fix(filters): aggresivity batch fix by aeppling · Pull Request #1895 · rtk-ai/rtk

aeppling · 2026-05-15T17:20:46Z

Summary

Some filter were too aggressive, lower filtering & more accurate results

…iltering Several filters dropped or rewrote information an agent needs, producing output that was misleading rather than merely compressed. This fixes the clear-cut cases surfaced by the TheDecipherist/rtk-test suite: - ls: stop hiding the `.env` file (removed `.env`/`env` from NOISE_DIRS). - git status: restore the explicit "HEAD detached at <sha>" line instead of the opaque "* HEAD (no branch)". - log/docker logs: recognise CRITICAL/FATAL/ALERT/EMERGENCY/SEVERE (and NOTICE) as severities so those lines are no longer silently filtered; expose `log` as a `rtk pipe` filter and accept a positional filter name. - wc: forward stdin to the child so `cat file | rtk wc` counts the pipe instead of reporting 0 (new RunOptions::inherit_stdin). - jest/vitest: include skipped/pending test count in the compact summary. - pytest: surface xfailed/xpassed counts and list XFAIL/XPASS entries with their reasons (adds `-rxX`). - ruff/eslint: list individual violations with file:line:col, not just rule/file group counts. - pnpm/npm list: render the actual package list instead of the false-positive "All packages up-to-date" when there's no upgrade info. - git add: stay silent on a no-op instead of printing an ambiguous "ok". - git stash: pass git's own message through instead of collapsing to "ok". - docker ps: keep the Status column (health/restart info) and list stopped/exited containers; docker images: show all images with full names; docker compose ps: use `-a` so crashed services stay visible. All 1870 unit tests pass; cargo fmt + clippy clean.

…ogs tail arg

KuSh

Minor nitpicks and questions. Otherwise LGTM!

KuSh · 2026-05-15T21:56:03Z

+        }
+    }
+    if running.len() > 20 {
+        rtk.push_str(&format!("  ... +{} more\n", running.len() - 20));


nitpick, prefer utf8 character …

Suggested change

rtk.push_str(&format!(" ... +{} more\n", running.len() - 20));

rtk.push_str(&format!(" … +{} more\n", running.len() - 20));

KuSh · 2026-05-15T21:56:29Z

-    if count > 15 {
-        rtk.push_str(&format!("  ... +{} more", count - 15));
+        if stopped.len() > 20 {
+            rtk.push_str(&format!("  ... +{} more\n", stopped.len() - 20));


Suggested change

rtk.push_str(&format!(" ... +{} more\n", stopped.len() - 20));

rtk.push_str(&format!(" … +{} more\n", stopped.len() - 20));

KuSh · 2026-05-15T21:57:21Z

-    if lines.len() > 15 {
-        rtk.push_str(&format!("  ... +{} more", lines.len() - 15));
+    if lines.len() > MAX_IMAGES {
+        rtk.push_str(&format!("  ... +{} more\n", lines.len() - MAX_IMAGES));


Suggested change

rtk.push_str(&format!(" ... +{} more\n", lines.len() - MAX_IMAGES));

rtk.push_str(&format!(" … +{} more\n", lines.len() - MAX_IMAGES));

KuSh · 2026-05-15T22:01:06Z

+
+    // Porcelain `-b` reduces a detached HEAD to "## HEAD (no branch)"; restore
+    // the explicit "HEAD detached at <sha>" from the plain status we captured.
+    if let Some(detached) = extract_detached_head(&raw_output) {


Wouldn't it be better to check that before calling format_status_output and pass it detached instead of replacing the result afterward?

KuSh · 2026-05-15T22:03:13Z

-        if pkgs.len() > 10 {
-            result.push_str(&format!("  ... +{} more\n", pkgs.len() - 10));
+        if pkgs.len() > MAX_PER_LETTER {
+            result.push_str(&format!("  ... +{} more\n", pkgs.len() - MAX_PER_LETTER));


while you're at it:

Suggested change

result.push_str(&format!(" ... +{} more\n", pkgs.len() - MAX_PER_LETTER));

result.push_str(&format!(" … +{} more\n", pkgs.len() - MAX_PER_LETTER));

KuSh · 2026-05-15T22:05:45Z

+    }
+    if diagnostics.len() > MAX_VIOLATIONS {
+        result.push_str(&format!(
+            "  ... +{} more violations\n",


Suggested change

" ... +{} more violations\n",

" … +{} more violations\n",

KuSh · 2026-05-15T22:08:32Z

+                lines.push(format!("  {} {}{}", dep.name, dep.current_version, dev));
+            }
+            if self.dependencies.len() > 50 {
+                lines.push(format!("  ... +{} more", self.dependencies.len() - 50));


Suggested change

lines.push(format!(" ... +{} more", self.dependencies.len() - 50));

lines.push(format!(" … +{} more", self.dependencies.len() - 50));

KuSh · 2026-05-15T22:10:51Z

+            for dep in self.dependencies.iter().take(50) {
+                let dev = if dep.dev_dependency { " (dev)" } else { "" };
+                lines.push(format!("  {} {}{}", dep.name, dep.current_version, dev));
+            }


Isn't there a hierarchy in deps? dep > dev dep > opt dep > ... shouldn't we first print dependencies, then if < 50 pick dev dependencies until we reach 50?

pszymkowiak · 2026-05-16T12:48:09Z

Tested the branch build against real commands — solid work, the filters no longer hide crashed containers, skipped/xfail tests, ruff locations, or report a false up-to-date. 11 commands verified end-to-end. Two points before merge:

1. git stash no-op is mis-reported on non-English locales (git.rs:1510)

result.combined().contains("No local changes") matches an English-only string. On a FR locale git stash prints "Pas de modifications locales à sauver", so the check fails and a no-op stash prints ok stashed — telling the agent its changes were saved when nothing was. (Verified on a FR machine.)

This is the tip of a larger issue: git.rs matches localized git output in ~9 places (rebase/merge detection L750-762, nothing to commit L846/L1078, Everything up-to-date L1123, Already up to date L1183, shortstat parsing L1194-1210…). The robust fix is to run rtk's internal git invocations with LC_ALL=C so git emits stable English — worth a dedicated change. At minimum, L1510 should not regress a no-op into ok stashed.

2. constants.rs — removing env/.env from NOISE_DIRS is the only change with no test and no rationale. env/ is a very common Python virtualenv dir; dropping it can flood rtk tree/find/grep with venv noise. Could you add a line explaining the intent (and ideally a test)?

Everything else is good to go.

- fix signal truncation gaps + kush reviews - added a new tee and hints function to give an hint with tail (avoid re-ingest of the head) - extracted patterns in constants

- docker ps reverts to plain (no -a augmentation) - docker ps -a / compose ps -a wired to the running/stopped split filter - ruff Violations: capped at 50 + force_tee_tail_hint recovery - tests for the ruff and pytest xfail caps

aeppling and others added 3 commits May 15, 2026 19:12

chore(filters): remove filter-level annotations and restore compose l…

16803a6

…ogs tail arg

fix(filters): add test for aggressive filter batch fix

f6b28c2

aeppling changed the title ~~fix: filter aggresivity batch fix~~ fix(filters): aggresivity batch fix May 15, 2026

KuSh approved these changes May 15, 2026

View reviewed changes

aeppling added 3 commits May 16, 2026 15:00

feat(hints): add tail hints for tee & hints + address reviews

46fe7c4

- fix signal truncation gaps + kush reviews - added a new tee and hints function to give an hint with tail (avoid re-ingest of the head) - extracted patterns in constants

fix: re-add env python as noisy dir

4eefe2f

fix: '...' ascii to unicode, remove some comments

3571d52

aeppling mentioned this pull request May 17, 2026

fix(tee): safe truncation caps and tee/hint coverage #1928

Merged

aeppling merged commit 90c285c into develop May 19, 2026
17 checks passed

aeppling mentioned this pull request May 19, 2026

Next Release #1879

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(filters): aggresivity batch fix#1895

fix(filters): aggresivity batch fix#1895
aeppling merged 7 commits into
developfrom
fix/aggressive-filters-batch

aeppling commented May 15, 2026

Uh oh!

KuSh left a comment

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

KuSh May 15, 2026

Uh oh!

pszymkowiak commented May 16, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

	rtk.push_str(&format!(" ... +{} more\n", running.len() - 20));
	rtk.push_str(&format!(" … +{} more\n", running.len() - 20));

	rtk.push_str(&format!(" ... +{} more\n", stopped.len() - 20));
	rtk.push_str(&format!(" … +{} more\n", stopped.len() - 20));

	rtk.push_str(&format!(" ... +{} more\n", lines.len() - MAX_IMAGES));
	rtk.push_str(&format!(" … +{} more\n", lines.len() - MAX_IMAGES));

	result.push_str(&format!(" ... +{} more\n", pkgs.len() - MAX_PER_LETTER));
	result.push_str(&format!(" … +{} more\n", pkgs.len() - MAX_PER_LETTER));

	lines.push(format!(" ... +{} more", self.dependencies.len() - 50));
	lines.push(format!(" … +{} more", self.dependencies.len() - 50));

Conversation

aeppling commented May 15, 2026

Summary

Uh oh!

KuSh left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

pszymkowiak commented May 16, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants