import gradio as gr from pathlib import Path import datetime import re import requests 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 --- 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") PREVIEW_DIR = Path("previews") FONT_DIR = Path("fonts") EMOJI_FONT_NAME = "NotoColorEmoji" # Create necessary directories OUTPUT_DIR.mkdir(exist_ok=True) PREVIEW_DIR.mkdir(exist_ok=True) FONT_DIR.mkdir(exist_ok=True) # --- Font & Emoji Handling --- def download_and_register_fonts(): """Downloads default fonts if needed, then finds and registers all .ttf files.""" fonts_to_download = { "DejaVuSans.ttf": "https://github.com/dejavu-fonts/dejavu-fonts/blob/main/ttf/DejaVuSans.ttf?raw=true", "NotoColorEmoji.ttf": "https://github.com/googlefonts/noto-emoji/blob/main/fonts/NotoColorEmoji.ttf?raw=true" } for filename, url in fonts_to_download.items(): font_path = FONT_DIR / filename if not font_path.exists(): print(f"Downloading {filename}...") try: r = requests.get(url, allow_redirects=True, timeout=20) r.raise_for_status() with open(font_path, "wb") as f: f.write(r.content) print(f"{filename} downloaded.") except Exception as e: print(f"Failed to download {filename}: {e}") font_names = [] for font_path in FONT_DIR.glob("*.ttf"): try: font_name = font_path.stem pdfmetrics.registerFont(TTFont(font_name, str(font_path))) # CRITICAL FIX for bold/italic error: # Register dummy bold/italic variants for all fonts, including emoji fonts. # ReportLab will use these fallbacks instead of crashing if a true variant is not found. 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 "emoji" not in font_name.lower(): font_names.append(font_name) except Exception as e: print(f"Could not register font {font_path.name}: {e}") return sorted(font_names) def apply_emoji_font(text: str) -> str: """Wraps emoji characters in a tag to use the dedicated emoji font.""" 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))))}]+)") return emoji_pattern.sub(fr'\1', text) # --- PDF Generation & Handling --- def markdown_to_story(markdown_text: str, font_name: str): """Converts markdown to a ReportLab story, handling emojis and page breaks.""" # This function remains largely the same but relies on the global font registration fix. styles = getSampleStyleSheet() style_normal = ParagraphStyle('BodyText', fontName=font_name, spaceAfter=6, leading=14) style_h1 = ParagraphStyle('h1', parent=styles['h1'], fontName=font_name, spaceBefore=12, fontSize=20, leading=24) style_code = ParagraphStyle('Code', fontName='Courier', backColor=colors.whitesmoke) story = [] # Process content line-by-line for line in markdown_text.split('\n'): line_with_emoji = apply_emoji_font(line) formatted_line = re.sub(r'_(.*?)_', r'\1', re.sub(r'\*\*(.*?)\*\*', r'\1', line_with_emoji)) if line.startswith("# "): story.append(Paragraph(formatted_line[2:], style_h1)) # ... other markdown elements would be handled similarly else: story.append(Paragraph(formatted_line, style_normal)) return story def create_pdf_preview(pdf_path: Path): """Creates a PNG preview of the first page of a PDF.""" 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 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, 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 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() # Group files by a common stem (e.g., AARON_1.md and AARON_2.png are grouped under "AARON") 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() # --- Process Markdown --- if assets['md']: md_content = "\n\n---PAGE_BREAK---\n\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) pagesize = LAYOUTS[layout_name]["size"] 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) # --- Process Images --- for img_path in assets['img']: with Image.open(img_path) as img: img_width, img_height = img.size img_buffer = io.BytesIO() # Page size is determined by the image itself 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) # --- Save Merged PDF --- 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(' ','-')}_{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" # --- Create Previews for Gallery --- 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 # --- Gradio UI Definition --- AVAILABLE_FONTS = download_and_register_fonts() SAMPLE_MARKDOWN = "# Sample Document 🚀\n\nThis shows **bold**, _italic_, and emojis like 😊 and 💻.\n\n### A Table\n| Flavor | Rating |\n|---|---|\n| Chocolate 🍫| 10/10 |\n| Vanilla 🍦| 9/10 |\n\n```python\n# Code blocks too!\ndef hello():\n return 'Hello from PDF!'\n```" with open("sample.md", "w") as f: f.write(SAMPLE_MARKDOWN) with gr.Blocks(theme=gr.themes.Soft(), title="Advanced PDF Generator") as demo: gr.Markdown("# 📄 Advanced PDF Generator") gr.Markdown("Upload Markdown and Image files. Group assets with a common name (e.g., `Doc_part1.md`, `Doc_img1.png`). The app will generate PDFs for all selected variations and display them below.") with gr.Row(): with gr.Column(scale=1): gr.Markdown("### ⚙️ Generation Settings") uploaded_files = gr.File(label="Upload Markdown & Image Files", file_count="multiple", file_types=[".md", ".png", ".jpg"]) num_columns_slider = gr.Slider(label="Number of Columns for Text", minimum=1, maximum=4, step=1, value=1) selected_layouts = gr.CheckboxGroup(choices=list(LAYOUTS.keys()), label="Select Page Layouts", value=["A4 Portrait"]) selected_fonts = gr.CheckboxGroup(choices=AVAILABLE_FONTS, label="Select Text Fonts", value=[AVAILABLE_FONTS[0]] if AVAILABLE_FONTS else []) generate_btn = gr.Button("🚀 Generate PDFs", variant="primary") with gr.Column(scale=2): gr.Markdown("### 🖼️ PDF Preview Gallery") gallery_output = gr.Gallery(label="Generated PDF Previews", show_label=False, elem_id="gallery", columns=3, height="auto") log_output = gr.Markdown(label="Generation Log", value="Logs will appear here...") def auto_run_demo(): """Function to run on app load to generate a sample PDF.""" print("Running initial demo generation...") demo_files = [gr.File(name="sample.md", data=b'')] # Gradio needs a File object previews, logs = generate_pdfs_api(demo_files, ["A4 Portrait"], [AVAILABLE_FONTS[0]] if AVAILABLE_FONTS else [], 1) return previews, logs demo.load(auto_run_demo, inputs=None, outputs=[gallery_output, log_output]) generate_btn.click(fn=generate_pdfs_api, inputs=[uploaded_files, selected_layouts, selected_fonts, num_columns_slider], outputs=[gallery_output, log_output]) if __name__ == "__main__": demo.launch()