pptx2pdf_web_en.py

import gradio as gr
import os
import shutil
import subprocess
import sys
import tempfile
from zipfile import ZipFile
from pathlib import Path

# 程序根目录
BASE_DIR = os.path.dirname(__file__)

# 可执行文件路径：根据操作系统确定
if sys.platform.startswith('win'):
    BINARY_PATH = os.path.join(BASE_DIR, "PptxToPDF.exe")
else:
    BINARY_PATH = os.path.join(BASE_DIR, "PptxToPDF")

# 启动时确保二进制有执行权限（在 Hugging Face Spaces 里无法手动 chmod）
try:
    if not sys.platform.startswith('win'):
        os.chmod(BINARY_PATH, 0o755)
except Exception as e:
    # 如果权限设置失败，也无需中断，只要后续能调用即可
    print(f"⚠️ 警告：设置执行权限失败：{e}")

def install_fonts_from_repository():
    """
    将仓库中的 fonts 目录下的所有字体文件复制到 ~/.fonts 目录，并刷新字体缓存。
    """
    try:
        # 获取当前用户的主目录路径
        home_dir = Path.home()

        # 定义目标字体目录路径：~/.fonts
        fonts_dir = home_dir / ".fonts"

        # 如果 ~/.fonts 目录不存在，则创建该目录
        if not fonts_dir.exists():
            fonts_dir.mkdir(parents=True)
            print(f"已创建字体目录：{fonts_dir}")
        else:
            print(f"字体目录已存在：{fonts_dir}")

        # 定义仓库中的 fonts 目录路径
        repo_fonts_dir = Path(__file__).parent / "fonts"

        # 检查仓库中的 fonts 目录是否存在
        if not repo_fonts_dir.exists():
            print(f"错误：仓库中的 fonts 目录不存在：{repo_fonts_dir}")
            return

        # 支持的字体文件扩展名
        font_extensions = [".ttf", ".otf", ".ttc"]

        # 统计已复制的字体文件数量
        copied_count = 0

        # 遍历 fonts 目录中的所有文件
        for font_file in repo_fonts_dir.iterdir():
            if font_file.suffix.lower() in font_extensions:
                destination = fonts_dir / font_file.name
                try:
                    shutil.copy(font_file, destination)
                    print(f"已复制字体文件：{font_file.name} 到 {destination}")
                    copied_count += 1
                except Exception as e:
                    print(f"复制字体文件时出错：{font_file.name}，错误信息：{e}")

        if copied_count == 0:
            print("未找到任何可复制的字体文件。")
            return

        # 刷新字体缓存（仅在 Linux/Mac 下）
        if not sys.platform.startswith('win'):
            try:
                print("正在刷新字体缓存...")
                subprocess.run(["fc-cache", "-f", "-v"], check=True)
                print("字体缓存刷新完成。")
            except subprocess.CalledProcessError as e:
                print(f"刷新字体缓存时出错：{e}")
    except Exception as e:
        print(f"安装字体时出错：{e}")

# 启动时安装字体
install_fonts_from_repository()

def convert_pptx_to_pdf(pptx_path, output_dir):
    # 检查可执行文件是否存在
    if not os.path.exists(BINARY_PATH):
        raise FileNotFoundError(f"PptxToPDF 可执行文件未找到: {BINARY_PATH}")
    
    pdf_path = os.path.join(output_dir, os.path.splitext(os.path.basename(pptx_path))[0] + ".pdf")
    result = subprocess.run([
        BINARY_PATH,
        pptx_path
    ], cwd=output_dir, capture_output=True, text=True)
    if result.returncode != 0 or not os.path.exists(pdf_path):
        raise RuntimeError(f"Conversion failed: {result.stderr}")
    return pdf_path

def make_zip_file(files, zip_path):
    with ZipFile(zip_path, 'w') as zipf:
        for f in files:
            zipf.write(f, os.path.basename(f))
    return zip_path

def pptx2pdf_web(pptx_file): # pptx_file is the Gradio file object
    # Create a temporary directory that won't be auto-deleted
    tmpdir = tempfile.mkdtemp()
    
    try:
        gradio_temp_file_path = pptx_file.name # Path to the file uploaded by Gradio
        
        # Create a path for our copy of the PPTX inside our own tmpdir
        # Use the original basename of the uploaded file
        target_pptx_filename = os.path.basename(gradio_temp_file_path)
        our_copy_of_pptx_path = os.path.join(tmpdir, target_pptx_filename)

        # Copy the uploaded PPTX from Gradio's temp location to our temp location
        shutil.copyfile(gradio_temp_file_path, our_copy_of_pptx_path)
        
        # Now, our_copy_of_pptx_path is the path to the PPTX file within tmpdir.
        # convert_pptx_to_pdf will create the PDF also within tmpdir.
        pdf_path = convert_pptx_to_pdf(our_copy_of_pptx_path, tmpdir)
        
        # Create persistent copies of the output files for Gradio
        import tempfile as tf
        
        # Create persistent PDF file
        pdf_fd, persistent_pdf_path = tf.mkstemp(suffix='.pdf', prefix='converted_')
        os.close(pdf_fd)
        shutil.copy2(pdf_path, persistent_pdf_path)
        
        # Create persistent ZIP file
        zip_fd, persistent_zip_path = tf.mkstemp(suffix='.zip', prefix='converted_')
        os.close(zip_fd)
        zip_path = os.path.join(tmpdir, 'converted.zip')
        make_zip_file([pdf_path], zip_path)
        shutil.copy2(zip_path, persistent_zip_path)
        
        return persistent_pdf_path, persistent_zip_path
        
    finally:
        # Clean up our temporary directory
        try:
            shutil.rmtree(tmpdir)
        except:
            pass  # Ignore cleanup errors

with gr.Blocks(title="PPTX to PDF Converter - Easily Convert PowerPoint to PDF") as demo:
    gr.Markdown("# PPTX to PDF Converter") # Added main title here
    
    # Application Introduction and Instructions - 直接显示，不使用手风琴
    gr.Markdown("""
    ## Application Introduction and Instructions
    
    Welcome to the PPTX to PDF Converter! This tool allows you to easily convert your PowerPoint (PPTX) files into PDF documents.
    
    **Features:**
    - **Simple Upload:** Easily upload your PPTX file through the web interface.
    - **One-Click Conversion:** Convert your file to PDF with a single click.
    - **Direct PDF Download:** Download the converted PDF file directly.
    - **Zipped Archive:** Optionally, download a ZIP file containing the converted PDF for convenience.
    
    **How to Use:**
    1.  **Upload PPTX File:** Click on the "Upload PPTX File" area or drag and drop your `.pptx` file.
    2.  **Click Convert:** Press the "Convert" button to start the conversion process.
    3.  **Download Files:** Once the conversion is complete, download links for the PDF file and a ZIP archive (containing the PDF) will appear. Click on your preferred option to download.
    """)
    
    gr.Markdown("---")
    with gr.Row():
        with gr.Column():
            pptx_input = gr.File(label="Upload PPTX File", file_types=[".pptx"])
            convert_btn = gr.Button("Convert", variant="primary", size="lg")
        with gr.Column():
            pdf_output = gr.File(label="Download PDF File")
            zip_output = gr.File(label="Download ZIP Archive")
            # 添加漂亮的下载按钮
            with gr.Row():
                download_btn = gr.DownloadButton(
                    label="📦 Download ZIP Package", 
                    variant="secondary",
                    size="lg",
                    visible=False
                )
    
    def handle_conversion(pptx_file):
        if pptx_file is None:
            return None, None, gr.update(visible=False)
        
        pdf_path, zip_path = pptx2pdf_web(pptx_file)
        
        # 更新下载按钮，使其可见并设置下载文件
        return pdf_path, zip_path, gr.update(visible=True, value=zip_path)
    
    convert_btn.click(
        handle_conversion, 
        inputs=[pptx_input], 
        outputs=[pdf_output, zip_output, download_btn]
    )

demo.queue(default_concurrency_limit=10, max_size=50)
demo.launch(show_api=False)