Repository navigation
rbindlist could use an "ignore.attributes" argument #3911
Description
Activity
Since data.table 1.13.0 and fread reading dates as IDate I come across this issue all the time.
rbindlist(list(data.table(x=1, y = Sys.Date()), data.table(x=1, y = as.IDate(Sys.Date())))) Error in rbindlist(list(data.table(x = 1, y = Sys.Date()), data.table(x = 1, : Class attribute on column 2 of item 2 does not match with column 2 of item 1.I think rbindlist should indeed ignore attributes in cases like this
Reacted by Rose Martinez, Peter Kress, Jonathan Rosenblatt, Shape, py9mrg and Daniel JostI just wanted to bump this issue as well - it's a real pain to have to specify
dt[, date.col:=as.Date(date.col)]before binding data.tables together.Reacted by Anand Mayakonda, therosko, Mr. Steve Charlesworth, vintro, Ashleych and dandnguyenI can only agree with this. Seems like a completely unnecessary step to add to a process, if the attributes could have just been ignored.
I m working on this issue, will raise a PR soon.
Binding
integer64with integer/numeric works silently and correctly, whereas bindingIDatewithDatefails and so does binding of attribute cols where one/more tables are empty (0-rows).integer 64 case:
The 2nd element's col
bseems to be implicitly converted tointeger64type without a message/warning.IDate case
First element's
bcolumn isIDateand 2nd isDateand there's an error.0-row data.table case:
Even if the empty table has a wrong type, it could be skipped as it doesn't really make it to the final table. (This used to work in previous versions when
fill=TRUE-- probably accidental).In essence, I'd think it'd be immensely convenient to have an argument to ignore strict attributes check, e.g.,
ignore.attr=FALSE(together with a global option so that code that has been written with old+lenient behaviour work continue to work) which would simply retain the first element's class.And if the default is to be strict, then, the first case should also error, for consistency.