Add filtering joins, with documentation and tests. #120
Conversation
Codecov Report
@@ Coverage Diff @@
## main #120 +/- ##
==========================================
+ Coverage 78.65% 78.99% +0.34%
==========================================
Files 20 21 +1
Lines 890 914 +24
==========================================
+ Hits 700 722 +22
- Misses 190 192 +2
Continue to review full report at Codecov.
|
R/join.R
Outdated
| na_matches = na_matches, | ||
| ...) | ||
|
|
||
| x <- mutate(x, `___retained` = `___row_number` %in% filtered_vars[['___row_number']]) |
There was a problem hiding this comment.
Can you skip the ___retained variable and just filter on the expression? (If not, I thnk you need to de-select the ___row_number variable, don't you?)
There was a problem hiding this comment.
Yeah, that's a good idea. I didn't de-select ___row_number because it was actually removed internally in the call to filter(). But just to be careful, in f240676 added a conditional select to remove that column if it still exists after using filter().
…so conditionally remove the `___row_number` variable if it's not already removed internally by `filter()`.
|
thanks! |
This would partially resolve #65, adding two types of joins (filtering rather than mutating) which won't accidentally mess up the survey design.
I've tested the use for in-memory datasets, but I'd like to add a couple tests for databases once I can fully wrap my head around how to get those tests running.