File size: 12,000 Bytes
9deebf2 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 | """
Encoding detection and normalization for text files.
Handles the messy reality of 20 years of web-generated data with
mixed encodings, mojibake, and partial writes containing null bytes.
"""
from pathlib import Path
from typing import Optional, Tuple, List
import chardet
import ftfy
from ..core.base_classes import BaseCleaner
from ..core.config import Config
from ..core.errata_logger import ErrataLogger
from ..core.exceptions import EncodingError
class EncodingCleaner(BaseCleaner):
"""
Detects and normalizes file encodings.
Handles:
- Automatic encoding detection using chardet
- Fallback to candidate encodings
- Mojibake correction using ftfy
- Null byte removal from partial writes
- Normalization to UTF-8
"""
def __init__(self, config: Config, errata_logger: Optional[ErrataLogger] = None):
"""
Initialize encoding cleaner.
Args:
config: Configuration object
errata_logger: Optional errata logger for error tracking
"""
super().__init__(config, errata_logger)
# Get configuration
self.candidate_encodings: List[str] = config.get(
'encoding.candidate_encodings',
['utf-8', 'latin-1', 'windows-1252', 'iso-8859-1']
)
self.confidence_threshold: float = config.get(
'encoding.confidence_threshold',
0.7
)
self.fix_mojibake: bool = config.get('encoding.fix_mojibake', True)
def detect_encoding(
self,
file_path: str | Path,
sample_size: int = 100000
) -> Tuple[str, float]:
"""
Detect file encoding using chardet.
Args:
file_path: Path to file
sample_size: Number of bytes to sample for detection
Returns:
Tuple of (encoding, confidence)
Raises:
EncodingError: If encoding cannot be detected
"""
file_path = Path(file_path)
try:
with open(file_path, 'rb') as f:
raw_data = f.read(sample_size)
# Remove null bytes before detection
raw_data = raw_data.replace(b'\x00', b'')
if not raw_data:
raise EncodingError(f"File is empty or contains only null bytes: {file_path}")
result = chardet.detect(raw_data)
encoding = result['encoding']
confidence = result['confidence']
if encoding is None:
raise EncodingError(f"Could not detect encoding for {file_path}")
return encoding, confidence
except IOError as e:
raise EncodingError(f"Failed to read file {file_path}: {e}")
def read_with_encoding(
self,
file_path: str | Path,
encoding: str,
remove_null_bytes: bool = True
) -> str:
"""
Read file with specified encoding.
Args:
file_path: Path to file
encoding: Encoding to use
remove_null_bytes: Whether to remove null bytes
Returns:
Decoded text content
Raises:
EncodingError: If reading with encoding fails
"""
file_path = Path(file_path)
try:
with open(file_path, 'rb') as f:
raw_data = f.read()
# Remove null bytes from partial writes
if remove_null_bytes:
raw_data = raw_data.replace(b'\x00', b'')
# Decode with specified encoding
text = raw_data.decode(encoding, errors='strict')
return text
except UnicodeDecodeError as e:
raise EncodingError(
f"Failed to decode {file_path} with {encoding}: {e}"
)
except IOError as e:
raise EncodingError(f"Failed to read file {file_path}: {e}")
def clean(
self,
data: str | Path,
file_path: Optional[str] = None,
**kwargs
) -> str:
"""
Clean encoding of text data or file.
Args:
data: Either file path or text string to clean
file_path: Optional file path for error logging
**kwargs: Additional parameters
Returns:
Cleaned UTF-8 text
Raises:
EncodingError: If encoding cannot be resolved
"""
# If data is a path, read the file
if isinstance(data, (str, Path)) and Path(data).is_file():
file_path = str(data)
data, was_utf8 = self._read_file_smart_tagged(Path(data))
elif not isinstance(data, str):
raise EncodingError(f"Invalid data type: {type(data)}")
else:
was_utf8 = False
# Fast path: if file decoded as UTF-8, skip ftfy and round-trip
# Normalize all line endings: CRLF -> LF, then bare CR -> LF
# (some files have mixed line endings, e.g. card190309.dat, loc020520.dat)
if was_utf8:
return data.replace('\r\n', '\n').replace('\r', '\n')
# Slow path: fix mojibake and normalize
if self.fix_mojibake:
try:
data = ftfy.fix_text(data)
except Exception as e:
self.log_error(
'mojibake_fix_failed',
f"Failed to fix mojibake: {e}",
file_path=file_path
)
# Normalize to UTF-8, logging any characters that get replaced
encoded = data.encode('utf-8', errors='replace')
replacement_count = encoded.count(b'\xef\xbf\xbd') # UTF-8 bytes of U+FFFD
if replacement_count > 0:
self.log_error(
'utf8_replacement_chars',
f"{replacement_count} character(s) replaced with U+FFFD during UTF-8 normalization",
file_path=file_path
)
data = encoded.decode('utf-8')
# Normalize line endings (same as fast path)
data = data.replace('\r\n', '\n').replace('\r', '\n')
return data
def _read_file_smart_tagged(self, file_path: Path) -> Tuple[str, bool]:
"""
Read file and return (text, was_utf8).
When was_utf8=True, the file decoded cleanly as UTF-8 and the caller
can skip expensive ftfy/mojibake processing (just strip CR).
"""
try:
with open(file_path, 'rb') as f:
raw_data = f.read()
raw_data = raw_data.replace(b'\x00', b'')
if not raw_data:
raise EncodingError(f"File is empty or contains only null bytes: {file_path}")
try:
return raw_data.decode('utf-8'), True
except UnicodeDecodeError:
pass # Fall through — need slow path for non-UTF-8
except IOError:
pass
# Non-UTF-8 or IO error: use full smart reader
return self._read_file_smart(file_path), False
def _read_file_smart(self, file_path: Path) -> str:
"""
Read file with intelligent encoding detection and fallback.
Tries:
1. Fast path: UTF-8 decode (covers ~99% of files)
2. Fast fallback: latin-1 decode (covers remaining)
3. Slow path: chardet detection (only if fast paths produce issues)
Args:
file_path: Path to file
Returns:
Decoded text
Raises:
EncodingError: If all attempts fail
"""
last_error = None
# Fast path: try UTF-8 directly, fallback to latin-1
# Profiling shows ~99% of files are ASCII/UTF-8 compatible.
# This avoids expensive chardet detection for the common case.
try:
with open(file_path, 'rb') as f:
raw_data = f.read()
raw_data = raw_data.replace(b'\x00', b'')
if not raw_data:
raise EncodingError(f"File is empty or contains only null bytes: {file_path}")
try:
return raw_data.decode('utf-8')
except UnicodeDecodeError:
return raw_data.decode('latin-1')
except (IOError, EncodingError):
pass # Fall through to slow path
# Slow path: Try automatic detection with chardet
try:
encoding, confidence = self.detect_encoding(file_path)
# Skip ASCII if detected - it's too restrictive, try UTF-8 or latin-1 instead
if encoding and encoding.lower() != 'ascii':
if confidence >= self.confidence_threshold:
text = self.read_with_encoding(file_path, encoding)
return text
else:
self.log_error(
'low_confidence_encoding',
f"Low confidence ({confidence:.2f}) for detected encoding {encoding}",
file_path=str(file_path)
)
except EncodingError as e:
last_error = str(e)
self.log_error(
'encoding_detection_failed',
str(e),
file_path=str(file_path)
)
# Try candidate encodings in order
# Prioritize more permissive encodings for problematic files
for encoding in self.candidate_encodings:
try:
text = self.read_with_encoding(file_path, encoding)
# Log which fallback encoding worked
if last_error:
self.log_error(
'encoding_fallback_success',
f"Successfully read with {encoding} after detection failed",
file_path=str(file_path)
)
return text
except EncodingError as e:
last_error = str(e)
continue
# Last resort: latin-1 (never fails, so validate the result)
self.log_error(
'encoding_fallback',
f"All encodings failed, falling back to latin-1",
file_path=str(file_path)
)
try:
text = self.read_with_encoding(file_path, 'latin-1')
except EncodingError as e:
raise EncodingError(
f"Fatal: Could not read {file_path} even with latin-1: {e}"
)
# Sanity check: latin-1 accepts any byte sequence, so a binary/corrupt
# file will decode without error into garbage. Reject if the result
# has a high ratio of control characters (excluding \t, \n, \r).
if text:
control_chars = sum(
1 for ch in text
if ch < ' ' and ch not in ('\t', '\n', '\r')
)
ratio = control_chars / len(text)
if ratio > 0.05:
self.log_error(
'encoding_garbage_detected',
f"latin-1 fallback produced {ratio:.1%} control characters — file appears corrupt",
file_path=str(file_path)
)
raise EncodingError(
f"File appears corrupt (latin-1 decoded to {ratio:.1%} control chars): {file_path}"
)
return text
def clean_batch(
self,
file_paths: List[str | Path]
) -> List[Tuple[str, str]]:
"""
Clean encoding for a batch of files.
Args:
file_paths: List of file paths to process
Returns:
List of tuples (file_path, cleaned_text)
"""
results = []
for file_path in file_paths:
try:
cleaned = self.clean(file_path)
results.append((str(file_path), cleaned))
except EncodingError as e:
self.log_error(
'batch_encoding_failed',
str(e),
file_path=str(file_path)
)
# Continue processing other files
return results |