File size: 15,286 Bytes
9aa0c5c | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 | from __future__ import annotations
import hashlib
import json
import mimetypes
import re
import unicodedata
from dataclasses import dataclass, field
from pathlib import Path
from typing import Any, Iterable
DOI_RE = re.compile(r"10\.\d{4,9}/[-._;()/:A-Z0-9]+", re.IGNORECASE)
HEADING_RE = re.compile(
r"\\(?P<kind>part|chapter|section|subsection|subsubsection)\*?\s*\{",
re.IGNORECASE,
)
ENV_TOKEN_RE = re.compile(r"\\(?P<op>begin|end)\s*\{(?P<env>[^{}]+)\}")
PROTECTED_ENVS = {
"equation",
"equation*",
"align",
"align*",
"alignat",
"alignat*",
"gather",
"gather*",
"multline",
"multline*",
"displaymath",
"math",
"theorem",
"lemma",
"proposition",
"corollary",
"definition",
"assumption",
"remark",
"example",
"proof",
"axiom",
"verbatim",
"lstlisting",
}
HEADING_LEVELS = {
"part": 0,
"chapter": 0,
"section": 1,
"subsection": 2,
"subsubsection": 3,
}
SECRET_PATTERNS = {
"openai_key": re.compile(r"(?<![A-Za-z0-9])sk-[A-Za-z0-9_-]{20,}"),
"huggingface_token": re.compile(r"(?<![A-Za-z0-9])hf_[A-Za-z0-9]{20,}"),
"github_token": re.compile(r"(?<![A-Za-z0-9])gh[pousr]_[A-Za-z0-9]{20,}"),
"private_key": re.compile(
r"-----BEGIN (?:RSA |OPENSSH |EC )?PRIVATE KEY-----"
),
}
def sha256_bytes(data: bytes) -> str:
return hashlib.sha256(data).hexdigest()
def sha256_file(path: Path, chunk_size: int = 1024 * 1024) -> str:
digest = hashlib.sha256()
with path.open("rb") as stream:
while chunk := stream.read(chunk_size):
digest.update(chunk)
return digest.hexdigest()
def canonical_json(data: Any) -> str:
return json.dumps(
data,
ensure_ascii=False,
indent=2,
sort_keys=True,
separators=(",", ": "),
) + "\n"
def write_text_lf(path: Path, text: str) -> None:
path.parent.mkdir(parents=True, exist_ok=True)
path.write_text(text.replace("\r\n", "\n"), encoding="utf-8", newline="\n")
def strip_tex_comments(text: str) -> str:
output: list[str] = []
for line in text.splitlines(keepends=True):
cut = None
for index, char in enumerate(line):
if char != "%":
continue
backslashes = 0
cursor = index - 1
while cursor >= 0 and line[cursor] == "\\":
backslashes += 1
cursor -= 1
if backslashes % 2 == 0:
cut = index
break
if cut is None:
output.append(line)
else:
newline = "\n" if line.endswith("\n") else ""
output.append(line[:cut] + newline)
return "".join(output)
def extract_braced_command(text: str, command: str) -> str:
match = re.search(rf"\\{re.escape(command)}\s*\{{", text)
if not match:
return ""
start = match.end()
cursor = start
depth = 1
while cursor < len(text) and depth:
char = text[cursor]
escaped = cursor > 0 and text[cursor - 1] == "\\"
if char == "{" and not escaped:
depth += 1
elif char == "}" and not escaped:
depth -= 1
cursor += 1
if depth:
return ""
return text[start : cursor - 1]
def clean_tex_label(text: str) -> str:
cleaned = strip_tex_comments(text)
cleaned = re.sub(r"\\\\(?:\[[^\]]*\])?", " ", cleaned)
cleaned = re.sub(r"\\(?:href)\s*\{[^{}]*\}\s*\{([^{}]*)\}", r"\1", cleaned)
cleaned = re.sub(r"\\(?:url)\s*\{([^{}]*)\}", r"\1", cleaned)
cleaned = re.sub(r"\\[A-Za-z@]+\*?(?:\[[^\]]*\])?", " ", cleaned)
cleaned = cleaned.replace("{", " ").replace("}", " ").replace("~", " ")
cleaned = re.sub(r"\s+", " ", cleaned)
return cleaned.strip()
def normalize_title(text: str) -> str:
normalized = unicodedata.normalize("NFKD", clean_tex_label(text)).lower()
normalized = normalized.replace("–", "-").replace("—", "-")
normalized = re.sub(r"[^a-z0-9]+", " ", normalized)
return " ".join(normalized.split())
def archive_stem_title(name: str) -> str:
stem = Path(name).stem
stem = re.sub(r"\s+\(\d+\)$", "", stem)
return stem.replace("__", " ").replace("_", " ").strip()
def extract_document_body(text: str) -> str:
begin = re.search(r"\\begin\s*\{document\}", text)
end_matches = list(re.finditer(r"\\end\s*\{document\}", text))
if not begin:
return strip_tex_comments(text).strip()
end = end_matches[-1].start() if end_matches else len(text)
return strip_tex_comments(text[begin.end() : end]).strip()
def brace_balance(text: str) -> int:
balance = 0
stripped = strip_tex_comments(text)
for index, char in enumerate(stripped):
if char not in "{}":
continue
backslashes = 0
cursor = index - 1
while cursor >= 0 and stripped[cursor] == "\\":
backslashes += 1
cursor -= 1
if backslashes % 2:
continue
balance += 1 if char == "{" else -1
return balance
def mime_type_for(name: str) -> str:
extension = Path(name).suffix.lower()
overrides = {
".tex": "application/x-tex",
".bib": "application/x-bibtex",
".json": "application/json",
".py": "text/x-python",
".zip": "application/zip",
}
return overrides.get(extension) or mimetypes.guess_type(name)[0] or "application/octet-stream"
def find_secret_patterns(text: str) -> list[str]:
return sorted(name for name, pattern in SECRET_PATTERNS.items() if pattern.search(text))
def find_unsafe_tex_references(text: str) -> list[str]:
unsafe: list[str] = []
pattern = re.compile(
r"\\(?:input|include|includegraphics|lstinputlisting)\s*"
r"(?:\[[^\]]*\])?\s*\{([^}]+)\}"
)
for match in pattern.finditer(text):
candidate = match.group(1)
if (
re.match(r"^[A-Za-z]:", candidate)
or candidate.startswith(("/", "\\\\"))
or ".." in Path(candidate).parts
):
unsafe.append(candidate)
return sorted(set(unsafe))
def readable_tex(text: str) -> tuple[str, list[str]]:
flags: list[str] = []
try:
from pylatexenc.latex2text import LatexNodes2Text
converter = LatexNodes2Text(
math_mode="verbatim",
keep_comments=False,
strict_latex_spaces=False,
)
rendered = converter.latex_to_text(text)
except Exception:
flags.append("latex_to_text_fallback")
rendered = clean_tex_label(text)
rendered = re.sub(r"[ \t]+\n", "\n", rendered)
rendered = re.sub(r"\n{3,}", "\n\n", rendered)
return rendered.strip(), flags
@dataclass(slots=True)
class TexEntry:
path: str
size: int
compressed_size: int
crc32: str
sha256: str
text: str
title_raw: str
title_clean: str
author_raw: str
author_clean: str
date_raw: str
doi_candidates: list[str]
content_status: str
quality_flags: list[str] = field(default_factory=list)
@dataclass(slots=True)
class Archive:
filename: str
source_path: Path
raw_path: str
size: int
sha256: str
entries: list[dict[str, Any]]
tex_entries: list[TexEntry]
primary_tex_index: int | None
mapped_dois: list[str] = field(default_factory=list)
mapping_method: str = "unresolved"
mapping_score: float = 0.0
mapping_status: str = "archive_only"
candidate_dois: list[str] = field(default_factory=list)
duplicate_of_archive: str = ""
quality_flags: list[str] = field(default_factory=list)
@property
def primary_tex(self) -> TexEntry | None:
if self.primary_tex_index is None:
return None
return self.tex_entries[self.primary_tex_index]
@property
def content_status(self) -> str:
primary = self.primary_tex
if primary is None:
return "invalid_source"
if self.duplicate_of_archive:
return "exact_duplicate"
return primary.content_status
@dataclass(slots=True)
class Block:
start: int
end: int
text: str
section_path: tuple[str, ...]
kind: str
def _protected_spans(text: str) -> list[tuple[int, int]]:
spans: list[tuple[int, int]] = []
stack: list[tuple[str, int]] = []
for match in ENV_TOKEN_RE.finditer(text):
env = match.group("env").strip().lower()
if env not in PROTECTED_ENVS:
continue
if match.group("op") == "begin":
stack.append((env, match.start()))
continue
for index in range(len(stack) - 1, -1, -1):
open_env, open_start = stack[index]
if open_env != env:
continue
is_outer = index == 0
del stack[index:]
if is_outer:
spans.append((open_start, match.end()))
break
return sorted(spans)
def _heading_title(block_text: str) -> tuple[int | None, str]:
match = HEADING_RE.search(block_text)
if not match:
return None, ""
command = match.group("kind").lower()
start = match.end()
cursor = start
depth = 1
while cursor < len(block_text) and depth:
char = block_text[cursor]
escaped = cursor > 0 and block_text[cursor - 1] == "\\"
if char == "{" and not escaped:
depth += 1
elif char == "}" and not escaped:
depth -= 1
cursor += 1
raw = block_text[start : cursor - 1] if depth == 0 else ""
return HEADING_LEVELS[command], clean_tex_label(raw)
def latex_blocks(body: str) -> list[Block]:
spans = _protected_spans(body)
raw_parts: list[tuple[int, int, str]] = []
cursor = 0
for start, end in spans:
if start > cursor:
raw_parts.append((cursor, start, "text"))
raw_parts.append((start, end, "environment"))
cursor = end
if cursor < len(body):
raw_parts.append((cursor, len(body), "text"))
pieces: list[tuple[int, int, str]] = []
for start, end, kind in raw_parts:
if kind == "environment":
pieces.append((start, end, kind))
continue
segment = body[start:end]
paragraph_starts = [0]
for match in re.finditer(r"\n\s*\n", segment):
paragraph_starts.append(match.end())
paragraph_starts.append(len(segment))
for left, right in zip(paragraph_starts, paragraph_starts[1:]):
absolute_left = start + left
absolute_right = start + right
content = body[absolute_left:absolute_right]
if not content.strip():
continue
heading_positions = [m.start() for m in HEADING_RE.finditer(content)]
if not heading_positions:
pieces.append((absolute_left, absolute_right, "text"))
continue
split_points = sorted(set([0, *heading_positions, len(content)]))
for local_left, local_right in zip(split_points, split_points[1:]):
if local_right <= local_left:
continue
piece_start = absolute_left + local_left
piece_end = absolute_left + local_right
if body[piece_start:piece_end].strip():
pieces.append((piece_start, piece_end, "heading_or_text"))
section_stack: list[str] = []
blocks: list[Block] = []
for start, end, kind in sorted(pieces):
text = body[start:end]
level, title = _heading_title(text)
block_kind = kind
if level is not None:
section_stack = section_stack[:level]
while len(section_stack) < level:
section_stack.append("")
if level == 0:
section_stack = [title]
else:
section_stack.append(title)
block_kind = "heading"
blocks.append(
Block(
start=start,
end=end,
text=text,
section_path=tuple(item for item in section_stack if item),
kind=block_kind,
)
)
return blocks
def make_chunks(
body: str,
target_chars: int = 4000,
max_chars: int = 6000,
overlap_chars: int = 400,
min_chars: int = 1500,
) -> list[dict[str, Any]]:
blocks = latex_blocks(body)
if not blocks and body.strip():
blocks = [Block(0, len(body), body, tuple(), "text")]
chunks: list[dict[str, Any]] = []
current: list[Block] = []
def emit(selected: list[Block]) -> None:
if not selected:
return
start = selected[0].start
end = selected[-1].end
tex = body[start:end].strip()
if not tex:
return
flags: list[str] = []
if len(tex) > max_chars:
flags.append("oversize_block")
plain, render_flags = readable_tex(tex)
flags.extend(render_flags)
path = selected[0].section_path
chunks.append(
{
"char_start": start,
"char_end": end,
"chunk_tex": tex,
"chunk_text": plain,
"section_path": list(path),
"section_title": path[-1] if path else "",
"quality_flags": sorted(set(flags)),
}
)
for block in blocks:
block_length = block.end - block.start
if block_length > max_chars:
emit(current)
current = []
emit([block])
continue
current_length = current[-1].end - current[0].start if current else 0
starts_new_section = block.kind == "heading" and bool(current)
would_exceed = current and block.end - current[0].start > max_chars
target_reached = current_length >= min_chars and (
starts_new_section or current_length >= target_chars
)
if current and (would_exceed or target_reached):
previous = list(current)
emit(previous)
overlap: list[Block] = []
overlap_size = 0
for candidate in reversed(previous):
candidate_size = candidate.end - candidate.start
if overlap and overlap_size + candidate_size > overlap_chars:
break
if candidate.kind == "heading" and overlap:
break
overlap.insert(0, candidate)
overlap_size += candidate_size
if overlap_size >= overlap_chars:
break
current = [] if starts_new_section else overlap
current.append(block)
emit(current)
return chunks
def content_size_category(count: int) -> str:
if count < 1000:
return "n<1K"
if count < 10_000:
return "1K<n<10K"
if count < 100_000:
return "10K<n<100K"
if count < 1_000_000:
return "100K<n<1M"
return "n>1M"
def batched(items: Iterable[Any], size: int) -> Iterable[list[Any]]:
batch: list[Any] = []
for item in items:
batch.append(item)
if len(batch) == size:
yield batch
batch = []
if batch:
yield batch
|