start refactoring process by setting up base + init

datafusion/sqllogictest/Cargo.toml

@@ -82,3 +82,7 @@ tokio = { workspace = true, features = ["rt-multi-thread"] }
 harness = false
 name = "sqllogictests"
 path = "bin/sqllogictests.rs"
+
+[[bin]]
+name = "diff"
+path = "bin/diff.rs"

datafusion/sqllogictest/bin/diff.rs

@@ -0,0 +1,191 @@
+// Licensed to the Apache Software Foundation (ASF) under one
+// or more contributor license agreements.  See the NOTICE file
+// distributed with this work for additional information
+// regarding copyright ownership.  The ASF licenses this file
+// to you under the Apache License, Version 2.0 (the
+// "License"); you may not use this file except in compliance
+// with the License.  You may obtain a copy of the License at
+//
+//   http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing,
+// software distributed under the License is distributed on an
+// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+// KIND, either express or implied.  See the License for the
+// specific language governing permissions and limitations
+// under the License.
+
+use sqllogictest::parser::parse_file;
+use sqllogictest::DefaultColumnType;
+use sqllogictest::Record;
+use sqllogictest::Record::{Query, Statement};
+use std::error::Error;
+use std::fs;
+use std::path::Path;
+
+// run inside sqllogictest or run with -p datafusion-sqllogictest
+// cargo run --bin diff -- path1.slt path2(.slt or folder)
+pub fn main() -> Result<(), Box<dyn Error>> {
+    let args: Vec<String> = std::env::args().collect();
+
+    // Check if we have exactly two file paths as arguments
+    if args.len() < 3 {
+        return Err("Please provide two arguments: a file path or folder, and a second file or folder.".into());
+    }
+
+    let path1 = Path::new(&args[1]);
+    let path2 = Path::new(&args[2]);
+
+    let result = diff(path1, path2);
+    println!("{:?}", result);
+    Ok(())
+}
+
+pub fn diff(path1: &Path, path2: &Path) -> Result<String, Box<dyn Error>> {
+    println!("Needles: {:?}", path1);
+    println!("Haystack: {:?}", path2);
+
+    // Parse the records from the first file
+    let records1 = parse_file(path1)?;
+
+    // let records2;
+
+    let records2 = if path2.is_dir() {
+        // If the second path is a directory, parse all files in that directory
+        parse_files_in_directory(path2)?
+    } else {
+        // If the second path is a file, just parse that file
+        parse_file(path2)?
+    };
+    let mut diffs = Vec::new();
+
+    // Check if each record in file 1 is contained in any record in file 2
+    for record1 in &records1 {
+        // let found;
+        let found = if !check_type(record1) {
+            false
+        } else {
+            records2
+                .iter()
+                .any(|record2| check_equality(record1, record2))
+        };
+        if check_equality(record1, record1) && !found {
+            diffs.push(format!(
+                "Record from Needles not found in Haystack: {:?}",
+                get_sql(record1)
+            ));
+        }
+    }
+    // If we have collected any differences, return them all at once
+    if !diffs.is_empty() {
+        Ok(diffs.join("\n"))
+    } else {
+        Ok("All records from Needles are present in Haystack.".to_string())
+    }
+}
+
+fn get_sql(record: &Record<DefaultColumnType>) -> String {
+    match record {
+        Query { sql, .. } => sql.clone(),
+        Statement { sql, .. } => sql.clone(),
+        _ => String::new(),
+    }
+}
+pub fn check_type(record1: &Record<DefaultColumnType>) -> bool {
+    // the type which is acceptable by check_equality(Query and Statement)
+    check_equality(record1, record1)
+}
+
+pub fn check_equality(
+    record1: &Record<DefaultColumnType>,
+    record2: &Record<DefaultColumnType>,
+) -> bool {
+    match (record1, record2) {
+        (
+            Query {
+                loc: _,
+                conditions: _,
+                connection: _,
+                sql: sql1,
+                expected: expected1,
+                retry: _,
+            },
+            Query {
+                loc: _,
+                conditions: _,
+                connection: _,
+                sql: sql2,
+                expected: expected2,
+                retry: _,
+            },
+        ) => sql1 == sql2 && expected1 == expected2,
+        (
+            Statement {
+                loc: _,
+                conditions: _,
+                connection: _,
+                sql: sql1,
+                expected: expected1,
+                retry: _,
+            },
+            Statement {
+                loc: _,
+                conditions: _,
+                connection: _,
+                sql: sql2,
+                expected: expected2,
+                retry: _,
+            },
+        ) => sql1 == sql2 && expected1 == expected2,
+        _ => false,
+    }
+}
+
+// Warning: This is not recursive, can be made recursive in future if needed.
+fn parse_files_in_directory(
+    directory: &Path,
+) -> Result<Vec<Record<DefaultColumnType>>, Box<dyn Error>> {
+    let mut all_records = Vec::new();
+
+    // Read all files in the directory
+    for entry in fs::read_dir(directory)? {
+        let entry = entry?;
+        let path = entry.path();
+
+        // Only process files (not directories)
+        if path.is_file() && path.extension().map(|ext| ext == "slt").unwrap_or(false) {
+            let records = parse_file(&path)?;
+            all_records.extend(records); // Add the records from this file
+        }
+    }
+
+    Ok(all_records)
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_files_aggregate_diff_archived() {
+        // Create two test files
+        let path1 = Path::new("./archive/complete_aggregate.slt");
+        let path2 = Path::new("./test_files/aggregate");
+        let result = diff(path1, path2);
+        assert_eq!(
+            result.unwrap(),
+            format!("All records from Needles are present in Haystack.")
+        )
+    }
+    #[test]
+    fn test_files_aggregate_diff_base() {
+        // Create two test files
+        let path1 = Path::new("./test_files/aggregate/base_aggregate.slt");
+        let path2 = Path::new("./archive/complete_aggregate.slt");
+        let result = diff(path1, path2);
+        assert_eq!(
+            result.unwrap(),
+            format!("All records from Needles are present in Haystack.")
+        )
+    }
+}

datafusion/sqllogictest/test_files/aggregate/README.md

@@ -0,0 +1,14 @@
+# Aggregate Tests
+
+##### History and Context:
+
+Aggregate used to be (perhaps still is depending on the progress on issue [#13723](https://github.com/apache/datafusion/issues/13723)).
+We have decided to refactor it for better navigation by using `extract and subtract` approach.
+
+Formally, `base_aggregate.slt` starts with all the test cases of original `aggregate.slt` which is currently in `datafusion/sqllogictest/archive`.
+
+Gradually, as we move out(`extract`) different portions of code we remove(`substract`) only that protion from `base_aggregate.slt`.
+
+It should be done in a manner that at all times set of all tests in aggregate folder is a superset of all tests covered in `datafusion/sqllogictest/archive/complete_aggregate.slt`
+
+Refer to [#14301](https://github.com/apache/datafusion/pull/14301) and [#14306](https://github.com/apache/datafusion/pull/14306) for more context and details.