Added support for sqlites json_type

Cargo.toml

@@ -4,10 +4,12 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
+anyhow = "1.0.75"
 datafusion = "33.0.0"
 ipnet = "2.7.2"
 serde = "1.0.192"
 serde_json = { version = "1.0.108", features = ["preserve_order"] }
+serde_json_path = "0.6.4"
 tokio = { version = "1.25.0", features = ["macros", "rt", "parking_lot"] }
 
 [features]

src/common/mod.rs

@@ -1 +1,71 @@
+use anyhow::anyhow;
+use serde_json_path::JsonPath;
+
 pub mod test_utils;
+
+pub(crate) fn get_value_at(
+    json: serde_json::Value,
+    path: &str,
+) -> anyhow::Result<serde_json::Value> {
+    let path = JsonPath::parse(path)?;
+    path.query(&json)
+        .exactly_one()
+        .map(|json| json.clone())
+        .map_err(|err| anyhow!(err))
+}
+
+pub(crate) fn get_value_at_string(
+    json_string: &str,
+    path: &str,
+) -> anyhow::Result<serde_json::Value> {
+    let json_value: serde_json::Value = serde_json::from_str(json_string)?;
+    get_value_at(json_value, path)
+}
+
+pub(crate) fn get_json_type(json_value: &serde_json::Value) -> anyhow::Result<String> {
+    let result = if json_value.is_boolean() {
+        match json_value.as_bool() {
+            Some(bool_value) => bool_value.to_string(),
+            None => return Err(anyhow!("Internal error".to_string())),
+        }
+    } else if json_value.is_number() {
+        if json_value.to_string().parse::<i64>().is_ok() {
+            "integer".to_string()
+        } else {
+            "real".to_string()
+        }
+    } else if json_value.is_f64() || json_value.is_i64() || json_value.is_u64() {
+        "real".to_string()
+    } else if json_value.is_string() {
+        "text".to_string()
+    } else if json_value.is_array() {
+        "array".to_string()
+    } else if json_value.is_object() {
+        "object".to_string()
+    } else if json_value.is_null() {
+        "null".to_string()
+    } else {
+        return Err(anyhow!("Unknown json type".to_string()));
+    };
+
+    Ok(result)
+}
+
+pub(crate) fn get_json_string_type(json_string: &str) -> anyhow::Result<String> {
+    let json_value: serde_json::Value =
+        serde_json::from_str(json_string).map_err(|err| anyhow!(err))?;
+    get_json_type(&json_value)
+}
+
+#[cfg(test)]
+mod test {
+    use crate::common::get_value_at;
+    use serde_json::json;
+
+    #[test]
+    fn test_get_value_at() {
+        let value = json!({"foo": ["bar", "baz"]});
+        let result = get_value_at(value, "$.foo").unwrap();
+        assert_eq!(result, json!(["bar", "baz"]));
+    }
+}

src/common/test_utils.rs

@@ -54,21 +54,42 @@ pub fn set_up_json_data_test() -> Result<SessionContext> {
     let batch = RecordBatch::try_new(
         schema,
         vec![
-            Arc::new(UInt8Array::from_iter_values([1, 2])),
+            Arc::new(UInt8Array::from_iter_values([1, 2, 3, 4, 5, 6, 7, 8])),
             Arc::new(StringArray::from(vec![
                 Some(r#" { "this" : "is", "a": [ "test" ] } "#),
-                // Some("172.16.0.0/20"),
-                // Some("10.0.0.0/16"),
-                // Some("2001:0db8::/32"),
-                // Some("2001:db8:abcd::/48"),
+                Some(r#"{"a":[2,3.5,true,false,null,"x"]}"#),
+                Some(r#"[ "one", "two" ]"#),
+                Some(r#"123"#),
+                Some(r#"12.3"#),
+                Some(r#"true"#),
+                Some(r#"false"#),
                 None,
             ])),
         ],
     )?;
 
     // declare a new context
     let ctx = SessionContext::new();
-    ctx.register_batch("json_table", batch)?;
     // declare a table in memory.
+    ctx.register_batch("json_table", batch)?;
+
+    // data for json_type
+    let schema = Arc::new(Schema::new(vec![
+        Field::new("index", DataType::UInt8, false),
+        Field::new("json_data", DataType::Utf8, true),
+    ]));
+
+    // define data.
+    let batch = RecordBatch::try_new(
+        schema,
+        vec![
+            Arc::new(UInt8Array::from_iter_values([1])),
+            Arc::new(StringArray::from(vec![Some(
+                r#"{"a":[2,3.5,true,false,null,"x"]}"#,
+            )])),
+        ],
+    )?;
+
+    ctx.register_batch("json_path_table", batch)?;
     Ok(ctx)
 }