Update app.py

app.py

@@ -1,40 +1,53 @@
 import gradio as gr
-import subprocess
-import os
 import tempfile
+from pdf2image import convert_from_path
+from PIL import Image
+import img2pdf
+import os
 
-def rasterize_with_pdftoppm(input_pdf):
+def downscale_pdf(pdf_file, dpi=150, max_width=1500):
     with tempfile.TemporaryDirectory() as tmpdir:
-        output_prefix = os.path.join(tmpdir, "page")
-        cmd1 = ["pdftoppm", "-png", "-r", "150", input_pdf, output_prefix]
-        res1 = subprocess.run(cmd1, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
-        if res1.returncode != 0:
-            return f"pdftoppm変換エラー: {res1.stderr.decode()}", None, None
-
-        # PNGファイル一覧取得
-        png_files = sorted([os.path.join(tmpdir, f) for f in os.listdir(tmpdir) if f.endswith(".png")])
-        if not png_files:
-            return "PNG変換に失敗しました", None, None
-
-        # PNG→PDF変換（ImageMagickのconvertが使えなければpypdfで合成も可能）
-        output_pdf = os.path.join(tmpdir, "output_rasterized.pdf")
-        cmd2 = ["convert"] + png_files + [output_pdf]
-        res2 = subprocess.run(cmd2, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
-        if res2.returncode != 0:
-            return f"PDF作成エラー: {res2.stderr.decode()}", None, None
-
-        size_bytes = os.path.getsize(output_pdf)
-        size_kb = size_bytes / 1024
-        size_str = f"{size_bytes} bytes ({size_kb:.1f} KB)"
-
-        return "pdftoppmでラスタライズ完了", output_pdf, size_str
-
-demo = gr.Interface(
-    fn=rasterize_with_pdftoppm,
-    inputs=gr.File(file_types=[".pdf"]),
-    outputs=[gr.Text(), gr.File(), gr.Text()],
-    title="PDFをpdftoppmでラスタライズ",
-    description="pdftoppmでPDFをPNGに変換し再度PDF化します。"
-)
+        # ステップ 1: PDF → 画像（dpiを指定）
+        images = convert_from_path(pdf_file.name, dpi=dpi, output_folder=tmpdir)
+
+        downscaled_paths = []
+
+        for i, img in enumerate(images):
+            # ステップ 2: サイズを制限
+            w, h = img.size
+            if w > max_width:
+                new_h = int(h * max_width / w)
+                img = img.resize((max_width, new_h), Image.LANCZOS)
+
+            # 一時保存
+            img_path = os.path.join(tmpdir, f"page_{i}.jpg")
+            img.save(img_path, "JPEG", quality=85)
+            downscaled_paths.append(img_path)
+
+        # ステップ 3: 画像をPDFに再結合
+        output_pdf_path = os.path.join(tmpdir, "downscaled.pdf")
+        with open(output_pdf_path, "wb") as f_out:
+            f_out.write(img2pdf.convert(downscaled_paths))
+
+        # 返却
+        return output_pdf_path
+
+with gr.Blocks() as demo:
+    gr.Markdown("# PDF 解像度ダウンサイザー\nPDFのページ画像を縮小して容量を減らします。")
+    with gr.Row():
+        with gr.Column():
+            pdf_input = gr.File(label="入力PDF", file_types=[".pdf"])
+            dpi_input = gr.Slider(72, 300, value=150, step=1, label="変換DPI")
+            width_input = gr.Slider(500, 3000, value=1500, step=50, label="最大幅(px)")
+            convert_button = gr.Button("変換")
+
+        with gr.Column():
+            pdf_output = gr.File(label="出力PDF")
+
+    convert_button.click(
+        fn=downscale_pdf,
+        inputs=[pdf_input, dpi_input, width_input],
+        outputs=pdf_output
+    )
 
 demo.launch()