Test: Add checks to sqllogictest temporary file creations (#17017)

* add checks to sqllogictest temporary file creation

* clippy

* review

* clean up

* clippy

* update cargo.lock

* update cargo.lock

---------

Co-authored-by: Andrew Lamb <[email protected]>

Author: 2010YOUY01

Cargo.lock

@@ -81,6 +81,7 @@ postgres = [
 
 [dev-dependencies]
 env_logger = { workspace = true }
+regex = { workspace = true }
 tokio = { workspace = true, features = ["rt-multi-thread"] }
 
 [[test]]

datafusion-testing

@@ -42,6 +42,7 @@ use crate::postgres_container::{
 };
 use datafusion::common::runtime::SpawnedTask;
 use std::ffi::OsStr;
+use std::fs;
 use std::path::{Path, PathBuf};
 
 #[cfg(feature = "postgres")]
@@ -124,6 +125,20 @@ async fn run_tests() -> Result<()> {
     let start = Instant::now();
 
     let test_files = read_test_files(&options)?;
+
+    // Perform scratch file sanity check
+    let scratch_errors = scratch_file_check(&test_files)?;
+    if !scratch_errors.is_empty() {
+        eprintln!("Scratch file sanity check failed:");
+        for error in &scratch_errors {
+            eprintln!("  {error}");
+        }
+
+        eprintln!("\nTemporary file check failed. Please ensure that within each test file, any scratch file created is placed under a folder with the same name as the test file (without extension).\nExample: inside `join.slt`, temporary files must be created under `.../scratch/join/`\n");
+
+        return exec_err!("sqllogictests scratch file check failed");
+    }
+
     let num_tests = test_files.len();
     let errors: Vec<_> = futures::stream::iter(test_files)
         .map(|test_file| {
@@ -738,3 +753,67 @@ impl Options {
         }
     }
 }
+
+/// Performs scratch file check for all test files.
+///
+/// Scratch file rule: In each .slt test file, the temporary file created must
+/// be under a folder that is has the same name as the test file.
+/// e.g. In `join.slt`, temporary files must be created under `.../scratch/join/`
+///
+/// See: <https://github.com/apache/datafusion/tree/main/datafusion/sqllogictest#running-tests-scratchdir>
+///
+/// This function searches for `scratch/[target]/...` patterns and verifies
+/// that the target matches the file name.
+///
+/// Returns a vector of error strings for incorrectly created scratch files.
+fn scratch_file_check(test_files: &[TestFile]) -> Result<Vec<String>> {
+    let mut errors = Vec::new();
+
+    // Search for any scratch/[target]/... patterns and check if they match the file name
+    let scratch_pattern = regex::Regex::new(r"scratch/([^/]+)/").unwrap();
+
+    for test_file in test_files {
+        // Get the file content
+        let content = match fs::read_to_string(&test_file.path) {
+            Ok(content) => content,
+            Err(e) => {
+                errors.push(format!(
+                    "Failed to read file {}: {}",
+                    test_file.path.display(),
+                    e
+                ));
+                continue;
+            }
+        };
+
+        // Get the expected target name (file name without extension)
+        let expected_target = match test_file.path.file_stem() {
+            Some(stem) => stem.to_string_lossy().to_string(),
+            None => {
+                errors.push(format!("File {} has no stem", test_file.path.display()));
+                continue;
+            }
+        };
+
+        let lines: Vec<&str> = content.lines().collect();
+
+        for (line_num, line) in lines.iter().enumerate() {
+            if let Some(captures) = scratch_pattern.captures(line) {
+                if let Some(found_target) = captures.get(1) {
+                    let found_target = found_target.as_str();
+                    if found_target != expected_target {
+                        errors.push(format!(
+                            "File {}:{}: scratch target '{}' does not match file name '{}'",
+                            test_file.path.display(),
+                            line_num + 1,
+                            found_target,
+                            expected_target
+                        ));
+                    }
+                }
+            }
+        }
+    }
+
+    Ok(errors)
+}

datafusion/sqllogictest/Cargo.toml

@@ -456,4 +456,4 @@ statement ok
 CREATE TABLE test0 AS VALUES ('foo',1), ('bar',2), ('foo',3);
 
 statement ok
-COPY (SELECT arrow_cast(column1, 'Dictionary(Int32, Utf8)') AS column1, column2 FROM test0) TO 'test_files/scratch/copy/part_dict_test' STORED AS PARQUET PARTITIONED BY (column1);
+COPY (SELECT arrow_cast(column1, 'Dictionary(Int32, Utf8)') AS column1, column2 FROM test0) TO 'test_files/scratch/dictionary/part_dict_test' STORED AS PARQUET PARTITIONED BY (column1);

datafusion/sqllogictest/bin/sqllogictests.rs

@@ -799,23 +799,23 @@ CREATE TABLE src_table (
 # File 1:
 query I
 COPY (SELECT * FROM src_table where part_key = 1)
-TO 'test_files/scratch/parquet/test_limit_with_partitions/part-0.parquet'
+TO 'test_files/scratch/limit/test_limit_with_partitions/part-0.parquet'
 STORED AS PARQUET;
 ----
 3
 
 # File 2:
 query I
 COPY (SELECT * FROM src_table where part_key = 2)
-TO 'test_files/scratch/parquet/test_limit_with_partitions/part-1.parquet'
+TO 'test_files/scratch/limit/test_limit_with_partitions/part-1.parquet'
 STORED AS PARQUET;
 ----
 4
 
 # File 3:
 query I
 COPY (SELECT * FROM src_table where part_key = 3)
-TO 'test_files/scratch/parquet/test_limit_with_partitions/part-2.parquet'
+TO 'test_files/scratch/limit/test_limit_with_partitions/part-2.parquet'
 STORED AS PARQUET;
 ----
 3
@@ -827,7 +827,7 @@ CREATE EXTERNAL TABLE test_limit_with_partitions
   value INT
 )
 STORED AS PARQUET
-LOCATION 'test_files/scratch/parquet/test_limit_with_partitions/';
+LOCATION 'test_files/scratch/limit/test_limit_with_partitions/';
 
 query TT
 explain
@@ -853,7 +853,7 @@ physical_plan
 01)ProjectionExec: expr=[1 as foo]
 02)--SortPreservingMergeExec: [part_key@0 ASC NULLS LAST], fetch=1
 03)----SortExec: TopK(fetch=1), expr=[part_key@0 ASC NULLS LAST], preserve_partitioning=[true]
-04)------DataSourceExec: file_groups={3 groups: [[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/parquet/test_limit_with_partitions/part-0.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/parquet/test_limit_with_partitions/part-1.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/parquet/test_limit_with_partitions/part-2.parquet]]}, projection=[part_key], file_type=parquet, predicate=DynamicFilterPhysicalExpr [ true ]
+04)------DataSourceExec: file_groups={3 groups: [[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/limit/test_limit_with_partitions/part-0.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/limit/test_limit_with_partitions/part-1.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/limit/test_limit_with_partitions/part-2.parquet]]}, projection=[part_key], file_type=parquet, predicate=DynamicFilterPhysicalExpr [ true ]
 
 query I
 with selection as (

datafusion/sqllogictest/test_files/dictionary.slt

@@ -17,10 +17,10 @@
 
 # Test file with different schema order but generating correct statistics for table
 statement ok
-COPY (SELECT * FROM values (1, 'a'), (2, 'b') t(int_col, str_col)) to 'test_files/scratch/table/1.parquet';
+COPY (SELECT * FROM values (1, 'a'), (2, 'b') t(int_col, str_col)) to 'test_files/scratch/listing_table_statistics/1.parquet';
 
 statement ok
-COPY (SELECT * FROM values ('c', 3), ('d', -1) t(str_col, int_col)) to 'test_files/scratch/table/2.parquet';
+COPY (SELECT * FROM values ('c', 3), ('d', -1) t(str_col, int_col)) to 'test_files/scratch/listing_table_statistics/2.parquet';
 
 statement ok
 set datafusion.execution.collect_statistics = true;
@@ -29,13 +29,13 @@ statement ok
 set datafusion.explain.show_statistics = true;
 
 statement ok
-create external table t stored as parquet location 'test_files/scratch/table';
+create external table t stored as parquet location 'test_files/scratch/listing_table_statistics';
 
 query TT
 explain format indent select * from t;
 ----
 logical_plan TableScan: t projection=[int_col, str_col]
-physical_plan DataSourceExec: file_groups={2 groups: [[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/table/1.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/table/2.parquet]]}, projection=[int_col, str_col], file_type=parquet, statistics=[Rows=Exact(4), Bytes=Exact(212), [(Col[0]: Min=Exact(Int64(-1)) Max=Exact(Int64(3)) Null=Exact(0)),(Col[1]: Min=Exact(Utf8View("a")) Max=Exact(Utf8View("d")) Null=Exact(0))]]
+physical_plan DataSourceExec: file_groups={2 groups: [[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/listing_table_statistics/1.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/listing_table_statistics/2.parquet]]}, projection=[int_col, str_col], file_type=parquet, statistics=[Rows=Exact(4), Bytes=Exact(212), [(Col[0]: Min=Exact(Int64(-1)) Max=Exact(Int64(3)) Null=Exact(0)),(Col[1]: Min=Exact(Utf8View("a")) Max=Exact(Utf8View("d")) Null=Exact(0))]]
 
 statement ok
 drop table t;

datafusion/sqllogictest/test_files/limit.slt

@@ -175,23 +175,23 @@ CREATE TABLE src_table (
 # File 1:
 query I
 COPY (SELECT * FROM src_table where part_key = 1)
-TO 'test_files/scratch/parquet/test_filter_with_limit/part-0.parquet'
+TO 'test_files/scratch/push_down_filter/test_filter_with_limit/part-0.parquet'
 STORED AS PARQUET;
 ----
 3
 
 # File 2:
 query I
 COPY (SELECT * FROM src_table where part_key = 2)
-TO 'test_files/scratch/parquet/test_filter_with_limit/part-1.parquet'
+TO 'test_files/scratch/push_down_filter/test_filter_with_limit/part-1.parquet'
 STORED AS PARQUET;
 ----
 4
 
 # File 3:
 query I
 COPY (SELECT * FROM src_table where part_key = 3)
-TO 'test_files/scratch/parquet/test_filter_with_limit/part-2.parquet'
+TO 'test_files/scratch/push_down_filter/test_filter_with_limit/part-2.parquet'
 STORED AS PARQUET;
 ----
 3
@@ -203,14 +203,14 @@ CREATE EXTERNAL TABLE test_filter_with_limit
   value INT
 )
 STORED AS PARQUET
-LOCATION 'test_files/scratch/parquet/test_filter_with_limit/';
+LOCATION 'test_files/scratch/push_down_filter/test_filter_with_limit/';
 
 query TT
 explain select * from test_filter_with_limit where value = 2 limit 1;
 ----
 physical_plan
 01)CoalescePartitionsExec: fetch=1
-02)--DataSourceExec: file_groups={3 groups: [[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/parquet/test_filter_with_limit/part-0.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/parquet/test_filter_with_limit/part-1.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/parquet/test_filter_with_limit/part-2.parquet]]}, projection=[part_key, value], limit=1, file_type=parquet, predicate=value@1 = 2, pruning_predicate=value_null_count@2 != row_count@3 AND value_min@0 <= 2 AND 2 <= value_max@1, required_guarantees=[value in (2)]
+02)--DataSourceExec: file_groups={3 groups: [[WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/push_down_filter/test_filter_with_limit/part-0.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/push_down_filter/test_filter_with_limit/part-1.parquet], [WORKSPACE_ROOT/datafusion/sqllogictest/test_files/scratch/push_down_filter/test_filter_with_limit/part-2.parquet]]}, projection=[part_key, value], limit=1, file_type=parquet, predicate=value@1 = 2, pruning_predicate=value_null_count@2 != row_count@3 AND value_min@0 <= 2 AND 2 <= value_max@1, required_guarantees=[value in (2)]
 
 query II
 select * from test_filter_with_limit where value = 2 limit 1;