Spaces:
Sleeping
Sleeping
File size: 12,834 Bytes
58e6885 | 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 | #!/usr/bin/env python3
from __future__ import annotations
import argparse
import csv
import json
import os
import sys
import time
from collections import Counter
from concurrent.futures import ThreadPoolExecutor, as_completed
from pathlib import Path
from typing import Any
from lxml import etree
ROOT = Path(__file__).resolve().parents[1]
sys.path.insert(0, str(ROOT))
os.chdir(ROOT)
from modules.slot_layout_planner.chart_sanitizer import ( # noqa: E402
DATA_BEARING_ATTRS,
DATA_BEARING_ROLE_VALUES,
sanitize_chart_svg,
)
from scripts.generate_template_samples import ( # noqa: E402
configure_chart_type_filter,
nested_chart_svg,
run_sample,
safe_slug,
)
GEOMETRY_ATTRS = (
"x",
"y",
"x1",
"y1",
"x2",
"y2",
"cx",
"cy",
"r",
"rx",
"ry",
"width",
"height",
"d",
)
def read_jsonl(path: Path) -> list[dict[str, Any]]:
with open(path, "r", encoding="utf-8") as f:
return [json.loads(line) for line in f if line.strip()]
def append_jsonl(path: Path, record: dict[str, Any]) -> None:
path.parent.mkdir(parents=True, exist_ok=True)
with open(path, "a", encoding="utf-8") as f:
f.write(json.dumps(record, ensure_ascii=False) + "\n")
def write_json(path: Path, payload: Any) -> None:
path.parent.mkdir(parents=True, exist_ok=True)
with open(path, "w", encoding="utf-8") as f:
json.dump(payload, f, indent=2, ensure_ascii=False)
def local_name(elem: etree._Element) -> str:
return etree.QName(elem).localname if isinstance(elem.tag, str) else ""
def normalized_role(elem: etree._Element) -> str:
role = str(elem.get("data-role") or "").strip().lower().replace("_", " ").replace("-", " ")
return " ".join(role.split())
def is_data_bearing(elem: etree._Element) -> bool:
for attr in DATA_BEARING_ATTRS:
value = elem.get(attr)
if value is not None and str(value).strip():
return True
role = normalized_role(elem)
return role in DATA_BEARING_ROLE_VALUES
def data_signature(elem: etree._Element) -> str:
attrs: dict[str, str] = {
"tag": local_name(elem),
"class": str(elem.get("class") or ""),
"id": str(elem.get("id") or ""),
}
for name, value in sorted(elem.attrib.items()):
if name.startswith("data-") or name in GEOMETRY_ATTRS:
attrs[name] = str(value)
text = "".join(elem.itertext()).strip()
if text:
attrs["text"] = " ".join(text.split())
return json.dumps(attrs, sort_keys=True, ensure_ascii=False)
def data_signatures(svg_path: Path) -> Counter[str]:
parser = etree.XMLParser(remove_blank_text=False, recover=True, huge_tree=True)
root = etree.parse(str(svg_path), parser).getroot()
signatures: Counter[str] = Counter()
for elem in root.xpath(".//*"):
if is_data_bearing(elem):
signatures[data_signature(elem)] += 1
return signatures
def metrics(svg_path: Path) -> dict[str, int]:
text = svg_path.read_text(encoding="utf-8", errors="ignore")
return {
"rect": text.count("<rect"),
"path": text.count("<path"),
"circle": text.count("<circle"),
"line": text.count("<line"),
"text": text.count("<text"),
"image": text.count("<image"),
}
def resolve_existing_chart_svg(record: dict[str, Any]) -> Path | None:
chart_svg = record.get("chart_svg")
if chart_svg:
path = Path(chart_svg)
if path.is_file():
return path
sample_dir = record.get("sample_dir")
if sample_dir:
nested = nested_chart_svg(Path(sample_dir))
if nested and nested.is_file():
return nested
return None
def render_chart(record: dict[str, Any], sample_dir: Path, output_png: bool) -> dict[str, Any]:
result = run_sample(
template_key=record["template_key"],
data_path=Path(record["data_source"]),
sample_dir=sample_dir,
output_png=output_png,
chart_only=True,
)
return result
def validate_record(
record: dict[str, Any],
output_root: Path,
render: bool,
output_png: bool,
) -> dict[str, Any]:
template_key = record["template_key"]
sample_index = int(record.get("sample_index", 0))
sample_dir = output_root / "charts" / safe_slug(template_key) / f"sample_{sample_index:02d}"
render_result: dict[str, Any] = {}
if render:
render_result = render_chart(record, sample_dir, output_png)
chart_svg = Path(render_result.get("chart_svg") or "") if render_result.get("chart_svg") else None
else:
chart_svg = resolve_existing_chart_svg(record)
if chart_svg is None or not chart_svg.is_file():
return {
"template_key": template_key,
"sample_index": sample_index,
"data_source": record.get("data_source", ""),
"success": False,
"error": "chart_svg_missing",
}
sanitize_dir = output_root / "sanitized" / safe_slug(template_key) / f"sample_{sample_index:02d}"
sanitized_svg = sanitize_dir / "sanitized_chart.svg"
report_path = sanitize_dir / "sanitized_chart.report.json"
before = data_signatures(chart_svg)
before_metrics = metrics(chart_svg)
report = sanitize_chart_svg(chart_svg, sanitized_svg, report_path=report_path)
after = data_signatures(sanitized_svg)
after_metrics = metrics(sanitized_svg)
lost = before - after
gained = after - before
row = {
"template_key": template_key,
"sample_index": sample_index,
"data_source": record.get("data_source", ""),
"success": True,
"render_success": render_result.get("success") if render else "",
"chart_svg": str(chart_svg),
"sanitized_svg": str(sanitized_svg),
"sanitizer_report": str(report_path),
"removed_count": report.removed_count,
"removed_by_reason": report.removed_by_reason,
"data_bearing_before": sum(before.values()),
"data_bearing_after": sum(after.values()),
"data_bearing_lost": sum(lost.values()),
"data_bearing_gained": sum(gained.values()),
"lost_examples": list(lost.keys())[:5],
"rect_before": before_metrics["rect"],
"rect_after": after_metrics["rect"],
"path_before": before_metrics["path"],
"path_after": after_metrics["path"],
"circle_before": before_metrics["circle"],
"circle_after": after_metrics["circle"],
"text_before": before_metrics["text"],
"text_after": after_metrics["text"],
"image_before": before_metrics["image"],
"image_after": after_metrics["image"],
"error": "",
}
return row
def parse_args() -> argparse.Namespace:
parser = argparse.ArgumentParser(
description="Batch validate that SVG sanitization does not remove data-bearing chart elements."
)
parser.add_argument("--records", type=Path, required=True, help="JSONL records with template_key/data_source/sample_index.")
parser.add_argument("--output-root", type=Path, required=True)
parser.add_argument("--render-chart", action="store_true", help="Render current chart SVGs before validation.")
parser.add_argument("--output-png", action="store_true", help="Generate chart PNGs during chart-only render.")
parser.add_argument("--workers", type=int, default=8)
parser.add_argument("--limit", type=int, default=0)
parser.add_argument("--resume", action="store_true")
return parser.parse_args()
def main() -> int:
args = parse_args()
configure_chart_type_filter(False)
args.output_root.mkdir(parents=True, exist_ok=True)
records = read_jsonl(args.records)
if args.limit:
records = records[: args.limit]
manifest_path = args.output_root / "manifest.jsonl"
done: set[tuple[str, int]] = set()
if args.resume and manifest_path.exists():
for row in read_jsonl(manifest_path):
done.add((row.get("template_key", ""), int(row.get("sample_index", -1))))
elif manifest_path.exists():
manifest_path.unlink()
pending = [
record
for record in records
if (record.get("template_key", ""), int(record.get("sample_index", -1))) not in done
]
write_json(
args.output_root / "run_config.json",
{
"records": str(args.records),
"output_root": str(args.output_root),
"render_chart": args.render_chart,
"output_png": args.output_png,
"workers": args.workers,
"limit": args.limit,
"selected_records": len(records),
"pending_records": len(pending),
},
)
started = time.time()
print(f"validating records={len(records)} pending={len(pending)} workers={args.workers}", flush=True)
if pending and args.workers > 1:
with ThreadPoolExecutor(max_workers=args.workers) as executor:
futures = [
executor.submit(validate_record, record, args.output_root, args.render_chart, args.output_png)
for record in pending
]
for index, future in enumerate(as_completed(futures), 1):
row = future.result()
append_jsonl(manifest_path, row)
elapsed = time.time() - started
rate = index / elapsed if elapsed else 0
remaining = (len(pending) - index) / rate if rate else 0
print(
f"[{index}/{len(pending)}] {row.get('template_key')} "
f"sample_{int(row.get('sample_index', 0)):02d} "
f"success={row.get('success')} lost={row.get('data_bearing_lost', '')} "
f"eta={remaining/60:.1f}m",
flush=True,
)
else:
for index, record in enumerate(pending, 1):
row = validate_record(record, args.output_root, args.render_chart, args.output_png)
append_jsonl(manifest_path, row)
elapsed = time.time() - started
rate = index / elapsed if elapsed else 0
remaining = (len(pending) - index) / rate if rate else 0
print(
f"[{index}/{len(pending)}] {row.get('template_key')} "
f"sample_{int(row.get('sample_index', 0)):02d} "
f"success={row.get('success')} lost={row.get('data_bearing_lost', '')} "
f"eta={remaining/60:.1f}m",
flush=True,
)
rows = read_jsonl(manifest_path) if manifest_path.exists() else []
successes = [row for row in rows if row.get("success")]
failures = [row for row in rows if not row.get("success")]
suspicious = [row for row in successes if int(row.get("data_bearing_lost") or 0) > 0]
removed_reasons: Counter[str] = Counter()
for row in successes:
for reason, count in (row.get("removed_by_reason") or {}).items():
removed_reasons[reason] += int(count)
summary = {
"total_records": len(rows),
"success": len(successes),
"failed": len(failures),
"templates": len({row.get("template_key") for row in rows}),
"suspicious_data_bearing_loss": len(suspicious),
"total_data_bearing_lost": sum(int(row.get("data_bearing_lost") or 0) for row in successes),
"total_removed_nodes": sum(int(row.get("removed_count") or 0) for row in successes),
"removed_by_reason": dict(removed_reasons.most_common()),
"failure_examples": failures[:20],
"suspicious_examples": suspicious[:20],
"manifest": str(manifest_path),
"csv": str(args.output_root / "summary.csv"),
}
write_json(args.output_root / "summary.json", summary)
fieldnames = [
"template_key",
"sample_index",
"success",
"render_success",
"removed_count",
"removed_by_reason",
"data_bearing_before",
"data_bearing_after",
"data_bearing_lost",
"data_bearing_gained",
"rect_before",
"rect_after",
"path_before",
"path_after",
"circle_before",
"circle_after",
"text_before",
"text_after",
"image_before",
"image_after",
"chart_svg",
"sanitized_svg",
"sanitizer_report",
"data_source",
"error",
]
with open(args.output_root / "summary.csv", "w", newline="", encoding="utf-8") as f:
writer = csv.DictWriter(f, fieldnames=fieldnames, extrasaction="ignore")
writer.writeheader()
writer.writerows(rows)
print(json.dumps(summary, indent=2, ensure_ascii=False), flush=True)
return 0 if not failures and not suspicious else 1
if __name__ == "__main__":
raise SystemExit(main())
|