fix(cubesql): Use writable streams with plain objects instead of JSON…

packages/cubejs-backend-native/js/index.ts

@@ -1,5 +1,6 @@
 import fs from 'fs';
 import path from 'path';
+import { Writable } from 'stream';
 // import { getEnv } from '@cubejs-backend/shared';
 
 export interface BaseMeta {
@@ -115,6 +116,9 @@ function wrapNativeFunctionWithChannelCallback(
     };
 };
 
+const errorString = (err: any) =>
+  err.message || err.stack?.toString() || typeof err === 'string' ? err.toString() : JSON.stringify(err);
+
 // TODO: Refactor - define classes
 function wrapNativeFunctionWithStream(
     fn: (extra: any) => unknown | Promise<unknown>
@@ -129,27 +133,48 @@ function wrapNativeFunctionWithStream(
             streamResponse = await fn(JSON.parse(extra));
             if (streamResponse && streamResponse.stream) {
                 writer.start();
-                let chunk: object[] = [];
-                streamResponse.stream.on('data', (c: object) => {
-                    chunk.push(c);
-                    if (chunk.length >= chunkLength) {
-                        if (!writer.chunk(JSON.stringify(chunk))) {
-                            // TODO replace with actual stream and high watermark implementation
-                            streamResponse.stream.destroy({
-                                stack: "Rejected by client"
-                            });
+
+                let chunkBuffer: any[] = [];
+                const writable = new Writable({
+                    objectMode: true,
+                    highWaterMark: chunkLength,
+                    write(row: any, encoding: BufferEncoding, callback: (error?: (Error | null)) => void) {
+                        chunkBuffer.push(row);
+                        if (chunkBuffer.length < chunkLength) {
+                            callback(null);
+                        } else {
+                            const toSend = chunkBuffer;
+                            chunkBuffer = [];
+                            writer.chunk(toSend, callback);
                         }
-                        chunk = [];
-                    }
-                });
-                streamResponse.stream.on('close', () => {
-                    if (chunk.length > 0) {
-                        writer.chunk(JSON.stringify(chunk));
+
+                    },
+                    final(callback: (error?: (Error | null)) => void) {
+                        const end = (err: any) => {
+                            if (err) {
+                                callback(err);
+                            } else {
+                                writer.end(callback);
+                            }
+                        }
+                        if (chunkBuffer.length > 0) {
+                            const toSend = chunkBuffer;
+                            chunkBuffer = [];
+                            writer.chunk(toSend, end);
+                        } else {
+                            end(null);
+                        }
+                    },
+                    destroy(error: Error | null, callback: (error: (Error | null)) => void) {
+                        if (error) {
+                            writer.reject(errorString(error));
+                        }
+                        callback(null);
                     }
-                    writer.end("");
                 });
+                streamResponse.stream.pipe(writable);
                 streamResponse.stream.on('error', (err: any) => {
-                    writer.reject(err.message || err.toString());
+                    writable.destroy(err);
                 });
             } else {
                 throw new Error(`Expected stream but nothing returned`);
@@ -158,7 +183,7 @@ function wrapNativeFunctionWithStream(
             if (!!streamResponse && !!streamResponse.stream) {
                 streamResponse.stream.destroy(e);
             }
-            writer.reject(e.message || e.toString());
+            writer.reject(errorString(e));
         }
     };
 };

packages/cubejs-backend-native/src/stream.rs

@@ -1,3 +1,7 @@
+use cubesql::compile::engine::df::scan::{
+    transform_response, FieldValue, MemberField, RecordBatch, SchemaRef, ValueObject,
+};
+use std::future::Future;
 use std::sync::{Arc, Mutex};
 
 use cubesql::CubeError;
@@ -10,11 +14,14 @@ use crate::utils::bind_method;
 use tokio::sync::mpsc::{channel as mpsc_channel, Receiver, Sender};
 use tokio::sync::oneshot;
 
-type Chunk = Result<String, CubeError>;
+type Chunk = Option<Result<RecordBatch, CubeError>>;
 
 pub struct JsWriteStream {
     sender: Sender<Chunk>,
     ready_sender: Mutex<Option<oneshot::Sender<Result<(), CubeError>>>>,
+    tokio_handle: tokio::runtime::Handle,
+    schema: SchemaRef,
+    member_fields: Vec<MemberField>,
 }
 
 impl Finalize for JsWriteStream {}
@@ -45,10 +52,13 @@ impl JsWriteStream {
         Ok(obj)
     }
 
-    fn push_chunk(&self, chunk: String) -> bool {
-        match self.sender.try_send(Ok(chunk)) {
-            Err(_) => false,
-            Ok(_) => true,
+    fn push_chunk(&self, chunk: RecordBatch) -> impl Future<Output = Result<(), CubeError>> {
+        let sender = self.sender.clone();
+        async move {
+            sender
+                .send(Some(Ok(chunk)))
+                .await
+                .map_err(|e| CubeError::user(format!("Can't send to channel: {}", e)))
         }
     }
 
@@ -58,29 +68,119 @@ impl JsWriteStream {
         }
     }
 
-    fn end(&self) {
-        self.push_chunk("".to_string());
+    fn end(&self) -> impl Future<Output = Result<(), CubeError>> {
+        let sender = self.sender.clone();
+        async move {
+            sender
+                .send(None)
+                .await
+                .map_err(|e| CubeError::user(format!("Can't send to channel: {}", e)))
+        }
     }
 
     fn reject(&self, err: String) {
         if let Some(ready_sender) = self.ready_sender.lock().unwrap().take() {
             let _ = ready_sender.send(Err(CubeError::internal(err.to_string())));
         }
-        let _ = self.sender.try_send(Err(CubeError::internal(err)));
+        let _ = self.sender.try_send(Some(Err(CubeError::internal(err))));
     }
 }
 
-fn js_stream_push_chunk(mut cx: FunctionContext) -> JsResult<JsBoolean> {
+fn wait_for_future_and_execute_callback(
+    tokio_handle: tokio::runtime::Handle,
+    channel: Channel,
+    callback: Root<JsFunction>,
+    future: impl Future<Output = Result<(), CubeError>> + Send + Sync + 'static,
+) {
+    tokio_handle.spawn(async move {
+        let push_result = future.await;
+        let send_result = channel.try_send(move |mut cx| {
+            let undefined = cx.undefined();
+            let result = match push_result {
+                Ok(()) => {
+                    let args = vec![cx.null().upcast::<JsValue>(), cx.null().upcast::<JsValue>()];
+                    callback.into_inner(&mut cx).call(&mut cx, undefined, args)
+                }
+                Err(e) => {
+                    let args = vec![cx.string(e.message).upcast::<JsValue>()];
+                    callback.into_inner(&mut cx).call(&mut cx, undefined, args)
+                }
+            };
+            if let Err(e) = result {
+                log::error!("Error during callback execution: {}", e);
+            }
+            Ok(())
+        });
+        if let Err(e) = send_result {
+            log::error!("Can't execute callback on node event loop: {}", e);
+        }
+    });
+}
+
+pub struct JsValueObject<'a> {
+    pub cx: FunctionContext<'a>,
+    pub handle: Handle<'a, JsArray>,
+}
+
+impl ValueObject for JsValueObject<'_> {
+    fn len(&mut self) -> Result<usize, CubeError> {
+        Ok(self.handle.len(&mut self.cx) as usize)
+    }
+
+    fn get(&mut self, index: usize, field_name: &str) -> Result<FieldValue, CubeError> {
+        let value = self
+            .handle
+            .get::<JsObject, _, _>(&mut self.cx, index as u32)
+            .map_err(|e| {
+                CubeError::user(format!("Can't get object at array index {}: {}", index, e))
+            })?
+            .get::<JsValue, _, _>(&mut self.cx, field_name)
+            .map_err(|e| {
+                CubeError::user(format!("Can't get '{}' field value: {}", field_name, e))
+            })?;
+        if let Ok(s) = value.downcast::<JsString, _>(&mut self.cx) {
+            Ok(FieldValue::String(s.value(&mut self.cx)))
+        } else if let Ok(n) = value.downcast::<JsNumber, _>(&mut self.cx) {
+            Ok(FieldValue::Number(n.value(&mut self.cx)))
+        } else if let Ok(b) = value.downcast::<JsBoolean, _>(&mut self.cx) {
+            Ok(FieldValue::Bool(b.value(&mut self.cx)))
+        } else if value.downcast::<JsUndefined, _>(&mut self.cx).is_ok()
+            || value.downcast::<JsNull, _>(&mut self.cx).is_ok()
+        {
+            Ok(FieldValue::Null)
+        } else {
+            Err(CubeError::user(format!(
+                "Expected primitive value but found: {:?}",
+                value
+            )))
+        }
+    }
+}
+
+fn js_stream_push_chunk(mut cx: FunctionContext) -> JsResult<JsUndefined> {
     #[cfg(build = "debug")]
     trace!("JsWriteStream.push_chunk");
 
     let this = cx
         .this()
         .downcast_or_throw::<JsBox<JsWriteStream>, _>(&mut cx)?;
-    let result = cx.argument::<JsString>(0)?;
-    let result = this.push_chunk(result.value(&mut cx));
-
-    Ok(cx.boolean(result))
+    let chunk_array = cx.argument::<JsArray>(0)?;
+    let callback = cx.argument::<JsFunction>(1)?.root(&mut cx);
+    let mut value_object = JsValueObject {
+        cx,
+        handle: chunk_array,
+    };
+    let value =
+        transform_response(&mut value_object, this.schema.clone(), &this.member_fields).unwrap();
+    let future = this.push_chunk(value);
+    wait_for_future_and_execute_callback(
+        this.tokio_handle.clone(),
+        value_object.cx.channel(),
+        callback,
+        future,
+    );
+
+    Ok(value_object.cx.undefined())
 }
 
 fn js_stream_start(mut cx: FunctionContext) -> JsResult<JsUndefined> {
@@ -102,7 +202,9 @@ fn js_stream_end(mut cx: FunctionContext) -> JsResult<JsUndefined> {
     let this = cx
         .this()
         .downcast_or_throw::<JsBox<JsWriteStream>, _>(&mut cx)?;
-    this.end();
+    let future = this.end();
+    let callback = cx.argument::<JsFunction>(0)?.root(&mut cx);
+    wait_for_future_and_execute_callback(this.tokio_handle.clone(), cx.channel(), callback, future);
 
     Ok(cx.undefined())
 }
@@ -123,16 +225,19 @@ pub async fn call_js_with_stream_as_callback(
     channel: Arc<Channel>,
     js_method: Arc<Root<JsFunction>>,
     query: Option<String>,
+    schema: SchemaRef,
+    member_fields: Vec<MemberField>,
 ) -> Result<Receiver<Chunk>, CubeError> {
-    let chunk_size = std::env::var("CUBEJS_DB_QUERY_STREAM_HIGH_WATER_MARK")
+    let channel_size = std::env::var("CUBEJS_DB_QUERY_STREAM_HIGH_WATER_MARK")
         .ok()
         .map(|v| v.parse::<usize>().unwrap())
         .unwrap_or(8192);
-    let channel_size = 1_000_000 / chunk_size;
 
     let (sender, receiver) = mpsc_channel::<Chunk>(channel_size);
     let (ready_sender, ready_receiver) = oneshot::channel();
 
+    let tokio_handle = tokio::runtime::Handle::current();
+
     channel
         .try_send(move |mut cx| {
             // https://github.com/neon-bindings/neon/issues/672
@@ -144,6 +249,9 @@ pub async fn call_js_with_stream_as_callback(
             let stream = JsWriteStream {
                 sender,
                 ready_sender: Mutex::new(Some(ready_sender)),
+                tokio_handle,
+                schema,
+                member_fields,
             };
             let this = cx.undefined();
             let args: Vec<Handle<_>> = vec![

packages/cubejs-backend-native/src/transport.rs

@@ -3,6 +3,7 @@ use neon::prelude::*;
 
 use async_trait::async_trait;
 use cubeclient::models::{V1Error, V1LoadRequestQuery, V1LoadResponse, V1MetaResponse};
+use cubesql::compile::engine::df::scan::{MemberField, SchemaRef};
 use cubesql::{
     di_service,
     sql::AuthContextRef,
@@ -175,6 +176,8 @@ impl TransportService for NodeBridgeTransport {
         query: V1LoadRequestQuery,
         ctx: AuthContextRef,
         meta: LoadRequestMeta,
+        schema: SchemaRef,
+        member_fields: Vec<MemberField>,
     ) -> Result<CubeStreamReceiver, CubeError> {
         trace!("[transport] Request ->");
 
@@ -201,11 +204,13 @@ impl TransportService for NodeBridgeTransport {
                 self.channel.clone(),
                 self.on_load_stream.clone(),
                 Some(extra),
+                schema.clone(),
+                member_fields.clone(),
             )
             .await;
 
             if let Err(e) = &res {
-                if e.message.to_lowercase() == "continue wait" {
+                if e.message.to_lowercase().contains("continue wait") {
                     continue;
                 }
             }

packages/cubejs-query-orchestrator/src/orchestrator/QueryCache.ts

@@ -601,10 +601,10 @@ export class QueryCache {
               .then(([client]) => client.streamQuery(req.query, req.values))
               .then((source) => {
                 const cleanup = (error) => {
-                  if (!source.destroyed) {
+                  if (error && !source.destroyed) {
                     source.destroy(error);
                   }
-                  if (!target.destroyed) {
+                  if (error && !target.destroyed) {
                     target.destroy(error);
                   }
                   if (!logged && source.destroyed && target.destroyed) {
@@ -625,13 +625,13 @@ export class QueryCache {
                   }
                 };
 
-                source.once('end', cleanup);
+                source.once('end', () => cleanup(undefined));
                 source.once('error', cleanup);
-                source.once('close', cleanup);
+                source.once('close', () => cleanup(undefined));
 
-                target.once('end', cleanup);
+                target.once('end', () => cleanup(undefined));
                 target.once('error', cleanup);
-                target.once('close', cleanup);
+                target.once('close', () => cleanup(undefined));
 
                 source.pipe(target);
               })