alamb commented on code in PR #17275:
URL: https://github.com/apache/datafusion/pull/17275#discussion_r2291910112
##########
datafusion/datasource-parquet/src/opener.rs:
##########
@@ -535,8 +535,8 @@ async fn load_page_index<T: AsyncFileReader>(
if missing_column_index || missing_offset_index {
let m = Arc::try_unwrap(Arc::clone(parquet_metadata))
.unwrap_or_else(|e| e.as_ref().clone());
- let mut reader =
-
ParquetMetaDataReader::new_with_metadata(m).with_page_indexes(true);
+ let mut reader = ParquetMetaDataReader::new_with_metadata(m)
Review Comment:
Due to this change from @kczimm
- https://github.com/apache/arrow-rs/issues/8070
##########
Cargo.toml:
##########
@@ -90,62 +90,62 @@ ahash = { version = "0.8", default-features = false,
features = [
"runtime-rng",
] }
apache-avro = { version = "0.17", default-features = false }
-arrow = { version = "56.0.0", features = [
+arrow = { version = "56.1.0", features = [
"prettyprint",
"chrono-tz",
] }
-arrow-buffer = { version = "56.0.0", default-features = false }
-arrow-flight = { version = "56.0.0", features = [
+arrow-buffer = { version = "56.1.0", default-features = false }
+arrow-flight = { version = "56.1.0", features = [
"flight-sql-experimental",
] }
-arrow-ipc = { version = "56.0.0", default-features = false, features = [
+arrow-ipc = { version = "56.1.0", default-features = false, features = [
"lz4",
] }
-arrow-ord = { version = "56.0.0", default-features = false }
-arrow-schema = { version = "56.0.0", default-features = false }
+arrow-ord = { version = "56.1.0", default-features = false }
+arrow-schema = { version = "56.1.0", default-features = false }
async-trait = "0.1.89"
bigdecimal = "0.4.8"
bytes = "1.10"
chrono = { version = "0.4.41", default-features = false }
criterion = "0.5.1"
ctor = "0.4.3"
dashmap = "6.0.1"
-datafusion = { path = "datafusion/core", version = "49.0.0", default-features
= false }
-datafusion-catalog = { path = "datafusion/catalog", version = "49.0.0" }
-datafusion-catalog-listing = { path = "datafusion/catalog-listing", version =
"49.0.0" }
-datafusion-common = { path = "datafusion/common", version = "49.0.0",
default-features = false }
-datafusion-common-runtime = { path = "datafusion/common-runtime", version =
"49.0.0" }
-datafusion-datasource = { path = "datafusion/datasource", version = "49.0.0",
default-features = false }
-datafusion-datasource-avro = { path = "datafusion/datasource-avro", version =
"49.0.0", default-features = false }
-datafusion-datasource-csv = { path = "datafusion/datasource-csv", version =
"49.0.0", default-features = false }
-datafusion-datasource-json = { path = "datafusion/datasource-json", version =
"49.0.0", default-features = false }
-datafusion-datasource-parquet = { path = "datafusion/datasource-parquet",
version = "49.0.0", default-features = false }
-datafusion-doc = { path = "datafusion/doc", version = "49.0.0" }
-datafusion-execution = { path = "datafusion/execution", version = "49.0.0" }
-datafusion-expr = { path = "datafusion/expr", version = "49.0.0" }
-datafusion-expr-common = { path = "datafusion/expr-common", version = "49.0.0"
}
-datafusion-ffi = { path = "datafusion/ffi", version = "49.0.0" }
-datafusion-functions = { path = "datafusion/functions", version = "49.0.0" }
-datafusion-functions-aggregate = { path = "datafusion/functions-aggregate",
version = "49.0.0" }
-datafusion-functions-aggregate-common = { path =
"datafusion/functions-aggregate-common", version = "49.0.0" }
-datafusion-functions-nested = { path = "datafusion/functions-nested", version
= "49.0.0" }
-datafusion-functions-table = { path = "datafusion/functions-table", version =
"49.0.0" }
-datafusion-functions-window = { path = "datafusion/functions-window", version
= "49.0.0" }
-datafusion-functions-window-common = { path =
"datafusion/functions-window-common", version = "49.0.0" }
-datafusion-macros = { path = "datafusion/macros", version = "49.0.0" }
-datafusion-optimizer = { path = "datafusion/optimizer", version = "49.0.0",
default-features = false }
-datafusion-physical-expr = { path = "datafusion/physical-expr", version =
"49.0.0", default-features = false }
-datafusion-physical-expr-adapter = { path =
"datafusion/physical-expr-adapter", version = "49.0.0", default-features =
false }
-datafusion-physical-expr-common = { path = "datafusion/physical-expr-common",
version = "49.0.0", default-features = false }
-datafusion-physical-optimizer = { path = "datafusion/physical-optimizer",
version = "49.0.0" }
-datafusion-physical-plan = { path = "datafusion/physical-plan", version =
"49.0.0" }
-datafusion-proto = { path = "datafusion/proto", version = "49.0.0" }
-datafusion-proto-common = { path = "datafusion/proto-common", version =
"49.0.0" }
-datafusion-pruning = { path = "datafusion/pruning", version = "49.0.0" }
-datafusion-session = { path = "datafusion/session", version = "49.0.0" }
-datafusion-spark = { path = "datafusion/spark", version = "49.0.0" }
-datafusion-sql = { path = "datafusion/sql", version = "49.0.0" }
-datafusion-substrait = { path = "datafusion/substrait", version = "49.0.0" }
+datafusion = { path = "datafusion/core", version = "49.0.1", default-features
= false }
Review Comment:
drive by change to update all versions in Cargo.toml to the latest
##########
datafusion-cli/src/main.rs:
##########
@@ -570,15 +570,15 @@ mod tests {
let df = ctx.sql(sql).await?;
let rbs = df.collect().await?;
- assert_snapshot!(batches_to_string(&rbs),@r#"
+ assert_snapshot!(batches_to_string(&rbs),@r"
+-----------------------------------+-----------------+---------------------+------+------------------+
| filename | file_size_bytes |
metadata_size_bytes | hits | extra |
+-----------------------------------+-----------------+---------------------+------+------------------+
| alltypes_plain.parquet | 1851 | 10181
| 2 | page_index=false |
- | alltypes_tiny_pages.parquet | 454233 | 881634
| 2 | page_index=true |
+ | alltypes_tiny_pages.parquet | 454233 | 881418
| 2 | page_index=true |
Review Comment:
I don't really know why the in-memory size of the ParquetMetadata has
decreased, but it seems like a good improvement to me
--
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]
