From dc80133ba0898a57680f1c68fe18bac28a93d44c Mon Sep 17 00:00:00 2001
From: leoca <leo.camus23@gmail.com>
Date: Wed, 14 Oct 2026 13:03:04 +0000
Subject: [PATCH 1/2] Escape the file name in the skip already transcribed glob

A media file whose name contains glob characters, for example
"Interview [Part 1].mp3", never matched its own result file because
Path.glob read "[Part 1]" as a character class, so the file was
transcribed again.
---
 buzz/plugins/skip_already_transcribed/plugin.py |  3 ++-
 tests/plugins/plugin_system_test.py             | 16 ++++++++++++++++
 2 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/buzz/plugins/skip_already_transcribed/plugin.py b/buzz/plugins/skip_already_transcribed/plugin.py
index ff9387553a..329ab5a269 100644
--- a/buzz/plugins/skip_already_transcribed/plugin.py
+++ b/buzz/plugins/skip_already_transcribed/plugin.py
@@ -1,3 +1,4 @@
+import glob
 import logging
 import os
 import re
@@ -89,7 +90,7 @@ def _find_result_file_segments(self, file_path: str, task, context: PluginContex
             if not search_dir.is_dir():
                 continue
             for fmt in OutputFormat:
-                for candidate in search_dir.glob(f"{stem}*.{fmt.value}"):
+                for candidate in search_dir.glob(f"{glob.escape(stem)}*.{fmt.value}"):
                     segments = self._parse_result_file(candidate, fmt, context)
                     if segments is not None:
                         return segments
diff --git a/tests/plugins/plugin_system_test.py b/tests/plugins/plugin_system_test.py
index 87db313060..20251304a3 100644
--- a/tests/plugins/plugin_system_test.py
+++ b/tests/plugins/plugin_system_test.py
@@ -788,6 +788,22 @@ def test_sat_skips_on_vtt_file(tmp_path):
     assert result[1].start == 3000
 
 
+def test_sat_skips_on_srt_file_with_brackets_in_name(tmp_path):
+    plugin = _sat_plugin()
+    audio = tmp_path / "Interview [Part 1].mp3"
+    audio.write_bytes(b"")
+    srt = tmp_path / "Interview [Part 1] (transcribed on 01-Jan-2024 12-00-00).srt"
+    srt.write_text(
+        "1\n00:00:01,000 --> 00:00:03,500\nHello there.\n\n",
+        encoding="utf-8",
+    )
+
+    ctx = _sat_context(config={"check_result_files": True, "check_database": False})
+    result = plugin.check_skip(_SatTask(file_path=str(audio)), ctx)
+    assert result is not None
+    assert result[0].text == "Hello there."
+
+
 def test_sat_returns_none_when_result_files_disabled(tmp_path):
     plugin = _sat_plugin()
     audio = tmp_path / "audio.mp3"

From 6e9a91973b12506906e7aa88c0b9ea94154718cf Mon Sep 17 00:00:00 2001
From: Raivis Dejus <raivisd@scandiweb.com>
Date: Sat, 17 Oct 2026 10:30:45 +0300
Subject: [PATCH 2/2] Fix for regression

---
 .../skip_already_transcribed/plugin.py        |  3 ++
 tests/plugins/plugin_system_test.py           | 46 +++++++++++++++++++
 2 files changed, 49 insertions(+)

diff --git a/buzz/plugins/skip_already_transcribed/plugin.py b/buzz/plugins/skip_already_transcribed/plugin.py
index 329ab5a269..9b331770fb 100644
--- a/buzz/plugins/skip_already_transcribed/plugin.py
+++ b/buzz/plugins/skip_already_transcribed/plugin.py
@@ -76,11 +76,14 @@ def check_skip(
         return None
 
     def _find_result_file_segments(self, file_path: str, task, context: PluginContext):
+        from buzz.paths import safe_filename_component
         from buzz.transcriber.transcriber import OutputFormat
 
         stem = Path(file_path).stem
         if stem.endswith("_speech"):
             stem = stem[:-7]
+        # Match the name get_output_file_path writes, e.g. "what?" -> "what_"
+        stem = safe_filename_component(stem)
 
         search_dirs = [Path(file_path).parent]
         if task.output_directory:
diff --git a/tests/plugins/plugin_system_test.py b/tests/plugins/plugin_system_test.py
index 20251304a3..f4dc30a8cb 100644
--- a/tests/plugins/plugin_system_test.py
+++ b/tests/plugins/plugin_system_test.py
@@ -804,6 +804,52 @@ def test_sat_skips_on_srt_file_with_brackets_in_name(tmp_path):
     assert result[0].text == "Hello there."
 
 
+@pytest.mark.parametrize(
+    "stem",
+    [
+        "plain",
+        "Interview [Part 1]",
+        "[2024] Talk",
+        "Interview (Part 1)",
+        "(2024) Talk",
+        "what?",
+        "a*b",
+        "Episode 3: Pilot",
+        'say "hi"',
+        "ends.",
+        "ends ",
+        "con",
+        "talk_speech",
+    ],
+)
+def test_sat_finds_result_file_written_by_get_output_file_path(tmp_path, stem):
+    from buzz.model_loader import TranscriptionModel
+    from buzz.transcriber.transcriber import (
+        OutputFormat,
+        Task,
+        get_output_file_path,
+    )
+
+    # The media file itself is not needed, so names invalid on Windows still work
+    audio = tmp_path / f"{stem}.mp3"
+    srt = get_output_file_path(
+        str(audio),
+        Task.TRANSCRIBE,
+        None,
+        TranscriptionModel(),
+        OutputFormat.SRT,
+        export_file_name_template="{{ input_file_name }} ({{ task }}d on {{ date_time }})",
+    )
+    with open(srt, "w", encoding="utf-8") as f:
+        f.write("1\n00:00:01,000 --> 00:00:03,500\nHello there.\n\n")
+
+    plugin = _sat_plugin()
+    ctx = _sat_context(config={"check_result_files": True, "check_database": False})
+    result = plugin.check_skip(_SatTask(file_path=str(audio)), ctx)
+    assert result is not None
+    assert result[0].text == "Hello there."
+
+
 def test_sat_returns_none_when_result_files_disabled(tmp_path):
     plugin = _sat_plugin()
     audio = tmp_path / "audio.mp3"