mkarbo commented on code in PR #7535: URL: https://github.com/apache/arrow-rs/pull/7535#discussion_r2102984307
########## parquet-variant/src/decoder.rs: ########## @@ -0,0 +1,199 @@ +// NOTE: Largely based on the implementation of @PinkCrow007 in https://github.com/apache/arrow-rs/pull/7452 +// And the feedback there. +use crate::variant::VariantType; +use arrow_schema::ArrowError; +use std::{array::TryFromSliceError, str}; + +#[derive(Debug, Clone, Copy)] +pub enum VariantBasicType { + Primitive = 0, + ShortString = 1, + Object = 2, + Array = 3, +} + +#[derive(Debug, Clone, Copy)] +pub enum VariantPrimitiveType { + Null = 0, + BooleanTrue = 1, + BooleanFalse = 2, + Int8 = 3, + // TODO: Add 'legs' for the rest of primitives, once API is agreed upon + String = 16, +} + +/// Extracts the basic type from a header byte +pub(crate) fn get_basic_type(header: u8) -> Result<VariantBasicType, ArrowError> { + // See https://github.com/apache/parquet-format/blob/master/VariantEncoding.md#value-encoding + let basic_type = header & 0x03; // Basic type is encoded in the first 2 bits + let basic_type = match basic_type { + 0 => VariantBasicType::Primitive, + 1 => VariantBasicType::ShortString, + 2 => VariantBasicType::Object, + 3 => VariantBasicType::Array, + _ => { + return Err(ArrowError::InvalidArgumentError(format!( + "unknown basic type: {}", + basic_type + ))) + } + }; + Ok(basic_type) +} + +/// Extracts the primitive type from a header byte +pub(crate) fn get_primitive_type(header: u8) -> Result<VariantPrimitiveType, ArrowError> { + // See https://github.com/apache/parquet-format/blob/master/VariantEncoding.md#value-encoding + //// Primitive type is encoded in the last 6 bits of the header byte + let primitive_type = (header >> 2) & 0x3F; + let primitive_type = match primitive_type { + 0 => VariantPrimitiveType::Null, + 1 => VariantPrimitiveType::BooleanTrue, + 2 => VariantPrimitiveType::BooleanFalse, + 3 => VariantPrimitiveType::Int8, Review Comment: Thanks for sharing! It's a bit annoying to duplicate, but it's only a few and it's a (hopefully) locked spec so I don't mind doing it, and prefer the readability over the "right" way you show here, so keeping it as it. -- 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-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
