Skip to content

Support parquet page filtering on min_max for decimal128 and string columns #4255

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

Merged
alamb merged 8 commits into from
Nov 22, 2022
Merged

Support parquet page filtering on min_max for decimal128 and string columns #4255

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
8a79f47
Support parquet page filtering for string columns
alamb Nov 7, 2022
fc84754
Support parquet page filtering on min_max for decimal128 columns
Ted-Jiang Nov 15, 2022
a719dff
Update datafusion/core/src/physical_plan/file_format/parquet/page_fil…
Ted-Jiang Nov 18, 2022
4e29644
Avoid unwarp
Ted-Jiang Nov 18, 2022
5b6c478
reorg test code
Ted-Jiang Nov 18, 2022
4c81dca
add test for page index
Ted-Jiang Nov 18, 2022
d0bea7b
fix commet
Ted-Jiang Nov 22, 2022
9ebe563
remove code
Ted-Jiang Nov 22, 2022
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
Next Next commit
Support parquet page filtering for string columns 
Signed-off-by: yangjiang <yangjiang@ebay.com>
  • Loading branch information
@alamb @Ted-Jiang
alamb authored and Ted-Jiang committed Nov 17, 2022
commit 8a79f475ee7a3b67b8e56a6ae9ece94e78a6c2a7
15 changes: 13 additions & 2 deletions datafusion/core/src/physical_plan/file_format/parquet/page_filter.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,9 @@

//! Contains code to filter entire pages

use arrow::array::{BooleanArray, Float32Array, Float64Array, Int32Array, Int64Array};
use arrow::array::{
BooleanArray, Float32Array, Float64Array, Int32Array, Int64Array, StringArray,
};
use arrow::{array::ArrayRef, datatypes::SchemaRef, error::ArrowError};
use datafusion_common::{Column, DataFusionError, Result};
use datafusion_optimizer::utils::split_conjunction;
Expand Down Expand Up @@ -419,7 +421,16 @@ macro_rules! get_min_max_values_for_page_index {
vec.iter().map(|x| x.$func().cloned()),
)))
}
Index::INT96(_) | Index::BYTE_ARRAY(_) | Index::FIXED_LEN_BYTE_ARRAY(_) => {
Index::BYTE_ARRAY(index) => {
let vec = &index.indexes;
Copy link
Contributor

@liukun4515 liukun4515 Nov 18, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

decimal should be supported for this logical type.

Copy link
Contributor

@liukun4515 liukun4515 Nov 18, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Arrow-rs contains the method of decoding decimal from byte array in ByteArrayReader

Copy link
Member Author

@Ted-Jiang Ted-Jiang Nov 18, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks, i prefer align with row group, do them together in other pr.

Copy link
Contributor

@alamb alamb Nov 21, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Addition additional support in a follow on PR sounds like a good idea to me -- maybe we can file a ticket to track the work

let array: StringArray = vec
.iter()
.map(|x| x.$func())
.map(|x| x.and_then(|x| std::str::from_utf8(x).ok()))
.collect();
Some(Arc::new(array))
}
Index::INT96(_) | Index::FIXED_LEN_BYTE_ARRAY(_) => {
//Todo support these type
None
}
Expand Down
25 changes: 11 additions & 14 deletions datafusion/core/tests/parquet/filter_pushdown.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -266,20 +266,17 @@ async fn single_file_small_data_pages() {
// page 3: DLE:RLE RLE:RLE VLE:RLE_DICTIONARY ST:[min: djzdyiecnumrsrcbizwlqzdhnpoiqdh, max: fktdcgtmzvoedpwhfevcvvrtaurzgex, num_nulls not defined] CRC:[none] SZ:7 VC:9216
// page 4: DLE:RLE RLE:RLE VLE:RLE_DICTIONARY ST:[min: fktdcgtmzvoedpwhfevcvvrtaurzgex, max: fwtdpgtxwqkkgtgvthhwycrvjiizdifyp, num_nulls not defined] CRC:[none] SZ:7 VC:9216
// page 5: DLE:RLE RLE:RLE VLE:RLE_DICTIONARY ST:[min: fwtdpgtxwqkkgtgvthhwycrvjiizdifyp, max: iadnalqpdzthpifrvewossmpqibgtsuin, num_nulls not defined] CRC:[none] SZ:7 VC:7739
//
Copy link
Contributor

@alamb alamb Nov 17, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🎉

// This test currently fails due to https://github.com/apache/arrow-datafusion/issues/3833
// (page index pruning not implemented for byte array)

// TestCase::new(&test_parquet_file)
// .with_name("selective")
// // predicate is chosen carefully to prune pages 0, 1, 2, 3, 4
// // pod = 'iadnalqpdzthpifrvewossmpqibgtsuin'
// .with_filter(col("pod").eq(lit("iadnalqpdzthpifrvewossmpqibgtsuin")))
// .with_pushdown_expected(PushdownExpected::Some)
// .with_page_index_filtering_expected(PageIndexFilteringExpected::Some)
// .with_expected_rows(2574)
// .run()
// .await;

TestCase::new(&test_parquet_file)
.with_name("selective")
// predicate is chosen carefully to prune pages 0, 1, 2, 3, 4
// pod = 'iadnalqpdzthpifrvewossmpqibgtsuin'
.with_filter(col("pod").eq(lit("iadnalqpdzthpifrvewossmpqibgtsuin")))
.with_pushdown_expected(PushdownExpected::Some)
.with_page_index_filtering_expected(PageIndexFilteringExpected::Some)
.with_expected_rows(2574)
.run()
.await;

// time TV=53819 RL=0 DL=0 DS: 7092 DE:PLAIN
// --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Expand Down