Skip to content

Speakr is a personal, self-hosted web application designed for transcribing audio recordings

Python

Fix JSON parser greedy extraction and newline handling, update video retention test patterns

Murtaza Nasir30f147823c59
diff --git a/src/utils/json_parser.py b/src/utils/json_parser.py
index f98533a6..116ac25d 100644
--- a/src/utils/json_parser.py
+++ b/src/utils/json_parser.py
@@ -131,6 +131,15 @@ def preprocess_json_escapes(json_string):
else:
# This is an inner quote that should be escaped
result.append('\\"')
+ elif char == '\n' and in_string:
+ # Literal newlines inside JSON strings must be escaped
+ result.append('\\n')
+ elif char == '\r' and in_string:
+ # Literal carriage returns inside JSON strings must be escaped
+ result.append('\\r')
+ elif char == '\t' and in_string:
+ # Literal tabs inside JSON strings must be escaped
+ result.append('\\t')
else:
result.append(char)
@@ -141,20 +150,56 @@ def preprocess_json_escapes(json_string):
def extract_json_object(text):
"""
- Extract the first complete JSON object or array from text using regex.
+ Extract the first complete JSON object or array from text using brace counting.
+ This is more robust than regex as it properly handles nested structures and
+ doesn't over-match when there's text after the JSON.
"""
- # Look for JSON object
- obj_match = re.search(r'\{.*\}', text, re.DOTALL)
- if obj_match:
- return obj_match.group(0)
-
- # Look for JSON array
- arr_match = re.search(r'\[.*\]', text, re.DOTALL)
- if arr_match:
- return arr_match.group(0)
-
- # Return original if no JSON structure found
- return text
+ # Find the first { or [ that starts a JSON structure
+ start = None
+ close_char = None
+ for i, ch in enumerate(text):
+ if ch == '{':
+ start = i
+ close_char = '}'
+ break
+ elif ch == '[':
+ start = i
+ close_char = ']'
+ break
+
+ if start is None:
+ return text
+
+ # Walk through counting braces/brackets, respecting strings
+ depth = 0
+ in_string = False
+ escape_next = False
+
+ for i in range(start, len(text)):
+ ch = text[i]
+
+ if escape_next:
+ escape_next = False
+ continue
+
+ if ch == '\\' and in_string:
+ escape_next = True
+ continue
+
+ if ch == '"':
+ in_string = not in_string
+ continue
+
+ if not in_string:
+ if ch in ('{', '['):
+ depth += 1
+ elif ch in ('}', ']'):
+ depth -= 1
+ if depth == 0:
+ return text[start:i + 1]
+
+ # If we never balanced, return from start to end (best effort)
+ return text[start:]
def safe_json_loads(json_string, fallback_value=None):
diff --git a/tests/test_video_retention.py b/tests/test_video_retention.py
index 5428d57b..c4a25e70 100644
--- a/tests/test_video_retention.py
+++ b/tests/test_video_retention.py
@@ -109,7 +109,7 @@ class TestUploadHandlerVideoRetention(unittest.TestCase):
def test_upload_handler_skips_conversion_for_video_retention(self):
"""Upload handler skips convert_if_needed for videos when retention is on."""
- self.assertIn('VIDEO_RETENTION and has_video', self.content)
+ self.assertIn('(VIDEO_RETENTION or VIDEO_PASSTHROUGH_ASR) and has_video', self.content)
self.assertIn('skipping conversion', self.content)
def test_upload_handler_has_video_from_codec_info(self):
@@ -138,8 +138,8 @@ class TestFileMonitorVideoRetention(unittest.TestCase):
def test_video_retention_skips_conversion(self):
"""When VIDEO_RETENTION=True and has_video=True, convert_if_needed is skipped."""
- # Should have the guard: if VIDEO_RETENTION and has_video: ... skip conversion
- self.assertIn('VIDEO_RETENTION and has_video', self.content)
+ # Should have the guard: if (VIDEO_PASSTHROUGH_ASR or VIDEO_RETENTION) and has_video: ... skip conversion
+ self.assertIn('VIDEO_PASSTHROUGH_ASR or VIDEO_RETENTION) and has_video', self.content)
self.assertIn('skipping conversion', self.content)
def test_no_double_extraction(self):
@@ -150,7 +150,7 @@ class TestFileMonitorVideoRetention(unittest.TestCase):
found_convert_in_else = False
for i, line in enumerate(lines):
- if 'VIDEO_RETENTION and has_video' in line and 'if' in line:
+ if 'VIDEO_PASSTHROUGH_ASR or VIDEO_RETENTION) and has_video' in line and 'if' in line:
in_retention_skip_block = True
elif in_retention_skip_block and 'else:' in line:
in_retention_skip_block = False
@@ -194,12 +194,12 @@ class TestSendFileConditional(unittest.TestCase):
"""Streaming send_file in recordings.py has conditional=True."""
content = self._read_file('src/api/recordings.py')
# Find the non-download send_file call
- self.assertIn('send_file(recording.audio_path, conditional=True)', content)
+ self.assertIn('send_file(recording.audio_path, mimetype=recording.mime_type, conditional=True)', content)
def test_recordings_download_has_conditional(self):
"""Download send_file in recordings.py has conditional=True."""
content = self._read_file('src/api/recordings.py')
- self.assertIn('as_attachment=True, download_name=filename, conditional=True', content)
+ self.assertIn('as_attachment=True, download_name=filename, mimetype=recording.mime_type, conditional=True', content)
def test_shares_has_conditional(self):
"""send_file in shares.py has conditional=True."""
@@ -331,7 +331,7 @@ class TestVideoRetentionMatrix(unittest.TestCase):
def test_file_monitor_has_both_branches(self):
"""file_monitor.py has both video retention skip and normal conversion paths."""
content = self._read_file('src/file_monitor.py')
- self.assertIn('VIDEO_RETENTION and has_video', content)
+ self.assertIn('VIDEO_PASSTHROUGH_ASR or VIDEO_RETENTION) and has_video', content)
# convert_if_needed should still exist in the else path
self.assertIn('convert_if_needed(', content)
@@ -348,8 +348,8 @@ class TestVideoRetentionMatrix(unittest.TestCase):
def test_all_three_entry_points_skip_for_video_retention(self):
"""All entry points (upload, file monitor, processing) handle VIDEO_RETENTION."""
for rel_path, marker in [
- ('src/api/recordings.py', 'VIDEO_RETENTION and has_video'),
- ('src/file_monitor.py', 'VIDEO_RETENTION and has_video'),
+ ('src/api/recordings.py', '(VIDEO_RETENTION or VIDEO_PASSTHROUGH_ASR) and has_video'),
+ ('src/file_monitor.py', '(VIDEO_PASSTHROUGH_ASR or VIDEO_RETENTION) and has_video'),
('src/tasks/processing.py', 'if VIDEO_RETENTION:'),
]:
content = self._read_file(rel_path)