import gradio as gr from pathlib import Path import datetime import re import os import shutil import fitz # PyMuPDF from PIL import Image from collections import defaultdict import io from pypdf import PdfWriter from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle, PageBreak, BaseDocTemplate, Frame, PageTemplate, Image as ReportLabImage 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 --- CWD = Path.cwd() 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 = CWD / "generated_pdfs" PREVIEW_DIR = CWD / "previews" FONT_DIR = CWD # Create necessary directories OUTPUT_DIR.mkdir(exist_ok=True) PREVIEW_DIR.mkdir(exist_ok=True) # --- Font & Emoji Handling --- def register_local_fonts(): """Finds and registers all .ttf files from the application's base directory.""" print("--- Font Registration Process Starting ---") text_font_names = [] emoji_font_name = None print(f"Scanning for fonts in: {FONT_DIR.absolute()}") font_files = list(FONT_DIR.glob("*.ttf")) print(f"Found {len(font_files)} .ttf files: {[f.name for f in font_files]}") for font_path in font_files: try: font_name = font_path.stem pdfmetrics.registerFont(TTFont(font_name, str(font_path))) pdfmetrics.registerFont(TTFont(f"{font_name}-Bold", str(font_path))) pdfmetrics.registerFont(TTFont(f"{font_name}-Italic", str(font_path))) pdfmetrics.registerFont(TTFont(f"{font_name}-BoldItalic", str(font_path))) pdfmetrics.registerFontFamily(font_name, normal=font_name, bold=f"{font_name}-Bold", italic=f"{font_name}-Italic", boldItalic=f"{font_name}-BoldItalic") if "notocoloremoji-regular" in font_name.lower(): emoji_font_name = font_name elif "notoemoji" not in font_name.lower(): # Exclude other symbol fonts from text selection text_font_names.append(font_name) except Exception as e: print(f"Could not register font {font_path.name}: {e}") if not text_font_names: print("WARNING: No text fonts found. Adding 'Helvetica' as a default.") text_font_names.append('Helvetica') print(f"Successfully registered user-selectable fonts: {text_font_names}") print(f"Emoji font set to: {emoji_font_name}") print("--- Font Registration Process Finished ---") return sorted(text_font_names), emoji_font_name def apply_emoji_font(text: str, emoji_font_name: str) -> str: """ Intelligently wraps emoji characters in a tag, preserving existing HTML-like tags. This prevents invalid nested tags like ... which ReportLab handles poorly. """ if not emoji_font_name: return text # Regex to find emojis emoji_pattern = re.compile(f"([{re.escape(''.join(map(chr, range(0x1f600, 0x1f650))))}" f"{re.escape(''.join(map(chr, range(0x1f300, 0x1f5ff))))}" f"{re.escape(''.join(map(chr, range(0x1f900, 0x1f9ff))))}" f"{re.escape(''.join(map(chr, range(0x2600, 0x26ff))))}" f"{re.escape(''.join(map(chr, range(0x2700, 0x27bf))))}]+)") # Regex to split the string by existing tags (, ) tag_pattern = re.compile(r"(<[^>]+>)") parts = tag_pattern.split(text) result = [] for part in parts: if tag_pattern.match(part): # It's a tag, append it as is result.append(part) else: # It's text, apply emoji font to any emojis within this segment result.append(emoji_pattern.sub(fr'\1', part)) return "".join(result) # --- PDF Generation & Handling --- def markdown_to_story(markdown_text: str, font_name: str, emoji_font: str): """ Converts markdown to a ReportLab story, with enhanced styling and page breaks. This version correctly separates structural parsing from content formatting. """ styles = getSampleStyleSheet() # Define styles for various markdown elements style_normal = ParagraphStyle('BodyText', fontName=font_name, spaceAfter=6, leading=14, fontSize=10) style_h1 = ParagraphStyle('h1', parent=styles['h1'], fontName=font_name, spaceBefore=12, fontSize=24, leading=28, textColor=colors.darkblue) style_h2 = ParagraphStyle('h2', parent=styles['h2'], fontName=font_name, fontSize=18, leading=22, spaceBefore=10) style_h3 = ParagraphStyle('h3', parent=styles['h3'], fontName=font_name, fontSize=14, leading=18, spaceBefore=8) style_code = ParagraphStyle('Code', fontName='Courier', backColor=colors.whitesmoke, textColor=colors.darkred, borderWidth=1, borderColor=colors.lightgrey, padding=8, leading=12, fontSize=9) style_table_header = ParagraphStyle('TableHeader', parent=style_normal, fontName=font_name + "-Bold" if font_name != 'Helvetica' else 'Helvetica-Bold') story = [] lines = markdown_text.split('\n') in_code_block, in_table = False, False code_block_text, table_data = "", [] first_heading = True for line in lines: stripped_line = line.strip() if stripped_line.startswith("```"): if in_code_block: story.append(Paragraph(code_block_text.replace('\n', '
'), style_code)); story.append(Spacer(1, 0.1 * inch)) in_code_block = False; code_block_text = "" else: in_code_block = True continue if in_code_block: code_block_text += line.replace('&', '&').replace('<', '<').replace('>', '>') + '\n' continue if stripped_line.startswith('|'): if not in_table: in_table = True if all(c in '-|: ' for c in stripped_line): continue cells = [cell.strip() for cell in stripped_line.strip('|').split('|')] table_data.append(cells) continue if in_table: in_table = False if table_data: header_content = [apply_emoji_font(re.sub(r'\*\*(.*?)\*\*', r'\1', cell), emoji_font) for cell in table_data[0]] header = [Paragraph(cell, style_table_header) for cell in header_content] formatted_rows = [] for row in table_data[1:]: formatted_cells = [apply_emoji_font(re.sub(r'\*\*(.*?)\*\*', r'\1', cell), emoji_font) for cell in row] formatted_rows.append([Paragraph(cell, style_normal) for cell in formatted_cells]) table = Table([header] + formatted_rows, hAlign='LEFT', repeatRows=1) table.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, 0), colors.lightgrey), ('GRID', (0, 0), (-1, -1), 1, colors.darkgrey), ('VALIGN', (0,0), (-1,-1), 'MIDDLE')])) story.append(table); story.append(Spacer(1, 0.2 * inch)) table_data = [] if not stripped_line: story.append(Spacer(1, 0.1 * inch)) continue # Default content is the whole stripped line content = stripped_line style = style_normal extra_args = {} # Detect structural elements and extract the raw content if stripped_line.startswith("# "): if not first_heading: story.append(PageBreak()) content = stripped_line.lstrip('# '); style = style_h1; first_heading = False elif stripped_line.startswith("## "): content = stripped_line.lstrip('## '); style = style_h2 elif stripped_line.startswith("### "): content = stripped_line.lstrip('### '); style = style_h3 elif stripped_line.startswith(("- ", "* ")): content = stripped_line[2:]; extra_args['bulletText'] = 'β€’' # Now, format the extracted content # Apply markdown formatting for bold/italic formatted_content = re.sub(r'_(.*?)_', r'\1', re.sub(r'\*\*(.*?)\*\*', r'\1', content)) # Then apply emoji font to the already formatted line final_content = apply_emoji_font(formatted_content, emoji_font) story.append(Paragraph(final_content, style, **extra_args)) return story def create_pdf_preview(pdf_path: Path): preview_path = PREVIEW_DIR / f"{pdf_path.stem}.png" try: doc = fitz.open(pdf_path); page = doc.load_page(0); pix = page.get_pixmap() pix.save(str(preview_path)); doc.close() return str(preview_path) except Exception as e: print(f"Could not create preview for {pdf_path.name}: {e}"); return None # --- Main API Function --- def generate_pdfs_api(files, layouts, fonts, num_columns, page_w_mult, page_h_mult, progress=gr.Progress(track_tqdm=True)): if not files: raise gr.Error("Please upload at least one Markdown or Image 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.") shutil.rmtree(OUTPUT_DIR, ignore_errors=True); shutil.rmtree(PREVIEW_DIR, ignore_errors=True) OUTPUT_DIR.mkdir(); PREVIEW_DIR.mkdir() grouped_files = defaultdict(lambda: {'md': [], 'img': []}) for f in files: file_path = Path(f.name) stem = file_path.stem.split('_')[0] if '_' in file_path.stem else file_path.stem if file_path.suffix.lower() == '.md': grouped_files[stem]['md'].append(file_path) elif file_path.suffix.lower() in ['.png', '.jpg', '.jpeg']: grouped_files[stem]['img'].append(file_path) log_updates, generated_pdf_paths = "", [] for stem, assets in progress.tqdm(grouped_files.items(), desc="Processing File Groups"): for layout_name in layouts: for font_name in fonts: merger = PdfWriter() if assets['md']: md_content = "\n".join([p.read_text(encoding='utf-8') for p in assets['md']]) md_buffer = io.BytesIO() story = markdown_to_story(md_content, font_name, EMOJI_FONT_NAME) base_w, base_h = LAYOUTS[layout_name]["size"] pagesize = (base_w * page_w_mult, base_h * page_h_mult) if num_columns > 1: doc = BaseDocTemplate(md_buffer, pagesize=pagesize, leftMargin=0.5*inch, rightMargin=0.5*inch, topMargin=0.5*inch, bottomMargin=0.5*inch) frame_width = (doc.width / num_columns) - (num_columns - 1) * 0.1*inch frames = [Frame(doc.leftMargin + i * (frame_width + 0.2*inch), doc.bottomMargin, frame_width, doc.height) for i in range(num_columns)] doc.addPageTemplates([PageTemplate(id='MultiCol', frames=frames)]) else: doc = SimpleDocTemplate(md_buffer, pagesize=pagesize) doc.build(story) merger.append(fileobj=md_buffer) for img_path in assets['img']: with Image.open(img_path) as img: img_width, img_height = img.size img_buffer = io.BytesIO() doc = SimpleDocTemplate(img_buffer, pagesize=(img_width, img_height), leftMargin=0, rightMargin=0, topMargin=0, bottomMargin=0) doc.build([ReportLabImage(img_path, width=img_width, height=img_height)]) merger.append(fileobj=img_buffer) if len(merger.pages) > 0: time_str = datetime.datetime.now().strftime('%m-%d-%a_%I%M%p').upper() filename = f"{stem}_{time_str}_{layout_name.replace(' ','-')}_{page_w_mult}x{page_h_mult}_{font_name}_Cols{num_columns}.pdf" output_path = OUTPUT_DIR / filename with open(output_path, "wb") as f: merger.write(f) generated_pdf_paths.append(output_path) log_updates += f"Generated: {filename}\n" gallery_previews = [create_pdf_preview(p) for p in generated_pdf_paths] final_gallery = [g for g in gallery_previews if g is not None] return final_gallery, log_updates, [str(p) for p in generated_pdf_paths] # --- Gradio UI Definition --- AVAILABLE_FONTS, EMOJI_FONT_NAME = register_local_fonts() SAMPLE_MARKDOWN = """# Deities Guide: Mythology and Moral Lessons 1. πŸ“œ **Introduction** - **Purpose**: Explore deities, spirits, saints, and beings with their epic stories and morals! - **Usage**: A guide for learning and storytelling across traditions. ️ - **Themes**: Justice βš–οΈ, faith πŸ™, hubris πŸ›οΈ, redemption ✨, cosmic order 🌌. 2. πŸ› οΈ **Core Concepts of Divinity** - **Powers**: Creation 🌍, omniscience πŸ‘οΈβ€πŸ—¨οΈ, shapeshifting πŸ¦‹ across entities. - **Life Cycle**: Mortality ⏳, immortality ♾️, transitions like saints and avatars 🌟. - **Communication**: Omens πŸ•ŠοΈ, visions πŸ‘οΈ, miracles ✨ from gods and spirits. # βš”οΈ Arthurian Legends - **Merlin, Morgan le Fay, Arthur**: Mentor πŸ§™, rival πŸ§™β€β™€οΈ, son πŸ‘‘. - **Relation**: Family tests loyalty 🀝. - **Lesson**: Honor πŸŽ–οΈ vs. betrayal πŸ—‘οΈ. # πŸ›οΈ Greek Mythology - **Zeus, Hera, Athena**: Father ⚑, mother πŸ‘‘, daughter πŸ¦‰. - **Relation**: Family rules with tension 🌩️. - **Lesson**: Hubris ΰ€…ΰ€Ήΰ€‚ΰ€•ΰ€Ύΰ€° meets wisdom 🧠. # πŸ•‰οΈ Hindu Trimurti - **Brahma, Vishnu, Shiva**: Creator Brahma, preserver Vishnu, destroyer Shiva. - **Relation**: Divine trio cycles existence πŸ”„. - **Lesson**: Balance βš–οΈ sustains life πŸ’–. """ with open(CWD / "sample.md", "w", encoding="utf-8") as f: f.write(SAMPLE_MARKDOWN) with gr.Blocks(theme=gr.themes.Soft(), title="Advanced PDF Generator") as demo: gr.Markdown("# πŸ“„ Advanced PDF Layout Engine") gr.Markdown("Upload Markdown/Image files. The app finds local `.ttf` fonts. Group assets with a common name (e.g., `Doc_part1.md`, `Doc_img1.png`) to combine them. `# Headers` create automatic page breaks.") with gr.Row(): with gr.Column(scale=1): gr.Markdown("### βš™οΈ Generation Settings") uploaded_files = gr.File(label="Upload Markdown & Image Files", file_count="multiple", file_types=[".md", ".png", ".jpg", ".jpeg"]) with gr.Row(): page_w_mult_slider = gr.Slider(label="Page Width Multiplier", minimum=1, maximum=5, step=1, value=1) page_h_mult_slider = gr.Slider(label="Page Height Multiplier", minimum=1, maximum=2, step=1, value=1) num_columns_slider = gr.Slider(label="Number of Text Columns", minimum=1, maximum=4, step=1, value=1) selected_layouts = gr.CheckboxGroup(choices=list(LAYOUTS.keys()), label="Select Base Page Layout", value=["A4 Portrait"]) selected_fonts = gr.CheckboxGroup(choices=AVAILABLE_FONTS, label="Select Text Font", value=[AVAILABLE_FONTS[0]] if AVAILABLE_FONTS else []) generate_btn = gr.Button("πŸš€ Generate PDFs", variant="primary") with gr.Column(scale=2): gr.Markdown("### πŸ–ΌοΈ PDF Preview Gallery") gallery_output = gr.Gallery(label="Generated PDF Previews", show_label=False, elem_id="gallery", columns=3, height="auto", object_fit="contain") log_output = gr.Markdown(label="Generation Log", value="Logs will appear here...") downloadable_files_output = gr.Files(label="Download Generated PDFs") generate_btn.click(fn=generate_pdfs_api, inputs=[uploaded_files, selected_layouts, selected_fonts, num_columns_slider, page_w_mult_slider, page_h_mult_slider], outputs=[gallery_output, log_output, downloadable_files_output]) if __name__ == "__main__": demo.launch()