ARROW-16415: [R] Update strptime binding signature with the tz argument

[dragosmg]

This PR enables the tz argument for the strptime binding by using the assume_timezone compute function.

Current behaviour:

library(arrow, warn.conflicts = FALSE)
library(dplyr, warn.conflicts = FALSE)

test_df <- tibble::tibble(
    timestamp_naive_string = c("2018-10-07 19:04:05", NA)
)

test_df %>% 
    arrow_table() %>% 
    mutate(
        timestamp = 
            strptime(
                timestamp_naive_string, 
                format = "%Y-%m-%d %H:%M:%S", 
                tz = "Pacific/Marquesas"
            )
    ) %>%
    collect()
#> Warning: In strptime(timestamp_naive_string, format = "%Y-%m-%d %H:%M:%S", ...,
#> Time zone argument not supported in Arrow; pulling data into R
#> # A tibble: 2 × 2
#>   timestamp_naive_string timestamp          
#>   <chr>                  <dttm>             
#> 1 2018-10-07 19:04:05    2018-10-07 19:04:05
#> 2 <NA>                   NA

^{Created on 2022-05-18 by the reprex package (v2.0.1)}

Future/desired behaviour:

library(arrow, warn.conflicts = FALSE)
library(dplyr, warn.conflicts = FALSE)

test_df <- tibble::tibble(
  timestamp_naive_string = c("2018-10-07 19:04:05", NA)
)

a <- test_df %>% 
  arrow_table() %>% 
  mutate(
    timestamp = 
      strptime(
        timestamp_naive_string, 
        format = "%Y-%m-%d %H:%M:%S", 
        tz = "Pacific/Marquesas"
      )
  ) %>%
  collect()

a
#> # A tibble: 2 × 2
#>   timestamp_naive_string timestamp          
#>   <chr>                  <dttm>             
#> 1 2018-10-07 19:04:05    2018-10-07 19:04:05
#> 2 <NA>                   NA
attributes(a$timestamp)
#> $class
#> [1] "POSIXlt" "POSIXt" 
#> 
#> $tzone
#> [1] "Pacific/Marquesas"

^{Created on 2022-05-19 by the reprex package (v2.0.1)}

[github-actions]

https://issues.apache.org/jira/browse/ARROW-16415

[thisisnic]

I think you might have some issues where the default argument of tz in strptime() in R is different to the one here. I tried running this test I wrote, but got an error:

> t_string %>%
+       arrow_table() %>%
+       mutate(
+         x = strptime(x, format = "%Y-%m-%d %H:%M:%S", tz = "")
+       ) %>%
+       collect() -> out
Error in `collect()`:
! Invalid: Cannot locate timezone '':  not found in timezone database
/home/nic2/arrow/cpp/src/arrow/compute/kernels/scalar_temporal_unary.cc:128  LocateZone(options.timezone)
/home/nic2/arrow/cpp/src/arrow/compute/exec.cc:701  kernel_->exec(kernel_ctx_, batch, &out)
/home/nic2/arrow/cpp/src/arrow/compute/exec.cc:642  ExecuteBatch(batch, listener)
/home/nic2/arrow/cpp/src/arrow/compute/exec/expression.cc:597  executor->Execute(arguments, &listener)
/home/nic2/arrow/cpp/src/arrow/compute/exec/project_node.cc:90  ExecuteScalarExpression(simplified_expr, target, plan()->exec_context())
/home/nic2/arrow/cpp/src/arrow/compute/exec/exec_plan.cc:467  iterator_.Next()
/home/nic2/arrow/cpp/src/arrow/record_batch.cc:337  ReadNext(&batch)
/home/nic2/arrow/cpp/src/arrow/record_batch.cc:351  ToRecordBatches()

[thisisnic]

Other than the comment it'd be great to see added, I think this is good to go. @jonkeane - mind taking a look when you have a chance, in case I've missed anything?

[jonkeane]

This is looking good — a comment about an additional test + a few more comments that would be helpful

[jonkeane]

Looks good, thanks for sticking with it