Add RecordReader trait and proc macro to implement it for a struct

parquet/src/record/mod.rs

@@ -19,6 +19,7 @@
 
 mod api;
 pub mod reader;
+mod record_reader;
 mod record_writer;
 mod triplet;
 
@@ -27,5 +28,6 @@ pub use self::{
         Field, List, ListAccessor, Map, MapAccessor, Row, RowAccessor, RowColumnIter,
         RowFormatter,
     },
+    record_reader::RecordReader,
     record_writer::RecordWriter,
 };

parquet/src/record/record_reader.rs

@@ -0,0 +1,27 @@
+// 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.
+
+use super::super::errors::ParquetError;
+use super::super::file::reader::RowGroupReader;
+
+pub trait RecordReader<T> {
+    fn read_from_row_group(
+        &mut self,
+        row_group_reader: &mut dyn RowGroupReader,
+        max_records: usize,
+    ) -> Result<(), ParquetError>;
+}

parquet_derive/README.md

@@ -19,9 +19,9 @@
 
 # Parquet Derive
 
-A crate for deriving `RecordWriter` for arbitrary, _simple_ structs. This does not generate writers for arbitrarily nested
-structures. It only works for primitives and a few generic structures and
-various levels of reference. Please see features checklist for what is currently
+A crate for deriving `RecordWriter` and `RecordReader` for arbitrary, _simple_ structs. This does not
+generate readers or writers for arbitrarily nested structures. It only works for primitives and a few
+generic structures and various levels of reference. Please see features checklist for what is currently
 supported.
 
 Derive also has some support for the chrono time library. You must must enable the `chrono` feature to get this support.
@@ -77,16 +77,55 @@ writer.close_row_group(row_group).unwrap();
 writer.close().unwrap();
 ```
 
+Example usage of deriving a `RecordReader` for your struct:
+
+```rust
+use parquet::file::{serialized_reader::SerializedFileReader, reader::FileReader};
+use parquet_derive::ParquetRecordReader;
+
+#[derive(ParquetRecordReader)]
+struct ACompleteRecord {
+    pub a_bool: bool,
+    pub a_string: String,
+    pub i16: i16,
+    pub i32: i32,
+    pub u64: u64,
+    pub isize: isize,
+    pub float: f32,
+    pub double: f64,
+    pub now: chrono::NaiveDateTime,
+    pub byte_vec: Vec<u8>,
+}
+
+// Initialize your parquet file
+let reader = SerializedFileReader::new(file).unwrap();
+let mut row_group = reader.get_row_group(0).unwrap();
+
+// create your records to read into
+let mut chunks = vec![ACompleteRecord{ ... }];
+
+// The derived `RecordReader` takes over here
+chunks.as_mut_slice().read_from_row_group(&mut *row_group, 2).unwrap();
+```
+
 ## Features
 
 - [x] Support writing `String`, `&str`, `bool`, `i32`, `f32`, `f64`, `Vec<u8>`
 - [ ] Support writing dictionaries
 - [x] Support writing logical types like timestamp
-- [x] Derive definition_levels for `Option`
-- [ ] Derive definition levels for nested structures
+- [x] Derive definition_levels for `Option` for writing
+- [ ] Derive definition levels for nested structures for writing
 - [ ] Derive writing tuple struct
 - [ ] Derive writing `tuple` container types
 
+- [x] Support reading `String`, `bool`, `i32`, `f32`, `f64`, `Vec<u8>`
+- [ ] Support reading/writing dictionaries
+- [x] Support reading/writing logical types like timestamp
+- [ ] Derive definition_levels for `Option` for reading
+- [ ] Derive definition levels for nested structures for reading
+- [ ] Derive reading/writing tuple struct
+- [ ] Derive reading/writing `tuple` container types
+
 ## Requirements
 
 - Same as `parquet-rs`
@@ -103,4 +142,4 @@ To compile and view in the browser, run `cargo doc --no-deps --open`.
 
 ## License
 
-Licensed under the Apache License, Version 2.0: http://www.apache.org/licenses/LICENSE-2.0.
+Licensed under the Apache License, Version 2.0: http://www.apache.org/licenses/LICENSE-2.0.

parquet_derive/src/lib.rs

@@ -44,7 +44,7 @@ mod parquet_field;
 /// use parquet::file::writer::SerializedFileWriter;
 ///
 /// use std::sync::Arc;
-//
+///
 /// #[derive(ParquetRecordWriter)]
 /// struct ACompleteRecord<'a> {
 ///   pub a_bool: bool,
@@ -137,3 +137,87 @@ pub fn parquet_record_writer(input: proc_macro::TokenStream) -> proc_macro::Toke
     }
   }).into()
 }
+
+/// Derive flat, simple RecordReader implementations. Works by parsing
+/// a struct tagged with `#[derive(ParquetRecordReader)]` and emitting
+/// the correct writing code for each field of the struct. Column readers
+/// are generated in the order they are defined.
+///
+/// It is up to the programmer to keep the order of the struct
+/// fields lined up with the schema.
+///
+/// Example:
+///
+/// ```ignore
+/// use parquet::file::{serialized_reader::SerializedFileReader, reader::FileReader};
+/// use parquet_derive::{ParquetRecordReader};
+///
+/// #[derive(ParquetRecordReader)]
+/// struct ACompleteRecord {
+///     pub a_bool: bool,
+///     pub a_string: String,
+/// }
+///
+/// pub fn read_some_records() -> ACompleteRecord {
+///   let mut samples = vec![
+///     ACompleteRecord {
+///       a_bool: true,
+///       a_string: String::from("I'm true");
+///     }
+///   ];
+///
+///   let reader = SerializedFileReader::new(file).unwrap();
+///   let mut row_group = reader.get_row_group(0).unwrap();
+///   samples.as_mut_slice().read_from_row_group(&mut *row_group, 2).unwrap();
+///   samples
+/// }
+/// ```
+///
+#[proc_macro_derive(ParquetRecordReader)]
+pub fn parquet_record_reader(input: proc_macro::TokenStream) -> proc_macro::TokenStream {
+    let input: DeriveInput = parse_macro_input!(input as DeriveInput);
+    let fields = match input.data {
+        Data::Struct(DataStruct { fields, .. }) => fields,
+        Data::Enum(_) => unimplemented!("Enum currently is not supported"),
+        Data::Union(_) => unimplemented!("Union currently is not supported"),
+    };
+
+    let field_infos: Vec<_> = fields.iter().map(parquet_field::Field::from).collect();
+    let reader_snippets: Vec<proc_macro2::TokenStream> =
+        field_infos.iter().map(|x| x.reader_snippet()).collect();
+    let i: Vec<_> = (0..reader_snippets.len()).collect();
+
+    let derived_for = input.ident;
+    let generics = input.generics;
+
+    let convertable = parquet_field::get_convertable_quote();
+
+    (quote! {
+    impl #generics ::parquet::record::RecordReader<#derived_for #generics> for &mut [#derived_for #generics] {
+      fn read_from_row_group(
+        &mut self,
+        row_group_reader: &mut dyn ::parquet::file::reader::RowGroupReader,
+        max_records: usize,
+      ) -> Result<(), ::parquet::errors::ParquetError> {
+        use ::parquet::column::reader::ColumnReader;
+
+        #convertable
+
+        let mut row_group_reader = row_group_reader;
+        let records = self; // Used by all the reader snippets to be more clear
+
+        #(
+          {
+              if let Ok(mut column_reader) = row_group_reader.get_column_reader(#i) {
+                  #reader_snippets
+              } else {
+                  return Err(::parquet::errors::ParquetError::General("Failed to get next column".into()))
+              }
+          }
+        );*
+
+        Ok(())
+      }
+    }
+  }).into()
+}