From bd7f7982a149b5f99eb21bf1800616679095fcf9 Mon Sep 17 00:00:00 2001
From: Dadepo Aderemi <dadepo@gmail.com>
Date: Sat, 18 Nov 2023 18:45:33 +0400
Subject: [PATCH] added test for json when json is invalid

---
 df_extras_sqlite/src/json_udfs.rs | 42 ++++++++++++++++++++++++++++---
 1 file changed, 38 insertions(+), 4 deletions(-)

diff --git a/df_extras_sqlite/src/json_udfs.rs b/df_extras_sqlite/src/json_udfs.rs
index 17376b9..78a388e 100644
--- a/df_extras_sqlite/src/json_udfs.rs
+++ b/df_extras_sqlite/src/json_udfs.rs
@@ -5,17 +5,20 @@ use datafusion::common::DataFusionError;
 use datafusion::error::Result;
 use serde_json::Value;
 
+/// The json(X) function verifies that its argument X is a valid JSON string and returns a minified
+/// version of that JSON string (with all unnecessary whitespace removed).
+/// If X is not a well-formed JSON string, then this routine throws an error.
 pub fn json(args: &[ArrayRef]) -> Result<ArrayRef> {
     let json_strings = datafusion::common::cast::as_string_array(&args[0])?;
 
     let mut string_builder = StringBuilder::with_capacity(json_strings.len(), u8::MAX as usize);
     json_strings.iter().try_for_each(|json_string| {
         if let Some(json_string) = json_string {
-            let value: Value = serde_json::from_str(json_string).map_err(|e| {
-                DataFusionError::Internal(format!("Parsing {json_string} failed with error {e}"))
+            let value: Value = serde_json::from_str(json_string).map_err(|_| {
+                DataFusionError::Internal("Runtime error: malformed JSON".to_string())
             })?;
-            let pretty_json = serde_json::to_string(&value).map_err(|e| {
-                DataFusionError::Internal(format!("Parsing {json_string} failed with error {e}"))
+            let pretty_json = serde_json::to_string(&value).map_err(|_| {
+                DataFusionError::Internal("Runtime error: malformed JSON".to_string())
             })?;
             string_builder.append_value(pretty_json);
             Ok::<(), DataFusionError>(())
@@ -28,6 +31,15 @@ pub fn json(args: &[ArrayRef]) -> Result<ArrayRef> {
     Ok(Arc::new(string_builder.finish()) as ArrayRef)
 }
 
+/// The json_valid(X) function return 1 if the argument X is well-formed canonical RFC-7159 JSON
+/// without any extensions, or return 0 if the argument X is not well-formed JSON or is
+/// JSON that includes JSON5 extensions.
+///
+/// Examples:
+///
+/// json_valid('{"x":35}') → 1
+/// json_valid('{"x":35') → 0
+/// json_valid(NULL) → NULL
 pub fn json_valid(args: &[ArrayRef]) -> Result<ArrayRef> {
     let json_strings = datafusion::common::cast::as_string_array(&args[0])?;
     let mut uint_builder = UInt8Array::builder(json_strings.len());
@@ -87,6 +99,28 @@ mod tests {
         Ok(())
     }
 
+    #[tokio::test]
+    async fn test_invalid_json() -> Result<()> {
+        let ctx = register_udfs_for_test()?;
+        let df = ctx
+            .sql(
+                r#"select index, json(' { "this" : "is", "a": [ "test" ]  ') as col_result FROM json_table ORDER BY index ASC"#,
+            )
+            .await?;
+
+        let result = df.clone().collect().await;
+
+        assert!(&result
+            .err()
+            .unwrap()
+            .find_root()
+            .find_root()
+            .to_string()
+            .contains("Internal error: Runtime error: malformed JSON"));
+
+        Ok(())
+    }
+
     #[tokio::test]
     async fn test_json_valid() -> Result<()> {
         let ctx = register_udfs_for_test()?;