feat: query engine impl on datafusion (#10)

* feat: query engine impl on datafusion

* feat: adds physical_optimizer, physical_planner and executor

* feat: impl adpaters between datafuion and greptime query engine core APIs.

* feat: impl PhysicalPlanAdapter and ExecutionPlanAdapter

* feat: rename table datafusion mod to adapter

* fix: clippy warning

* fix: conflicts with develop branch

* feat: add database mod

* fix: CR comment

* fix: by CR comments

* fix: conflicts with develop branch

* fix: by CR comments

src/common/query/Cargo.toml

@@ -0,0 +1,7 @@
+[package]
+name = "common-query"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+datafusion = { git = "https://github.com/apache/arrow-datafusion.git" , branch = "arrow2", features = ["simd"]}

src/common/query/src/lib.rs

@@ -0,0 +1 @@
+pub mod logical_plan;

src/common/query/src/logical_plan/expr.rs

@@ -0,0 +1,18 @@
+use datafusion::logical_plan::Expr as DfExpr;
+
+/// Central struct of query API.
+/// Represent logical expressions such as `A + 1`, or `CAST(c1 AS int)`.
+#[derive(Clone, PartialEq, Hash)]
+pub struct Expr {
+    df_expr: DfExpr,
+}
+
+impl Expr {
+    pub fn new(df_expr: DfExpr) -> Self {
+        Self { df_expr }
+    }
+
+    pub fn df_expr(&self) -> &DfExpr {
+        &self.df_expr
+    }
+}

src/common/query/src/logical_plan/mod.rs

@@ -0,0 +1,3 @@
+mod expr;
+
+pub use self::expr::Expr;

src/common/recordbatch/Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "common-recordbatch"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies.arrow]
+package = "arrow2"
+version="0.10"
+features = ["io_csv", "io_json", "io_parquet", "io_parquet_compression", "io_ipc", "ahash", "compute"]
+
+[dependencies]
+datafusion = { git = "https://github.com/apache/arrow-datafusion.git" , branch = "arrow2", features = ["simd"]}
+datafusion-common = { git = "https://github.com/apache/arrow-datafusion.git" , branch = "arrow2"}
+datatypes = {path ="../../datatypes" }
+futures = "0.3"
+snafu = "0.7.0"

src/common/recordbatch/src/error.rs

@@ -0,0 +1,10 @@
+use arrow::error::ArrowError;
+use snafu::Snafu;
+
+#[derive(Debug, Snafu)]
+#[snafu(visibility(pub))]
+pub enum Error {
+    #[snafu(display("Arrow error: {}", source))]
+    Arrow { source: ArrowError },
+}
+pub type Result<T> = std::result::Result<T, Error>;

src/common/recordbatch/src/lib.rs

@@ -0,0 +1,15 @@
+pub mod error;
+mod recordbatch;
+
+use std::pin::Pin;
+
+use datatypes::schema::SchemaRef;
+use error::Result;
+use futures::Stream;
+pub use recordbatch::RecordBatch;
+
+pub trait RecordBatchStream: Stream<Item = Result<RecordBatch>> {
+    fn schema(&self) -> SchemaRef;
+}
+
+pub type SendableRecordBatchStream = Pin<Box<dyn RecordBatchStream + Send>>;

src/common/recordbatch/src/recordbatch.rs

@@ -0,0 +1,10 @@
+use std::sync::Arc;
+
+use datafusion_common::record_batch::RecordBatch as DfRecordBatch;
+use datatypes::schema::Schema;
+
+#[derive(Clone, Debug)]
+pub struct RecordBatch {
+    pub schema: Arc<Schema>,
+    pub df_recordbatch: DfRecordBatch,
+}