File size: 19,446 Bytes
463fc7e | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 | """
Repository File Type Chunker - Universal chunker for all file types.
This module provides file-type-aware chunking for repositories, handling
everything from Python code to configuration files, documentation, and
special files. It's the universal interface that delegates to specialized
chunkers based on file type.
ARCHITECTURE POSITION:
- File Type Dispatcher: Routes files to appropriate chunkers
- Universal Interface: Single entry point for all file types
- Metadata Enricher: Adds repository context to all chunks
KEY FEATURES:
1. File type detection and intelligent routing
2. Hierarchical chunking for Python files
3. Documentation chunking for markdown/RST
4. Configuration file handling (JSON/YAML/TOML)
5. Special file handling (README, requirements.txt, Dockerfile)
6. Binary file detection and skipping
FILE TYPE SUPPORT:
- .py: HierarchicalChunker (AST + Tree-sitter)
- .md/.mdx/.rst: Documentation chunker
- .json/.yaml/.toml: Configuration chunker
- requirements.txt/Dockerfile: Special chunker
- .txt/.cfg/.ini: Text chunker
- README/LICENSE: Documentation chunker
- Others: Text chunker with binary detection
DATA FLOW:
File → Type detection → Route to specialized chunker →
Add repo metadata → Return chunks
USAGE:
chunker = RepoChunker()
chunks = chunker.chunk_file(Path("file.py"), repo_metadata)
"""
from pathlib import Path
from typing import List, Dict, Optional, cast
import json
import yaml
import re
import hashlib
from .hierarchical_chunker import HierarchicalChunker
from .chunk_schema import CodeChunk, ChunkAST, ChunkSpan, ChunkHierarchy, ChunkType, ASTSymbolType
from .doc_chunker import chunk_document as chunk_markdown_file
class RepoChunker:
"""
Repository chunker that handles ALL file types with proper structure
"""
def __init__(self, use_hierarchical: bool = True):
if use_hierarchical:
self.hierarchical_chunker = HierarchicalChunker()
self.use_hierarchical = use_hierarchical
def _generate_stable_id(self, content: str, prefix: str = "stable") -> str:
"""
Generate deterministic chunk ID using SHA256.
IMPORTANT: This ensures IDs are stable across runs, processes,
and Python versions - crucial for RAG reproducibility.
Args:
content: The text content to hash
prefix: ID prefix (config, doc, text, etc.)
Returns:
Deterministic ID like "config_8a3b2c1d"
"""
# Use SHA256 for consistency with id_utils.py
hash_digest = hashlib.sha256(content.encode("utf-8")).hexdigest()[:8]
return f"{prefix}_{hash_digest}"
def chunk_file(self, file_path: Path, repo_metadata: Optional[Dict] = None) -> List[CodeChunk]:
"""
Chunk ANY file type with repository context
Args:
file_path: Path to the file
repo_metadata: Optional dict with repo metadata
"""
suffix = file_path.suffix.lower()
# Python files - use your advanced hierarchical chunker
if suffix == '.py':
return self._chunk_python_file(file_path, repo_metadata)
# Markdown/RST documentation
elif suffix in ['.md', '.mdx', '.rst']:
return self._chunk_markdown_file_wrapper(file_path, repo_metadata)
# JSON config files
elif suffix == '.json':
return self._chunk_json_file(file_path, repo_metadata)
# YAML/TOML config files
elif suffix in ['.yaml', '.yml', '.toml']:
return self._chunk_config_file(file_path, repo_metadata)
# Requirements/Docker files
elif file_path.name.lower() in ['requirements.txt', 'dockerfile', 'docker-compose.yml']:
return self._chunk_special_file(file_path, repo_metadata)
# Text files
elif suffix in ['.txt', '.cfg', '.ini', '.conf']:
return self._chunk_text_file(file_path, repo_metadata)
# README/LICENSE files
elif file_path.name.lower() in ['readme', 'readme.md', 'license', 'license.txt', 'license.md']:
return self._chunk_readme_file(file_path, repo_metadata)
# All other files
else:
return self._chunk_other_file(file_path, repo_metadata)
def _chunk_python_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Use our hierarchical chunker for Python files"""
try:
if self.use_hierarchical:
chunks = self.hierarchical_chunker.chunk_file(file_path)
else:
# Fallback to basic text chunking instead of hybrid
return self._chunk_text_file(file_path, repo_metadata)
# Add repository metadata
if repo_metadata:
for chunk in chunks:
if "repo_info" not in chunk.metadata:
chunk.metadata["repo_info"] = {}
chunk.metadata["repo_info"].update(repo_metadata)
return chunks
except Exception as e:
print(f"[ERROR] Error chunking Python file {file_path}: {e}")
return self._chunk_text_file(file_path, repo_metadata)
def _chunk_markdown_file_wrapper(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Chunk markdown files using our doc_chunker"""
try:
content = file_path.read_text(encoding='utf-8', errors='ignore')
# Use your existing doc_chunker
doc_chunks = chunk_markdown_file(
content,
source_name=file_path.name,
source_url=f"file://{file_path}"
)
# Convert to CodeChunk schema
code_chunks = []
for doc_chunk in doc_chunks:
code_chunk = CodeChunk(
chunk_id=doc_chunk["chunk_id"], # Already uses SHA1 from doc_chunker.py
file_path=str(file_path),
language=doc_chunk.get("language", "markdown"),
chunk_type="documentation",
code=doc_chunk["content"],
ast=ChunkAST(
symbol_type="documentation",
name=file_path.name,
parent=None,
docstring=None
),
span=ChunkSpan(
start_line=doc_chunk.get("metadata", {}).get("line_start", 1),
end_line=doc_chunk.get("metadata", {}).get("line_end", 1)
),
metadata={
"doc_chunk_type": doc_chunk.get("chunk_type", "text"),
"repo_info": repo_metadata or {},
**doc_chunk.get("metadata", {})
},
hierarchy=ChunkHierarchy(
is_primary=True,
is_extracted=False,
depth=0
)
)
code_chunks.append(code_chunk)
return code_chunks
except Exception as e:
print(f"[ERROR] Error chunking markdown file {file_path}: {e}")
return self._chunk_text_file(file_path, repo_metadata)
def _chunk_json_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Chunk JSON config files"""
try:
content = file_path.read_text(encoding='utf-8', errors='ignore')
data = json.loads(content)
pretty_content = json.dumps(data, indent=2)
# FIXED: Use deterministic SHA256 instead of hash()
chunk = CodeChunk(
chunk_id=self._generate_stable_id(pretty_content, "config"),
file_path=str(file_path),
language="json",
chunk_type="configuration",
code=pretty_content,
ast=ChunkAST(
symbol_type="configuration",
name=file_path.name,
parent=None,
docstring=None
),
span=ChunkSpan(
start_line=1,
end_line=len(pretty_content.split('\n'))
),
metadata={
"file_type": "json_config",
"config_keys": list(data.keys()) if isinstance(data, dict) else [],
"repo_info": repo_metadata or {}
},
hierarchy=ChunkHierarchy(
is_primary=True,
is_extracted=False,
depth=0
)
)
return [chunk]
except Exception as e:
print(f"[ERROR] Error chunking JSON file {file_path}: {e}")
return self._chunk_text_file(file_path, repo_metadata)
def _chunk_config_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Chunk YAML/TOML config files"""
try:
content = file_path.read_text(encoding='utf-8', errors='ignore')
suffix = file_path.suffix.lower()
language = "yaml" if suffix in ['.yaml', '.yml'] else "toml"
# FIXED: Use deterministic SHA256 instead of hash()
chunk = CodeChunk(
chunk_id=self._generate_stable_id(content, "config"),
file_path=str(file_path),
language=language,
chunk_type="configuration",
code=content,
ast=ChunkAST(
symbol_type="configuration",
name=file_path.name,
parent=None,
docstring=None
),
span=ChunkSpan(
start_line=1,
end_line=len(content.split('\n'))
),
metadata={
"file_type": f"{language}_config",
"repo_info": repo_metadata or {}
},
hierarchy=ChunkHierarchy(
is_primary=True,
is_extracted=False,
depth=0
)
)
return [chunk]
except Exception as e:
print(f"[ERROR] Error chunking config file {file_path}: {e}")
return self._chunk_text_file(file_path, repo_metadata)
def _chunk_special_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Chunk special files (requirements.txt, Dockerfile, etc.)"""
try:
content = file_path.read_text(encoding='utf-8', errors='ignore')
file_name = file_path.name.lower()
if 'requirements' in file_name:
language = "requirements"
chunk_type = "configuration"
prefix = "config"
elif 'docker' in file_name:
language = "dockerfile"
chunk_type = "script"
prefix = "script"
else:
language = "text"
chunk_type = "text"
prefix = "text"
# FIXED: Use deterministic SHA256 instead of hash()
chunk = CodeChunk(
chunk_id=self._generate_stable_id(content, prefix),
file_path=str(file_path),
language=language,
chunk_type=chunk_type,
code=content,
ast=ChunkAST(
symbol_type=chunk_type,
name=file_path.name,
parent=None,
docstring=None
),
span=ChunkSpan(
start_line=1,
end_line=len(content.split('\n'))
),
metadata={
"file_type": file_name,
"repo_info": repo_metadata or {},
"dependencies": self._extract_dependencies(content) if "requirements" in file_name else []
},
hierarchy=ChunkHierarchy(
is_primary=True,
is_extracted=False,
depth=0
)
)
return [chunk]
except Exception as e:
print(f"[ERROR] Error chunking special file {file_path}: {e}")
return self._chunk_text_file(file_path, repo_metadata)
def _chunk_text_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Chunk plain text files"""
try:
content = file_path.read_text(encoding='utf-8', errors='ignore')
# Create a single chunk for small files, multiple for large ones
if len(content.split('\n')) <= 200:
chunks = [self._create_text_chunk(content, file_path, repo_metadata)]
else:
# Split large text files into reasonable chunks
chunks = []
lines = content.split('\n')
chunk_size = 100
for i in range(0, len(lines), chunk_size):
chunk_lines = lines[i:i + chunk_size]
chunk_content = '\n'.join(chunk_lines)
chunk = self._create_text_chunk(
chunk_content,
file_path,
repo_metadata,
chunk_index=i // chunk_size
)
chunks.append(chunk)
return chunks
except Exception as e:
print(f"[ERROR] Error reading text file {file_path}: {e}")
return []
def _chunk_readme_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Special handling for README/LICENSE files"""
try:
content = file_path.read_text(encoding='utf-8', errors='ignore')
file_name_lower = file_path.name.lower()
# Determine appropriate prefix
if 'readme' in file_name_lower:
prefix = "doc"
elif 'license' in file_name_lower:
prefix = "license"
else:
prefix = "doc"
# FIXED: Use deterministic SHA256 instead of hash()
chunk = CodeChunk(
chunk_id=self._generate_stable_id(content, prefix),
file_path=str(file_path),
language="markdown" if file_path.suffix in ['.md', '.mdx'] else "text",
chunk_type="documentation",
code=content,
ast=ChunkAST(
symbol_type="documentation",
name=file_path.name,
parent=None,
docstring=None
),
span=ChunkSpan(
start_line=1,
end_line=len(content.split('\n'))
),
metadata={
"file_type": "readme_license",
"is_readme": "readme" in file_name_lower,
"is_license": "license" in file_name_lower,
"repo_info": repo_metadata or {}
},
hierarchy=ChunkHierarchy(
is_primary=True,
is_extracted=False,
depth=0
)
)
return [chunk]
except Exception as e:
print(f"[ERROR] Error chunking README file {file_path}: {e}")
return self._chunk_text_file(file_path, repo_metadata)
def _chunk_other_file(self, file_path: Path, repo_metadata: Optional[Dict]) -> List[CodeChunk]:
"""Fallback for unknown file types (binary or unsupported)"""
try:
# Try to read as text first
content = file_path.read_text(encoding='utf-8', errors='ignore')
# If it looks like binary (mostly non-printable characters)
if self._looks_like_binary(content):
print(f"[SKIPPED] Skipping binary file: {file_path}")
return []
# If readable text, treat as text file
return self._chunk_text_file(file_path, repo_metadata)
except UnicodeDecodeError:
print(f"[SKIPPED] Skipping binary file: {file_path}")
return []
except Exception as e:
print(f"[ERROR] Error with file {file_path}: {e}")
return []
def _create_text_chunk(self, content: str, file_path: Path,
repo_metadata: Optional[Dict], chunk_index: int = 0) -> CodeChunk:
"""Helper to create a text chunk"""
lines = content.split('\n')
# ENHANCED: Use deterministic ID that includes chunk_index for uniqueness
id_payload = f"{content}_{chunk_index}"
return CodeChunk(
chunk_id=self._generate_stable_id(id_payload, "text"),
file_path=str(file_path),
language="text",
chunk_type="text",
code=content,
ast=ChunkAST(
symbol_type="text",
name=file_path.name,
parent=None,
docstring=None
),
span=ChunkSpan(
start_line=1,
end_line=len(lines)
),
metadata={
"file_type": "text",
"chunk_index": chunk_index,
"total_lines": len(lines),
"repo_info": repo_metadata or {}
},
hierarchy=ChunkHierarchy(
is_primary=True,
is_extracted=False,
depth=0
)
)
def _extract_dependencies(self, requirements_content: str) -> List[str]:
"""Extract package names from requirements.txt"""
dependencies = []
for line in requirements_content.split('\n'):
line = line.strip()
if line and not line.startswith('#'):
# Extract package name (before version specifiers)
package = line.split('==')[0].split('>=')[0].split('<=')[0].strip()
if package:
dependencies.append(package)
return dependencies
def _looks_like_binary(self, content: str, threshold: float = 0.3) -> bool:
"""Check if content looks like binary data"""
if not content:
return False
# Count printable vs non-printable characters
printable = sum(1 for c in content if 32 <= ord(c) <= 126 or c in '\n\r\t')
total = len(content)
if total == 0:
return False
ratio = printable / total
return ratio < threshold
|