ENH: read_stata return non-nano #55642
Conversation
mroeschke
added
IO Stata
|
Does the writer code automatically handle non-ns timestamps? Or does it need some work? I give some time if needed. |
IIUC _datetime_to_stata_elapsed_vec handles this correctly, but another pair of eyeballs would be a good idea |
| if fmt.startswith(("%tc", "tc")): | ||
| # Delta ms relative to base | ||
| td = np.timedelta64(stata_epoch - unix_epoch, "ms") | ||
| res = np.array(dates._values, dtype="M8[ms]") + td |
There was a problem hiding this comment.
Not very familiar with this part of the code base, but this looks like it would overflow when dates._values are close to the minimum storage value. Does numpy catch that for us or return junk values?
There was a problem hiding this comment.
good catch. numpy does not catch this
There was a problem hiding this comment.
@bashtage am i right in thinking stata doesn't support dates millions of years in the past/future so we don't need to worry about these overflows?
There was a problem hiding this comment.
I just did a check of type tc and found that the maximum date it would represent is 31Dec9999 21:41:39
There was a problem hiding this comment.
This value is stored as 253717911699456
|
@bashtage gentle ping |
|
Will take a look later today. |
|
This pull request is stale because it has been open for thirty days with no activity. Please update and respond to this comment if you're still interested in working on this. |
jbrockmendel
force-pushed
the
enh-stata-non-nano
branch
from
ce09bf0
to
7b802c7
Compare
|
Thanks @jbrockmendel |
* ENH: read_stata return non-nano * GH ref * mypy fixup * update doctest * simplify * avoid Series.view * dont go through Series * move whatsnew * remove outdated whatsnew
doc/source/whatsnew/vX.X.X.rstfile if fixing a bug or adding a new feature.Pre-emptively addresses issues that come up when implementing #55564
cc @bashtage