Use .tp$function_name() to get access to the functions that are translated
by tidypolars without loading the package these functions originally come
from.
This may be useful in cases where you want to benefit from the interface of
these functions but don't want to add some tidyverse dependencies to your
project (e.g. stringr because it might be slow to build the package in some
cases).
Note that the name of the package that originally provided the function must
be appended to the function name. For instance, if you want to use
stringr::str_extract() without loading stringr, you can do so with
.tp$str_extract_stringr(). This is because multiple packages may have a
function named str_extract(), so we need to inform tidypolars of which
translation we want exactly.
Note: using .tp will make it harder to convert tidypolars code to run
with other tidyverse-based backends because .tp will be unknown to those
backends. If you expect to switch between tidypolars, the original tidyverse,
and tidyverse-based backends, you should avoid using .tpand load the original
packages in the session instead.
This is similar to the dd object in duckplyr and to the .sql object in
dbplyr.
Examples
# List of all functions stored in this object
sort(names(.tp))
#> [1] "ISOdatetime" "abs"
#> [3] "acos" "acosh"
#> [5] "all" "am_lubridate"
#> [7] "any" "as.Date"
#> [9] "as.character" "as.logical"
#> [11] "as.numeric" "asin"
#> [13] "asinh" "atan"
#> [15] "atanh" "between_dplyr"
#> [17] "case_match" "case_when"
#> [19] "ceiling" "coalesce_dplyr"
#> [21] "consecutive_id_dplyr" "cos"
#> [23] "cosh" "cumcount"
#> [25] "cummin" "cumprod"
#> [27] "cumsum" "date_lubridate"
#> [29] "day_lubridate" "days_in_month_lubridate"
#> [31] "days_lubridate" "ddays_lubridate"
#> [33] "dense_rank_dplyr" "desc_dplyr"
#> [35] "dhours_lubridate" "diff"
#> [37] "dmicroseconds_lubridate" "dmilliseconds_lubridate"
#> [39] "dminutes_lubridate" "dmy_lubridate"
#> [41] "dnanoseconds_lubridate" "dseconds_lubridate"
#> [43] "duplicated" "dweeks_lubridate"
#> [45] "dym_lubridate" "exp"
#> [47] "first_dplyr" "floor"
#> [49] "force_tz_lubridate" "grepl"
#> [51] "gsub" "hour_lubridate"
#> [53] "hours_lubridate" "ifelse"
#> [55] "infinite" "interpolate"
#> [57] "is.finite" "is.infinite"
#> [59] "is.nan" "is.null"
#> [61] "is_first" "isoyear_lubridate"
#> [63] "kurtosis" "lag_dplyr"
#> [65] "last_dplyr" "lead_dplyr"
#> [67] "leap_year_lubridate" "length"
#> [69] "log" "log10"
#> [71] "make_date_lubridate" "make_datetime_lubridate"
#> [73] "max" "mday_lubridate"
#> [75] "mdy_lubridate" "mean"
#> [77] "median" "microseconds_lubridate"
#> [79] "milliseconds_lubridate" "min"
#> [81] "min_rank_dplyr" "minute_lubridate"
#> [83] "minutes_lubridate" "mode"
#> [85] "month_lubridate" "my_lubridate"
#> [87] "myd_lubridate" "n_distinct_dplyr"
#> [89] "n_dplyr" "na_if_dplyr"
#> [91] "nanoseconds_lubridate" "nchar"
#> [93] "now_lubridate" "nth_dplyr"
#> [95] "paste" "paste0"
#> [97] "pm_lubridate" "quarter_lubridate"
#> [99] "rank" "replace_na_tidyr"
#> [101] "rev" "rollback_lubridate"
#> [103] "rollbackward_lubridate" "rollforward_lubridate"
#> [105] "round" "row_number_dplyr"
#> [107] "sample" "sd"
#> [109] "second_lubridate" "seconds_lubridate"
#> [111] "seq" "seq_len"
#> [113] "sign" "sin"
#> [115] "sinh" "skew"
#> [117] "sort" "sqrt"
#> [119] "str_count_stringr" "str_detect_stringr"
#> [121] "str_dup_stringr" "str_ends_stringr"
#> [123] "str_equal_stringr" "str_extract_all_stringr"
#> [125] "str_extract_stringr" "str_length_stringr"
#> [127] "str_pad_stringr" "str_remove_all_stringr"
#> [129] "str_remove_stringr" "str_replace_all_stringr"
#> [131] "str_replace_na_stringr" "str_replace_stringr"
#> [133] "str_split_i_stringr" "str_split_stringr"
#> [135] "str_squish_stringr" "str_starts_stringr"
#> [137] "str_sub_stringr" "str_to_lower_stringr"
#> [139] "str_to_title_stringr" "str_to_upper_stringr"
#> [141] "str_trim_stringr" "str_trunc_stringr"
#> [143] "strptime" "substr"
#> [145] "sum" "tan"
#> [147] "tanh" "toTitleCase"
#> [149] "tolower" "toupper"
#> [151] "trimws" "unique"
#> [153] "unique_counts" "var"
#> [155] "wday_lubridate" "week_lubridate"
#> [157] "weeks_lubridate" "which.max"
#> [159] "which.min" "with_tz_lubridate"
#> [161] "word_stringr" "yday_lubridate"
#> [163] "ydm_lubridate" "year_lubridate"
#> [165] "ym_lubridate" "ymd_hms_lubridate"
#> [167] "ymd_lubridate"
dat <- polars::pl$DataFrame(x = c("abc12", "def3"))
dat |>
mutate(y = .tp$str_extract_stringr(x, "\\d+"))
#> shape: (2, 2)
#> ┌───────┬─────┐
#> │ x ┆ y │
#> │ --- ┆ --- │
#> │ str ┆ str │
#> ╞═══════╪═════╡
#> │ abc12 ┆ 12 │
#> │ def3 ┆ 3 │
#> └───────┴─────┘