# common type failure uses error call and error arg (#1641, #1692)
Code
vec_c("x", 1, .error_call = call("foo"), .error_arg = "arg")
Condition
Error in `foo()`:
! Can't combine `arg[[1]]` <character> and `arg[[2]]` <double>.
---
Code
vec_c("x", .ptype = integer(), .error_call = call("foo"), .error_arg = "arg")
Condition
Error in `foo()`:
! Can't convert `arg[[1]]` <character> to <integer>.
# common type failure uses positional errors
Code
(expect_error(vec_c(1, a = "x", 2)))
Output
<error/vctrs_error_ptype2>
Error in `vec_c()`:
! Can't combine `..1` <double> and `a` <character>.
Code
(expect_error(vec_c(1, a = "x", 2, .ptype = double(), .error_arg = "arg")))
Output
<error/vctrs_error_cast>
Error in `vec_c()`:
! Can't convert `arg$a` <character> to <double>.
Code
(expect_error(vec_c(1, a = 2.5, .ptype = integer())))
Output
<error/vctrs_error_cast_lossy>
Error in `vec_c()`:
! Can't convert from `a` <double> to <integer> due to loss of precision.
* Locations: 1
# vec_c() includes index in argument tag
Code
vec_c(df1, df2)
Condition
Error in `vec_c()`:
! Can't combine `..1$x$y$z` <double> and `..2$x$y$z` <character>.
---
Code
vec_c(df1, df1, df2)
Condition
Error in `vec_c()`:
! Can't combine `..1$x$y$z` <double> and `..3$x$y$z` <character>.
---
Code
vec_c(foo = df1, bar = df2)
Condition
Error in `vec_c()`:
! Can't combine `foo$x$y$z` <double> and `bar$x$y$z` <character>.
# vec_c() can repair names quietly
Code
res_unique <- vec_c(x = TRUE, x = 0, .name_repair = "unique_quiet")
res_universal <- vec_c(`if` = TRUE, `in` = 0, .name_repair = "universal_quiet")
# vec_c() fails with complex foreign S3 classes
Code
x <- structure(foobar(1), attr_foo = "foo")
y <- structure(foobar(2), attr_bar = "bar")
(expect_error(vec_c(x, y), class = "vctrs_error_incompatible_type"))
Output
<error/vctrs_error_ptype2>
Error in `vec_c()`:
! Can't combine `..1` <vctrs_foobar> and `..2` <vctrs_foobar>.
x Some attributes are incompatible.
i The author of the class should implement vctrs methods.
i See <https://vctrs.r-lib.org/reference/faq-error-incompatible-attributes.html>.
Code
(expect_error(vec_c(x, y, .error_call = call("foo"), .error_arg = "arg"),
class = "vctrs_error_incompatible_type"))
Output
<error/vctrs_error_ptype2>
Error in `foo()`:
! Can't combine `arg[[1]]` <vctrs_foobar> and `arg[[2]]` <vctrs_foobar>.
x Some attributes are incompatible.
i The author of the class should implement vctrs methods.
i See <https://vctrs.r-lib.org/reference/faq-error-incompatible-attributes.html>.
# vec_c() fails with complex foreign S4 classes
Code
joe <- .Counts(c(1L, 2L), name = "Joe")
jane <- .Counts(3L, name = "Jane")
(expect_error(vec_c(joe, jane), class = "vctrs_error_incompatible_type"))
Output
<error/vctrs_error_ptype2>
Error in `vec_c()`:
! Can't combine `..1` <vctrs_Counts> and `..2` <vctrs_Counts>.
x Some attributes are incompatible.
i The author of the class should implement vctrs methods.
i See <https://vctrs.r-lib.org/reference/faq-error-incompatible-attributes.html>.
Code
(expect_error(vec_c(joe, jane, .error_call = call("foo"), .error_arg = "arg"),
class = "vctrs_error_incompatible_type"))
Output
<error/vctrs_error_ptype2>
Error in `foo()`:
! Can't combine `arg[[1]]` <vctrs_Counts> and `arg[[2]]` <vctrs_Counts>.
x Some attributes are incompatible.
i The author of the class should implement vctrs methods.
i See <https://vctrs.r-lib.org/reference/faq-error-incompatible-attributes.html>.
# vec_c() fallback doesn't support `name_spec` or `ptype`
Code
(expect_error(with_c_foobar(vec_c(foobar(1), foobar(2), .name_spec = "{outer}_{inner}")),
"name specification"))
Output
<error/rlang_error>
Error in `vec_c()`:
! Can't use a name specification with non-vctrs types.
vctrs methods must be implemented for class `vctrs_foobar`.
See <https://vctrs.r-lib.org/articles/s3-vector.html>.
Code
(expect_error(with_c_foobar(vec_c(foobar(1), foobar(2), .ptype = "")), class = "vctrs_error_incompatible_type")
)
Output
<error/vctrs_error_cast>
Error in `vec_c()`:
! Can't convert `..1` <vctrs_foobar> to <character>.
Code
(expect_error(with_c_foobar(vec_c(foobar(1), foobar(2), .error_call = call(
"foo"), .name_spec = "{outer}_{inner}"))))
Output
<error/rlang_error>
Error in `foo()`:
! Can't use a name specification with non-vctrs types.
vctrs methods must be implemented for class `vctrs_foobar`.
See <https://vctrs.r-lib.org/articles/s3-vector.html>.
# can ignore names in `vec_c()` by providing a `zap()` name-spec (#232)
Code
(expect_error(vec_c(a = c(b = letters), b = 1, .name_spec = zap()), class = "vctrs_error_incompatible_type")
)
Output
<error/vctrs_error_ptype2>
Error in `vec_c()`:
! Can't combine `a` <character> and `b` <double>.
# concatenation performs expected allocations
Code
ints <- rep(list(1L), 100)
dbls <- rep(list(1), 100)
# # `vec_c()`
# Integers
with_memory_prof(vec_c_list(ints))
Output
[1] 1.96KB
Code
# Doubles
with_memory_prof(vec_c_list(dbls))
Output
[1] 2.35KB
Code
# Integers to integer
with_memory_prof(vec_c_list(ints, ptype = int()))
Output
[1] 1.7KB
Code
# Doubles to integer
with_memory_prof(vec_c_list(dbls, ptype = int()))
Output
[1] 1.7KB
Code
# # `list_unchop()`
# Integers
with_memory_prof(list_unchop(ints))
Output
[1] 1.13KB
Code
# Doubles
with_memory_prof(list_unchop(dbls))
Output
[1] 1.52KB
Code
# Integers to integer
with_memory_prof(list_unchop(ints, ptype = int()))
Output
[1] 896B
Code
# Doubles to integer
with_memory_prof(list_unchop(dbls, ptype = int()))
Output
[1] 896B
Code
# # Concatenation with names
# Named integers
ints <- rep(list(set_names(1:3, letters[1:3])), 100)
with_memory_prof(list_unchop(ints))
Output
[1] 4.3KB
Code
# Named matrices
mat <- matrix(1:4, 2, dimnames = list(c("foo", "bar")))
mats <- rep(list(mat), 100)
with_memory_prof(list_unchop(mats))
Output
[1] 5.52KB
Code
# Data frame with named columns
df <- data_frame(x = set_names(as.list(1:2), c("a", "b")), y = set_names(1:2, c(
"A", "B")), z = data_frame(Z = set_names(1:2, c("Za", "Zb"))))
dfs <- rep(list(df), 100)
with_memory_prof(list_unchop(dfs))
Output
[1] 8.79KB
Code
# Data frame with rownames (non-repaired, non-recursive case)
df <- data_frame(x = 1:2)
dfs <- rep(list(df), 100)
dfs <- map2(dfs, seq_along(dfs), set_rownames_recursively)
with_memory_prof(list_unchop(dfs))
Output
[1] 6.02KB
Code
# Data frame with rownames (repaired, non-recursive case)
dfs <- map(dfs, set_rownames_recursively)
with_memory_prof(list_unchop(dfs))
Output
[1] 12.2KB
Code
# Data frame with rownames (non-repaired, recursive case) (#1217)
df <- data_frame(x = 1:2, y = data_frame(x = 1:2))
dfs <- rep(list(df), 100)
dfs <- map2(dfs, seq_along(dfs), set_rownames_recursively)
with_memory_prof(list_unchop(dfs))
Output
[1] 11.4KB
Code
# Data frame with rownames (repaired, recursive case) (#1217)
dfs <- map(dfs, set_rownames_recursively)
with_memory_prof(list_unchop(dfs))
Output
[1] 23.7KB
Code
# list-ofs (#1496)
make_list_of <- (function(n) {
df <- tibble::tibble(x = new_list_of(vec_chop(1:n), ptype = integer()))
vec_chop(df)
})
with_memory_prof(list_unchop(make_list_of(1000)))
Output
[1] 112KB
Code
with_memory_prof(list_unchop(make_list_of(2000)))
Output
[1] 221KB
Code
with_memory_prof(list_unchop(make_list_of(4000)))
Output
[1] 440KB
Generated by dwww version 1.15 on Sun Jun 16 07:35:46 CEST 2024.