All,
Now that 1.5.2 is on CRAN I've made this change which may introduce
incompatibilities for existing code.
This is in 1.5.3 on R-Forge to be tried out for a while, discussed, and
rolled back or done differently if need be. Please let us know if this
causes any difficulties.
Tests 104, 144 and 229 have been changed accordingly.
A side effect of this change is that it should be faster for two
reasons: i) the grp columns were redundant in .SD (repeated values) so
taking them out leaves less for dogroups to do and ii) less need to take
a subset of columns of .SD merely to remove the grp columns to do
something on the others (code that does that may be up to twice as fast
when the subset of .SD is removed).
>From NEWS :
o .SD no longer includes 'by' columns, FR#978. This resolves
the long standing annoyance of duplicated 'by' columns
when the j expression returns a subset of rows from .SD.
For example, the following query no longer contains
a redundant 'colA.1' duplicate.
DT[,.SD[2],by=colA] # 2nd row of each group
Any existing code that uses .SD may require simple
changes to remove workarounds. If columns have been
referenced by name, as recommended where possible, then
no changes to existing code should be required.
Matthew
_______________________________________________
datatable-help mailing list
[email protected]
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help
