From e1a96f18c0024b8548a39306c073800b9f4d5811 Mon Sep 17 00:00:00 2001
From: sora <2075279110@qq.com>
Date: Mon, 14 Sep 2026 03:51:25 +0000
Subject: [PATCH] Scoring phase retargets the same progress bar

The bar kept its stale 100% generation state during scoring with only
the description counter moving. Now set_scoring() refills the bar with
judged samples (0->100%, fresh clock/rate/eta for the phase), clears
the generation '+N new' marker, and the next bench retargets back to
generating cleanly.

Co-Authored-By: Claude <noreply@anthropic.com>
---
 evalharness/cli.py                    | 10 +++++++---
 evalharness/progress/rich_terminal.py | 24 ++++++++++++++++++++++++
 2 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/evalharness/cli.py b/evalharness/cli.py
index 1064197..7084092 100644
--- a/evalharness/cli.py
+++ b/evalharness/cli.py
@@ -720,9 +720,13 @@ def _cmd_eval_run(args) -> int:
                     for minutes with zero feedback otherwise (bar sits at
                     'generating 100%' and looks hung)."""
                     if _reporter is not None:
-                        # live terminal: the bar carries the counter -- log
-                        # lines here are pure noise (21 lines per bench)
-                        _reporter.set_phase(f'scoring {done}/{total_s}')
+                        # live terminal: RETARGET the bar to scoring -- it
+                        # refills 0->100% with judged samples
+                        _ss = getattr(_reporter, 'set_scoring', None)
+                        if _ss is not None:
+                            _ss(done, total_s)
+                        else:
+                            _reporter.set_phase(f'scoring {done}/{total_s}')
                         return
                     # pipes/redirects (no live bar): milestone lines instead,
                     # first completion immediately then every ~10%
diff --git a/evalharness/progress/rich_terminal.py b/evalharness/progress/rich_terminal.py
index c78d256..ee06ead 100644
--- a/evalharness/progress/rich_terminal.py
+++ b/evalharness/progress/rich_terminal.py
@@ -123,6 +123,7 @@ class RichTerminalProgress:
         self.inflight = 0
         self.bench_name = description
         self.restored = max(completed, 0)  # checkpoint head start this bench
+        self._scoring_for = None           # next scoring phase retargets anew
         desc = f'[green]{self.bench_tag}{description} · generating[/green]'
         if self.task_id is None:
             self.progress.start()
@@ -143,6 +144,29 @@ class RichTerminalProgress:
             self.heartbeat_task.cancel()
         self.heartbeat_task = asyncio.create_task(self._heartbeat())
 
+    def set_scoring(self, done: int, total: int):
+        """Retarget the SAME bar to the scoring phase: generation is finished
+        and its filled 100% state is stale -- now the bar refills with judged
+        samples (0% -> 100%), with a fresh clock/eta for this phase."""
+        if self.disabled or self.task_id is None:
+            return
+        if getattr(self, '_scoring_for', None) != total:
+            # first callback of this scoring phase: restart the clock, clear
+            # generation markers (the '+N new' tag is about generating)
+            self._scoring_for = total
+            self._last_phase = 'scoring'
+            self.started = time.monotonic()
+            self.inflight = 0
+            self.restored = 0
+        elapsed = max(time.monotonic() - self.started, 1e-6)
+        self.progress.update(
+            self.task_id,
+            description=f'[green]{self.bench_tag}{self.bench_name} · scoring[/green]',
+            total=total, completed=min(done, total), new='',
+            rate=f'{done / elapsed:.2f}', inflight=0, cur='0s',
+            elapsed=_fmt(elapsed),
+            eta=_fmt((total - done) * elapsed / done) if done and total > done else '-')
+
     def set_bench_tag(self, tag: str):
         if self.disabled:
             return