Remove functions from parquet crate deprecated in or before 54.0.0 (#7811)

# Which issue does this PR close?

Part of #7810

# Rationale for this change

Reducing code bloat.

# What changes are included in this PR?

Remove deprecated functions and structs.

# Are these changes tested?

Removed functions were not referenced so no tests necessary

# Are there any user-facing changes?

Yes, public functions are removed.

Author: etseidl

parquet/src/arrow/async_reader/metadata.rs

@@ -77,7 +77,7 @@ impl ParquetObjectReader {
     }
 
     /// Provide a hint as to the size of the parquet file's footer,
-    /// see [fetch_parquet_metadata](crate::arrow::async_reader::fetch_parquet_metadata)
+    /// see [`ParquetMetaDataReader::with_prefetch_hint`]
     pub fn with_footer_size_hint(self, hint: usize) -> Self {
         Self {
             metadata_size_hint: Some(hint),

parquet/src/arrow/async_reader/store.rs

@@ -202,9 +202,6 @@ pub use self::async_reader::ParquetRecordBatchStreamBuilder;
 pub use self::async_writer::AsyncArrowWriter;
 use crate::schema::types::{SchemaDescriptor, Type};
 use arrow_schema::{FieldRef, Schema};
-// continue to export deprecated methods until they are removed
-#[allow(deprecated)]
-pub use self::schema::arrow_to_parquet_schema;
 
 pub use self::schema::{
     add_encoded_arrow_schema_to_metadata, encode_arrow_schema, parquet_to_arrow_field_levels,

parquet/src/arrow/mod.rs

@@ -358,15 +358,6 @@ impl<'a> ArrowSchemaConverter<'a> {
     }
 }
 
-/// Convert arrow schema to parquet schema
-///
-/// The name of the root schema element defaults to `"arrow_schema"`, this can be
-/// overridden with [`ArrowSchemaConverter`]
-#[deprecated(since = "54.0.0", note = "Use `ArrowSchemaConverter` instead")]
-pub fn arrow_to_parquet_schema(schema: &Schema) -> Result<SchemaDescriptor> {
-    ArrowSchemaConverter::new().convert(schema)
-}
-
 fn parse_key_value_metadata(
     key_value_metadata: Option<&Vec<KeyValue>>,
 ) -> Option<HashMap<String, String>> {

parquet/src/arrow/schema/mod.rs

@@ -102,16 +102,6 @@ impl ColumnWriter<'_> {
     }
 }
 
-#[deprecated(
-    since = "54.0.0",
-    note = "Seems like a stray and nobody knows what's it for. Will be removed in the next release."
-)]
-#[allow(missing_docs)]
-pub enum Level {
-    Page,
-    Column,
-}
-
 /// Gets a specific column writer corresponding to column descriptor `descr`.
 pub fn get_column_writer<'a>(
     descr: ColumnDescPtr,

parquet/src/column/writer/mod.rs

@@ -74,12 +74,6 @@ impl Int96 {
         self.value = [elem0, elem1, elem2];
     }
 
-    /// Converts this INT96 into an i64 representing the number of MILLISECONDS since Epoch
-    #[deprecated(since = "54.0.0", note = "Use `to_millis` instead")]
-    pub fn to_i64(&self) -> i64 {
-        self.to_millis()
-    }
-
     /// Converts this INT96 into an i64 representing the number of SECONDS since EPOCH
     ///
     /// Will wrap around on overflow
@@ -1214,26 +1208,6 @@ pub trait DataType: 'static + Send {
         Self: Sized;
 }
 
-// Workaround bug in specialization
-#[deprecated(
-    since = "54.0.0",
-    note = "Seems like a stray and nobody knows what's it for. Will be removed in 55.0.0"
-)]
-#[allow(missing_docs)]
-pub trait SliceAsBytesDataType: DataType
-where
-    Self::T: SliceAsBytes,
-{
-}
-
-#[allow(deprecated)]
-impl<T> SliceAsBytesDataType for T
-where
-    T: DataType,
-    <T as DataType>::T: SliceAsBytes,
-{
-}
-
 macro_rules! make_type {
     ($name:ident, $reader_ident: ident, $writer_ident: ident, $native_ty:ty, $size:expr) => {
         #[doc = concat!("Parquet physical type: ", stringify!($name))]

parquet/src/data_type.rs

@@ -208,22 +208,6 @@ impl ParquetMetaData {
         self.file_decryptor = file_decryptor;
     }
 
-    /// Creates Parquet metadata from file metadata, a list of row
-    /// group metadata, and the column index structures.
-    #[deprecated(since = "53.1.0", note = "Use ParquetMetaDataBuilder")]
-    pub fn new_with_page_index(
-        file_metadata: FileMetaData,
-        row_groups: Vec<RowGroupMetaData>,
-        column_index: Option<ParquetColumnIndex>,
-        offset_index: Option<ParquetOffsetIndex>,
-    ) -> Self {
-        ParquetMetaDataBuilder::new(file_metadata)
-            .set_row_groups(row_groups)
-            .set_column_index(column_index)
-            .set_offset_index(offset_index)
-            .build()
-    }
-
     /// Convert this ParquetMetaData into a [`ParquetMetaDataBuilder`]
     pub fn into_builder(self) -> ParquetMetaDataBuilder {
         self.into()
@@ -1397,20 +1381,6 @@ impl ColumnChunkMetaDataBuilder {
         self
     }
 
-    /// Sets file offset in bytes.
-    ///
-    /// This field was meant to provide an alternate to storing `ColumnMetadata` directly in
-    /// the `ColumnChunkMetadata`. However, most Parquet readers assume the `ColumnMetadata`
-    /// is stored inline and ignore this field.
-    #[deprecated(
-        since = "53.0.0",
-        note = "The Parquet specification requires this field to be 0"
-    )]
-    pub fn set_file_offset(mut self, value: i64) -> Self {
-        self.0.file_offset = value;
-        self
-    }
-
     /// Sets number of values.
     pub fn set_num_values(mut self, value: i64) -> Self {
         self.0.num_values = value;

parquet/src/file/footer.rs

@@ -99,7 +99,6 @@
 //! ```
 #[cfg(feature = "encryption")]
 pub mod column_crypto_metadata;
-pub mod footer;
 pub mod metadata;
 pub mod page_encoding_stats;
 pub mod page_index;

parquet/src/file/metadata/mod.rs

@@ -15,7 +15,7 @@
 // specific language governing permissions and limitations
 // under the License.
 
-//! Support for reading [`Index`] and [`PageLocation`] from parquet metadata.
+//! Support for reading [`Index`] and [`OffsetIndex`] from parquet metadata.
 
 use crate::basic::Type;
 use crate::data_type::Int96;
@@ -24,7 +24,7 @@ use crate::file::metadata::ColumnChunkMetaData;
 use crate::file::page_index::index::{Index, NativeIndex};
 use crate::file::page_index::offset_index::OffsetIndexMetaData;
 use crate::file::reader::ChunkReader;
-use crate::format::{ColumnIndex, OffsetIndex, PageLocation};
+use crate::format::{ColumnIndex, OffsetIndex};
 use crate::thrift::{TCompactSliceInputProtocol, TSerializable};
 use std::ops::Range;
 
@@ -83,45 +83,6 @@ pub fn read_columns_indexes<R: ChunkReader>(
     .transpose()
 }
 
-/// Reads [`OffsetIndex`],  per-page [`PageLocation`] for all columns of a row
-/// group.
-///
-/// Returns a vector of `location[column_number][page_number]`
-///
-/// Return an empty vector if this row group does not contain an
-/// [`OffsetIndex]`.
-///
-/// See [Page Index Documentation] for more details.
-///
-/// [Page Index Documentation]: https://github.com/apache/parquet-format/blob/master/PageIndex.md
-#[deprecated(since = "53.0.0", note = "Use read_offset_indexes")]
-pub fn read_pages_locations<R: ChunkReader>(
-    reader: &R,
-    chunks: &[ColumnChunkMetaData],
-) -> Result<Vec<Vec<PageLocation>>, ParquetError> {
-    let fetch = chunks
-        .iter()
-        .fold(None, |range, c| acc_range(range, c.offset_index_range()));
-
-    let fetch = match fetch {
-        Some(r) => r,
-        None => return Ok(vec![]),
-    };
-
-    let bytes = reader.get_bytes(fetch.start as _, (fetch.end - fetch.start).try_into()?)?;
-
-    chunks
-        .iter()
-        .map(|c| match c.offset_index_range() {
-            Some(r) => decode_page_locations(
-                &bytes[usize::try_from(r.start - fetch.start)?
-                    ..usize::try_from(r.end - fetch.start)?],
-            ),
-            None => Err(general_err!("missing offset index")),
-        })
-        .collect()
-}
-
 /// Reads per-column [`OffsetIndexMetaData`] for all columns of a row group by
 /// decoding [`OffsetIndex`] .
 ///
@@ -172,12 +133,6 @@ pub(crate) fn decode_offset_index(data: &[u8]) -> Result<OffsetIndexMetaData, Pa
     OffsetIndexMetaData::try_new(offset)
 }
 
-pub(crate) fn decode_page_locations(data: &[u8]) -> Result<Vec<PageLocation>, ParquetError> {
-    let mut prot = TCompactSliceInputProtocol::new(data);
-    let offset = OffsetIndex::read_from_in_protocol(&mut prot)?;
-    Ok(offset.page_locations)
-}
-
 pub(crate) fn decode_column_index(data: &[u8], column_type: Type) -> Result<Index, ParquetError> {
     let mut prot = TCompactSliceInputProtocol::new(data);