bump sqllogictest to 0.9.0 (#4547)

* bump sqllogictest to 0.9.0

* cargo update datafusion-cli

* fix clippy

* fix Cargo.lock

* fix test case

Author: xxchan

datafusion-cli/Cargo.lock

@@ -109,7 +109,7 @@ doc-comment = "0.3"
 env_logger = "0.10"
 parquet-test-utils = { path = "../../parquet-test-utils" }
 rstest = "0.16.0"
-sqllogictest = "0.8.0"
+sqllogictest = "0.9.0"
 sqlparser = "0.27"
 test-utils = { path = "../../test-utils" }
 

datafusion/core/Cargo.toml

@@ -25,10 +25,11 @@ use datafusion::prelude::SessionContext;
 use datafusion_common::{DFSchema, DataFusionError};
 use datafusion_expr::Expr as DFExpr;
 use datafusion_sql::planner::{object_name_to_table_reference, PlannerContext, SqlToRel};
+use sqllogictest::DBOutput;
 use sqlparser::ast::{Expr, SetExpr, Statement as SQLStatement};
 use std::sync::Arc;
 
-pub async fn insert(ctx: &SessionContext, insert_stmt: SQLStatement) -> Result<String> {
+pub async fn insert(ctx: &SessionContext, insert_stmt: SQLStatement) -> Result<DBOutput> {
     // First, use sqlparser to get table name and insert values
     let table_reference;
     let insert_values: Vec<Vec<Expr>>;
@@ -60,6 +61,7 @@ pub async fn insert(ctx: &SessionContext, insert_stmt: SQLStatement) -> Result<S
     // Third, transfer insert values to `RecordBatch`
     // Attention: schema info can be ignored. (insert values don't contain schema info)
     let sql_to_rel = SqlToRel::new(&LogicTestContextProvider {});
+    let num_rows = insert_values.len();
     for row in insert_values.into_iter() {
         let logical_exprs = row
             .into_iter()
@@ -85,5 +87,5 @@ pub async fn insert(ctx: &SessionContext, insert_stmt: SQLStatement) -> Result<S
     let new_provider = MemTable::try_new(schema, vec![origin_batches])?;
     ctx.register_table(&table_reference, Arc::new(new_provider))?;
 
-    Ok("".to_string())
+    Ok(DBOutput::StatementComplete(num_rows as u64))
 }

datafusion/core/tests/sqllogictests/src/insert/mod.rs

@@ -22,6 +22,7 @@ use datafusion::prelude::{SessionConfig, SessionContext};
 use datafusion_sql::parser::{DFParser, Statement};
 use log::info;
 use normalize::normalize_batch;
+use sqllogictest::{ColumnType, DBOutput};
 use sqlparser::ast::Statement as SQLStatement;
 use std::path::{Path, PathBuf};
 use std::time::Duration;
@@ -46,7 +47,7 @@ pub struct DataFusion {
 impl sqllogictest::AsyncDB for DataFusion {
     type Error = DFSqlLogicTestError;
 
-    async fn run(&mut self, sql: &str) -> Result<String> {
+    async fn run(&mut self, sql: &str) -> Result<DBOutput> {
         println!("[{}] Running query: \"{}\"", self.file_name, sql);
         let result = run_query(&self.ctx, sql).await?;
         Ok(result)
@@ -172,19 +173,42 @@ async fn context_for_test_file(file_name: &str) -> SessionContext {
     }
 }
 
-fn format_batches(batches: Vec<RecordBatch>) -> Result<String> {
+fn convert_batches(batches: Vec<RecordBatch>) -> Result<DBOutput> {
     let mut bytes = vec![];
+    if batches.is_empty() {
+        return Ok(DBOutput::StatementComplete(0));
+    }
+    // TODO: use the actual types
+    let types = vec![ColumnType::Any; batches[0].num_columns()];
+
     {
-        let builder = WriterBuilder::new().has_headers(false).with_delimiter(b' ');
+        let builder = WriterBuilder::new()
+            .has_headers(false)
+            .with_delimiter(b'\t');
         let mut writer = builder.build(&mut bytes);
         for batch in batches {
             writer.write(&normalize_batch(batch)).unwrap();
         }
     }
-    Ok(String::from_utf8(bytes).unwrap())
+    let res = String::from_utf8(bytes).unwrap();
+    let rows = res
+        .lines()
+        .map(|s| {
+            s.split('\t')
+                .map(|s| {
+                    if s.is_empty() {
+                        "NULL".to_string()
+                    } else {
+                        s.to_string()
+                    }
+                })
+                .collect()
+        })
+        .collect();
+    Ok(DBOutput::Rows { types, rows })
 }
 
-async fn run_query(ctx: &SessionContext, sql: impl Into<String>) -> Result<String> {
+async fn run_query(ctx: &SessionContext, sql: impl Into<String>) -> Result<DBOutput> {
     let sql = sql.into();
     // Check if the sql is `insert`
     if let Ok(mut statements) = DFParser::parse_sql(&sql) {
@@ -198,6 +222,6 @@ async fn run_query(ctx: &SessionContext, sql: impl Into<String>) -> Result<Strin
     }
     let df = ctx.sql(sql.as_str()).await?;
     let results: Vec<RecordBatch> = df.collect().await?;
-    let formatted_batches = format_batches(results)?;
+    let formatted_batches = convert_batches(results)?;
     Ok(formatted_batches)
 }

datafusion/core/tests/sqllogictests/src/main.rs

@@ -29,11 +29,9 @@ use arrow::{
 ///
 /// This is to make the output comparable to the semi-standard .slt format
 ///
-/// Normalizations applied:
-/// 1. Null Values (TODO)
-/// 2. [Empty Strings]
+/// Normalizations applied to [NULL Values and empty strings]
 ///
-/// [Empty Strings]: https://duckdb.org/dev/sqllogictest/result_verification#null-values-and-empty-strings
+/// [NULL Values and empty strings]: https://duckdb.org/dev/sqllogictest/result_verification#null-values-and-empty-strings
 pub fn normalize_batch(batch: RecordBatch) -> RecordBatch {
     let new_columns = batch
         .columns()