hyperstreamdb 0.2.3__tar.gz → 0.2.4__tar.gz

{hyperstreamdb-0.2.3 → hyperstreamdb-0.2.4}/.gitignore

@@ -103,11 +103,7 @@ groq_api_key.txt
 rag_db/
 news_db/
 
-<<<<<<< HEAD
 .hypothesis/
-=======
-.hypothesis/
-venv_demo/
 venv_demo/
 .env
->>>>>>> 91645c4f4d113f606607c9c4edf6635702e26b87
+

{hyperstreamdb-0.2.3 → hyperstreamdb-0.2.4}/Cargo.lock

@@ -3375,7 +3375,7 @@ dependencies = [
 
 [[package]]
 name = "hyperstreamdb"
-version = "0.2.3"
+version = "0.2.4"
 dependencies = [
  "ahash 0.8.12",
  "anyhow",

{hyperstreamdb-0.2.3 → hyperstreamdb-0.2.4}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "hyperstreamdb"
-version = "0.2.3"
+version = "0.2.4"
 edition = "2021"
 license = "MIT AND Apache-2.0"
 description = "HyperStreamDB - Serverless Index-Streaming Database with Overlay Indexing and Vector Search"

{hyperstreamdb-0.2.3 → hyperstreamdb-0.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hyperstreamdb
-Version: 0.2.3
+Version: 0.2.4
 Classifier: Development Status :: 3 - Alpha
 Classifier: Intended Audience :: Developers
 Classifier: Programming Language :: Rust

{hyperstreamdb-0.2.3 → hyperstreamdb-0.2.4}/src/core/iceberg.rs

@@ -1247,7 +1247,8 @@ impl IcebergWriter {
                                  "Int64" | "long" => r#"["null", "long"]"#,
                                  "Float32" | "float" => r#"["null", "float"]"#,
                                  "Float64" | "double" => r#"["null", "double"]"#,
-                                 "Boolean" | "bool" => r#"["null", "boolean"]"#,
+                                 "Boolean" | "bool" | "boolean" => r#"["null", "boolean"]"#,
+                                 "string" | "utf8" | "utf-8" | "String" | "Utf8" => r#"["null", "string"]"#,
                                  _ => r#"["null", "string"]"#,
                              }
                          } else {

{hyperstreamdb-0.2.3 → hyperstreamdb-0.2.4}/src/core/manifest.rs

@@ -10,6 +10,7 @@ use futures::StreamExt;
 use chrono::Utc;
 use tracing;
 use arrow::record_batch::RecordBatch;
+use arrow::array::Array;
 
 pub type SegmentId = String;
 
@@ -1659,16 +1660,37 @@ impl PartitionSpec {
                         if col.is_null(i) {
                             serde_json::Value::Null
                         } else {
-                            // Basic extraction (Handle Utf8, Int64, Int32 for common identity partitioning)
-                            if let Some(s) = col.as_any().downcast_ref::<arrow::array::StringArray>() {
-                                serde_json::Value::String(s.value(i).to_string())
-                            } else if let Some(n) = col.as_any().downcast_ref::<arrow::array::Int64Array>() {
-                                serde_json::Value::Number(serde_json::Number::from(n.value(i)))
-                            } else if let Some(n) = col.as_any().downcast_ref::<arrow::array::Int32Array>() {
-                                serde_json::Value::Number(serde_json::Number::from(n.value(i)))
-                            } else {
-                                serde_json::to_value(format!("{:?}", col.slice(i, 1))).unwrap_or(serde_json::Value::Null)
-                            }
+                             if let Some(s) = col.as_any().downcast_ref::<arrow::array::StringArray>() {
+                                 serde_json::Value::String(s.value(i).to_string())
+                             } else if let Some(s) = col.as_any().downcast_ref::<arrow::array::LargeStringArray>() {
+                                 serde_json::Value::String(s.value(i).to_string())
+                             } else if let Some(dict) = col.as_any().downcast_ref::<arrow::array::DictionaryArray<arrow::datatypes::Int32Type>>() {
+                                 let values = dict.values().as_any().downcast_ref::<arrow::array::StringArray>().unwrap();
+                                 if dict.is_null(i) {
+                                     serde_json::Value::Null
+                                 } else {
+                                     let key = dict.key(i).unwrap();
+                                     serde_json::Value::String(values.value(key as usize).to_string())
+                                 }
+                             } else if let Some(dict) = col.as_any().downcast_ref::<arrow::array::DictionaryArray<arrow::datatypes::Int64Type>>() {
+                                 let values = dict.values().as_any().downcast_ref::<arrow::array::StringArray>().unwrap();
+                                 if dict.is_null(i) {
+                                     serde_json::Value::Null
+                                 } else {
+                                     let key = dict.key(i).unwrap() as usize;
+                                     serde_json::Value::String(values.value(key).to_string())
+                                 }
+                             } else if let Some(n) = col.as_any().downcast_ref::<arrow::array::Int64Array>() {
+                                 serde_json::Value::Number(serde_json::Number::from(n.value(i)))
+                             } else if let Some(n) = col.as_any().downcast_ref::<arrow::array::Int32Array>() {
+                                 serde_json::Value::Number(serde_json::Number::from(n.value(i)))
+                             } else if let Some(n) = col.as_any().downcast_ref::<arrow::array::Float64Array>() {
+                                 serde_json::Number::from_f64(n.value(i)).map(serde_json::Value::Number).unwrap_or(serde_json::Value::Null)
+                             } else if let Some(n) = col.as_any().downcast_ref::<arrow::array::Float32Array>() {
+                                 serde_json::Number::from_f64(n.value(i) as f64).map(serde_json::Value::Number).unwrap_or(serde_json::Value::Null)
+                             } else {
+                                 serde_json::to_value(format!("{:?}", col.slice(i, 1))).unwrap_or(serde_json::Value::Null)
+                             }
                         }
                     },
                     _ => {