Release 0.66.0-alpha.5

## Summary
This PR introduces an End to End test suite for apply-patch, so we can
easily validate behavior against other implementations as well.

## Testing
- [x] These are tests

codex-rs/Cargo.toml

@@ -47,7 +47,7 @@ members = [
 resolver = "2"
 
 [workspace.package]
-version = "0.0.0"
+version = "0.66.0-alpha.5"
 # Track the edition for all workspace crates in one place. Individual
 # crates can still override this value, but keeping it here means new
 # crates created with `cargo new -w ...` automatically inherit the 2024

codex-rs/app-server-protocol/src/protocol/v2.rs

@@ -942,6 +942,9 @@ pub struct TurnError {
 #[ts(export_to = "v2/")]
 pub struct ErrorNotification {
     pub error: TurnError,
+    // Set to true if the error is transient and the app-server process will automatically retry.
+    // If true, this will not interrupt a turn.
+    pub will_retry: bool,
     pub thread_id: String,
     pub turn_id: String,
 }
@@ -1141,6 +1144,9 @@ pub enum ThreadItem {
         arguments: JsonValue,
         result: Option<McpToolCallResult>,
         error: Option<McpToolCallError>,
+        /// The duration of the MCP tool call in milliseconds.
+        #[ts(type = "number | null")]
+        duration_ms: Option<i64>,
     },
     #[serde(rename_all = "camelCase")]
     #[ts(rename_all = "camelCase")]

codex-rs/app-server/src/bespoke_event_handling.rs

@@ -333,6 +333,7 @@ pub(crate) async fn apply_bespoke_event_handling(
             outgoing
                 .send_server_notification(ServerNotification::Error(ErrorNotification {
                     error: turn_error,
+                    will_retry: false,
                     thread_id: conversation_id.to_string(),
                     turn_id: event_turn_id.clone(),
                 }))
@@ -348,6 +349,7 @@ pub(crate) async fn apply_bespoke_event_handling(
             outgoing
                 .send_server_notification(ServerNotification::Error(ErrorNotification {
                     error: turn_error,
+                    will_retry: true,
                     thread_id: conversation_id.to_string(),
                     turn_id: event_turn_id.clone(),
                 }))
@@ -1178,6 +1180,7 @@ async fn construct_mcp_tool_call_notification(
         arguments: begin_event.invocation.arguments.unwrap_or(JsonValue::Null),
         result: None,
         error: None,
+        duration_ms: None,
     };
     ItemStartedNotification {
         thread_id,
@@ -1197,6 +1200,7 @@ async fn construct_mcp_tool_call_end_notification(
     } else {
         McpToolCallStatus::Failed
     };
+    let duration_ms = i64::try_from(end_event.duration.as_millis()).ok();
 
     let (result, error) = match &end_event.result {
         Ok(value) => (
@@ -1222,6 +1226,7 @@ async fn construct_mcp_tool_call_end_notification(
         arguments: end_event.invocation.arguments.unwrap_or(JsonValue::Null),
         result,
         error,
+        duration_ms,
     };
     ItemCompletedNotification {
         thread_id,
@@ -1598,6 +1603,7 @@ mod tests {
                 arguments: serde_json::json!({"server": ""}),
                 result: None,
                 error: None,
+                duration_ms: None,
             },
         };
 
@@ -1751,6 +1757,7 @@ mod tests {
                 arguments: JsonValue::Null,
                 result: None,
                 error: None,
+                duration_ms: None,
             },
         };
 
@@ -1804,6 +1811,7 @@ mod tests {
                     structured_content: None,
                 }),
                 error: None,
+                duration_ms: Some(0),
             },
         };
 
@@ -1845,6 +1853,7 @@ mod tests {
                 error: Some(McpToolCallError {
                     message: "boom".to_string(),
                 }),
+                duration_ms: Some(1),
             },
         };
 

codex-rs/apply-patch/tests/fixtures/scenarios/.gitattributes

@@ -0,0 +1 @@
+** text eol=lf

codex-rs/apply-patch/tests/fixtures/scenarios/001_add_file/expected/bar.md

@@ -0,0 +1 @@
+This is a new file

codex-rs/apply-patch/tests/fixtures/scenarios/001_add_file/patch.txt

@@ -0,0 +1,4 @@
+*** Begin Patch
+*** Add File: bar.md
++This is a new file
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/002_multiple_operations/expected/modify.txt

@@ -0,0 +1,2 @@
+line1
+changed

codex-rs/apply-patch/tests/fixtures/scenarios/002_multiple_operations/expected/nested/new.txt

@@ -0,0 +1 @@
+created

codex-rs/apply-patch/tests/fixtures/scenarios/002_multiple_operations/input/delete.txt

@@ -0,0 +1 @@
+obsolete

codex-rs/apply-patch/tests/fixtures/scenarios/002_multiple_operations/input/modify.txt

@@ -0,0 +1,2 @@
+line1
+line2

codex-rs/apply-patch/tests/fixtures/scenarios/002_multiple_operations/patch.txt

@@ -0,0 +1,9 @@
+*** Begin Patch
+*** Add File: nested/new.txt
++created
+*** Delete File: delete.txt
+*** Update File: modify.txt
+@@
+-line2
++changed
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/003_multiple_chunks/expected/multi.txt

@@ -0,0 +1,4 @@
+line1
+changed2
+line3
+changed4

codex-rs/apply-patch/tests/fixtures/scenarios/003_multiple_chunks/input/multi.txt

@@ -0,0 +1,4 @@
+line1
+line2
+line3
+line4

codex-rs/apply-patch/tests/fixtures/scenarios/003_multiple_chunks/patch.txt

@@ -0,0 +1,9 @@
+*** Begin Patch
+*** Update File: multi.txt
+@@
+-line2
++changed2
+@@
+-line4
++changed4
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/004_move_to_new_directory/expected/old/other.txt

@@ -0,0 +1 @@
+unrelated file

codex-rs/apply-patch/tests/fixtures/scenarios/004_move_to_new_directory/expected/renamed/dir/name.txt

@@ -0,0 +1 @@
+new content

codex-rs/apply-patch/tests/fixtures/scenarios/004_move_to_new_directory/input/old/name.txt

@@ -0,0 +1 @@
+old content

codex-rs/apply-patch/tests/fixtures/scenarios/004_move_to_new_directory/input/old/other.txt

@@ -0,0 +1 @@
+unrelated file

codex-rs/apply-patch/tests/fixtures/scenarios/004_move_to_new_directory/patch.txt

@@ -0,0 +1,7 @@
+*** Begin Patch
+*** Update File: old/name.txt
+*** Move to: renamed/dir/name.txt
+@@
+-old content
++new content
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/005_rejects_empty_patch/patch.txt

@@ -0,0 +1,2 @@
+*** Begin Patch
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/006_rejects_missing_context/expected/modify.txt

@@ -0,0 +1,2 @@
+line1
+line2

codex-rs/apply-patch/tests/fixtures/scenarios/006_rejects_missing_context/input/modify.txt

@@ -0,0 +1,2 @@
+line1
+line2

codex-rs/apply-patch/tests/fixtures/scenarios/006_rejects_missing_context/patch.txt

@@ -0,0 +1,6 @@
+*** Begin Patch
+*** Update File: modify.txt
+@@
+-missing
++changed
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/007_rejects_missing_file_delete/patch.txt

@@ -0,0 +1,3 @@
+*** Begin Patch
+*** Delete File: missing.txt
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/008_rejects_empty_update_hunk/patch.txt

@@ -0,0 +1,3 @@
+*** Begin Patch
+*** Update File: foo.txt
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/009_requires_existing_file_for_update/patch.txt

@@ -0,0 +1,6 @@
+*** Begin Patch
+*** Update File: missing.txt
+@@
+-old
++new
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/010_move_overwrites_existing_destination/expected/old/other.txt

@@ -0,0 +1 @@
+unrelated file

codex-rs/apply-patch/tests/fixtures/scenarios/010_move_overwrites_existing_destination/expected/renamed/dir/name.txt

@@ -0,0 +1 @@
+new

codex-rs/apply-patch/tests/fixtures/scenarios/010_move_overwrites_existing_destination/input/old/name.txt

@@ -0,0 +1 @@
+from

codex-rs/apply-patch/tests/fixtures/scenarios/010_move_overwrites_existing_destination/input/old/other.txt

@@ -0,0 +1 @@
+unrelated file

codex-rs/apply-patch/tests/fixtures/scenarios/010_move_overwrites_existing_destination/input/renamed/dir/name.txt

@@ -0,0 +1 @@
+existing

codex-rs/apply-patch/tests/fixtures/scenarios/010_move_overwrites_existing_destination/patch.txt

@@ -0,0 +1,7 @@
+*** Begin Patch
+*** Update File: old/name.txt
+*** Move to: renamed/dir/name.txt
+@@
+-from
++new
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/011_add_overwrites_existing_file/expected/duplicate.txt

@@ -0,0 +1 @@
+new content

codex-rs/apply-patch/tests/fixtures/scenarios/011_add_overwrites_existing_file/input/duplicate.txt

@@ -0,0 +1 @@
+old content

codex-rs/apply-patch/tests/fixtures/scenarios/011_add_overwrites_existing_file/patch.txt

@@ -0,0 +1,4 @@
+*** Begin Patch
+*** Add File: duplicate.txt
++new content
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/012_delete_directory_fails/patch.txt

@@ -0,0 +1,3 @@
+*** Begin Patch
+*** Delete File: dir
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/013_rejects_invalid_hunk_header/patch.txt

@@ -0,0 +1,3 @@
+*** Begin Patch
+*** Frobnicate File: foo
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/014_update_file_appends_trailing_newline/expected/no_newline.txt

@@ -0,0 +1,2 @@
+first line
+second line

codex-rs/apply-patch/tests/fixtures/scenarios/014_update_file_appends_trailing_newline/input/no_newline.txt

@@ -0,0 +1 @@
+no newline at end

codex-rs/apply-patch/tests/fixtures/scenarios/014_update_file_appends_trailing_newline/patch.txt

@@ -0,0 +1,7 @@
+*** Begin Patch
+*** Update File: no_newline.txt
+@@
+-no newline at end
++first line
++second line
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/015_failure_after_partial_success_leaves_changes/expected/created.txt

@@ -0,0 +1 @@
+hello

codex-rs/apply-patch/tests/fixtures/scenarios/015_failure_after_partial_success_leaves_changes/patch.txt

@@ -0,0 +1,8 @@
+*** Begin Patch
+*** Add File: created.txt
++hello
+*** Update File: missing.txt
+@@
+-old
++new
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/016_pure_addition_update_chunk/expected/input.txt

@@ -0,0 +1,4 @@
+line1
+line2
+added line 1
+added line 2

codex-rs/apply-patch/tests/fixtures/scenarios/016_pure_addition_update_chunk/input/input.txt

@@ -0,0 +1,2 @@
+line1
+line2

codex-rs/apply-patch/tests/fixtures/scenarios/016_pure_addition_update_chunk/patch.txt

@@ -0,0 +1,6 @@
+*** Begin Patch
+*** Update File: input.txt
+@@
++added line 1
++added line 2
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/017_whitespace_padded_hunk_header/expected/foo.txt

@@ -0,0 +1 @@
+new

codex-rs/apply-patch/tests/fixtures/scenarios/017_whitespace_padded_hunk_header/input/foo.txt

@@ -0,0 +1 @@
+old

codex-rs/apply-patch/tests/fixtures/scenarios/017_whitespace_padded_hunk_header/patch.txt

@@ -0,0 +1,6 @@
+*** Begin Patch
+  *** Update File: foo.txt
+@@
+-old
++new
+*** End Patch

codex-rs/apply-patch/tests/fixtures/scenarios/018_whitespace_padded_patch_markers/expected/file.txt

@@ -0,0 +1 @@
+two

codex-rs/apply-patch/tests/fixtures/scenarios/018_whitespace_padded_patch_markers/input/file.txt

@@ -0,0 +1 @@
+one

codex-rs/apply-patch/tests/fixtures/scenarios/018_whitespace_padded_patch_markers/patch.txt

@@ -0,0 +1,6 @@
+ *** Begin Patch
+*** Update File: file.txt
+@@
+-one
++two
+*** End Patch 

codex-rs/apply-patch/tests/fixtures/scenarios/README.md

@@ -0,0 +1,18 @@
+# Overview
+This directory is a collection of end to end tests for the apply-patch specification, meant to be easily portable to other languages or platforms.
+
+
+# Specification
+Each test case is one directory, composed of input state (input/), the patch operation (patch.txt), and the expected final state (expected/). This structure is designed to keep tests simple (i.e. test exactly one patch at a time) while still providing enough flexibility to test any given operation across files.
+
+Here's what this would look like for a simple test apply-patch test case to create a new file:
+
+```
+001_add/
+  input/
+    foo.md
+  expected/
+    foo.md
+    bar.md
+  patch.txt
+```

codex-rs/apply-patch/tests/suite/mod.rs

@@ -1,3 +1,4 @@
 mod cli;
+mod scenarios;
 #[cfg(not(target_os = "windows"))]
 mod tool;

codex-rs/apply-patch/tests/suite/scenarios.rs

@@ -0,0 +1,114 @@
+use assert_cmd::prelude::*;
+use pretty_assertions::assert_eq;
+use std::collections::BTreeMap;
+use std::fs;
+use std::path::Path;
+use std::path::PathBuf;
+use std::process::Command;
+use tempfile::tempdir;
+
+#[test]
+fn test_apply_patch_scenarios() -> anyhow::Result<()> {
+    for scenario in fs::read_dir("tests/fixtures/scenarios")? {
+        let scenario = scenario?;
+        let path = scenario.path();
+        if path.is_dir() {
+            run_apply_patch_scenario(&path)?;
+        }
+    }
+    Ok(())
+}
+
+/// Reads a scenario directory, copies the input files to a temporary directory, runs apply-patch,
+/// and asserts that the final state matches the expected state exactly.
+fn run_apply_patch_scenario(dir: &Path) -> anyhow::Result<()> {
+    let tmp = tempdir()?;
+
+    // Copy the input files to the temporary directory
+    let input_dir = dir.join("input");
+    if input_dir.is_dir() {
+        copy_dir_recursive(&input_dir, tmp.path())?;
+    }
+
+    // Read the patch.txt file
+    let patch = fs::read_to_string(dir.join("patch.txt"))?;
+
+    // Run apply_patch in the temporary directory. We intentionally do not assert
+    // on the exit status here; the scenarios are specified purely in terms of
+    // final filesystem state, which we compare below.
+    Command::cargo_bin("apply_patch")?
+        .arg(patch)
+        .current_dir(tmp.path())
+        .output()?;
+
+    // Assert that the final state matches the expected state exactly
+    let expected_dir = dir.join("expected");
+    let expected_snapshot = snapshot_dir(&expected_dir)?;
+    let actual_snapshot = snapshot_dir(tmp.path())?;
+
+    assert_eq!(
+        actual_snapshot,
+        expected_snapshot,
+        "Scenario {} did not match expected final state",
+        dir.display()
+    );
+
+    Ok(())
+}
+
+#[derive(Debug, Clone, PartialEq, Eq)]
+enum Entry {
+    File(Vec<u8>),
+    Dir,
+}
+
+fn snapshot_dir(root: &Path) -> anyhow::Result<BTreeMap<PathBuf, Entry>> {
+    let mut entries = BTreeMap::new();
+    if root.is_dir() {
+        snapshot_dir_recursive(root, root, &mut entries)?;
+    }
+    Ok(entries)
+}
+
+fn snapshot_dir_recursive(
+    base: &Path,
+    dir: &Path,
+    entries: &mut BTreeMap<PathBuf, Entry>,
+) -> anyhow::Result<()> {
+    for entry in fs::read_dir(dir)? {
+        let entry = entry?;
+        let path = entry.path();
+        let Some(stripped) = path.strip_prefix(base).ok() else {
+            continue;
+        };
+        let rel = stripped.to_path_buf();
+        let file_type = entry.file_type()?;
+        if file_type.is_dir() {
+            entries.insert(rel.clone(), Entry::Dir);
+            snapshot_dir_recursive(base, &path, entries)?;
+        } else if file_type.is_file() {
+            let contents = fs::read(&path)?;
+            entries.insert(rel, Entry::File(contents));
+        }
+    }
+    Ok(())
+}
+
+fn copy_dir_recursive(src: &Path, dst: &Path) -> anyhow::Result<()> {
+    for entry in fs::read_dir(src)? {
+        let entry = entry?;
+        let path = entry.path();
+        let file_type = entry.file_type()?;
+        let dest_path = dst.join(entry.file_name());
+        if file_type.is_dir() {
+            fs::create_dir_all(&dest_path)?;
+            copy_dir_recursive(&path, &dest_path)?;
+        } else if file_type.is_file() {
+            if let Some(parent) = dest_path.parent() {
+                fs::create_dir_all(parent)?;
+            }
+            fs::copy(&path, &dest_path)?;
+        }
+    }
+    Ok(())
+}

codex-rs/codex-api/src/sse/chat.rs

@@ -161,8 +161,10 @@ pub async fn process_chat_sse<S>(
                         }
 
                         if let Some(func) = tool_call.get("function") {
-                            if let Some(fname) = func.get("name").and_then(|n| n.as_str()) {
-                                call_state.name = Some(fname.to_string());
+                            if let Some(fname) = func.get("name").and_then(|n| n.as_str())
+                                && !fname.is_empty()
+                            {
+                                call_state.name.get_or_insert_with(|| fname.to_string());
                             }
                             if let Some(arguments) = func.get("arguments").and_then(|a| a.as_str())
                             {
@@ -432,6 +434,47 @@ mod tests {
         );
     }
 
+    #[tokio::test]
+    async fn preserves_tool_call_name_when_empty_deltas_arrive() {
+        let delta_with_name = json!({
+            "choices": [{
+                "delta": {
+                    "tool_calls": [{
+                        "id": "call_a",
+                        "function": { "name": "do_a" }
+                    }]
+                }
+            }]
+        });
+
+        let delta_with_empty_name = json!({
+            "choices": [{
+                "delta": {
+                    "tool_calls": [{
+                        "id": "call_a",
+                        "function": { "name": "", "arguments": "{}" }
+                    }]
+                }
+            }]
+        });
+
+        let finish = json!({
+            "choices": [{
+                "finish_reason": "tool_calls"
+            }]
+        });
+
+        let body = build_body(&[delta_with_name, delta_with_empty_name, finish]);
+        let events = collect_events(&body).await;
+        assert_matches!(
+            &events[..],
+            [
+                ResponseEvent::OutputItemDone(ResponseItem::FunctionCall { name, arguments, .. }),
+                ResponseEvent::Completed { .. }
+            ] if name == "do_a" && arguments == "{}"
+        );
+    }
+
     #[tokio::test]
     async fn emits_tool_calls_even_when_content_and_reasoning_present() {
         let delta_content_and_tools = json!({

codex-rs/core/src/codex.rs

@@ -11,6 +11,7 @@ use crate::compact;
 use crate::compact::run_inline_auto_compact_task;
 use crate::compact::should_use_remote_compact_task;
 use crate::compact_remote::run_inline_remote_auto_compact_task;
+use crate::exec_policy::load_exec_policy_for_features;
 use crate::features::Feature;
 use crate::features::Features;
 use crate::openai_models::models_manager::ModelsManager;
@@ -174,9 +175,10 @@ impl Codex {
 
         let user_instructions = get_user_instructions(&config).await;
 
-        let exec_policy = crate::exec_policy::exec_policy_for(&config.features, &config.codex_home)
+        let exec_policy = load_exec_policy_for_features(&config.features, &config.codex_home)
             .await
             .map_err(|err| CodexErr::Fatal(format!("failed to load execpolicy: {err}")))?;
+        let exec_policy = Arc::new(RwLock::new(exec_policy));
 
         let config = Arc::new(config);
 

codex-rs/core/src/exec_policy.rs

@@ -73,14 +73,18 @@ pub enum ExecPolicyUpdateError {
     FeatureDisabled,
 }
 
-pub(crate) async fn exec_policy_for(
+pub(crate) async fn load_exec_policy_for_features(
     features: &Features,
     codex_home: &Path,
-) -> Result<Arc<RwLock<Policy>>, ExecPolicyError> {
+) -> Result<Policy, ExecPolicyError> {
     if !features.enabled(Feature::ExecPolicy) {
-        return Ok(Arc::new(RwLock::new(Policy::empty())));
+        Ok(Policy::empty())
+    } else {
+        load_exec_policy(codex_home).await
     }
+}
 
+pub async fn load_exec_policy(codex_home: &Path) -> Result<Policy, ExecPolicyError> {
     let policy_dir = codex_home.join(POLICY_DIR_NAME);
     let policy_paths = collect_policy_files(&policy_dir).await?;
 
@@ -102,7 +106,7 @@ pub(crate) async fn exec_policy_for(
             })?;
     }
 
-    let policy = Arc::new(RwLock::new(parser.build()));
+    let policy = parser.build();
     tracing::debug!(
         "loaded execpolicy from {} files in {}",
         policy_paths.len(),
@@ -306,7 +310,7 @@ mod tests {
         features.disable(Feature::ExecPolicy);
         let temp_dir = tempdir().expect("create temp dir");
 
-        let policy = exec_policy_for(&features, temp_dir.path())
+        let policy = load_exec_policy_for_features(&features, temp_dir.path())
             .await
             .expect("policy result");
 
@@ -319,10 +323,7 @@ mod tests {
                     decision: Decision::Allow
                 }],
             },
-            policy
-                .read()
-                .await
-                .check_multiple(commands.iter(), &|_| Decision::Allow)
+            policy.check_multiple(commands.iter(), &|_| Decision::Allow)
         );
         assert!(!temp_dir.path().join(POLICY_DIR_NAME).exists());
     }
@@ -350,7 +351,7 @@ mod tests {
         )
         .expect("write policy file");
 
-        let policy = exec_policy_for(&Features::with_defaults(), temp_dir.path())
+        let policy = load_exec_policy(temp_dir.path())
             .await
             .expect("policy result");
         let command = [vec!["rm".to_string()]];
@@ -362,10 +363,7 @@ mod tests {
                     decision: Decision::Forbidden
                 }],
             },
-            policy
-                .read()
-                .await
-                .check_multiple(command.iter(), &|_| Decision::Allow)
+            policy.check_multiple(command.iter(), &|_| Decision::Allow)
         );
     }
 
@@ -378,7 +376,7 @@ mod tests {
         )
         .expect("write policy file");
 
-        let policy = exec_policy_for(&Features::with_defaults(), temp_dir.path())
+        let policy = load_exec_policy(temp_dir.path())
             .await
             .expect("policy result");
         let command = [vec!["ls".to_string()]];
@@ -390,10 +388,7 @@ mod tests {
                     decision: Decision::Allow
                 }],
             },
-            policy
-                .read()
-                .await
-                .check_multiple(command.iter(), &|_| Decision::Allow)
+            policy.check_multiple(command.iter(), &|_| Decision::Allow)
         );
     }
 

codex-rs/tui/src/bottom_pane/chat_composer.rs

@@ -1504,14 +1504,9 @@ impl ChatComposer {
             return false;
         }
 
-        let toggles = matches!(
-            key_event,
-            KeyEvent {
-                code: KeyCode::Char('?'),
-                modifiers: KeyModifiers::NONE,
-                ..
-            } if self.is_empty()
-        );
+        let toggles = matches!(key_event.code, KeyCode::Char('?'))
+            && !has_ctrl_or_alt(key_event.modifiers)
+            && self.is_empty();
 
         if !toggles {
             return false;