From d3abd8774ed2caa13db253ee232791e5a2d2d8c5 Mon Sep 17 00:00:00 2001
From: jif <jif@openai.com>
Date: Wed, 10 Jun 2026 18:23:42 +0100
Subject: [PATCH] Fix compressed rollout search path matching (#27407)

## Why

`thread/search` found content inside compressed rollouts but could drop
the result when joining it with SQLite-backed thread metadata. Search
returned the physical `.jsonl.zst` path while SQLite retained the
logical `.jsonl` path, so exact path matching failed.

## What changed

- Key rollout search matches by their canonical logical `.jsonl` path,
independent of the on-disk representation.
- Canonicalize thread-list paths before joining them with content-search
matches.
- Update compressed-rollout coverage to assert the logical-path
contract.

## Validation

- Ran `just fmt`.
- Ran `git diff --check`.
- Tests and Clippy were intentionally left to CI.
---
 codex-rs/rollout/src/compression_tests.rs         |  5 ++---
 codex-rs/rollout/src/search.rs                    | 11 +++++++++--
 codex-rs/thread-store/src/local/search_threads.rs |  3 ++-
 3 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/codex-rs/rollout/src/compression_tests.rs b/codex-rs/rollout/src/compression_tests.rs
index eb8111296..682361ea9 100644
--- a/codex-rs/rollout/src/compression_tests.rs
+++ b/codex-rs/rollout/src/compression_tests.rs
@@ -106,14 +106,13 @@ async fn append_rollout_item_materializes_compressed_rollout() -> anyhow::Result
 }
 
 #[tokio::test]
-async fn search_rollout_matches_returns_compressed_snippet() -> anyhow::Result<()> {
+async fn search_rollout_matches_uses_logical_path_for_compressed_rollout() -> anyhow::Result<()> {
     let home = TempDir::new()?;
     let uuid = Uuid::from_u128(15);
     let thread_id = ThreadId::from_string(&uuid.to_string())?;
     let rollout_path = rollout_path(home.path(), "2025-01-03T12-00-00", uuid);
     write_rollout(&rollout_path, thread_id, "targeted search term")?;
     compress_now(&rollout_path)?;
-    let compressed_path = compressed_rollout_path(&rollout_path);
 
     let matches = search_rollout_matches(
         std::path::Path::new("missing-rg-for-test"),
@@ -124,7 +123,7 @@ async fn search_rollout_matches_returns_compressed_snippet() -> anyhow::Result<(
     .await?;
 
     assert_eq!(
-        matches.get(compressed_path.as_path()),
+        matches.get(rollout_path.as_path()),
         Some(&Some("targeted search term".to_string()))
     );
     Ok(())
diff --git a/codex-rs/rollout/src/search.rs b/codex-rs/rollout/src/search.rs
index 705d24e7b..622fac8b0 100644
--- a/codex-rs/rollout/src/search.rs
+++ b/codex-rs/rollout/src/search.rs
@@ -21,6 +21,7 @@ use super::compression;
 const MATCH_CONTEXT_BEFORE_CHARS: usize = 48;
 const MATCH_CONTEXT_AFTER_CHARS: usize = 96;
 
+/// Search matches keyed by the canonical `.jsonl` path for each rollout.
 pub type RolloutSearchMatches = HashMap<PathBuf, Option<String>>;
 
 pub async fn search_rollout_paths(
@@ -145,7 +146,10 @@ async fn scan_rollout_matches(
                 if let Some(snippet) =
                     first_rollout_content_match_snippet(rollout_file.path(), search_term).await?
                 {
-                    matches.insert(rollout_file.into_path(), Some(snippet));
+                    matches.insert(
+                        compression::plain_rollout_path(rollout_file.path()),
+                        Some(snippet),
+                    );
                 }
                 continue;
             }
@@ -217,7 +221,10 @@ async fn scan_compressed_rollout_matches(
             if let Some(snippet) =
                 first_rollout_content_match_snippet(rollout_file.path(), search_term).await?
             {
-                matches.insert(rollout_file.into_path(), Some(snippet));
+                matches.insert(
+                    compression::plain_rollout_path(rollout_file.path()),
+                    Some(snippet),
+                );
             }
         }
     }
diff --git a/codex-rs/thread-store/src/local/search_threads.rs b/codex-rs/thread-store/src/local/search_threads.rs
index 20f5da824..c9394c396 100644
--- a/codex-rs/thread-store/src/local/search_threads.rs
+++ b/codex-rs/thread-store/src/local/search_threads.rs
@@ -109,7 +109,8 @@ pub(super) async fn search_threads(
         )
         .await?;
         for item in page.items {
-            let Some(snippet) = (match remaining_rollouts.remove(item.path.as_path()) {
+            let logical_path = codex_rollout::plain_rollout_path(item.path.as_path());
+            let Some(snippet) = (match remaining_rollouts.remove(logical_path.as_path()) {
                 Some(Some(snippet)) => Some(snippet),
                 Some(None) => first_rollout_content_match_snippet(item.path.as_path(), search_term)
                     .await