You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I just learned that (some) databases seem to (silently?) change invalid column names (see tidyverse/dbplyr#1016). This causes issues with the new optimisation in dbplyr::copy_to() where column names do not match.
To handle this correctly it would be necessary for copy_to() to revert to the old behaviour and query the column names again.
To avoid this extra roundtrip with the database it would be great if dbWriteTable() would warn (or even error) if the output column names are different from the input column names. @krlmlr What do you think?
The text was updated successfully, but these errors were encountered:
There is no possibility to directly get back the column names when creating the table, is there?
dbplyr queries them via SELECT * FROM <tbl> WHERE 0 = 1 but of course it would be nice if we could avoid this extra query to the database. Maybe don't query the names if they are very simple (ASCII + lowercase)?
I just learned that (some) databases seem to (silently?) change invalid column names (see tidyverse/dbplyr#1016). This causes issues with the new optimisation in
dbplyr::copy_to()
where column names do not match.To handle this correctly it would be necessary for
copy_to()
to revert to the old behaviour and query the column names again.To avoid this extra roundtrip with the database it would be great if
dbWriteTable()
would warn (or even error) if the output column names are different from the input column names.@krlmlr What do you think?
The text was updated successfully, but these errors were encountered: