#!/usr/bin/env python3 """ RDLC Multi-Deliverable Generator — TEMPLATE Ported scaffold from states-project-research/scripts/generate_pdf.py (claude-rdlc-wizard v0.1.0). One source markdown research base feeds multiple HTML/PDF deliverables, each with its own audience boundaries. The generator enforces audience-firewall rules at render time — content the public deliverable must not contain is stripped before HTML is written. Run: python3 scripts/generate_deliverable.py # all deliverables python3 scripts/generate_deliverable.py interview # one deliverable Customize: 1. Replace the DELIVERABLE_CONFIGS dict with your project's deliverables. 2. Each config picks: source markdown, output filename, audience, sections to include, and forbidden_patterns (audience-firewall regex list). 3. Add domain-specific CSS/JS at the bottom. """ import re import sys from pathlib import Path ROOT = Path(__file__).resolve().parent.parent RESEARCH_DIR = ROOT / "research" OUTPUT_DIR = ROOT / "output" # --- Configuration --- DELIVERABLE_CONFIGS = { # TODO: Replace these examples with your project's deliverables. "research_document": { "source_md": OUTPUT_DIR / "research_document.md", "output_html": OUTPUT_DIR / "research_document.html", "title": "Research Document", "audience": "internal", "include_sections": "all", # or list of section header strings "forbidden_patterns": [], # audience-firewall regex list "footer": "Internal — do not distribute", }, # Example public-facing deliverable with audience firewall: # "strategic_analysis": { # "source_md": OUTPUT_DIR / "research_document.md", # "output_html": OUTPUT_DIR / "strategic_analysis.html", # "title": "Strategic Analysis", # "audience": "external", # "include_sections": ["Org Overview", "Strategy", "Competitive Landscape"], # "forbidden_patterns": [ # r"salary range", # r"\$1[01]\d,\d{3}", # any salary number # r"mock interview", # r"interviewer strategy", # ], # "footer": "Public-facing analysis", # }, } # --- Audience firewall enforcement --- def enforce_audience_firewall(content: str, forbidden: list[str], deliverable_name: str) -> str: """Strip lines matching any forbidden pattern. Loud failure if pattern hits. The point: PDFs and HTML get forwarded without their context. The boundary must hold at render time, not just at the source page. """ if not forbidden: return content violations = [] kept_lines = [] for i, line in enumerate(content.split("\n"), start=1): line_violated = False for pattern in forbidden: if re.search(pattern, line, re.IGNORECASE): violations.append((i, pattern, line.strip()[:80])) line_violated = True break if not line_violated: kept_lines.append(line) if violations: print( f"WARN [{deliverable_name}]: stripped {len(violations)} forbidden lines:", file=sys.stderr, ) for ln, pat, snippet in violations[:5]: print(f" line {ln} matched /{pat}/: {snippet}", file=sys.stderr) if len(violations) > 5: print(f" ... and {len(violations) - 5} more", file=sys.stderr) return "\n".join(kept_lines) # --- Section filter --- def filter_sections(content: str, include_sections) -> str: """Keep only specified top-level sections.""" if include_sections == "all": return content if not isinstance(include_sections, list): return content keep = set(s.lower() for s in include_sections) out = [] in_section = True for line in content.split("\n"): m = re.match(r"^#+\s+(.+)$", line) if m: heading = m.group(1).strip().lower() in_section = any(target in heading for target in keep) if in_section: out.append(line) return "\n".join(out) # --- Markdown to HTML (lightweight, no dependencies) --- def md_to_html(md: str, title: str, footer: str) -> str: """Minimal markdown-to-HTML renderer. Replace this with your preferred renderer (markdown2, mistune, pandoc) when the project's deliverables justify the dependency. Kept inline at v0.1 so the template installs without `pip install`. """ html_body = md # Headers html_body = re.sub(r"^# (.+)$", r"

\1

", html_body, flags=re.MULTILINE) html_body = re.sub(r"^## (.+)$", r"

\1

", html_body, flags=re.MULTILINE) html_body = re.sub(r"^### (.+)$", r"

\1

", html_body, flags=re.MULTILINE) # Bold html_body = re.sub(r"\*\*(.+?)\*\*", r"\1", html_body) # Inline code html_body = re.sub(r"`([^`]+)`", r"\1", html_body) # Paragraphs (very basic) paragraphs = [] for block in re.split(r"\n\n+", html_body.strip()): if block.startswith("{block}

") html_body = "\n".join(paragraphs) return f""" {title} {html_body}
{footer}
""" # --- Generate --- def generate(name: str, config: dict) -> None: src = config["source_md"] if not src.exists(): print(f"ERROR [{name}]: source markdown not found at {src}", file=sys.stderr) sys.exit(1) content = src.read_text(encoding="utf-8") content = filter_sections(content, config.get("include_sections", "all")) content = enforce_audience_firewall( content, config.get("forbidden_patterns", []), name ) html = md_to_html(content, config.get("title", name), config.get("footer", "")) out = config["output_html"] out.parent.mkdir(parents=True, exist_ok=True) out.write_text(html, encoding="utf-8") print(f"Wrote {out}") def main() -> None: target = sys.argv[1] if len(sys.argv) > 1 else "all" if target == "all": for name, config in DELIVERABLE_CONFIGS.items(): generate(name, config) elif target in DELIVERABLE_CONFIGS: generate(target, DELIVERABLE_CONFIGS[target]) else: print( f"Unknown target '{target}'. Available: {list(DELIVERABLE_CONFIGS.keys()) + ['all']}", file=sys.stderr, ) sys.exit(2) if __name__ == "__main__": main()