mkleen commented on code in PR #22613:
URL: https://github.com/apache/datafusion/pull/22613#discussion_r3329859014
##########
datafusion/execution/src/cache/cache_manager.rs:
##########
Review Comment:
Yes, thanks.
##########
datafusion/execution/src/cache/cache_manager.rs:
##########
@@ -78,37 +106,23 @@ impl CachedFileMetadata {
}
}
-/// A cache for file statistics and orderings.
-///
-/// This cache stores [`CachedFileMetadata`] which includes:
-/// - File metadata for validation (size, last_modified)
-/// - Statistics for the file
-/// - Ordering information for the file
-///
-/// If enabled via [`CacheManagerConfig::with_file_statistics_cache`] this
-/// cache avoids inferring the same file statistics repeatedly during the
-/// session lifetime.
-///
-/// The typical usage pattern is:
-/// 1. Call `get(path)` to check for cached value
-/// 2. If `Some(cached)`, validate with `cached.is_valid_for(¤t_meta)`
-/// 3. If invalid or missing, compute new value and call `put(path, new_value)`
-///
-/// See [`crate::runtime_env::RuntimeEnv`] for more details
-pub trait FileStatisticsCache:
- CacheAccessor<TableScopedPath, CachedFileMetadata>
-{
- /// Cache memory limit in bytes.
- fn cache_limit(&self) -> usize;
+impl Value for CachedFileMetadata {
+ fn size(&self) -> usize {
+ DFHeapSize::heap_size(self, &mut DFHeapSizeCtx::default())
+ }
+}
- /// Updates the cache with a new memory limit in bytes.
- fn update_cache_limit(&self, limit: usize);
+pub const DEFAULT_LIST_FILES_CACHE_MEMORY_LIMIT: usize = 1024 * 1024; // 1MiB
- /// Retrieves the information about the entries currently cached.
- fn list_entries(&self) -> HashMap<TableScopedPath,
FileStatisticsCacheEntry>;
+pub const DEFAULT_LIST_FILES_CACHE_TTL: Option<Duration> = None; // Infinite
- fn drop_table_entries(&self, table_ref: &Option<TableReference>) ->
Result<()>;
-}
+pub const DEFAULT_FILE_STATISTICS_MEMORY_LIMIT: usize = 20 * 1024 * 1024; //
20MiB
+
+pub const DEFAULT_METADATA_CACHE_LIMIT: usize = 50 * 1024 * 1024; // 50M
Review Comment:
Yes.
--
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]
