mbutrovich commented on code in PR #3224: URL: https://github.com/apache/iceberg-rust/pull/3224#discussion_r4259822125
########## crates/iceberg/tests/manifest_list_memory.rs: ########## @@ -0,0 +1,200 @@ +// 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. + +//! Manifest-list compatibility and peak live-allocation regression. + +use std::alloc::{GlobalAlloc, Layout, System}; +use std::sync::atomic::AtomicUsize; +use std::sync::atomic::Ordering::Relaxed; +use std::time::Instant; + +use apache_avro::types::Value; +use apache_avro::{Codec, Reader, Writer}; +use iceberg::io::FileIO; +use iceberg::spec::{ + FieldSummary, FormatVersion, ManifestContentType, ManifestFile, ManifestList, + ManifestListWriter, +}; + +struct CountingAllocator; +static LIVE: AtomicUsize = AtomicUsize::new(0); +static PEAK: AtomicUsize = AtomicUsize::new(0); + +fn allocated(bytes: usize) { + PEAK.fetch_max(LIVE.fetch_add(bytes, Relaxed) + bytes, Relaxed); +} + +// The allocator is confined to this test executable. Every allocation is tracked, +// including those made before measurement, so deallocation cannot underflow LIVE. +unsafe impl GlobalAlloc for CountingAllocator { + unsafe fn alloc(&self, layout: Layout) -> *mut u8 { + let ptr = unsafe { System.alloc(layout) }; + if !ptr.is_null() { + allocated(layout.size()); + } + ptr + } + + unsafe fn alloc_zeroed(&self, layout: Layout) -> *mut u8 { + let ptr = unsafe { System.alloc_zeroed(layout) }; + if !ptr.is_null() { + allocated(layout.size()); + } + ptr + } + + unsafe fn dealloc(&self, ptr: *mut u8, layout: Layout) { + unsafe { System.dealloc(ptr, layout) }; + LIVE.fetch_sub(layout.size(), Relaxed); + } + + unsafe fn realloc(&self, ptr: *mut u8, layout: Layout, size: usize) -> *mut u8 { + let new_ptr = unsafe { System.realloc(ptr, layout, size) }; + if !new_ptr.is_null() { + LIVE.fetch_sub(layout.size(), Relaxed); + allocated(size); + } + new_ptr + } +} + +#[global_allocator] +static ALLOCATOR: CountingAllocator = CountingAllocator; Review Comment: > Those cases aren't covered there today, and as separate tests a failure in one wouldn't hide the others. I got part of this wrong, sorry for the churn. `test_parse_manifest_list_v1`, `test_parse_manifest_list_v2` and `test_parse_manifest_list_v3` in the same module already write a list with `ManifestListWriter` and parse it back. [`test_manifest_list_writer_v1_as_v2`](https://github.com/apache/iceberg-rust/blob/3db218ec660658ca33acfebcf849006d727065c9/crates/iceberg/src/spec/manifest_list/writer.rs#L460-L503) in `writer.rs` writes a V1 list and parses it as V2, which is what `test_parse_manifest_list_v1_as_v2` checks. Only the empty and truncated cases are new. What do you think about dropping `test_parse_manifest_list_round_trip` and `test_parse_manifest_list_v1_as_v2`? The round trip does add inputs the existing tests don't have: a non-empty `key_metadata`, and a null `contains_nan` and `upper_bound`, which go through the union handling in `Resolved`. Setting those on an entry in the existing `test_parse_manifest_list_v*` tests would keep that coverage without a second set of round trips. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
