Skip to content

Commit 011512a

Browse files
committed
fix: support reading compressed metadata
The spec mentions that metadata files "may be compressed with GZIP", here: https://iceberg.apache.org/spec/#table-metadata-and-snapshots
1 parent d3d3127 commit 011512a

File tree

4 files changed

+47
-2
lines changed

4 files changed

+47
-2
lines changed

Cargo.lock

Lines changed: 1 addition & 0 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

Cargo.toml

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -71,6 +71,7 @@ enum-ordinalize = "4.3.0"
7171
env_logger = "0.11.8"
7272
expect-test = "1"
7373
faststr = "0.2.31"
74+
flate2 = "1.1.5"
7475
fnv = "1.0.7"
7576
fs-err = "3.1.0"
7677
futures = "0.3"

crates/iceberg/Cargo.toml

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -63,6 +63,7 @@ bytes = { workspace = true }
6363
chrono = { workspace = true }
6464
derive_builder = { workspace = true }
6565
expect-test = { workspace = true }
66+
flate2 = { workspace = true }
6667
fnv = { workspace = true }
6768
futures = { workspace = true }
6869
itertools = { workspace = true }

crates/iceberg/src/spec/table_metadata.rs

Lines changed: 44 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -22,10 +22,12 @@ use std::cmp::Ordering;
2222
use std::collections::HashMap;
2323
use std::fmt::{Display, Formatter};
2424
use std::hash::Hash;
25+
use std::io::Read as _;
2526
use std::sync::Arc;
2627

2728
use _serde::TableMetadataEnum;
2829
use chrono::{DateTime, Utc};
30+
use flate2::read::GzDecoder;
2931
use serde::{Deserialize, Serialize};
3032
use serde_repr::{Deserialize_repr, Serialize_repr};
3133
use uuid::Uuid;
@@ -413,9 +415,24 @@ impl TableMetadata {
413415
file_io: &FileIO,
414416
metadata_location: impl AsRef<str>,
415417
) -> Result<TableMetadata> {
416-
let input_file = file_io.new_input(metadata_location)?;
418+
let input_file = file_io.new_input(metadata_location.as_ref())?;
417419
let metadata_content = input_file.read().await?;
418-
let metadata = serde_json::from_slice::<TableMetadata>(&metadata_content)?;
420+
421+
// Check if the file is compressed by looking for the gzip "magic number".
422+
let metadata = if metadata_content.len() > 2
423+
&& metadata_content[0] == 0x1F
424+
&& metadata_content[1] == 0x8B
425+
{
426+
let mut decoder = GzDecoder::new(metadata_content.as_ref());
427+
let mut decompressed_data = Vec::new();
428+
decoder
429+
.read_to_end(&mut decompressed_data)
430+
.map_err(|e| Error::new(ErrorKind::DataInvalid, e.to_string()))?;
431+
serde_json::from_slice(&decompressed_data)?
432+
} else {
433+
serde_json::from_slice(&metadata_content)?
434+
};
435+
419436
Ok(metadata)
420437
}
421438

@@ -1318,6 +1335,7 @@ impl SnapshotLog {
13181335
mod tests {
13191336
use std::collections::HashMap;
13201337
use std::fs;
1338+
use std::io::Write as _;
13211339
use std::sync::Arc;
13221340

13231341
use anyhow::Result;
@@ -3053,6 +3071,30 @@ mod tests {
30533071
assert_eq!(read_metadata, original_metadata);
30543072
}
30553073

3074+
#[tokio::test]
3075+
async fn test_table_metadata_read_compressed() {
3076+
let temp_dir = TempDir::new().unwrap();
3077+
let metadata_location = temp_dir.path().join("v1.gz.metadata.json");
3078+
3079+
let original_metadata: TableMetadata = get_test_table_metadata("TableMetadataV2Valid.json");
3080+
let json = serde_json::to_string(&original_metadata).unwrap();
3081+
3082+
let mut encoder = flate2::write::GzEncoder::new(Vec::new(), flate2::Compression::default());
3083+
encoder.write_all(json.as_bytes()).unwrap();
3084+
std::fs::write(&metadata_location, encoder.finish().unwrap())
3085+
.expect("failed to write metadata");
3086+
3087+
// Read the metadata back
3088+
let file_io = FileIOBuilder::new_fs_io().build().unwrap();
3089+
let metadata_location = metadata_location.to_str().unwrap();
3090+
let read_metadata = TableMetadata::read_from(&file_io, metadata_location)
3091+
.await
3092+
.unwrap();
3093+
3094+
// Verify the metadata matches
3095+
assert_eq!(read_metadata, original_metadata);
3096+
}
3097+
30563098
#[tokio::test]
30573099
async fn test_table_metadata_read_nonexistent_file() {
30583100
// Create a FileIO instance

0 commit comments

Comments
 (0)