SHIMA Tatsuya created ARROW-17414:
-------------------------------------
Summary: [R]: Lack of `assume_timezone` binding
Key: ARROW-17414
URL: https://issues.apache.org/jira/browse/ARROW-17414
Project: Apache Arrow
Issue Type: Improvement
Components: R
Affects Versions: 9.0.0
Reporter: SHIMA Tatsuya
If we run the following code in R, we will get a C++ derived error message
telling us to use {{assume_timezone}}.
However, this error message is not helpful because there is no binding for the
{{assume_timezone}} function in R.
{code:r}
tf <- tempfile()
writeLines("2004-04-01 12:00", tf)
arrow::read_csv_arrow(tf, schema = arrow::schema(col1 = arrow::timestamp("s",
"UTC")))
#> Error:
#> ! Invalid: In CSV column #0: CSV conversion error to timestamp[s, tz=UTC]:
expected a zone offset in '2004-04-01 12:00'. If these timestamps are in local
time, parse them as timestamps without timezone, then call assume_timezone.
#> ℹ If you have supplied a schema and your data contains a header row, you
should supply the argument `skip = 1` to prevent the header being read in as
data.
{code}
It would be useful to improve the error message or to allow {{assume_timezone}}
to be used from R as well.
(although {{lubridate::with_tz()}} and {{lubridate::force_tz()}} could be more
useful within a dplyr query)
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
