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 --- # Use the current working directory as the base for all paths for robustness. 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 is now the main directory of the app. 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()}") # Use glob to find all .ttf files in the base directory font_files = list(FONT_DIR.glob("*.ttf")) print(f"Found {len(font_files)} .ttf files: {[f.name for f in font_files]}") if not font_files: print("WARNING: No .ttf files found. Please add font files to the application directory.") for font_path in font_files: try: font_name = font_path.stem print(f"Registering font: '{font_name}' from {font_path.name}") pdfmetrics.registerFont(TTFont(font_name, str(font_path))) # CRITICAL FIX for bold/italic error: Register dummy variants for all fonts. 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") # Correctly identify the special color emoji font for rendering. # All other fonts, including other NotoEmoji variants, will be selectable for text. if "notocoloremoji-regular" in font_name.lower(): emoji_font_name = font_name # Add all non-color-emoji fonts to the list for the UI. if "notocoloremoji-regular" not in font_name.lower(): 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: """Wraps emoji characters in a tag to use the dedicated emoji font.""" if not emoji_font_name: return text # Return original text if no emoji font is available 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, emoji_font: str): """Converts markdown to a ReportLab story, with improved handling of elements.""" 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_h2 = ParagraphStyle('h2', parent=styles['h2'], fontName=font_name, fontSize=16, leading=20) style_h3 = ParagraphStyle('h3', parent=styles['h3'], fontName=font_name, fontSize=14, leading=18) style_code = ParagraphStyle('Code', fontName='Courier', backColor=colors.whitesmoke, textColor=colors.darkblue, borderWidth=1, borderColor=colors.lightgrey, padding=5) 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 = "", [] for line in lines: if line.strip().startswith("```"): if in_code_block: story.append(Paragraph(code_block_text.replace('\n', '
'), style_code)) 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 line.strip().startswith('|'): if not in_table: in_table = True if all(c in '-|: ' for c in line.strip()): continue cells = [apply_emoji_font(cell.strip(), emoji_font) for cell in line.strip().strip('|').split('|')] table_data.append(cells) continue if in_table: in_table = False if table_data: header = [Paragraph(cell, style_table_header) for cell in table_data[0]] formatted_rows = [[Paragraph(cell, style_normal) for cell in row] for row in table_data[1:]] 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 = [] line_with_emoji = apply_emoji_font(line, emoji_font) 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)) elif line.startswith("## "): story.append(Paragraph(formatted_line[3:], style_h2)) elif line.startswith("### "): story.append(Paragraph(formatted_line[4:], style_h3)) elif line.strip().startswith(("* ", "- ")): story.append(Paragraph(formatted_line.strip()[2:], style_normal, bulletText='•')) elif re.match(r'^\d+\.\s', line.strip()): story.append(Paragraph(formatted_line.strip(), style_normal)) elif line.strip() == "": story.append(Spacer(1, 0.1 * inch)) 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 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, 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\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, EMOJI_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) 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(' ','-')}_{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 = "# 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(CWD / "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. The app will find any `.ttf` fonts in its directory. Group assets with a common name (e.g., `Doc_part1.md`, `Doc_img1.png`) to combine them.") 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", ".jpeg"]) 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", 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], outputs=[gallery_output, log_output, downloadable_files_output]) if __name__ == "__main__": demo.launch()