import gradio as gr from pathlib import Path import datetime import re import requests import os import shutil from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle from reportlab.lib.pagesizes import letter, A4, legal, landscape from reportlab.lib.units import inch from reportlab.lib import colors from reportlab.pdfbase import pdfmetrics from reportlab.pdfbase.ttfonts import TTFont # --- Configuration & Setup --- # These settings define the available layouts and directories for fonts and outputs. LAYOUTS = { "A4 Portrait": {"size": A4}, "A4 Landscape": {"size": landscape(A4)}, "Letter Portrait": {"size": letter}, "Letter Landscape": {"size": landscape(letter)}, "Legal Portrait": {"size": legal}, "Legal Landscape": {"size": landscape(legal)}, } OUTPUT_DIR = Path("generated_pdfs") OUTPUT_DIR.mkdir(exist_ok=True) FONT_DIR = Path("fonts") FONT_DIR.mkdir(exist_ok=True) # --- Font Handling --- def download_default_font(): """Downloads DejaVuSans as a default font if no fonts are available in the 'fonts' directory.""" dejavu_path = FONT_DIR / "DejaVuSans.ttf" if not dejavu_path.exists(): print("No fonts found. Downloading default font (DejaVuSans)...") try: url = "https://github.com/dejavu-fonts/dejavu-fonts/blob/main/ttf/DejaVuSans.ttf?raw=true" r = requests.get(url, allow_redirects=True) r.raise_for_status() with open(dejavu_path, "wb") as f: f.write(r.content) print("Default font downloaded successfully.") return True except Exception as e: print(f"Failed to download default font: {e}") return False return True def discover_and_register_fonts(): """Finds .ttf files in the font directory, registers them with ReportLab, and returns their names.""" font_files = list(FONT_DIR.glob("*.ttf")) if not font_files: if not download_default_font(): return [] # Return empty if download fails font_files = list(FONT_DIR.glob("*.ttf")) font_names = [] for font_path in font_files: try: font_name = font_path.stem pdfmetrics.registerFont(TTFont(font_name, str(font_path))) font_names.append(font_name) except Exception as e: print(f"Could not register font {font_path.name}: {e}") return sorted(font_names) # --- ReportLab PDF Generation (Core Logic) --- def markdown_to_story(markdown_text: str, font_name: str): """Converts a markdown string into a list of ReportLab Flowables ('story') with enhanced styling.""" styles = getSampleStyleSheet() style_normal = ParagraphStyle('BodyText', parent=styles['BodyText'], fontName=font_name, spaceAfter=6) style_h1 = ParagraphStyle('h1', parent=styles['h1'], fontName=font_name) style_h2 = ParagraphStyle('h2', parent=styles['h2'], fontName=font_name) style_h3 = ParagraphStyle('h3', parent=styles['h3'], fontName=font_name) style_code = ParagraphStyle('Code', parent=styles['Code'], fontName=font_name, backColor=colors.whitesmoke, borderColor=colors.lightgrey, borderWidth=1, padding=(5, 5)) story = [] lines = markdown_text.split('\n') in_code_block, in_table = False, False code_block_text, table_data = "", [] for line in lines: if line.strip().startswith('|') and line.strip().endswith('|'): # Table handler if not in_table: in_table = True table_data = [] if all(c in '-|: ' for c in line.strip()): continue cells = [cell.strip() for cell in line.strip().strip('|').split('|')] table_data.append([Paragraph(cell, style_normal) for cell in cells]) continue elif in_table: in_table = False if table_data: table = Table(table_data, hAlign='LEFT', colWidths=[1.5*inch]*len(table_data[0])) table.setStyle(TableStyle([('BACKGROUND', (0,0), (-1,0), colors.lightgrey), ('GRID', (0,0), (-1,-1), 1, colors.black)])) story.append(table) story.append(Spacer(1, 0.2 * inch)) table_data = [] if line.strip().startswith("```"): # Code block handler in_code_block = not in_code_block if not in_code_block: story.append(Paragraph(code_block_text.replace('\n', '
'), style_code)) code_block_text = "" continue if in_code_block: code_block_text += line.replace('&', '&').replace('<', '<').replace('>', '>') + '\n' continue # Markdown elements to ReportLab Flowables if line.startswith("# "): story.append(Paragraph(line[2:], style_h1)) elif line.startswith("## "): story.append(Paragraph(line[3:], style_h2)) elif line.startswith("### "): story.append(Paragraph(line[4:], style_h3)) elif line.strip().startswith(("* ", "- ")): story.append(Paragraph(line.strip()[2:], style_normal, bulletText='•')) elif re.match(r'^\d+\.\s', line.strip()): story.append(Paragraph(line.strip(), style_normal)) elif line.strip() == "": story.append(Spacer(1, 0.1 * inch)) else: story.append(Paragraph(re.sub(r'_(.*?)_', r'\1', re.sub(r'\*\*(.*?)\*\*', r'\1', line)), style_normal)) return story # --- Gradio API Function --- def generate_pdfs_api(files, layouts, fonts, progress=gr.Progress(track_tqdm=True)): """Main function to drive PDF generation from the Gradio UI.""" if not files: raise gr.Error("Please upload at least one Markdown file.") if not layouts: raise gr.Error("Please select at least one page layout.") if not fonts: raise gr.Error("Please select at least one font.") # Clean output directory before generation if OUTPUT_DIR.exists(): for f in OUTPUT_DIR.glob('*'): if f.is_file(): os.remove(f) log_updates = "Starting PDF generation...\n" total_pdfs = len(files) * len(layouts) * len(fonts) for md_file_obj in progress.tqdm(files, desc="Processing Files"): original_md_path = Path(md_file_obj.name) log_updates += f"\nProcessing: **{original_md_path.name}**\n" md_content = "" with open(original_md_path, 'r', encoding='utf-8') as f: md_content = f.read() for layout_name in layouts: for font_name in fonts: status_text = f" - Generating: {layout_name}, {font_name}" log_updates += status_text + "\n" try: date_str = datetime.datetime.now().strftime("%Y-%m-%d") output_filename = f"{original_md_path.stem}_{layout_name.replace(' ', '-')}_{font_name}_{date_str}.pdf" output_path = OUTPUT_DIR / output_filename doc = SimpleDocTemplate(str(output_path), pagesize=LAYOUTS[layout_name]["size"], rightMargin=inch, leftMargin=inch, topMargin=inch, bottomMargin=inch) story = markdown_to_story(md_content, font_name) doc.build(story) except Exception as e: log_updates += f" - **ERROR**: Failed to process with font {font_name}: {e}\n" log_updates += "\nāœ… PDF generation complete!" generated_files = [str(f) for f in OUTPUT_DIR.glob("*.pdf")] return generated_files, log_updates # --- Gradio UI Definition --- AVAILABLE_FONTS = discover_and_register_fonts() SAMPLE_MARKDOWN = "# Sample Document\n\nThis is a sample markdown file. **ReportLab** is now creating the PDF with dynamic fonts and layouts.\n\n### Features\n- Item 1\n- Item 2\n\n| Header 1 | Header 2 |\n|----------|----------|\n| Cell 1 | Cell 2 |\n\n```python\ndef hello():\n print(\"Hello, PDF!\")\n```\n" with gr.Blocks(theme=gr.themes.Soft(), title="Advanced PDF Generator") as demo: gr.Markdown("# šŸ“„ Advanced Markdown to PDF Generator") gr.Markdown("Upload Markdown files, select layouts and fonts, and generate multiple PDF variations.") with gr.Row(): with gr.Column(scale=1): gr.Markdown("### āš™ļø Generation Settings") gr.Textbox(value=SAMPLE_MARKDOWN, label="Sample Markdown (for reference)", lines=10) uploaded_files = gr.File(label="Upload Markdown Files (.md)", file_count="multiple", file_types=[".md"]) selected_layouts = gr.CheckboxGroup(choices=list(LAYOUTS.keys()), label="Select Page Layouts", value=list(LAYOUTS.keys())[0]) if not AVAILABLE_FONTS: gr.Warning("No fonts found in 'fonts' directory. Using default.") AVAILABLE_FONTS = ["Helvetica"] # ReportLab's default selected_fonts = gr.CheckboxGroup(choices=AVAILABLE_FONTS, label="Select Fonts to Use", value=AVAILABLE_FONTS[0] if AVAILABLE_FONTS else None) generate_btn = gr.Button("šŸš€ Generate PDFs", variant="primary") with gr.Column(scale=2): gr.Markdown("### šŸ“„ Results") log_output = gr.Markdown(label="Generation Log", value="Logs will appear here...") file_output = gr.Files(label="Download Generated PDFs") generate_btn.click(fn=generate_pdfs_api, inputs=[uploaded_files, selected_layouts, selected_fonts], outputs=[file_output, log_output]) if __name__ == "__main__": demo.launch()