R/util.R

In nanoarrow: Interface to the 'nanoarrow' 'C' Library

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.


arrow_installed <- function() {
  opt <- Sys.getenv(
    "R_NANOARROW_WITHOUT_ARROW",
    getOption("nanoarrow.without_arrow", FALSE)
  )

  if (identical(tolower(opt), "true")) {
    FALSE
  } else {
    requireNamespace("arrow", quietly = TRUE)
  }
}

assert_arrow_installed <- function(reason) {
  if (!arrow_installed()) {
    stop(
      sprintf("Package 'arrow' required for %s", reason),
      call. = FALSE
    )
  }
}

warn_lossy_conversion <- function(count, msg) {
  cnd <- simpleWarning(
    sprintf("%d value(s) %s", count, msg),
    call = sys.call(-1)
  )
  class(cnd) <- union("nanoarrow_warning_lossy_conversion", class(cnd))

  warning(cnd)
}

# Internally we use R_PreserveObject() and R_ReleaseObject() to manage R objects
# that must be kept alive for ArrowArray buffers to stay valid. This count
# should be zero after tests have run in a fresh session and both gc() and
# preserved_empty() have been run. If this isn't the case, compile with
# -DNANOARROW_DEBUG_PRESERVE and run preserved_empty() to get verbose output
# about which objects didn't get released (including an R traceback to where
# they were preserved).
preserved_count <- function() {
  .Call(nanoarrow_c_preserved_count)
}

# Most objects are both preserved and released on the R main thread; however
# when sending objects into the wild there is no guarantee that they will be
# deleted on the R main thread (even though they usually are). The R package
# handles this by keeping a list of objects that couldn't be released: calling
# this function will release them and return how many were released.
preserved_empty <- function() {
  .Call(nanoarrow_c_preserved_empty)
}

# To test the "release from another thread" mechanism, this preserves obj,
# releases it from another thread and returns.
preserve_and_release_on_other_thread <- function(obj) {
  invisible(.Call(nanoarrow_c_preserve_and_release_on_other_thread, obj))
}

# This is used by bookkeeping infrastructure when debugging an imbalance in
# preserved/released SEXPs.
current_stack_trace_chr <- function() {
  tb <- rlang::trace_back()
  paste0(utils::capture.output(print(tb)), collapse = "\n")
}

# Consolidate places we should call vctrs::vec_slice()
# if/when a vctrs dependency is added
vec_slice2 <- function(x, i) {
  if (is.data.frame(x)) {
    x[i, , drop = FALSE]
  } else {
    x[i]
  }
}

`%||%` <- function(rhs, lhs) {
  if (is.null(rhs)) lhs else rhs
}

new_data_frame <- function(x, nrow) {
  structure(x, row.names = c(NA, nrow), class = "data.frame")
}

vec_gen <- function(ptype, n = 1e3, prop_true = 0.5,  prop_na = 0,
                    chr_len = function(n) ceiling(25 * stats::runif(n))) {
  vec <- switch(
    class(ptype)[1],
    logical = stats::runif(n) < prop_true,
    integer = as.integer(stats::runif(n, min = -1, max = 1) * .Machine$integer.max),
    numeric = stats::runif(n),
    character = strrep(rep_len(letters, n), chr_len(n)),
    data.frame = new_data_frame(
      lapply(
        ptype,
        vec_gen,
        n = n,
        prop_true = prop_true,
        prop_na = prop_na,
        chr_len = chr_len
      ),
      n
    ),
    stop(sprintf("Don't know how to generate vector for type %s", class(ptype)[1]))
  )

  if (!is.data.frame(vec) && prop_na > 0) {
    is_na <- stats::runif(n) < prop_na
    vec[is_na] <- ptype[NA_integer_]
  }

  vec
}

vec_shuffle <- function(x) {
  if (is.data.frame(x)) {
    i <- sample(seq_len(nrow(x)), replace = FALSE)
  } else {
    i <- sample(seq_along(x), replace = FALSE)
  }

  vec_slice2(x, i)
}