Skip to content

Delay Iceberg manifest initialization until after pruning #996

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
Kuinox wants to merge 3 commits into from
+11 −13
Closed

Delay Iceberg manifest initialization until after pruning #996

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
a7a18e3
Delay Iceberg manifest initialization until after pruning
Kuinox May 19, 2026
883c4ef
Update Iceberg request logging test
Kuinox May 27, 2026
f147557
Merge remote-tracking branch 'upstream/main' into delay-init
Tishj Jul 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 9 additions & 11 deletions src/planning/iceberg_multi_file_list.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -353,10 +353,6 @@ void IcebergMultiFileList::Bind(vector<LogicalType> &return_types, vector<Identi
std::move(temp_data), snapshot_info, *schema);
}

if (!view_initialized) {
InitializeFiles(guard);
}

auto &schema = GetSchema().columns;
for (auto &schema_entry : schema) {
names.push_back(Identifier(schema_entry->name));
Expand Down Expand Up @@ -466,10 +462,6 @@ vector<OpenFileInfo> IcebergMultiFileList::GetAllFiles() const {
}

FileExpandResult IcebergMultiFileList::GetExpandResult() const {
// GetFileInternal(1) will ensure files with index 0 and index 1 are expanded if they are available
lock_guard<mutex> guard(shared_state->lock);
GetFileInternal(1, guard);

// always return multiple files, In the case there is only 1 data file,
// we only lose performance if it is small
return FileExpandResult::MULTIPLE_FILES;
Expand All @@ -493,9 +485,10 @@ unique_ptr<NodeStatistics> IcebergMultiFileList::GetCardinality(ClientContext &c
return nullptr;
}

//! Make sure we have fetched all manifests
(void)GetTotalFileCount();
D_ASSERT(view_initialized);
if (!view_initialized) {
//! Avoid forcing manifest initialization before filter pushdown has had a chance to narrow the table.
return nullptr;
}

idx_t cardinality = 0;
for (idx_t i = 0; i < data_manifests.size(); i++) {
Expand Down Expand Up @@ -545,6 +538,11 @@ void IcebergMultiFileList::GetStatistics(vector<PartitionStatistics> &result) co
return;
}

if (!view_initialized) {
//! Avoid forcing manifest initialization before filter pushdown has had a chance to narrow the table.
return;
}

for (idx_t i = 0; i < delete_manifests.size(); i++) {
if (delete_manifest_matches[i]) {
//! if a matching delete manifest exists, return;
Expand Down
4 changes: 2 additions & 2 deletions ...cal/catalog_custom_setup/fixture/enable_logging_http/test_table_information_requests.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -78,8 +78,6 @@ with cte as (
)
select CASE WHEN contains(val, '.') THEN (split(val, '.')[2]) ELSE NULL END FROM CTE ORDER BY ALL
----
avro
avro
NULL

statement ok
Expand Down Expand Up @@ -126,6 +124,8 @@ with cte as (
)
select CASE WHEN contains(val, '.') THEN (split(val, '.')[2]) ELSE NULL END FROM CTE ORDER BY ALL
----
avro
avro
parquet
NULL

Expand Down
Loading