File size: 10,635 Bytes
eaaeb1b |
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 |
#!/usr/bin/env python3
from __future__ import annotations
import argparse
import difflib
import hashlib
import os
import textwrap
from dataclasses import dataclass
from pathlib import Path
def sha256_short(data: bytes, n: int = 12) -> str:
return hashlib.sha256(data).hexdigest()[:n]
def is_probably_text(path: Path) -> bool:
ext = path.suffix.lower()
return ext in {".tex", ".bib", ".txt", ".md", ".rst", ".csv", ".json", ".yaml", ".yml"}
def escape_pdf_string(text: str) -> str:
return (
text.replace("\\", "\\\\")
.replace("(", "\\(")
.replace(")", "\\)")
.replace("\r", "\\r")
.replace("\t", "\\t")
)
@dataclass(frozen=True)
class PdfLine:
text: str
rgb: tuple[float, float, float]
class SimplePdf:
def __init__(
self,
page_width: int = 612, # US Letter
page_height: int = 792,
margin_left: int = 40,
margin_top: int = 760,
margin_bottom: int = 40,
font_size: int = 8,
leading: int = 10,
max_chars: int = 100,
) -> None:
self.page_width = page_width
self.page_height = page_height
self.margin_left = margin_left
self.margin_top = margin_top
self.margin_bottom = margin_bottom
self.font_size = font_size
self.leading = leading
self.max_chars = max_chars
def _lines_per_page(self) -> int:
usable = self.margin_top - self.margin_bottom
return max(1, usable // self.leading)
def _chunk_lines(self, lines: list[PdfLine]) -> list[list[PdfLine]]:
chunks: list[list[PdfLine]] = []
current: list[PdfLine] = []
for line in lines:
parts = self._wrap_preserve(line.text, self.max_chars)
for part in parts:
current.append(PdfLine(part, line.rgb))
if len(current) >= self._lines_per_page():
chunks.append(current)
current = []
if current:
chunks.append(current)
return chunks
@staticmethod
def _wrap_preserve(text: str, width: int) -> list[str]:
if len(text) <= width:
return [text]
# Preserve leading whitespace for diff readability.
leading_ws_len = len(text) - len(text.lstrip(" "))
prefix = " " * min(leading_ws_len, width - 1)
rest = text.lstrip(" ")
wrapped = textwrap.wrap(
rest,
width=width - len(prefix),
break_long_words=True,
break_on_hyphens=False,
drop_whitespace=False,
)
if not wrapped:
return [text[:width], prefix + text[width:]]
out = [prefix + wrapped[0]]
out.extend(prefix + w for w in wrapped[1:])
return out
def render(self, lines: list[PdfLine], out_path: Path) -> None:
pages = self._chunk_lines(lines)
objects: list[bytes] = []
def add_object(payload: bytes) -> int:
objects.append(payload)
return len(objects)
# Font (Courier)
font_obj = add_object(b"<< /Type /Font /Subtype /Type1 /BaseFont /Courier >>")
content_objs: list[int] = []
page_objs: list[int] = []
for page_lines in pages:
stream = self._content_stream(page_lines).encode("utf-8")
content_obj = add_object(
b"<< /Length "
+ str(len(stream)).encode("ascii")
+ b" >>\nstream\n"
+ stream
+ b"\nendstream"
)
content_objs.append(content_obj)
# Pages object created after page objects, but we need its object number for /Parent.
# Reserve a slot.
pages_obj_placeholder = add_object(b"")
pages_obj_num = pages_obj_placeholder
for content_obj in content_objs:
page_obj = add_object(
(
f"<< /Type /Page /Parent {pages_obj_num} 0 R "
f"/MediaBox [0 0 {self.page_width} {self.page_height}] "
f"/Resources << /Font << /F1 {font_obj} 0 R >> >> "
f"/Contents {content_obj} 0 R >>"
).encode("ascii")
)
page_objs.append(page_obj)
kids = " ".join(f"{n} 0 R" for n in page_objs)
pages_obj = (
f"<< /Type /Pages /Kids [{kids}] /Count {len(page_objs)} >>".encode("ascii")
)
objects[pages_obj_num - 1] = pages_obj
catalog_obj = add_object(
f"<< /Type /Catalog /Pages {pages_obj_num} 0 R >>".encode("ascii")
)
self._write_pdf(objects, catalog_obj, out_path)
def _content_stream(self, page_lines: list[PdfLine]) -> str:
x = self.margin_left
y = self.margin_top
parts = [
"BT",
f"/F1 {self.font_size} Tf",
f"{x} {y} Td",
]
prev_rgb: tuple[float, float, float] | None = None
first = True
for line in page_lines:
if first:
first = False
else:
parts.append(f"0 -{self.leading} Td")
if prev_rgb != line.rgb:
r, g, b = line.rgb
parts.append(f"{r:.3f} {g:.3f} {b:.3f} rg")
prev_rgb = line.rgb
parts.append(f"({escape_pdf_string(line.text)}) Tj")
parts.append("ET")
return "\n".join(parts)
@staticmethod
def _write_pdf(objects: list[bytes], root_obj_num: int, out_path: Path) -> None:
# PDF header: include binary chars line per spec recommendation.
out = bytearray()
out += b"%PDF-1.4\n%\xe2\xe3\xcf\xd3\n"
xref_positions = [0] # object 0
for i, payload in enumerate(objects, start=1):
xref_positions.append(len(out))
out += f"{i} 0 obj\n".encode("ascii")
out += payload + b"\nendobj\n"
xref_start = len(out)
out += f"xref\n0 {len(objects) + 1}\n".encode("ascii")
out += b"0000000000 65535 f \n"
for pos in xref_positions[1:]:
out += f"{pos:010d} 00000 n \n".encode("ascii")
out += (
b"trailer\n"
+ f"<< /Size {len(objects) + 1} /Root {root_obj_num} 0 R >>\n".encode(
"ascii"
)
+ b"startxref\n"
+ str(xref_start).encode("ascii")
+ b"\n%%EOF\n"
)
out_path.parent.mkdir(parents=True, exist_ok=True)
out_path.write_bytes(out)
def diff_to_pdf_lines(old_root: Path, new_root: Path) -> list[PdfLine]:
old_files = {p.relative_to(old_root) for p in old_root.rglob("*") if p.is_file()}
new_files = {p.relative_to(new_root) for p in new_root.rglob("*") if p.is_file()}
added = sorted(new_files - old_files)
removed = sorted(old_files - new_files)
common = sorted(old_files & new_files)
changed_text: list[Path] = []
changed_bin: list[Path] = []
for rel in common:
p1 = old_root / rel
p2 = new_root / rel
b1 = p1.read_bytes()
b2 = p2.read_bytes()
if b1 == b2:
continue
if is_probably_text(rel):
changed_text.append(rel)
else:
changed_bin.append(rel)
lines: list[PdfLine] = []
def add_line(text: str, rgb: tuple[float, float, float] = (0, 0, 0)) -> None:
lines.append(PdfLine(text, rgb))
add_line(f"Diff: {old_root} -> {new_root}", (0.1, 0.1, 0.1))
add_line("", (0, 0, 0))
add_line("Summary", (0.1, 0.1, 0.6))
add_line(f"- Changed text files: {len(changed_text)}", (0, 0, 0))
for rel in changed_text:
add_line(f" - {rel}", (0, 0, 0))
add_line(f"- Changed binary files: {len(changed_bin)}", (0, 0, 0))
for rel in changed_bin:
p1 = old_root / rel
p2 = new_root / rel
b1 = p1.read_bytes()
b2 = p2.read_bytes()
add_line(
f" - {rel} (old {len(b1)}B {sha256_short(b1)}, new {len(b2)}B {sha256_short(b2)})",
(0, 0, 0),
)
add_line(f"- Added files: {len(added)}", (0, 0, 0))
for rel in added:
p = new_root / rel
b = p.read_bytes()
add_line(f" - {rel} ({len(b)}B {sha256_short(b)})", (0, 0, 0))
add_line(f"- Removed files: {len(removed)}", (0, 0, 0))
for rel in removed:
add_line(f" - {rel}", (0, 0, 0))
add_line("", (0, 0, 0))
add_line("Unified diffs (text)", (0.1, 0.1, 0.6))
add_line("", (0, 0, 0))
def color_for_diff_line(s: str) -> tuple[float, float, float]:
if s.startswith("diff "):
return (0.2, 0.2, 0.2)
if s.startswith("--- ") or s.startswith("+++ "):
return (0.2, 0.2, 0.2)
if s.startswith("@@"):
return (0.1, 0.1, 0.6)
if s.startswith("+") and not s.startswith("+++"):
return (0.0, 0.45, 0.0)
if s.startswith("-") and not s.startswith("---"):
return (0.75, 0.0, 0.0)
return (0, 0, 0)
for rel in changed_text:
p1 = old_root / rel
p2 = new_root / rel
old_lines = p1.read_text(encoding="utf-8", errors="replace").splitlines()
new_lines = p2.read_text(encoding="utf-8", errors="replace").splitlines()
add_line(f"File: {rel}", (0.2, 0.2, 0.2))
diff_lines = difflib.unified_diff(
old_lines,
new_lines,
fromfile=str(old_root / rel),
tofile=str(new_root / rel),
lineterm="",
n=3,
)
for s in diff_lines:
add_line(s, color_for_diff_line(s))
add_line("", (0, 0, 0))
return lines
def main() -> int:
parser = argparse.ArgumentParser(
description="Generate a PDF containing unified diffs between two paper versions."
)
parser.add_argument("old_dir", type=Path)
parser.add_argument("new_dir", type=Path)
parser.add_argument(
"-o",
"--out",
type=Path,
default=Path("paper-sources/paper-301-vs-757-diff.pdf"),
help="Output PDF path (default: paper-sources/paper-301-vs-757-diff.pdf)",
)
args = parser.parse_args()
old_dir = args.old_dir.resolve()
new_dir = args.new_dir.resolve()
if not old_dir.is_dir():
raise SystemExit(f"Not a directory: {old_dir}")
if not new_dir.is_dir():
raise SystemExit(f"Not a directory: {new_dir}")
pdf = SimplePdf()
lines = diff_to_pdf_lines(old_dir, new_dir)
pdf.render(lines, args.out)
print(args.out)
return 0
if __name__ == "__main__":
raise SystemExit(main())
|