Applied suggestions

src/io/parquet/read/deserialize/binary/basic.rs

@@ -311,28 +311,28 @@ impl<'a, O: Offset> utils::Decoder<'a> for BinaryDecoder<O> {
                 // this is an upper bound - we may not consume the whole page.
                 // we do not know how many nulls are there, so we do not know how many
                 // valid items are there to discount over the sequence.
-                Binary::<O>::with_capacity(capacity, values.num_bytes()),
+                Binary::<O>::with_capacity(capacity, Some(values.num_bytes())),
                 MutableBitmap::with_capacity(capacity),
             ),
             State::Required(values) => (
                 // this is an upper bound - we may not consume the whole page.
-                Binary::<O>::with_capacity(capacity, values.values.num_bytes()),
+                Binary::<O>::with_capacity(capacity, Some(values.values.num_bytes())),
                 MutableBitmap::new(),
             ),
             State::FilteredRequiredDictionary(_)
             | State::FilteredRequired(_)
             | State::RequiredDictionary(_)
             | State::Delta(_)
             | State::FilteredDelta(_) => (
-                Binary::<O>::with_capacity(capacity, 0),
+                Binary::<O>::with_capacity(capacity, None),
                 MutableBitmap::new(),
             ),
             State::FilteredOptionalDictionary(_, _)
             | State::FilteredOptional(_, _)
             | State::OptionalDictionary(_, _)
             | State::OptionalDelta(_, _)
             | State::FilteredOptionalDelta(_, _) => (
-                Binary::<O>::with_capacity(capacity, 0),
+                Binary::<O>::with_capacity(capacity, None),
                 MutableBitmap::with_capacity(capacity),
             ),
         }

src/io/parquet/read/deserialize/binary/dictionary.rs

@@ -59,7 +59,7 @@ fn read_dict<O: Offset>(data_type: DataType, dict: &DictPage) -> Box<dyn Array>
 
     let values = SizedBinaryIter::new(&dict.buffer, dict.num_values);
 
-    let mut data = Binary::<O>::with_capacity(dict.num_values, 0);
+    let mut data = Binary::<O>::with_capacity(dict.num_values, None);
     data.values = Vec::with_capacity(dict.buffer.len() - 4 * dict.num_values);
     for item in values {
         data.push(item)

src/io/parquet/read/deserialize/binary/nested.rs

@@ -85,7 +85,7 @@ impl<'a, O: Offset> NestedDecoder<'a> for BinaryDecoder<O> {
 
     fn with_capacity(&self, capacity: usize) -> Self::DecodedState {
         (
-            Binary::<O>::with_capacity(capacity, 0),
+            Binary::<O>::with_capacity(capacity, None),
             MutableBitmap::with_capacity(capacity),
         )
     }

src/io/parquet/read/deserialize/binary/utils.rs

@@ -36,7 +36,8 @@ impl<O: Offset> Pushable<usize> for Offsets<O> {
 
 impl<O: Offset> Binary<O> {
     #[inline]
-    pub fn with_capacity(capacity: usize, values_capacity: usize) -> Self {
+    pub fn with_capacity(capacity: usize, values_capacity: Option<usize>) -> Self {
+        let values_capacity = values_capacity.unwrap_or(capacity.min(100) * 24);
         Self {
             offsets: Offsets::with_capacity(capacity),
             values: Vec::with_capacity(values_capacity),