You are viewing a plain text version of this content. The canonical link for it is here.
Posted to github@arrow.apache.org by GitBox <gi...@apache.org> on 2022/11/21 17:47:31 UTC

[GitHub] [arrow-rs] viirya commented on a diff in pull request #3122: Add GenericByteBuilder (#2969)

viirya commented on code in PR #3122:
URL: https://github.com/apache/arrow-rs/pull/3122#discussion_r1028355055


##########
arrow-array/src/builder/generic_bytes_builder.rs:
##########
@@ -17,70 +17,81 @@
 
 use crate::builder::null_buffer_builder::NullBufferBuilder;
 use crate::builder::{ArrayBuilder, BufferBuilder, UInt8BufferBuilder};
-use crate::{ArrayRef, GenericBinaryArray, OffsetSizeTrait};
+use crate::types::{ByteArrayType, GenericBinaryType, GenericStringType};
+use crate::{ArrayRef, GenericByteArray, OffsetSizeTrait};
+use arrow_buffer::ArrowNativeType;
 use arrow_data::ArrayDataBuilder;
 use std::any::Any;
 use std::sync::Arc;
 
-///  Array builder for [`GenericBinaryArray`]
-#[derive(Debug)]
-pub struct GenericBinaryBuilder<OffsetSize: OffsetSizeTrait> {
+///  Array builder for [`GenericByteArray`]
+pub struct GenericByteBuilder<T: ByteArrayType> {
     value_builder: UInt8BufferBuilder,
-    offsets_builder: BufferBuilder<OffsetSize>,
+    offsets_builder: BufferBuilder<T::Offset>,
     null_buffer_builder: NullBufferBuilder,
 }
 
-impl<OffsetSize: OffsetSizeTrait> GenericBinaryBuilder<OffsetSize> {
-    /// Creates a new [`GenericBinaryBuilder`].
+impl<T: ByteArrayType> GenericByteBuilder<T> {
+    /// Creates a new [`GenericByteBuilder`].
     pub fn new() -> Self {
         Self::with_capacity(1024, 1024)
     }
 
-    /// Creates a new [`GenericBinaryBuilder`].
+    /// Creates a new [`GenericByteBuilder`].
     ///
     /// - `item_capacity` is the number of items to pre-allocate.
     ///   The size of the preallocated buffer of offsets is the number of items plus one.
     /// - `data_capacity` is the total number of bytes of string data to pre-allocate

Review Comment:
   ```suggestion
       /// - `data_capacity` is the total number of bytes of data to pre-allocate
   ```



##########
arrow-array/src/builder/generic_bytes_builder.rs:
##########
@@ -17,70 +17,81 @@
 
 use crate::builder::null_buffer_builder::NullBufferBuilder;
 use crate::builder::{ArrayBuilder, BufferBuilder, UInt8BufferBuilder};
-use crate::{ArrayRef, GenericBinaryArray, OffsetSizeTrait};
+use crate::types::{ByteArrayType, GenericBinaryType, GenericStringType};
+use crate::{ArrayRef, GenericByteArray, OffsetSizeTrait};
+use arrow_buffer::ArrowNativeType;
 use arrow_data::ArrayDataBuilder;
 use std::any::Any;
 use std::sync::Arc;
 
-///  Array builder for [`GenericBinaryArray`]
-#[derive(Debug)]
-pub struct GenericBinaryBuilder<OffsetSize: OffsetSizeTrait> {
+///  Array builder for [`GenericByteArray`]
+pub struct GenericByteBuilder<T: ByteArrayType> {
     value_builder: UInt8BufferBuilder,
-    offsets_builder: BufferBuilder<OffsetSize>,
+    offsets_builder: BufferBuilder<T::Offset>,
     null_buffer_builder: NullBufferBuilder,
 }
 
-impl<OffsetSize: OffsetSizeTrait> GenericBinaryBuilder<OffsetSize> {
-    /// Creates a new [`GenericBinaryBuilder`].
+impl<T: ByteArrayType> GenericByteBuilder<T> {
+    /// Creates a new [`GenericByteBuilder`].
     pub fn new() -> Self {
         Self::with_capacity(1024, 1024)
     }
 
-    /// Creates a new [`GenericBinaryBuilder`].
+    /// Creates a new [`GenericByteBuilder`].
     ///
     /// - `item_capacity` is the number of items to pre-allocate.
     ///   The size of the preallocated buffer of offsets is the number of items plus one.
     /// - `data_capacity` is the total number of bytes of string data to pre-allocate
     ///   (for all items, not per item).
     pub fn with_capacity(item_capacity: usize, data_capacity: usize) -> Self {
-        let mut offsets_builder = BufferBuilder::<OffsetSize>::new(item_capacity + 1);
-        offsets_builder.append(OffsetSize::zero());
+        let mut offsets_builder = BufferBuilder::<T::Offset>::new(item_capacity + 1);
+        offsets_builder.append(T::Offset::from_usize(0).unwrap());
         Self {
             value_builder: UInt8BufferBuilder::new(data_capacity),
             offsets_builder,
             null_buffer_builder: NullBufferBuilder::new(item_capacity),
         }
     }
 
-    /// Appends a byte slice into the builder.
+    /// Appends a value into the builder.
     #[inline]
-    pub fn append_value(&mut self, value: impl AsRef<[u8]>) {
-        self.value_builder.append_slice(value.as_ref());
+    pub fn append_value(&mut self, value: impl AsRef<T::Native>) {
+        self.value_builder.append_slice(value.as_ref().as_ref());
         self.null_buffer_builder.append(true);
         self.offsets_builder
-            .append(OffsetSize::from_usize(self.value_builder.len()).unwrap());
+            .append(T::Offset::from_usize(self.value_builder.len()).unwrap());
+    }
+
+    /// Append an `Option` value into the builder.
+    #[inline]
+    pub fn append_option(&mut self, value: Option<impl AsRef<T::Native>>) {
+        match value {
+            None => self.append_null(),
+            Some(v) => self.append_value(v),
+        };
     }
 
     /// Append a null value into the builder.
     #[inline]
     pub fn append_null(&mut self) {
         self.null_buffer_builder.append(false);
         self.offsets_builder
-            .append(OffsetSize::from_usize(self.value_builder.len()).unwrap());
+            .append(T::Offset::from_usize(self.value_builder.len()).unwrap());
     }
 
-    /// Builds the [`GenericBinaryArray`] and reset this builder.
-    pub fn finish(&mut self) -> GenericBinaryArray<OffsetSize> {
-        let array_type = GenericBinaryArray::<OffsetSize>::DATA_TYPE;
+    /// Builds the [`GenericByteBuilder`] and reset this builder.

Review Comment:
   ```suggestion
       /// Builds the [`GenericByteArray`] and reset this builder.
   ```



-- 
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.

To unsubscribe, e-mail: github-unsubscribe@arrow.apache.org

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