Spaces:
Running
Running
unknown
Claude Opus 5.5
Journal Formatter: footnotes, web references, books and reports, author marks, cleaner LaTeX
2eb3609 | """Compile a test manuscript's LaTeX output (texlive.net, test content only) and save its pages as PNGs, to look at. | |
| python devtools/fmt_pages.py ieee_journal [numeric] [pages] -> devtools/logs/pages/<preset>_<doc>_p<n>.png""" | |
| import io | |
| import os | |
| import subprocess | |
| import sys | |
| import tempfile | |
| import zipfile | |
| from pathlib import Path | |
| ROOT = Path(__file__).resolve().parent.parent | |
| sys.path.insert(0, str(ROOT)) | |
| sys.path.insert(0, str(Path(__file__).resolve().parent)) | |
| os.environ.setdefault("SKIP_MODEL_LOAD", "1") | |
| import fmt_corpus as C # noqa: E402 | |
| from formatmatch.engine import FormatSource, build_spec, format_manuscript # noqa: E402 | |
| preset = sys.argv[1] | |
| kind = sys.argv[2] if len(sys.argv) > 2 else "numeric" | |
| pages = int(sys.argv[3]) if len(sys.argv) > 3 else 2 | |
| b = io.BytesIO() | |
| C.build(kind).save(b) | |
| r = format_manuscript(f"{kind}.docx", b.getvalue(), build_spec(FormatSource(preset=preset))) | |
| z = zipfile.ZipFile(io.BytesIO(r.latex_zip)) | |
| out_dir = ROOT / "devtools" / "logs" / "pages" | |
| out_dir.mkdir(parents=True, exist_ok=True) | |
| with tempfile.TemporaryDirectory() as td: | |
| args = ["curl", "-s", "-L", "--max-time", "200", "-o", str(Path(td) / "out.pdf")] | |
| for n in z.namelist(): | |
| if n.endswith((".tex", ".bib", ".cls", ".sty", ".bst")): | |
| body = z.read(n).decode("utf-8") | |
| if n == "main.tex": | |
| body, n = "\\PassOptionsToPackage{demo}{graphicx}\n" + body, "document.tex" | |
| p = Path(td) / n.replace("/", "_") | |
| p.write_text(body, encoding="utf-8", newline="\n") | |
| args += ["-F", f"filecontents[]=<{p}", "-F", f"filename[]={n}"] | |
| args += ["-F", "engine=pdflatex", "-F", "return=pdf", "https://texlive.net/cgi-bin/latexcgi"] | |
| subprocess.run(args, capture_output=True, timeout=260) | |
| data = (Path(td) / "out.pdf").read_bytes() | |
| if data[:4] != b"%PDF": | |
| print("no PDF:", data[-600:].decode("utf-8", "replace")) | |
| sys.exit(1) | |
| import pymupdf # noqa: E402 | |
| d = pymupdf.open(stream=data, filetype="pdf") | |
| print(preset, kind, len(d), "pages") | |
| for i in range(min(pages, len(d))): | |
| out = out_dir / f"{preset}_{kind}_p{i + 1}.png" | |
| d[i].get_pixmap(dpi=70).save(str(out)) | |
| print(out) | |