cherry-pick d1f4b6e5
This commit is contained in:
parent
1a167fb7f7
commit
e73248eb24
1 changed files with 5 additions and 51 deletions
|
|
@ -927,68 +927,22 @@ def load_json(file_name):
|
||||||
return json.load(f)
|
return json.load(f)
|
||||||
|
|
||||||
|
|
||||||
def _sanitize_string_for_json(text: str) -> str:
|
|
||||||
"""Remove characters that cannot be encoded in UTF-8 for JSON serialization.
|
|
||||||
|
|
||||||
This is a simpler sanitizer specifically for JSON that directly removes
|
|
||||||
problematic characters without attempting to encode first.
|
|
||||||
|
|
||||||
Args:
|
|
||||||
text: String to sanitize
|
|
||||||
|
|
||||||
Returns:
|
|
||||||
Sanitized string safe for UTF-8 encoding in JSON
|
|
||||||
"""
|
|
||||||
if not text:
|
|
||||||
return text
|
|
||||||
|
|
||||||
# Directly filter out problematic characters without pre-validation
|
|
||||||
sanitized = ""
|
|
||||||
for char in text:
|
|
||||||
code_point = ord(char)
|
|
||||||
# Skip surrogate characters (U+D800 to U+DFFF) - main cause of encoding errors
|
|
||||||
if 0xD800 <= code_point <= 0xDFFF:
|
|
||||||
continue
|
|
||||||
# Skip other non-characters in Unicode
|
|
||||||
elif code_point == 0xFFFE or code_point == 0xFFFF:
|
|
||||||
continue
|
|
||||||
else:
|
|
||||||
sanitized += char
|
|
||||||
|
|
||||||
return sanitized
|
|
||||||
|
|
||||||
|
|
||||||
def _sanitize_json_data(data: Any) -> Any:
|
def _sanitize_json_data(data: Any) -> Any:
|
||||||
"""Recursively sanitize all string values in data structure for safe UTF-8 encoding
|
"""Recursively sanitize all string values in data structure for safe UTF-8 encoding
|
||||||
|
|
||||||
Handles all JSON-serializable types including:
|
|
||||||
- Dictionary keys and values
|
|
||||||
- Lists and tuples (preserves type)
|
|
||||||
- Nested structures
|
|
||||||
- Strings at any level
|
|
||||||
|
|
||||||
Args:
|
Args:
|
||||||
data: Data to sanitize (dict, list, tuple, str, or other types)
|
data: Data to sanitize (dict, list, str, or other types)
|
||||||
|
|
||||||
Returns:
|
Returns:
|
||||||
Sanitized data with all strings cleaned of problematic characters
|
Sanitized data with all strings cleaned of problematic characters
|
||||||
"""
|
"""
|
||||||
if isinstance(data, dict):
|
if isinstance(data, dict):
|
||||||
# Sanitize both keys and values
|
return {k: _sanitize_json_data(v) for k, v in data.items()}
|
||||||
return {
|
elif isinstance(data, list):
|
||||||
_sanitize_string_for_json(k)
|
return [_sanitize_json_data(item) for item in data]
|
||||||
if isinstance(k, str)
|
|
||||||
else k: _sanitize_json_data(v)
|
|
||||||
for k, v in data.items()
|
|
||||||
}
|
|
||||||
elif isinstance(data, (list, tuple)):
|
|
||||||
# Handle both lists and tuples, preserve original type
|
|
||||||
sanitized = [_sanitize_json_data(item) for item in data]
|
|
||||||
return type(data)(sanitized)
|
|
||||||
elif isinstance(data, str):
|
elif isinstance(data, str):
|
||||||
return _sanitize_string_for_json(data)
|
return sanitize_text_for_encoding(data, replacement_char="")
|
||||||
else:
|
else:
|
||||||
# Numbers, booleans, None, etc. - return as-is
|
|
||||||
return data
|
return data
|
||||||
|
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Add table
Reference in a new issue