fread with col.names and drop #7493
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #7493 +/- ##
==========================================
- Coverage 98.97% 98.96% -0.01%
==========================================
Files 87 87
Lines 16741 16757 +16
==========================================
+ Hits 16569 16584 +15
- Misses 172 173 +1 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Generated via commit 4197bec Download link for the artifact containing the test results: ↓ atime-results.zip
|
inst/tests/tests.Rraw
Outdated
| # drop works with user override colnames #3459 | ||
| DT = data.table(a=c(1L, 4L), c=c(3L, 6L)) | ||
| test(2352.1, fread("a,b,c\n1,2,3\n4,5,6", drop='b'), DT) | ||
| test(2352.2, fread(col.names=c("a","b","c"),"1,2,3\n4,5,6", drop='b'), DT) |
There was a problem hiding this comment.
what happens here?
fread(text = "
a,b,c
1,2,3
4,5,6
", col.names=c("b", "c", "d"), drop='b')There was a problem hiding this comment.
Questionable! But current implementation favors header in read data > renaming col.names header.
There was a problem hiding this comment.
I think either can be justified; I might find it more natural to prefer col.names.
either way we'll want to document it in ?fread and with a regression test.
cc @Rdatatable/committers to get consensus on the preferred approach.
2 participants
Closes #3459