Add arXiv submission packaging script.

Zip arxiv.tex with lean-listings, mermaid figures, and 00README.json for pdfLaTeX upload.

Co-authored-by: Cursor <cursoragent@cursor.com>

Author: catskillsresearch

.cursorignore

@@ -13,3 +13,6 @@ arxiv.tex
 arxiv.pdf
 lean-listings/
 figures/
+
+dist/
+00README.json

.gitignore

@@ -16,6 +16,8 @@ arxiv.tex
 arxiv.pdf
 lean-listings/
 figures/
+dist/
+00README.json
 
 # LaTeX build artifacts (the .tex/.pdf deliverables are kept)
 *.aux

arxiv.md

@@ -1096,6 +1096,7 @@ lake exe cache get
 lake build Scott1972
 bash scripts/generate_arxiv_with_code.sh   # optional: arxiv_with_code.md (gitignored)
 bash scripts/build_arxiv_tex.sh            # optional: arxiv.tex + lean-listings/ (gitignored)
+bash scripts/package_arxiv_submit.sh       # optional: dist/arxiv_submit.zip (gitignored)
 ```
 
 ---

scripts/package_arxiv_submit.sh

@@ -0,0 +1,73 @@
+#!/usr/bin/env bash
+# Build arxiv.tex and zip everything arXiv needs to compile it (pdfLaTeX).
+set -euo pipefail
+
+ROOT="$(cd "$(dirname "$0")/.." && pwd)"
+cd "$ROOT"
+
+TEX="arxiv.tex"
+LISTINGS_DIR="lean-listings"
+FIGURES_DIR="figures"
+OUT_DIR="dist"
+ZIP="${OUT_DIR}/arxiv_submit.zip"
+
+echo "==> Regenerating arxiv_with_code.md, arxiv.tex, listings, and figures"
+bash scripts/build_arxiv_tex.sh
+
+missing=0
+if [[ ! -f "$TEX" ]]; then
+  echo "error: missing $TEX" >&2
+  missing=1
+fi
+if [[ ! -d "$LISTINGS_DIR" ]]; then
+  echo "error: missing $LISTINGS_DIR" >&2
+  missing=1
+fi
+lean_count="$(find "$LISTINGS_DIR" -maxdepth 1 -type f 2>/dev/null | wc -l)"
+if [[ "$lean_count" -eq 0 ]]; then
+  echo "error: no listing files in $LISTINGS_DIR" >&2
+  missing=1
+fi
+fig_count="$(find "$FIGURES_DIR" -maxdepth 1 -name '*.pdf' 2>/dev/null | wc -l)"
+if [[ "$fig_count" -eq 0 ]]; then
+  echo "error: no mermaid figure PDFs in $FIGURES_DIR" >&2
+  missing=1
+fi
+if [[ "$missing" -ne 0 ]]; then
+  exit 1
+fi
+
+mkdir -p "$OUT_DIR"
+rm -f "$ZIP"
+
+echo "==> Writing 00README.json (mark listing files as include so arXiv does not drop them)"
+python3 - <<'PY'
+import json
+from pathlib import Path
+
+sources = [{"filename": "arxiv.tex", "usage": "toplevel"}]
+for path in sorted(p for p in Path("lean-listings").iterdir() if p.is_file()):
+    sources.append({"filename": path.as_posix(), "usage": "include"})
+for path in sorted(Path("figures").glob("*.pdf")):
+    sources.append({"filename": path.as_posix(), "usage": "include"})
+readme = {"process": {"compiler": "pdflatex"}, "sources": sources}
+Path("00README.json").write_text(json.dumps(readme, indent=2) + "\n")
+print(f"  {len(sources)} sources")
+PY
+
+echo "==> Packaging"
+zip -r "$ZIP" \
+  00README.json \
+  "$TEX" \
+  "$LISTINGS_DIR" \
+  "$FIGURES_DIR"/*.pdf
+
+echo "wrote $ZIP ($(du -h "$ZIP" | cut -f1))"
+echo "Contents:"
+zipinfo -1 "$ZIP" | sed 's/^/  /' | head -40
+echo
+echo "Upload $ZIP to arXiv (pdfLaTeX; UTF-8 Lean listings render via the listings literate"
+echo "table; mermaid diagrams ship as pre-rendered figures/*.pdf since AutoTeX cannot run mmdc)."
+echo "On arXiv Add Files: Delete All before uploading (uploads merge, they do not replace)."
+echo "On arXiv Review Files: if any lean-listings/*.lean or figures/*.pdf are marked for"
+echo "deletion, UNCHECK them."