This is an automated email from the ASF dual-hosted git repository.
thisisnic pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/arrow.git
The following commit(s) were added to refs/heads/main by this push:
new acd8e44293 GH-40886: [R] Cryptic error when creating Arrow array from
POSIXct with invalid time zones (#49714)
acd8e44293 is described below
commit acd8e442939b5fbf90d0cd65ca506514ec3851c4
Author: Nic Crane <[email protected]>
AuthorDate: Tue Apr 21 10:56:21 2026 -0400
GH-40886: [R] Cryptic error when creating Arrow array from POSIXct with
invalid time zones (#49714)
### Rationale for this change
Horrible error with invalid timezones
### What changes are included in this PR?
Do more checking of type
### Are these changes tested?
Yup
### Are there any user-facing changes?
No
### AI usage
Done with Codex, but I went through it myself and am happy with it.
* GitHub Issue: #40886
Authored-by: Nic Crane <[email protected]>
Signed-off-by: Nic Crane <[email protected]>
---
r/src/type_infer.cpp | 4 ++++
r/tests/testthat/test-type.R | 19 +++++++++++++++++++
2 files changed, 23 insertions(+)
diff --git a/r/src/type_infer.cpp b/r/src/type_infer.cpp
index a8334387c5..76a1499305 100644
--- a/r/src/type_infer.cpp
+++ b/r/src/type_infer.cpp
@@ -72,6 +72,8 @@ std::shared_ptr<arrow::DataType>
InferArrowTypeFromVector<INTSXP>(SEXP x) {
if (Rf_isNull(tzone_sexp)) {
auto systzone_sexp = cpp11::package("base")["Sys.timezone"];
return timestamp(TimeUnit::MICRO, CHAR(STRING_ELT(systzone_sexp(), 0)));
+ } else if (TYPEOF(tzone_sexp) != STRSXP) {
+ cpp11::stop("`tzone` attribute of a `POSIXct` vector must be a character
vector");
} else {
return timestamp(TimeUnit::MICRO, CHAR(STRING_ELT(tzone_sexp, 0)));
}
@@ -89,6 +91,8 @@ std::shared_ptr<arrow::DataType>
InferArrowTypeFromVector<REALSXP>(SEXP x) {
if (Rf_isNull(tzone_sexp)) {
auto systzone_sexp = cpp11::package("base")["Sys.timezone"];
return timestamp(TimeUnit::MICRO, CHAR(STRING_ELT(systzone_sexp(), 0)));
+ } else if (TYPEOF(tzone_sexp) != STRSXP) {
+ cpp11::stop("`tzone` attribute of a `POSIXct` vector must be a character
vector");
} else {
return timestamp(TimeUnit::MICRO, CHAR(STRING_ELT(tzone_sexp, 0)));
}
diff --git a/r/tests/testthat/test-type.R b/r/tests/testthat/test-type.R
index 96968830f3..74c868faac 100644
--- a/r/tests/testthat/test-type.R
+++ b/r/tests/testthat/test-type.R
@@ -53,6 +53,25 @@ test_that("infer_type() infers from R type", {
)
})
+test_that("infer_type() errors clearly for POSIXct with invalid tzone", {
+ x <- as.POSIXct("2019-02-14 13:55:05", tz = "UTC")
+ attr(x, "tzone") <- 123
+
+ expect_error(
+ infer_type(x),
+ "`tzone` attribute of a `POSIXct` vector must be a character vector"
+ )
+
+ # Also check zero-length POSIXct
+ x <- as.POSIXct(x = NULL)
+ attr(x, "tzone") <- 123
+
+ expect_error(
+ infer_type(x),
+ "`tzone` attribute of a `POSIXct` vector must be a character vector"
+ )
+})
+
test_that("infer_type() default method errors for unknown classes", {
vec <- structure(list(), class = "class_not_supported")
