖
id html_url issue_url node_id user created_at updated_at ▲ author_association body reactions performed_via_github_app issue
411222843 https://github.com/pydata/xarray/issues/1875#issuecomment-411222843 https://api.github.com/repos/pydata/xarray/issues/1875 MDEyOklzc3VlQ29tbWVudDQxMTIyMjg0Mw== shoyer 1217238 2018-08-07T22:25:16Z 2018-08-07T22:25:16Z MEMBER

Yes, it would be nice to change this to behavior (1) above.

There are a few options here:

  • Just change how roll works (as a breaking change) in xarray v0.11, with the argument that the current behavior is broken/useless. This is easiest if we aren't concerned about backwards compatibility, but unfortunately I'm pretty sure at least some people rely on this, e.g., based this StackOverflow post.
  • Deprecate the current behavior, e.g., by inserting a new argument coords=None which by default shifts coordinates (equivalent to coords=True) and raises a warning. In a future version of xarray, switch the default to coords=False.
  • Implement the desired behavior in a new function (roll_axis?) and deprecate roll. This is the preferred option for mature projects like NumPy.

I am loathe to loose the name roll so this second option is probably the best choice. Any interest in putting together a pull request?

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   roll doesn't handle periodic boundary conditions well 293345254
377655780 https://github.com/pydata/xarray/issues/1875#issuecomment-377655780 https://api.github.com/repos/pydata/xarray/issues/1875 MDEyOklzc3VlQ29tbWVudDM3NzY1NTc4MA== shoyer 1217238 2018-03-31T01:22:02Z 2018-03-31T01:22:02Z MEMBER

@mathause @rabernat any opinions here? My sense is that the current version of roll() is not very useful for anyone, but I'd like to check before we change it unilaterally...

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   roll doesn't handle periodic boundary conditions well 293345254
362104114 https://github.com/pydata/xarray/issues/1875#issuecomment-362104114 https://api.github.com/repos/pydata/xarray/issues/1875 MDEyOklzc3VlQ29tbWVudDM2MjEwNDExNA== fujiisoup 6815844 2018-01-31T23:19:27Z 2018-01-31T23:19:27Z MEMBER

I like the option 1. I agree that the circular coordinate is used in many physics fields (actually I'm not from geoscience and I also sometimes use the circular coordinate), but one confusing sene I can think of is when using it in radian, (0, 0.5 pi, pi ...). Manual overwite would be more explicit

Another option could be to add a circular cooredinate class after #1603 is implemented?

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   roll doesn't handle periodic boundary conditions well 293345254

Advanced export

JSON shape: default, array, newline-delimited, object

CSV options: 

CREATE TABLE [issue_comments] (
   [html_url] TEXT,
   [issue_url] TEXT,
   [id] INTEGER PRIMARY KEY,
   [node_id] TEXT,
   [user] INTEGER REFERENCES [users]([id]),
   [created_at] TEXT,
   [updated_at] TEXT,
   [author_association] TEXT,
   [body] TEXT,
   [reactions] TEXT,
   [performed_via_github_app] TEXT,
   [issue] INTEGER REFERENCES [issues]([id])
);
CREATE INDEX [idx_issue_comments_issue]
    ON [issue_comments] ([issue]);
CREATE INDEX [idx_issue_comments_user]
    ON [issue_comments] ([user]);