Update app.py

app.py

@@ -20,33 +20,30 @@ from pipelines.utils import detect_filetype, load_doc_text
 APP_TITLE = "候補者インテーク & レジュメ標準化（OpenAI版）"
 
 
-def process_resumes(filepaths, candidate_id: str, additional_notes: str = ""):
-    if not filepaths:
+def process_resumes(files, candidate_id: str, additional_notes: str = ""):
+    if not files:
         raise gr.Error("少なくとも1ファイルをアップロードしてください。")
 
     partial_records = []
     raw_texts = []
 
-    # ⚠️ gr.Files(type="filepath") のため、ここは「パス文字列」の配列
-    for path in filepaths:
-        filepath = str(path)
-        filename = os.path.basename(filepath)
-        with open(filepath, "rb") as fp:
-            raw_bytes = fp.read()
+    for f in files:
+        raw_bytes = f.read()
+        filetype = detect_filetype(f.name, raw_bytes)
 
-        filetype = detect_filetype(filename, raw_bytes)
-
-        # 1) テキスト抽出：画像/PDF→Vision OCR、docx/txt→生文面+整形
+        # 1) テキスト抽出：画像/PDFはOpenAI Vision OCR、docx/txtは生文面＋OpenAI整形
         if filetype in {"pdf", "image"}:
-            text = extract_text_with_openai(raw_bytes, filename=filename, filetype=filetype)
+            text = extract_text_with_openai(raw_bytes, filename=f.name, filetype=filetype)
         else:
             base_text = load_doc_text(filetype, raw_bytes)
-            text = extract_text_with_openai(base_text.encode("utf-8"), filename=filename, filetype="txt")
+            # 生テキストをそのままOpenAIへ渡し、軽く整形した全文を返す
+            text = extract_text_with_openai(base_text.encode("utf-8"), filename=f.name, filetype="txt")
 
-        raw_texts.append({"filename": filename, "text": text})
+        raw_texts.append({"filename": f.name, "text": text})
 
-        # 2) 構造化 → 正規化
+        # 2) OpenAIでセクション構造化
         structured = structure_with_openai(text)
+        # 念のためルールベース正規化も適用（期間抽出など補助）
         normalized = normalize_resume({
             "work_experience": structured.get("work_experience_raw", ""),
             "education": structured.get("education_raw", ""),
@@ -54,7 +51,7 @@ def process_resumes(filepaths, candidate_id: str, additional_notes: str = ""):
             "skills": ", ".join(structured.get("skills_list", [])),
         })
         partial_records.append({
-            "source": filename,
+            "source": f.name,
             "text": text,
             "structured": structured,
             "normalized": normalized,
@@ -63,7 +60,7 @@ def process_resumes(filepaths, candidate_id: str, additional_notes: str = ""):
     # 3) 統合（複数ファイル→1候補者）
     merged = merge_normalized_records([r["normalized"] for r in partial_records])
 
-    # 4) スキル抽出
+    # 4) スキル抽出（辞書/正規表現）
     merged_text = "\n\n".join([r["text"] for r in partial_records])
     skills = extract_skills(merged_text, {
         "work_experience": merged.get("raw_sections", {}).get("work_experience", ""),
@@ -72,20 +69,20 @@ def process_resumes(filepaths, candidate_id: str, additional_notes: str = ""):
         "skills": ", ".join(merged.get("skills", [])),
     })
 
-    # 5) 匿名化 → PDF化
+    # 5) 匿名化
     anonymized_text, anon_map = anonymize_text(merged_text)
     anon_pdf_bytes = render_anonymized_pdf(anonymized_text)
 
     # 6) 品質スコア
     score = compute_quality_score(merged_text, merged)
 
-    # 7) 要約
+    # 7) 要約（300/100/1文）
     summaries = summarize_with_openai(merged_text)
 
     # 8) 構造化出力
     result_json = {
         "candidate_id": candidate_id or hashlib.sha256(merged_text.encode("utf-8")).hexdigest()[:16],
-        "files": [os.path.basename(p) for p in filepaths],
+        "files": [f.name for f in files],
         "merged": merged,
         "skills": skills,
         "quality_score": score,
@@ -94,7 +91,7 @@ def process_resumes(filepaths, candidate_id: str, additional_notes: str = ""):
         "notes": additional_notes,
     }
 
-    # 9) HF Datasets 保存（任意）
+    # 9) HF Datasets 保存
     dataset_repo = os.environ.get("DATASET_REPO")
     commit_info = None
     if dataset_repo:
@@ -110,10 +107,9 @@ def process_resumes(filepaths, candidate_id: str, additional_notes: str = ""):
 
     anon_pdf = (result_json["candidate_id"] + ".anon.pdf", anon_pdf_bytes)
 
-    # ← GradioのAPIスキーマ生成バグ回避のため、JSONは文字列で返す（gr.Codeに表示）
     return (
         json.dumps(result_json, ensure_ascii=False, indent=2),
-        json.dumps(skills, ensure_ascii=False, indent=2),
+        skills,
         json.dumps(score, ensure_ascii=False, indent=2),
         summaries["300chars"],
         summaries["100chars"],
@@ -131,7 +127,7 @@ with gr.Blocks(title=APP_TITLE) as demo:
             label="レジュメ類 (PDF/画像/Word/テキスト) 複数可",
             file_count="multiple",
             file_types=[".pdf", ".png", ".jpg", ".jpeg", ".tiff", ".bmp", ".docx", ".txt"],
-            type="filepath",  # ★ 重要: 'file' ではなく 'filepath'
+            type="file"  # ※読み取りコードが .read() 前提のため file のまま
         )
         candidate_id = gr.Textbox(label="候補者ID（任意。未入力なら自動生成）")
     notes = gr.Textbox(label="補足メモ（任意）", lines=3)
@@ -142,7 +138,8 @@ with gr.Blocks(title=APP_TITLE) as demo:
         out_json = gr.Code(label="統合出力 (JSON)")
 
     with gr.Tab("抽出スキル"):
-        out_skills = gr.Code(label="スキル一覧 (JSON)")  # ★ gr.JSON は使わない
+        # gr.JSON → gr.Code に変更（Gradio 4.44.0 の schema 解析バグ回避）
+        out_skills = gr.Code(label="スキル一覧(JSON表示)")
 
     with gr.Tab("品質スコア"):
         out_score = gr.Code(label="品質評価")
@@ -166,5 +163,5 @@ with gr.Blocks(title=APP_TITLE) as demo:
 
 
 if __name__ == "__main__":
-    # HF Spaces で share=True は未対応。明示的に 0.0.0.0 を指定。
+    # HF Spaces 上で localhost 非アクセス時の ValueError を回避
     demo.launch(server_name="0.0.0.0", server_port=7860)