|Next release:||Version 0.7.0|
|Previous release:||Version 0.5.0|
DT[i, j, by]is now supported.
fread()will detect a feather file and issue an appropriate error message.
fread()extracts data from archives into memory, it will now display the size of the extracted data in verbose mode.
Multiple reduction operators can now be performed at once.
In groupby, reduction columns can now be combined with regular or computed columns.
During grouping, group keys are now added automatically to the select list.
The equality operator
==now works for string columns too.
Improved performance of groupby operations.
fread()will no longer emit an error if there is an NA string in the header.
If the input contains excessively long lines,
fread()will no longer waste time printing a sample of first 5 lines in verbose mode.
Fixed wrong calculation of mean / standard deviation of line length in
fread()if the sample contained broken lines.
Frame view will no longer get stuck in a Jupyter notebook.
This release was created with the help of 3 people who contributed code and documentation, and 3 more people who submitted bug reports and feature requests.
Code & documentation contributors: