Chore: update conversions and error-handling

NEWS.md

@@ -38,6 +38,8 @@
   columns to these functions will now compute the min/max/sum in each column
   separately. Use `pl$min_horizontal()` `pl$max_horizontal()`, and
   `pl$sum_horizontal()` instead for rowwise computation (#508).
+- `pl$concat_list()`: elements being strings are now interpreted as column names. Use `pl$lit` to
+  concat with a string.
 - `$is_not()` is deprecated and will be removed in 0.12.0. Use `$not()` instead
   (#511, #531).
 - `$is_first()` is deprecated and will be removed in 0.12.0. Use `$is_first_distinct()`

R/expr__expr.R

@@ -1712,7 +1712,8 @@ Expr_fill_nan = function(expr = NULL) {
 #' @examples
 #' pl$select(pl$lit(1:5)$std())
 Expr_std = function(ddof = 1) {
-  unwrap(.pr$Expr$std(self, ddof))
+  .pr$Expr$std(self, ddof) |>
+    unwrap("in $std():")
 }
 
 #' Get variance
@@ -1722,7 +1723,8 @@ Expr_std = function(ddof = 1) {
 #' @examples
 #' pl$select(pl$lit(1:5)$var())
 Expr_var = function(ddof = 1) {
-  unwrap(.pr$Expr$var(self, ddof))
+  .pr$Expr$var(self, ddof) |>
+    unwrap("in $var():")
 }
 
 #' Get maximum value
@@ -2657,6 +2659,7 @@ Expr_rank = function(method = "average", descending = FALSE) {
   unwrap(.pr$Expr$rank(self, method, descending))
 }
 
+
 #' Difference
 #'
 #' Calculate the n-th discrete difference.
@@ -2669,8 +2672,9 @@ Expr_rank = function(method = "average", descending = FALSE) {
 #'   diff_default = pl$col("a")$diff(),
 #'   diff_2_ignore = pl$col("a")$diff(2, "ignore")
 #' )
-Expr_diff = function(n = 1, null_behavior = "ignore") {
-  unwrap(.pr$Expr$diff(self, n, null_behavior))
+  Expr_diff = function(n = 1, null_behavior = c("ignore", "drop")) {
+  .pr$Expr$diff(self, n, null_behavior) |>
+    unwrap("in $diff():")
 }
 
 #' Percentage change
@@ -3072,7 +3076,8 @@ Expr_ewm_mean = function(
     com = NULL, span = NULL, half_life = NULL, alpha = NULL,
     adjust = TRUE, min_periods = 1L, ignore_nulls = TRUE) {
   alpha = prepare_alpha(com, span, half_life, alpha)
-  unwrap(.pr$Expr$ewm_mean(self, alpha, adjust, min_periods, ignore_nulls))
+  .pr$Expr$ewm_mean(self, alpha, adjust, min_periods, ignore_nulls) |>
+    unwrap("in $ewm_mean()")
 }
 
 #' Exponentially-weighted moving standard deviation
@@ -3087,7 +3092,8 @@ Expr_ewm_std = function(
     com = NULL, span = NULL, half_life = NULL, alpha = NULL,
     adjust = TRUE, bias = FALSE, min_periods = 1L, ignore_nulls = TRUE) {
   alpha = prepare_alpha(com, span, half_life, alpha)
-  unwrap(.pr$Expr$ewm_std(self, alpha, adjust, bias, min_periods, ignore_nulls))
+  .pr$Expr$ewm_std(self, alpha, adjust, bias, min_periods, ignore_nulls) |>
+    unwrap("in $ewm_std()")
 }
 
 #' Exponentially-weighted moving variance
@@ -3102,7 +3108,8 @@ Expr_ewm_var = function(
     com = NULL, span = NULL, half_life = NULL, alpha = NULL,
     adjust = TRUE, bias = FALSE, min_periods = 1L, ignore_nulls = TRUE) {
   alpha = prepare_alpha(com, span, half_life, alpha)
-  unwrap(.pr$Expr$ewm_var(self, alpha, adjust, bias, min_periods, ignore_nulls))
+  .pr$Expr$ewm_var(self, alpha, adjust, bias, min_periods, ignore_nulls)  |>
+    unwrap("in $ewm_var()")
 }
 
 #' Extend Series with a constant
@@ -3116,7 +3123,8 @@ Expr_ewm_var = function(
 #' pl$select(pl$lit(1:4)$extend_constant(10.1, 2))
 #' pl$select(pl$lit(1:4)$extend_constant(NULL, 2))
 Expr_extend_constant = function(value, n) {
-  unwrap(.pr$Expr$extend_constant(self, wrap_e(value), n))
+  .pr$Expr$extend_constant(self, wrap_e(value), n) |>
+    unwrap("in $extend_constant()")
 }
 
 #' Repeat a Series
@@ -3134,7 +3142,8 @@ Expr_extend_constant = function(value, n) {
 #' pl$select(pl$lit("alice")$rep(n = 3))
 #' pl$select(pl$lit(1:3)$rep(n = 2))
 Expr_rep = function(n, rechunk = TRUE) {
-  unwrap(.pr$Expr$rep(self, n, rechunk))
+  .pr$Expr$rep(self, n, rechunk)  |>
+    unwrap("in $rep()")
 }
 
 #' Extend a Series by repeating values

R/expr__list.R

@@ -121,9 +121,9 @@ ExprList_unique = function() .pr$Expr$list_unique(self)
 #' )
 #' df$select(pl$col("a")$list$concat(pl$col("b")))
 #'
-#' df$select(pl$col("a")$list$concat("hello from R"))
+#' df$select(pl$col("a")$list$concat(pl$lit("hello from R")))
 #'
-#' df$select(pl$col("a")$list$concat(list("hello", c("hello", "world"))))
+#' df$select(pl$col("a")$list$concat(pl$lit(list("hello", c("hello", "world")))))
 ExprList_concat = function(other) {
   pl$concat_list(list(self, other))
 }
@@ -283,9 +283,10 @@ ExprList_arg_max = function() .pr$Expr$list_arg_max(self)
 #' @aliases Expr_list_diff
 #' @examples
 #' df = pl$DataFrame(list(s = list(1:4, c(10L, 2L, 1L))))
-#' df$select(pl$col("s")$list$diff())
-ExprList_diff = function(n = 1, null_behavior = "ignore") {
-  unwrap(.pr$Expr$list_diff(self, n, null_behavior))
+#' df$select(pl$col("s")$list$diff(1))
+ExprList_diff = function(n = 1, null_behavior = c("ignore", "drop")) {
+  .pr$Expr$list_diff(self, n, null_behavior) |>
+    unwrap("in $list$diff()")
 }
 
 #' Shift sublists
@@ -364,7 +365,9 @@ ExprList_tail = function(n = 5L) {
 
 #' List to Struct
 #' @param n_field_strategy Strategy to determine the number of fields of the struct.
-#'  default = 'first_non_null' else 'max_width'
+#'  default = "first_non_null": set number of fields equal to the length of the
+#'  first non zero-length sublist. else 'max_width'; else  "max_width":
+#'  set number of fields as max length of all sublists.
 #' @param name_generator an R function that takes an R scalar double and outputs
 #' a string value. It is a f64 because i32 might not be a big enough enumerate all.
 #' The default (`NULL`) is equivalent to the R function
@@ -390,7 +393,10 @@ ExprList_tail = function(n = 5L) {
 #'
 #' df2$to_list()
 ExprList_to_struct = function(
-    n_field_strategy = "first_non_null", name_generator = NULL, upper_bound = 0) {
+    n_field_strategy = c("first_non_null","max_width"),
+    name_generator = NULL,
+    upper_bound = 0
+) {
   .pr$Expr$list_to_struct(self, n_field_strategy, name_generator, upper_bound) |>
     unwrap("in <List>$to_struct():")
 }
@@ -410,7 +416,7 @@ ExprList_to_struct = function(
 #' @return Expr
 #' @aliases list_eval
 #' @examples
-#' df = pl$DataFrame(a = list(c(1, 8, 3), b = c(4, 5, 2)))
+#' df = pl$DataFrame(a = c(1, 8, 3), b = c(4, 5, 2))
 #' df$select(pl$all()$cast(pl$dtypes$Int64))$with_columns(
 #'   pl$concat_list(c("a", "b"))$list$eval(pl$element()$rank())$alias("rank")
 #' )

R/extendr-wrappers.R

@@ -659,7 +659,7 @@ Expr$list_slice <- function(offset, length) .Call(wrap__Expr__list_slice, self,
 
 Expr$list_eval <- function(expr, parallel) .Call(wrap__Expr__list_eval, self, expr, parallel)
 
-Expr$list_to_struct <- function(width_strat, name_gen, upper_bound) .Call(wrap__Expr__list_to_struct, self, width_strat, name_gen, upper_bound)
+Expr$list_to_struct <- function(n_field_strategy, name_gen, upper_bound) .Call(wrap__Expr__list_to_struct, self, n_field_strategy, name_gen, upper_bound)
 
 Expr$str_to_date <- function(format, strict, exact, cache, ambiguous) .Call(wrap__Expr__str_to_date, self, format, strict, exact, cache, ambiguous)
 

R/functions__lazy.R

@@ -476,7 +476,6 @@ pl$approx_n_unique = function(column) { #-> int or Expr
 #' df = pl$DataFrame(a = 1:2, b = 3:4, c = 5:6)
 #'
 #' # Compute sum in several columns
-#' df$with_columns(pl$sum("a", "c"))
 #' df$with_columns(pl$sum("*"))
 pl$sum = function(..., verbose = TRUE) {
   column = list2(...)
@@ -519,9 +518,6 @@ pl$sum = function(..., verbose = TRUE) {
 #' )
 #' df
 #'
-#' df$with_columns(
-#'   pl$min("a", "b", "c")
-#' )
 pl$min = function(..., verbose = TRUE) {
   column = list2(...)
   if (length(column) == 1L) column <- column[[1L]]
@@ -567,9 +563,6 @@ pl$min = function(..., verbose = TRUE) {
 #' )
 #' df
 #'
-#' df$with_columns(
-#'   pl$max("a", "b", "c")
-#' )
 pl$max = function(..., verbose = TRUE) {
   column = list2(...)
   if (length(column) == 1L) column <- column[[1L]]
@@ -667,7 +660,7 @@ pl$var = function(column, ddof = 1) {
 #' Concat the arrays in a Series dtype List in linear time.
 #' @description Folds the expressions from left to right, keeping the first non-null value.
 #' @name pl_concat_list
-#' @param exprs list of Expr or Series or strings or a mix, or a char vector
+#' @param exprs list of Into<Expr>, strings interpreted as column names
 #' @return Expr
 #'
 #' @keywords Expr_new
@@ -692,9 +685,8 @@ pl$var = function(column, ddof = 1) {
 #' ))$alias("alice")$lit_to_s()
 #'
 pl$concat_list = function(exprs) {
-  l_expr = lapply(as.list(exprs), wrap_e)
-  pra = do.call(construct_ProtoExprArray, l_expr)
-  unwrap(concat_list(pra))
+  concat_list(as.list(exprs)) |>
+    unwrap(" in pl$concat_list():")
 }
 
 

inst/misc/develop_polars.R

@@ -281,7 +281,7 @@ find_missing_return = function() {
 #' @export
 #'
 #' @examples
-run_all_examples_collect_errors = \(skip_these = character()) {
+run_all_examples_collect_errors = \(skip_these = character(), time_examples = FALSE) {
   paths = list.files(full.names = TRUE, path = "./man/.")
   fnames = list.files(full.names = FALSE, path = "./man/.")
   names(paths) = fnames
@@ -290,10 +290,16 @@ run_all_examples_collect_errors = \(skip_these = character()) {
 
 
   out = lapply(paths, \(path) {
-    print(path)
+    cat("\n",path)
+    if(time_examples) t1 = Sys.time()
     txt = capture.output({
       err = polars:::result(pkgload::run_example(path = path))$err
     })
+    if(time_examples) {
+      t2 = Sys.time()
+      duration = difftime(t2,t1, units = "secs")
+      if(duration >.1) cat(" ", duration,"s")
+    }
     if (!is.null(err)) list(err = err, txt = txt)
   })