Refuse an unreadable OCR frame rate instead of assuming 30 fps

src/docgen/ocr.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import math
 import re
 from dataclasses import dataclass, field
 from pathlib import Path
@@ -26,6 +27,19 @@ class OCRReport:
     passed: bool = True
 
 
+def _require_positive_fps(cap: Any, raw: float) -> float:
+    """A zero or missing frame rate must not be treated as 30 fps."""
+    try:
+        fps = float(raw)
+    except (TypeError, ValueError) as exc:
+        cap.release()
+        raise ValueError("video frame rate is not a number") from exc
+    if not math.isfinite(fps) or fps <= 0:
+        cap.release()
+        raise ValueError("video frame rate must be a positive finite number")
+    return fps
+
+
 class OCRScanner:
     def __init__(self, config: Config) -> None:
         self.config = config
@@ -44,7 +58,7 @@ def scan_video(self, video_path: str | Path) -> OCRReport:
             report.frames.append(OCRFrame(0, "", issues=["Could not open video"]))
             return report
 
-        fps = cap.get(cv2.CAP_PROP_FPS) or 30
+        fps = _require_positive_fps(cap, cap.get(cv2.CAP_PROP_FPS))
         total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
         duration = total_frames / fps if fps > 0 else 0
         interval = self.ocr_cfg.get("sample_interval_sec", 2)

tests/test_ocr.py

@@ -0,0 +1,62 @@
+"""OCR scanning must not invent a frame rate when the video does not report one."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import cv2
+import pytest
+import yaml
+
+from docgen.config import Config
+from docgen.ocr import OCRScanner
+
+
+def _cfg(tmp_path: Path) -> Config:
+    raw = {"segments": {"all": ["01"], "default": ["01"]}}
+    path = tmp_path / "docgen.yaml"
+    path.write_text(yaml.dump(raw), encoding="utf-8")
+    return Config.from_yaml(path)
+
+
+class _Cap:
+    def __init__(self, *_args: object, **_kwargs: object) -> None:
+        self.fps = 0.0
+        self.frames = 0.0
+        self.sampled = False
+
+    def isOpened(self) -> bool:
+        return True
+
+    def get(self, prop: int) -> float:
+        if prop == cv2.CAP_PROP_FRAME_COUNT:
+            return self.frames
+        return self.fps
+
+    def set(self, *_args: object) -> None:
+        self.sampled = True
+
+    def read(self) -> tuple[bool, None]:
+        self.sampled = True
+        return False, None
+
+    def release(self) -> None:
+        return None
+
+
+def test_zero_fps_is_refused(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
+    cap = _Cap()
+    monkeypatch.setattr(cv2, "VideoCapture", lambda *_a, **_k: cap)
+    with pytest.raises(ValueError, match="positive finite"):
+        OCRScanner(_cfg(tmp_path)).scan_video(tmp_path / "clip.mp4")
+    assert cap.sampled is False
+
+
+def test_positive_fps_still_samples(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
+    cap = _Cap()
+    cap.fps = 30.0
+    cap.frames = 30.0
+    monkeypatch.setattr(cv2, "VideoCapture", lambda *_a, **_k: cap)
+    report = OCRScanner(_cfg(tmp_path)).scan_video(tmp_path / "clip.mp4")
+    assert cap.sampled is True
+    assert report.passed is True