issue_comments

Table actions

5 rows where author_association = "MEMBER" and issue = 203999231 sorted by updated_at descending

✎ View and edit SQL

This data as json, CSV (advanced)

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

user 3

issue 1

  • `set_index` converts string-dtype to object-dtype · 5 ✖

author_association 1

  • MEMBER · 5 ✖
id html_url issue_url node_id user created_at updated_at ▲ author_association body reactions performed_via_github_app issue
751254357 https://github.com/pydata/xarray/issues/1238#issuecomment-751254357 https://api.github.com/repos/pydata/xarray/issues/1238 MDEyOklzc3VlQ29tbWVudDc1MTI1NDM1Nw== mathause 10194086 2020-12-25T14:07:05Z 2020-12-25T14:07:05Z MEMBER

still relevant

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   `set_index` converts string-dtype to object-dtype 203999231
457220713 https://github.com/pydata/xarray/issues/1238#issuecomment-457220713 https://api.github.com/repos/pydata/xarray/issues/1238 MDEyOklzc3VlQ29tbWVudDQ1NzIyMDcxMw== shoyer 1217238 2019-01-24T14:42:21Z 2019-01-24T14:42:21Z MEMBER

I think this will eventually get fixed "for free" as part of the explicit indexes refactor.

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   `set_index` converts string-dtype to object-dtype 203999231
457207288 https://github.com/pydata/xarray/issues/1238#issuecomment-457207288 https://api.github.com/repos/pydata/xarray/issues/1238 MDEyOklzc3VlQ29tbWVudDQ1NzIwNzI4OA== max-sixty 5635139 2019-01-24T14:03:22Z 2019-01-24T14:03:22Z MEMBER

Thanks @gerritholl

More broadly: is this reasonable to fix, given how much we rely on pandas' indexing, given that pandas converts string arrays to object dtype?

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   `set_index` converts string-dtype to object-dtype 203999231
276142223 https://github.com/pydata/xarray/issues/1238#issuecomment-276142223 https://api.github.com/repos/pydata/xarray/issues/1238 MDEyOklzc3VlQ29tbWVudDI3NjE0MjIyMw== shoyer 1217238 2017-01-30T18:09:14Z 2017-01-30T18:09:14Z MEMBER

No, pandas uses dtype=object to represent strings. We paper over this in some cases in xarray because it's convenient to get objects with the original dtype back from array.data.

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   `set_index` converts string-dtype to object-dtype 203999231
276138760 https://github.com/pydata/xarray/issues/1238#issuecomment-276138760 https://api.github.com/repos/pydata/xarray/issues/1238 MDEyOklzc3VlQ29tbWVudDI3NjEzODc2MA== shoyer 1217238 2017-01-30T17:59:29Z 2017-01-30T17:59:29Z MEMBER

PandasIndexAdapter contains some logic for preserving dtypes but it looks like didn't hook that up to set_index machinery.

 
{
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
}
   `set_index` converts string-dtype to object-dtype 203999231

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]);