fin-paper-convert
Compile LaTeX manuscripts to publication-ready PDFs and generate submission variants (anonymous, arxiv, word) for the target journal.
Step 0: Pre-compilation Check
Before running LaTeX compilation, verify all components are ready:
bash# Check required files exist ls -la output/fin-manuscript/draft_v1/ ls -la output/fin-manuscript/draft_v1/sections/ 2>/dev/null || echo "No sections dir" ls -la output/fin-manuscript/draft_v1/figures/ 2>/dev/null || echo "No figures dir" # Check LaTeX installation which xelatex pdflatex latexmk 2>/dev/null latexmk --version 2>/dev/null || echo "latexmk not found" # Check bibliography tool which bibtex biber 2>/dev/null
python#!/usr/bin/env python3 """Pre-compilation validation for fin-paper-convert.""" import os import re from pathlib import Path from dataclasses import dataclass from typing import List, Optional, Tuple @dataclass class ValidationResult: ok: bool errors: List[str] warnings: List[str] missing_files: List[str] def validate_latex_project(base_dir: str) -> ValidationResult: """ Validate all required files and structure before compilation. """ base = Path(base_dir) errors = [] warnings = [] missing = [] # Check main.tex main_tex = base / "main.tex" if not main_tex.exists(): missing.append(str(main_tex)) errors.append("main.tex not found") # Check references.bib bib_file = base / "references.bib" if not bib_file.exists(): missing.append(str(bib_file)) errors.append("references.bib not found") else: # Count references bib_content = bib_file.read_text(encoding="utf-8") ref_count = len(re.findall(r'@\w+\{', bib_content)) warnings.append(f"Found {ref_count} references in references.bib") # Check figures directory fig_dir = base / "figures" if not fig_dir.exists(): warnings.append("figures/ directory not found (no figures will be included)") else: fig_files = list(fig_dir.glob("*.pdf")) + list(fig_dir.glob("*.png")) if not fig_files: warnings.append("No figure files (.pdf/.png) found in figures/") # Check sections directory sec_dir = base / "sections" if sec_dir.exists(): tex_files = list(sec_dir.glob("*.tex")) if tex_files: warnings.append(f"Found {len(tex_files)} section .tex files") # Validate main.tex structure if main_tex.exists(): content = main_tex.read_text(encoding="utf-8") # Check required packages if r'\documentclass' not in content: errors.append("main.tex missing \\documentclass") # Check bibliography commands if r'\bibliography' not in content and r'\addbibresource' not in content: warnings.append("No bibliography command found in main.tex") # Check input commands for sections if sec_dir.exists() and not any(r'\input{' in content for _ in [1]): warnings.append("No \\input{} commands found — sections may not be included") return ValidationResult( ok=len(errors) == 0, errors=errors, warnings=warnings, missing_files=missing, )
Step 1: Journal Template Selection
pythonJOURNAL_TEMPLATE_CONFIG = { "经济研究": { "class": "ctexart", "font": "xeCJK", "cite_style": "gbt-7714-2015", "compile_cmd": "xelatex main.tex && bibtex main && xelatex main.tex && xelatex main.tex", "compile_sequence": ["xelatex", "bibtex", "xelatex", "xelatex"], "required_packages": [ "ctex", "amsmath", "amssymb", "amsfonts", "booktabs", "threeline", "graphicx", "hyperref", "geometry", "setspace", "titlesec", ], "anonymous": False, "arxiv_compatible": False, "word_compatible": True, "submission_format": "pdf", "notes": "使用CTeX宏包，中文参考文献GB/T 7714格式", }, "金融研究": { "class": "ctexart", "font": "xeCJK", "cite_style": "gbt-7714-2015", "compile_cmd": "xelatex main.tex && bibtex main && xelatex main.tex && xelatex main.tex", "compile_sequence": ["xelatex", "bibtex", "xelatex", "xelatex"], "required_packages": ["ctex", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref"], "anonymous": False, "arxiv_compatible": False, "word_compatible": True, "submission_format": "pdf", }, "管理世界": { "class": "ctexart", "font": "xeCJK", "cite_style": "gbt-7714-2015", "compile_cmd": "xelatex main.tex && bibtex main && xelatex main.tex && xelatex main.tex", "compile_sequence": ["xelatex", "bibtex", "xelatex", "xelatex"], "required_packages": ["ctex", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref"], "anonymous": False, "arxiv_compatible": False, "word_compatible": True, "submission_format": "pdf", }, "JF": { "class": "aea", "font": "times", "cite_style": "aer", "compile_cmd": "pdflatex main.tex && bibtex main && pdflatex main.tex && pdflatex main.tex", "compile_sequence": ["pdflatex", "bibtex", "pdflatex", "pdflatex"], "required_packages": ["aertt", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref", "natbib"], "anonymous": True, "arxiv_compatible": True, "word_compatible": True, "submission_format": "pdf", }, "JFE": { "class": "jfe", "font": "times", "cite_style": "jfe", "compile_cmd": "pdflatex main.tex && bibtex main && pdflatex main.tex && pdflatex main.tex", "compile_sequence": ["pdflatex", "bibtex", "pdflatex", "pdflatex"], "required_packages": ["jfecls", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref", "natbib"], "anonymous": True, "arxiv_compatible": True, "word_compatible": True, "submission_format": "pdf", }, "RFS": { "class": "rfs", "font": "times", "cite_style": "rfs", "compile_cmd": "pdflatex main.tex && biber main && pdflatex main.tex && pdflatex main.tex", "compile_sequence": ["pdflatex", "biber", "pdflatex", "pdflatex"], "required_packages": ["rfs", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref", "biblatex"], "anonymous": True, "arxiv_compatible": True, "word_compatible": True, "submission_format": "pdf", "online_appendix": True, }, "JAE": { "class": "aea", "font": "times", "cite_style": "aer", "compile_cmd": "pdflatex main.tex && bibtex main && pdflatex main.tex && pdflatex main.tex", "compile_sequence": ["pdflatex", "bibtex", "pdflatex", "pdflatex"], "required_packages": ["aertt", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref", "natbib"], "anonymous": True, "arxiv_compatible": True, "word_compatible": True, "submission_format": "pdf", }, "AER": { "class": "aea", "font": "times", "cite_style": "aer", "compile_cmd": "pdflatex main.tex && bibtex main && pdflatex main.tex && pdflatex main.tex", "compile_sequence": ["pdflatex", "bibtex", "pdflatex", "pdflatex"], "required_packages": ["aertt", "amsmath", "amssymb", "booktabs", "graphicx", "hyperref", "natbib"], "anonymous": True, "arxiv_compatible": True, "word_compatible": True, "submission_format": "pdf", }, } def get_template(journal_name: str) -> dict: """Get the template configuration for a journal.""" return JOURNAL_TEMPLATE_CONFIG.get(journal_name, JOURNAL_TEMPLATE_CONFIG["经济研究"]) def format_latex_preamble(template: dict, language: str = "zh") -> str: """Generate LaTeX preamble based on template configuration.""" if language == "zh": return f"""\\documentclass[12pt,twocolumn]{{{template['class']}}} \\usepackage{{{template['font']}}} \\usepackage{amsmath,amssymb,amsfonts}} \\usepackage{{booktabs}} \\usepackage{{graphicx}} \\usepackage{{hyperref}} \\usepackage{{geometry}} \\geometry{{a4paper,left=2.5cm,right=2.5cm,top=2.5cm,bottom=2.5cm}} \\usepackage{{setspace}} \\setlength{{\\parindent}}{{2em}} \\setlength{{\\parskip}}{{6pt}} """ else: return f"""\\documentclass[american]{{{template['class']}}} \\usepackage{{times}} \\usepackage{{amsmath,amssymb}} \\usepackage{{booktabs}} \\usepackage{{graphicx}} \\usepackage{{hyperref}} \\usepackage{{natbib}} \\bibliographystyle{{{template['cite_style']}}} """
Step 2: Generate main.tex
Create a complete main.tex file from existing content:
python#!/usr/bin/env python3 """Generate main.tex from PAPER_OUTLINE.md and section files.""" import os import re from pathlib import Path from typing import Optional class ReportGenerator: """ Generate complete LaTeX manuscripts from structured content. Supports Chinese (CTeX) and English (AEA) journal formats. """ def __init__(self, output_dir: str, language: str = "zh"): self.output_dir = Path(output_dir) self.language = language self.output_dir.mkdir(parents=True, exist_ok=True) def generate_main_tex(self, title: str, authors: str, abstract: str, keywords: list, journal_config: dict, sections: dict) -> str: """ Generate complete main.tex. Parameters ---------- title : str Paper title authors : str Author information abstract : str Abstract content keywords : list List of keywords journal_config : dict From get_template() sections : dict Mapping of section names to .tex file paths or content """ if self.language == "zh": return self._generate_chinese_tex( title, authors, abstract, keywords, journal_config, sections ) else: return self._generate_english_tex( title, authors, abstract, keywords, journal_config, sections ) def _generate_chinese_tex(self, title, authors, abstract, keywords, config, sections) -> str: """Generate Chinese (CTeX) LaTeX manuscript.""" # Keywords string kw_str = "；".join(keywords) # Section includes section_includes = [] for sec_name, sec_path in sections.items(): # Try to include as file path if isinstance(sec_path, str) and Path(sec_path).exists(): section_includes.append(f"\\input{{{sec_path}}}") elif isinstance(sec_path, str) and sec_path.startswith("sections/"): section_includes.append(f"\\input{{{sec_path}}}") if not section_includes: # Generate inline sections section_includes = [ "\\input{sections/01_introduction}", "\\input{sections/02_literature}", "\\input{sections/03_data}", "\\input{sections/04_results}", "\\input{sections/05_robustness}", "\\input{sections/06_conclusion}", ] tex = f"""% !TEX program = xelatex \\documentclass[12pt]{{ctexart}} \\usepackage{{xeCJK}} \\usepackage{{amsmath,amssymb,amsfonts}} \\usepackage{{booktabs}} \\usepackage{{threeline}} \\usepackage{{graphicx}} \\usepackage{{hyperref}} \\usepackage{{geometry}} \\geometry{{a4paper,left=2.5cm,right=2.5cm,top=2.5cm,bottom=2.5cm}} \\usepackage{{setspace}} \\setlength{{\\parindent}}{{2em}} \\setlength{{\\parskip}}{{6pt}} % Title \\title{{{title}}} \\author{{{authors}}} % Abstract \\renewcommand{{\\abstractname}}{{摘要}} \\begin{{document}} \\maketitle \\begin{{abstract}}} {{{abstract}}} \\end{{abstract}} \\ paragraph*{{{keywords}:}} {kw_str} \\newpage \\input{{sections/01_introduction}} \\input{{sections/02_literature}} \\input{{sections/03_data}} \\input{{sections/04_results}} \\input{{sections/05_robustness}} \\input{{sections/06_conclusion}} \\bibliographystyle{{gbt-7714-2015}} \\bibliography{{references}} \\end{{document}} """ return tex def _generate_english_tex(self, title, authors, abstract, keywords, config, sections) -> str: """Generate English (AEA) LaTeX manuscript.""" kw_str = ", ".join(keywords) jel_str = "" # JEL codes if provided section_includes = [] for sec_name, sec_path in sections.items(): if isinstance(sec_path, str) and Path(sec_path).exists(): section_includes.append(f"\\input{{{sec_path}}}") if not section_includes: section_includes = [ "\\input{sections/01_introduction}", "\\input{sections/02_literature}", "\\input{sections/03_data}", "\\input{sections/04_results}", "\\input{sections/05_robustness}", "\\input{sections/06_conclusion}", ] tex = f"""% !TEX program = pdflatex \\documentclass[american]{{aea}} \\usepackage{{times}} \\usepackage{{amsmath,amssymb}} \\usepackage{{booktabs}} \\usepackage{{graphicx}} \\usepackage{{hyperref}} \\usepackage{{natbib}} \\bibliographystyle{{aer}} % Title \\title{{{title}}} \\author{{{authors}}} \\date{{\\today}} \\begin{{document}} \\maketitle \\begin{{abstract}}}} {{{abstract}}} \\end{{abstract}} \\begin{{JEL-classification}} {jel_str} \\end{{JEL-classification}} \\begin{{keywords}} {kw_str} \\end{{keywords}} {"".join(f"\\input{{{s}}}\n" for s in section_includes)} \\bibliography{{references}} \\end{{document}} """ return tex def write_tex(self, filename: str, content: str): """Write LaTeX content to file.""" path = self.output_dir / filename path.write_text(content, encoding="utf-8") print(f" Written: {path}")
Step 3: Compilation Pipeline
python#!/usr/bin/env python3 """LaTeX compilation pipeline for fin-paper-convert.""" import subprocess import os import re import time from pathlib import Path from dataclasses import dataclass from typing import List, Optional, Tuple @dataclass class CompileResult: success: bool log_content: str error_count: int warning_count: int reference_warnings: List[str] page_count: Optional[int] = None compilation_time: Optional[float] = None class LatexCompiler: """ Robust LaTeX compilation with multiple passes and error detection. """ def __init__(self, working_dir: str, project_name: str = "main"): self.working_dir = Path(working_dir) self.project_name = project_name self.log_file = self.working_dir / f"{project_name}.log" def compile(self, compile_sequence: List[str], timeout: int = 120) -> CompileResult: """ Run a full compilation sequence (e.g., pdflatex → bibtex → pdflatex → pdflatex). Parameters ---------- compile_sequence : list List of commands in order, e.g. ["pdflatex", "bibtex", "pdflatex", "pdflatex"] timeout : int Timeout per command in seconds Returns ------- CompileResult """ start_time = time.time() for i, cmd_name in enumerate(compile_sequence): print(f"\n Pass {i+1}/{len(compile_sequence)}: {cmd_name}") # Build command if cmd_name == "bibtex": cmd = ["bibtex", self.project_name] elif cmd_name == "biber": cmd = ["biber", self.project_name] elif cmd_name == "xelatex": cmd = ["xelatex", "-interaction=nonstopmode", f"\\input{{{self.project_name}.tex}}"] elif cmd_name == "pdflatex": cmd = ["pdflatex", "-interaction=nonstopmode", f"\\input{{{self.project_name}.tex}}"] else: print(f" ⚠ Unknown command: {cmd_name}") continue # Execute try: result = subprocess.run( cmd, cwd=self.working_dir, capture_output=True, text=True, timeout=timeout, ) # Check for errors if result.returncode != 0: # Extract error lines from log error_lines = self._extract_errors() print(f" 🔴 Error in {cmd_name} pass") for line in error_lines[:5]: print(f" {line}") # Try to continue anyway (some errors are non-fatal) # Count warnings log_content = self.log_file.read_text(errors="ignore") if self.log_file.exists() else "" warnings = self._count_warnings(log_content) if warnings > 0: print(f" ⚠ {warnings} warnings") except subprocess.TimeoutExpired: print(f" 🔴 Timeout ({timeout}s) for {cmd_name}") return CompileResult( success=False, log_content="", error_count=1, warning_count=0, reference_warnings=[], ) except Exception as e: print(f" 🔴 Exception: {e}") elapsed = time.time() - start_time # Final check pdf_file = self.working_dir / f"{self.project_name}.pdf" log_content = self.log_file.read_text(errors="ignore") if self.log_file.exists() else "" error_count = self._count_errors(log_content) warning_count = self._count_warnings(log_content) ref_warnings = self._extract_reference_warnings(log_content) page_count = self._extract_page_count(log_content) success = pdf_file.exists() and error_count == 0 return CompileResult( success=success, log_content=log_content[-5000:], # Last 5000 chars error_count=error_count, warning_count=warning_count, reference_warnings=ref_warnings, page_count=page_count, compilation_time=elapsed, ) def _extract_errors(self) -> List[str]: """Extract error lines from the log file.""" if not self.log_file.exists(): return ["Log file not found"] content = self.log_file.read_text(errors="ignore") errors = [] for line in content.split('\n'): if 'Error' in line or 'error' in line or '!' in line: errors.append(line.strip()) return errors[:10] def _count_errors(self, log_content: str) -> int: """Count compilation errors in log.""" error_lines = [ line for line in log_content.split('\n') if line.strip().startswith('! ') ] return len(error_lines) def _count_warnings(self, log_content: str) -> int: """Count compilation warnings in log.""" return log_content.count('LaTeX Warning') def _extract_reference_warnings(self, log_content: str) -> List[str]: """Extract citation/label reference warnings.""" warnings = [] for line in log_content.split('\n'): if 'Citation' in line or 'Reference' in line or 'Label' in line: if 'Warning' in line or '?' in line: warnings.append(line.strip()) return warnings[:10] def _extract_page_count(self, log_content: str) -> Optional[int]: """Extract page count from log.""" match = re.search(r'\[(\d+)\s+pages?\]', log_content) if match: return int(match.group(1)) # Try alternative pattern match = re.search(r'Total pages?:\s*(\d+)', log_content) if match: return int(match.group(1)) return None
Step 4: Run Full Compilation
python#!/usr/bin/env python3 """Main compilation script for fin-paper-convert.""" import os import sys import shutil import subprocess from pathlib import Path from typing import Optional # Add project scripts to path sys.path.insert(0, str(Path(__file__).parent.parent.parent)) from scripts.research_framework.latex_compiler import LatexCompiler, CompileResult from scripts.research_framework.report_generator import ReportGenerator from scripts.research_framework.journal_templates import get_template def compile_paper(project_dir: str, journal_name: str = "经济研究", project_name: str = "main") -> CompileResult: """ Compile LaTeX paper to PDF for the target journal. Parameters ---------- project_dir : str Directory containing main.tex and all source files journal_name : str Target journal name (经济研究 / JF / JFE / etc.) project_name : str Project file name (without .tex extension) Returns ------- CompileResult """ project_path = Path(project_dir) template = get_template(journal_name) print("=" * 60) print(f"Compiling: {journal_name}") print(f"Directory: {project_dir}") print(f"Template: {template['class']}") print(f"Compile sequence: {' → '.join(template['compile_sequence'])}") print("=" * 60) # Step 1: Validate from scripts.research_framework.latex_validator import validate_latex_project validation = validate_latex_project(project_dir) if not validation.ok: print("\n🔴 Validation FAILED:") for err in validation.errors: print(f" • {err}") return CompileResult( success=False, log_content="", error_count=len(validation.errors), warning_count=0, reference_warnings=[], ) if validation.warnings: print("\n⚠ Warnings:") for w in validation.warnings[:5]: print(f" • {w}") # Step 2: Clean auxiliary files print("\n[Clean] Removing auxiliary files...") for ext in ['.aux', '.bbl', '.blg', '.log', '.out', '.toc', '.lof', '.lot', '.fls', '.fdb_latexmk']: aux_file = project_path / f"{project_name}{ext}" if aux_file.exists(): aux_file.unlink() # Step 3: Compile compiler = LatexCompiler(project_dir, project_name) result = compiler.compile(template['compile_sequence']) # Step 4: Report print("\n" + "=" * 60) print("COMPILATION RESULT") print("=" * 60) print(f"Success: {'✅ Yes' if result.success else '🔴 No'}") print(f"Errors: {result.error_count}") print(f"Warnings: {result.warning_count}") print(f"Pages: {result.page_count or 'unknown'}") print(f"Time: {result.compilation_time:.1f}s" if result.compilation_time else "") if result.reference_warnings: print(f"\n⚠ Reference warnings ({len(result.reference_warnings)}):") for w in result.reference_warnings[:3]: print(f" {w}") if result.error_count > 0 and result.log_content: print("\n📋 Last log entries (errors):") for line in result.log_content.split('\n'): if line.strip().startswith('!'): print(f" {line.strip()}") return result
bash# Direct execution python -c " import sys sys.path.insert(0, 'scripts') from research_framework.latex_compiler import compile_paper result = compile_paper( project_dir='output/fin-manuscript/draft_v1', journal_name='经济研究', project_name='main' ) sys.exit(0 if result.success else 1) "
Step 5: Generate Submission Variants
After successful PDF compilation, generate submission variants:
python#!/usr/bin/env python3 """Generate submission variants: anonymous, arxiv, word.""" import os import shutil import subprocess from pathlib import Path from typing import Optional def generate_anonymous_pdf(source_dir: str, output_dir: str, remove_patterns: list = None) -> str: """ Generate anonymous version by removing author info and acknowledgments. Parameters ---------- source_dir : str Directory with main.pdf output_dir : str Output directory for anonymous version remove_patterns : list Regex patterns to remove (author names, acknowledgments, etc.) Returns ------- str Path to anonymous PDF """ if remove_patterns is None: remove_patterns = [ r'\\thanks\{[^}]*\}', r'\\author\{[^}]*\}', r'\\acker\{[^}]*\}', r'\\section\*\{[^}]*致谢[^}]*\}', r'\\section\*\{[^}]*Acknowledgment[^}]*\}', r'\\footnote\{[^}]*[^)]*基金[^)]*\}', ] source_path = Path(source_dir) output_path = Path(output_dir) output_path.mkdir(parents=True, exist_ok=True) # Copy PDF (for PDF-based anonymization) src_pdf = source_path / "main.pdf" dst_pdf = output_path / "anonymous.pdf" if src_pdf.exists(): shutil.copy2(src_pdf, dst_pdf) print(f" Copied: {dst_pdf}") # TODO: For true anonymization, use PDF redaction tools # e.g., pdf редактор or python-pdf-tools print(f" Note: Full PDF redaction requires specialized tools") else: print(f" ⚠ Source PDF not found: {src_pdf}") return str(dst_pdf) def generate_arxiv_pdf(source_dir: str, output_dir: str, remove_funding: bool = True, remove_author: bool = True) -> str: """ Generate arXiv-compatible version. arXiv-specific requirements: - No author acknowledgments - No funding disclosure - LaTeX source compiles with pdflatex - main.pdf included - No .tex files > 10MB """ source_path = Path(source_dir) output_path = Path(output_dir) / "arxiv_source" output_path.mkdir(parents=True, exist_ok=True) # Copy essential files files_to_copy = ["main.tex", "references.bib", "figures/"] for item in files_to_copy: src = source_path / item dst = output_path / item if src.exists(): if src.is_dir(): shutil.copytree(src, dst, dirs_exist_ok=True) else: shutil.copy2(src, dst) # Copy main.pdf as arxiv.pdf src_pdf = source_path / "main.pdf" if src_pdf.exists(): shutil.copy2(src_pdf, output_path / "arxiv.pdf") # Create submission README readme = """arXiv Submission Package ======================== Files: - main.tex: Main manuscript - references.bib: Bibliography - figures/: Figure files - arxiv.pdf: Compiled PDF Submission Instructions: 1. Upload this entire directory to arXiv 2. Verify compilation: pdflatex main.tex && bibtex main && pdflatex main.tex && pdflatex main.tex 3. Check that all figures are embedded """ (output_path / "README.txt").write_text(readme) print(f" arXiv package: {output_path}") return str(output_path) def generate_word_docx(source_dir: str, output_dir: str, use_pandoc: bool = True) -> str: """ Generate Word (.docx) version for journals requiring Word submission. Requires: pandoc (install via: brew install pandoc) """ source_path = Path(source_dir) output_path = Path(output_dir) output_path.mkdir(parents=True, exist_ok=True) tex_file = source_path / "main.tex" docx_file = output_path / "main.docx" if not use_pandoc: print(" ⚠ pandoc not available, creating placeholder") docx_file.write_text("") return str(docx_file) try: # Check pandoc availability subprocess.run(["pandoc", "--version"], capture_output=True, check=True) # Convert cmd = [ "pandoc", str(tex_file), "-o", str(docx_file), "--reference-doc", "templates/journal_ref.docx", # Optional "-f", "latex", "-t", "docx", "--wrap=none", ] result = subprocess.run(cmd, capture_output=True, text=True) if result.returncode == 0 and docx_file.exists(): print(f" Converted: {docx_file}") else: print(f" ⚠ pandoc conversion failed: {result.stderr[:200]}") # Fallback: try without reference doc cmd[-2] = "" # Remove reference-doc flag result = subprocess.run(cmd, capture_output=True, text=True) if result.returncode == 0: print(f" Converted (basic): {docx_file}") except (subprocess.CalledProcessError, FileNotFoundError): print(" ⚠ pandoc not installed. Install via: brew install pandoc") print(" Creating placeholder .docx file") docx_file.write_text("Placeholder - install pandoc and re-run") return str(docx_file) def create_submission_package(project_dir: str, journal_name: str, output_dir: str) -> str: """ Create a complete submission package with all required files. """ import zipfile project_path = Path(project_dir) output_path = Path(output_dir) pkg_path = output_path / f"submission_package_{journal_name}.zip" print("\n[Package] Creating submission package...") with zipfile.ZipFile(pkg_path, 'w', zipfile.ZIP_DEFLATED) as zf: # Add main PDF main_pdf = project_path / "main.pdf" if main_pdf.exists(): zf.write(main_pdf, "main.pdf") # Add anonymous PDF anon_pdf = project_path / "anonymous.pdf" if anon_pdf.exists(): zf.write(anon_pdf, "anonymous.pdf") # Add main.tex source main_tex = project_path / "main.tex" if main_tex.exists(): zf.write(main_tex, "main.tex") # Add references bib_file = project_path / "references.bib" if bib_file.exists(): zf.write(bib_file, "references.bib") # Add figures directory fig_dir = project_path / "figures" if fig_dir.exists(): for fig in fig_dir.glob("*"): if fig.is_file(): zf.write(fig, f"figures/{fig.name}") print(f" Package: {pkg_path}") return str(pkg_path)
Step 6: PDF Validation
pythondef validate_pdf(pdf_path: str) -> dict: """ Validate compiled PDF for submission. """ import re result = { "exists": False, "file_size_kb": 0, "pages": 0, "font_embedding": None, "issues": [], } pdf_file = Path(pdf_path) if not pdf_file.exists(): result["issues"].append("PDF file not found") return result result["exists"] = True result["file_size_kb"] = pdf_file.stat().st_size / 1024 # Read PDF header to check validity with open(pdf_file, 'rb') as f: header = f.read(8) if not header.startswith(b'%PDF'): result["issues"].append("Invalid PDF header") # Extract page count with open(pdf_file, 'rb') as f: content = f.read().decode('latin-1', errors='ignore') page_match = re.search(r'/Type\s*/Page[^s]', content) page_count = len(re.findall(r'/Type\s*/Page[^s]', content)) result["pages"] = page_count # Check for embedded fonts (basic check) if '/Font' in content or '/Type1' in content or '/TrueType' in content: result["font_embedding"] = "embedded" else: result["font_embedding"] = "unknown" # File size check if result["file_size_kb"] < 10: result["issues"].append(f"PDF very small ({result['file_size_kb']:.0f}KB) - may be incomplete") elif result["file_size_kb"] > 50000: result["issues"].append(f"PDF very large ({result['file_size_kb']:.0f}KB) - may need compression") return result
Step 7: Full Pipeline Execution
python#!/usr/bin/env python3 """Complete fin-paper-convert pipeline.""" import sys import os from pathlib import Path # Add project scripts sys.path.insert(0, str(Path(__file__).parent.parent.parent)) from scripts.research_framework.latex_compiler import compile_paper, validate_pdf from scripts.research_framework.latex_compiler import generate_anonymous_pdf, generate_arxiv_pdf, generate_word_docx, create_submission_package from scripts.research_framework.latex_validator import validate_latex_project def main(): project_dir = "output/fin-manuscript/draft_v1" journal_name = "经济研究" # or "JF", "JFE", etc. print("=" * 60) print("fin-paper-convert: LaTeX Compilation & Format Conversion") print("=" * 60) # Step 1: Validate project print("\n[Step 1/4] Validating project structure...") validation = validate_latex_project(project_dir) if not validation.ok: print("🔴 Validation failed:") for e in validation.errors: print(f" {e}") return # Step 2: Compile to PDF print("\n[Step 2/4] Compiling LaTeX to PDF...") result = compile_paper(project_dir, journal_name, "main") if not result.success: print("\n🔴 Compilation failed!") print(f"Errors: {result.error_count}") if result.log_content: print("\nLast 2000 chars of log:") print(result.log_content[-2000:]) return # Validate PDF print("\n[Step 3/4] Validating PDF...") pdf_path = f"{project_dir}/main.pdf" pdf_info = validate_pdf(pdf_path) print(f" File size: {pdf_info['file_size_kb']:.0f} KB") print(f" Pages: {pdf_info['pages']}") print(f" Font embedding: {pdf_info['font_embedding']}") if pdf_info['issues']: for issue in pdf_info['issues']: print(f" ⚠ {issue}") # Step 4: Generate variants print("\n[Step 4/4] Generating submission variants...") # Anonymous version try: generate_anonymous_pdf(project_dir, project_dir) except Exception as e: print(f" ⚠ Anonymous PDF failed: {e}") # arXiv version (English journals only) if journal_name in ["JF", "JFE", "RFS", "AER", "JAE"]: try: generate_arxiv_pdf(project_dir, project_dir) except Exception as e: print(f" ⚠ arXiv package failed: {e}") # Word version try: generate_word_docx(project_dir, project_dir) except Exception as e: print(f" ⚠ Word conversion failed: {e}") # Submission package try: create_submission_package(project_dir, journal_name, project_dir) except Exception as e: print(f" ⚠ Package creation failed: {e}") print("\n" + "=" * 60) print("✅ COMPILATION COMPLETE") print("=" * 60) print(f"\nOutput files in {project_dir}/:") print(f" 📄 main.pdf — Main manuscript") print(f" 📄 anonymous.pdf — Anonymous version") if journal_name in ["JF", "JFE", "RFS", "AER", "JAE"]: print(f" 📁 arxiv_source/ — arXiv submission") print(f" 📄 main.docx — Word version") print(f" 📦 submission_package_{journal_name}.zip") print("=" * 60) if __name__ == "__main__": main()
Step 8: Command-Line Usage
bash# Direct command-line execution python scripts/research_framework/latex_compiler.py \ --dir output/fin-manuscript/draft_v1 \ --journal 经济研究 \ --project main # Check compilation errors python scripts/research_framework/latex_compiler.py \ --dir output/fin-manuscript/draft_v1 \ --check-log # Compile with latexmk (if available) latexmk -pdf -interaction=nonstopmode main.tex
Step 9: Error Handling Reference
| Error | Cause | Fix |
|---|---|---|
! LaTeX Error: File 'xxx.sty' not found | Missing package | Install: tlmgr install xxx or check \usepackage{} |
! Undefined control sequence. | Typo or undefined command | Check .log for line number |
[?] Reference | Citation/label not found | Run pdflatex again (3 passes) |
Font ... not loadable | Font not installed | Install font or use alternative |
PDF inclusion: impossible | Figure format issue | Convert to PDF: pdftk fig.png output fig.pdf |
Float too large | Figure/table exceeds page | Reduce figure size or use \FloatBarrier |
Overfull \hbox | Line overflow | Adjust \hbox_to or add \small |
Underfull \hbox | Underfilled line | Usually OK, can ignore |
Package natbib Error: Bibliography not compatible with author-year | Bib style mismatch | Use matching bib style |
Output Summary
After compilation:
✅ LATEX COMPILATION COMPLETE Journal: 经济研究 Version: draft_v1 Compilation: 4 passes (xelatex → bibtex → xelatex → xelatex) Time: 12.3s Pages: 28 Errors: 0 Warnings: 15 Output Files: output/fin-manuscript/draft_v1/ ├── main.pdf — 28 pages, 2.1 MB ├── main.tex — LaTeX source ├── references.bib — Bibliography ├── anonymous.pdf — Anonymized version ├── main.docx — Word format └── submission_package_经济研究.zip Next Step: → Run: fin-submit-check Verify format compliance before submission
