[GitHub] [arrow] lidavidm commented on a change in pull request #8648: ARROW-7906: [C++] [Python] Add ORC write support

[GitBox <gi...@apache.org>]

lidavidm commented on a change in pull request #8648:
URL: https://github.com/apache/arrow/pull/8648#discussion_r613181812



##########
File path: cpp/src/arrow/adapters/orc/adapter_test.cc
##########
@@ -58,6 +82,182 @@ class MemoryOutputStream : public liborc::OutputStream {
   uint64_t length_, natural_write_size_;
 };
 
+std::shared_ptr<Buffer> GenerateFixedDifferenceBuffer(int32_t fixed_length,
+                                                      int64_t length) {
+  BufferBuilder builder;
+  int32_t offsets[length];
+  ARROW_EXPECT_OK(builder.Resize(4 * length));
+  for (int32_t i = 0; i < length; i++) {
+    offsets[i] = fixed_length * i;
+  }
+  ARROW_EXPECT_OK(builder.Append(offsets, 4 * length));
+  std::shared_ptr<Buffer> buffer;
+  ARROW_EXPECT_OK(builder.Finish(&buffer));
+  return buffer;
+}
+
+std::shared_ptr<Array> CastFixedSizeBinaryArrayToBinaryArray(
+    std::shared_ptr<Array> array) {
+  auto fixed_size_binary_array = std::static_pointer_cast<FixedSizeBinaryArray>(array);
+  std::shared_ptr<Buffer> value_offsets = GenerateFixedDifferenceBuffer(
+      fixed_size_binary_array->byte_width(), array->length() + 1);
+  return std::make_shared<BinaryArray>(array->length(), value_offsets,
+                                       array->data()->buffers[1],
+                                       array->data()->buffers[0]);
+}
+
+template <typename TargetArrayType>
+std::shared_ptr<Array> CastInt64ArrayToTemporalArray(
+    const std::shared_ptr<DataType>& type, std::shared_ptr<Array> array) {
+  std::shared_ptr<ArrayData> new_array_data =
+      ArrayData::Make(type, array->length(), array->data()->buffers);
+  return std::make_shared<TargetArrayType>(new_array_data);
+}
+
+template <typename T, typename U>
+void randintpartition(int64_t n, T sum, std::vector<U>* out) {
+  const int random_seed = 0;
+  std::default_random_engine gen(random_seed);
+  out->resize(n, static_cast<T>(0));
+  T remaining_sum = sum;
+  std::generate(out->begin(), out->end() - 1, [&gen, &remaining_sum] {
+    std::uniform_int_distribution<T> d(static_cast<T>(0), remaining_sum);
+    auto res = d(gen);
+    remaining_sum -= res;
+    return static_cast<U>(res);
+  });
+  (*out)[n - 1] += remaining_sum;
+  std::random_shuffle(out->begin(), out->end());
+}
+
+Result<std::shared_ptr<Array>> GenerateRandomDate64Array(int64_t size,
+                                                         double null_probability) {
+  arrow::random::RandomArrayGenerator rand(kRandomSeed);
+  return CastInt64ArrayToTemporalArray<Date64Array>(
+      date64(), rand.Int64(size, kMilliMin, kMilliMax, null_probability));
+}
+
+Result<std::shared_ptr<Array>> GenerateRandomTimestampArray(int64_t size,

Review comment:
       You could do something like `GenerateArray(field("", timestamp(TimeUnit::SECOND), key_value_metadata({{"min", kSecondMin}, {"max", kSecondMax}})))` but that isn't really much more concise than what you have here at this point, right, since your constraint is that they all have to be castable to nanoseconds without overflowing.




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org