@@ -14,12 +14,10 @@ Pre-1.0.0 numbering scheme: 0.x will indicate releases, while 0.x.y will indicat
14
14
with ` covid ` . The data set previously named ` jhu_confirmed_cumulative_num ` has
15
15
been removed from the package, but a renamed version is has been removed from
16
16
the package, but a renamed version is still available in ` epidatasets ` .
17
-
18
- ## Bug fixes
19
-
20
- - Removed ` .window_size = 1 ` default from ` epi_slide_{mean,sum,opt} ` ; this
21
- argument is now mandatory, and should nearly always be greater than 1 except
22
- for testing purposes.
17
+ - ` epi_slide_{sum,mean,opt} ` have improved default output column names, and
18
+ additional arguments for specifying names: ` .prefix ` , ` .suffix ` ,
19
+ ` .new_col_names ` . To obtain the old naming behavior, use `.prefix =
20
+ "slide_value_ "`.
23
21
24
22
## Improvements
25
23
@@ -29,6 +27,18 @@ Pre-1.0.0 numbering scheme: 0.x will indicate releases, while 0.x.y will indicat
29
27
- Improved validation of ` .window_size ` arguments.
30
28
- Rewrote a lot of the package documentation to be more consistent and
31
29
informative. Simplified and streamlined the vignettes.
30
+ - ` epi_slide_{sum,mean,opt} ` on ungrouped ` epi_df ` s will now temporarily group
31
+ by ` geo_value ` and any ` other_keys ` for the slide operation rather than raise
32
+ an error about duplicated time values. ` epi_slide ` 's analogous automatic
33
+ grouping has been made temporary in order to match.
34
+
35
+ ## Bug fixes
36
+
37
+ - Removed ` .window_size = 1 ` default from ` epi_slide_{mean,sum,opt} ` ; this
38
+ argument is now mandatory, and should nearly always be greater than 1 except
39
+ for testing purposes.
40
+ - Fixed ` epi_slide_{sum,mean,opt} ` raising an error on certain tidyselect
41
+ expressions.
32
42
33
43
## Cleanup
34
44
0 commit comments