Maximilian Paulisch
Maximilian Paulisch
Hera a not so short reprex ``` r con [1] 0 DBI::dbExecute(con, "CREATE SEQUENCE IF NOT EXISTS dbi_test_sequence;") #> [1] 0 DBI::dbExecute( con, "CREATE TABLE dbi_test ( id bigint DEFAULT...
To support this case here is a benchmarks from my original issue in tibble: ``` r f 1 tibble 523.22ms 523.22ms 1.91 211.5KB 15.3 #> 2 dataframe 78.31ms 80.04ms 12.0...
Comparing the performance of the "all false" case to the PR in vctrs it seems that we might pay quite a bit for the `has_names` check in every iteration.
I'm not sure how the SQL looks like in SQL Server but I quickly hacked something in Postgres ```sql select json_agg(row_to_json(row(, , ...))) ``` This would allow for a relatively...
Already fixed in the dev version of dbplyr via #1014.
When you see this, you really, really want dbplyr to handle this for you 😄
In the R package `tidyr` they also have the parameter `names_sep` which I find very useful. In polars this could be something like: ```python df = pl.DataFrame( {"a": { "x":...
As I have written in the other issue which was closed: Not so long ago in the R language the tidyverse team had similar issues and they came up with...
I agree it would be very nice and I missed it a couple of times myself. Therefore, I opened an issue in dplyr: https://github.com/tidyverse/dplyr/issues/6905. I guess the discussion should continue...
There is already a PR #984 open for this. If you want to test it out you can install it via `install_github("tidyverse/dbplyr#984")`. It would be great to get feedback in...