[R] Ensure that schema metadata can actually be set as a named character vectory #35952
Comments
thisisnic
pushed a commit
that referenced
this issue
Jun 6, 2023
…med character vector (#35954) This wasn't necessarily a regression (reprex fails in 12.0.0 as well), although the comments suggest that assigning a named character vector will work when assigning schema metadata and this feature appears to be used by at least one of our dependencies (sfarrow). Given that the sfarrow check passes on 12.0.0, there is possibly also a place in our code that returns a named character vector rather than a list. I've confirmed that this fix solves the reverse dependency failure building the sfarrow example vignette. ``` r library(arrow, warn.conflicts = FALSE) schema <- schema(x = int32()) schema$metadata <- c("name" = "value") schema$metadata #> $name #> [1] "value" ``` <sup>Created on 2023-06-06 with [reprex v2.0.2](https://reprex.tidyverse.org)</sup> * Closes: #35952 Authored-by: Dewey Dunnington <dewey@voltrondata.com> Signed-off-by: Nic Crane <thisisnic@gmail.com>
thisisnic
pushed a commit
to thisisnic/arrow
that referenced
this issue
Jun 6, 2023
…s a named character vector (apache#35954) This wasn't necessarily a regression (reprex fails in 12.0.0 as well), although the comments suggest that assigning a named character vector will work when assigning schema metadata and this feature appears to be used by at least one of our dependencies (sfarrow). Given that the sfarrow check passes on 12.0.0, there is possibly also a place in our code that returns a named character vector rather than a list. I've confirmed that this fix solves the reverse dependency failure building the sfarrow example vignette. ``` r library(arrow, warn.conflicts = FALSE) schema <- schema(x = int32()) schema$metadata <- c("name" = "value") schema$metadata #> $name #> [1] "value" ``` <sup>Created on 2023-06-06 with [reprex v2.0.2](https://reprex.tidyverse.org)</sup> * Closes: apache#35952 Authored-by: Dewey Dunnington <dewey@voltrondata.com> Signed-off-by: Nic Crane <thisisnic@gmail.com>
thisisnic
pushed a commit
to thisisnic/arrow
that referenced
this issue
Jun 13, 2023
…s a named character vector (apache#35954) This wasn't necessarily a regression (reprex fails in 12.0.0 as well), although the comments suggest that assigning a named character vector will work when assigning schema metadata and this feature appears to be used by at least one of our dependencies (sfarrow). Given that the sfarrow check passes on 12.0.0, there is possibly also a place in our code that returns a named character vector rather than a list. I've confirmed that this fix solves the reverse dependency failure building the sfarrow example vignette. ``` r library(arrow, warn.conflicts = FALSE) schema <- schema(x = int32()) schema$metadata <- c("name" = "value") schema$metadata #> $name #> [1] "value" ``` <sup>Created on 2023-06-06 with [reprex v2.0.2](https://reprex.tidyverse.org)</sup> * Closes: apache#35952 Authored-by: Dewey Dunnington <dewey@voltrondata.com> Signed-off-by: Nic Crane <thisisnic@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Describe the bug, including details regarding any error messages, version, and platform.
Before #35860, setting schema metadata as a named character vector was possible. The changes in the R bindings made this error, as exposed in a reverse dependency check on the 12.0.1 release branch.
Created on 2023-06-06 with reprex v2.0.2
Component(s)
R
The text was updated successfully, but these errors were encountered: