From 92228bd7b3c36687c6b20b29ff92a17acaad0616 Mon Sep 17 00:00:00 2001
From: Ion Koutsouris <15728914+ion-elgreco@users.noreply.github.com>
Date: Mon, 22 Apr 2024 18:08:23 +0200
Subject: [PATCH] adds remove actions during overwrite create on existing table

---
 crates/core/src/operations/create.rs |  9 ++++++++-
 python/tests/test_create.py          | 13 ++++++++++++-
 2 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/crates/core/src/operations/create.rs b/crates/core/src/operations/create.rs
index b2092cafe8..9ad5bcfb9b 100644
--- a/crates/core/src/operations/create.rs
+++ b/crates/core/src/operations/create.rs
@@ -310,6 +310,7 @@ impl CreateBuilder {
         };
 
         let mut actions = vec![Action::Protocol(protocol), Action::Metadata(metadata)];
+
         actions.extend(
             self.actions
                 .into_iter()
@@ -329,7 +330,7 @@ impl std::future::IntoFuture for CreateBuilder {
         Box::pin(async move {
             let mode = this.mode;
             let app_metadata = this.metadata.clone().unwrap_or_default();
-            let (mut table, actions, operation) = this.into_table_and_actions()?;
+            let (mut table, mut actions, operation) = this.into_table_and_actions()?;
             let log_store = table.log_store();
 
             let table_state = if log_store.is_delta_table_location().await? {
@@ -342,6 +343,12 @@ impl std::future::IntoFuture for CreateBuilder {
                     }
                     SaveMode::Overwrite => {
                         table.load().await?;
+                        let remove_actions = table
+                            .snapshot()?
+                            .log_data()
+                            .into_iter()
+                            .map(|p| p.remove_action(true).into());
+                        actions.extend(remove_actions);
                         Some(table.snapshot()?)
                     }
                 }
diff --git a/python/tests/test_create.py b/python/tests/test_create.py
index 3852fc2bab..ceca8178c3 100644
--- a/python/tests/test_create.py
+++ b/python/tests/test_create.py
@@ -3,7 +3,7 @@
 import pyarrow as pa
 import pytest
 
-from deltalake import DeltaTable
+from deltalake import DeltaTable, write_deltalake
 from deltalake.exceptions import DeltaError
 
 
@@ -54,3 +54,14 @@ def test_create_schema(tmp_path: pathlib.Path, sample_data: pa.Table):
     )
 
     assert dt.schema().to_pyarrow() == sample_data.schema
+
+
+def test_create_or_replace_existing_table(
+    tmp_path: pathlib.Path, sample_data: pa.Table
+):
+    write_deltalake(table_or_uri=tmp_path, data=sample_data)
+    dt = DeltaTable.create(
+        tmp_path, sample_data.schema, partition_by=["utf8"], mode="overwrite"
+    )
+
+    assert dt.files() == []