Update app.py

app.py

@@ -19,6 +19,41 @@ 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）"
 
+# Gradio v4 Filesの入力を安全にハンドリング
+def _iter_files(files):
+    """
+    Gradio v4:
+      - type="filepath": files は 文字列パスのリスト
+      - type="binary"  : files は {name: str, data: bytes} の辞書リスト
+    互換目的で file-like も許容。
+    """
+    for f in files:
+        # filepath
+        if isinstance(f, str):
+            path = f
+            name = os.path.basename(path)
+            with open(path, "rb") as fh:
+                data = fh.read()
+            yield name, data
+            continue
+        # binary
+        if isinstance(f, dict) and "name" in f and "data" in f:
+            name = os.path.basename(f["name"] or "uploaded")
+            data = f["data"]
+            yield name, data
+            continue
+        # file-like
+        if hasattr(f, "read"):
+            try:
+                name = os.path.basename(getattr(f, "name", "uploaded"))
+            except Exception:
+                name = "uploaded"
+            data = f.read()
+            yield name, data
+            continue
+        # 不明形式はスキップ
+        continue
+
 
 def process_resumes(files, candidate_id: str, additional_notes: str = ""):
     if not files:
@@ -27,21 +62,19 @@ def process_resumes(files, candidate_id: str, additional_notes: str = ""):
     partial_records = []
     raw_texts = []
 
-    for f in files:
-        raw_bytes = f.read()
-        filetype = detect_filetype(f.name, raw_bytes)
+    for fname, raw_bytes in _iter_files(files):
+        filetype = detect_filetype(fname, raw_bytes)
 
         # 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=f.name, filetype=filetype)
+            text = extract_text_with_openai(raw_bytes, filename=fname, filetype=filetype)
         else:
             base_text = load_doc_text(filetype, raw_bytes)
-            # 生テキストをOpenAIへ渡し、整形本文を返す
-            text = extract_text_with_openai(base_text.encode("utf-8"), filename=f.name, filetype="txt")
+            text = extract_text_with_openai(base_text.encode("utf-8"), filename=fname, filetype="txt")
 
-        raw_texts.append({"filename": f.name, "text": text})
+        raw_texts.append({"filename": fname, "text": text})
 
-        # 2) OpenAIでセクション構造化 → ルールベース正規化
+        # 2) OpenAIでセクション構造化 → ルール正規化
         structured = structure_with_openai(text)
         normalized = normalize_resume({
             "work_experience": structured.get("work_experience_raw", ""),
@@ -50,7 +83,7 @@ def process_resumes(files, candidate_id: str, additional_notes: str = ""):
             "skills": ", ".join(structured.get("skills_list", [])),
         })
         partial_records.append({
-            "source": f.name,
+            "source": fname,
             "text": text,
             "structured": structured,
             "normalized": normalized,
@@ -68,20 +101,21 @@ def process_resumes(files, candidate_id: str, additional_notes: str = ""):
         "skills": ", ".join(merged.get("skills", [])),
     })
 
-    # 5) 匿名化
+    # 5) 匿名化 → 匿名PDF生成
     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（300/100/1文）
+    # 7) 要約
     summaries = summarize_with_openai(merged_text)
 
-    # 8) 構造化出力（UIバグ回避のため **すべて文字列** で返す）
+    # 8) 構造化出力
+    cid = candidate_id or hashlib.sha256(merged_text.encode("utf-8")).hexdigest()[:16]
     result_json = {
-        "candidate_id": candidate_id or hashlib.sha256(merged_text.encode("utf-8")).hexdigest()[:16],
-        "files": [f.name for f in files],
+        "candidate_id": cid,
+        "files": [r["source"] for r in partial_records],
         "merged": merged,
         "skills": skills,
         "quality_score": score,
@@ -90,29 +124,28 @@ def process_resumes(files, candidate_id: str, additional_notes: str = ""):
         "notes": additional_notes,
     }
 
-    # 9) HF Datasets 保存
+    # 9) HF Datasets 保存（任意）
     dataset_repo = os.environ.get("DATASET_REPO")
     commit_info = None
     if dataset_repo:
-        file_hash = result_json["candidate_id"]
         commit_info = persist_to_hf(
             dataset_repo=dataset_repo,
             record=result_json,
             anon_pdf_bytes=anon_pdf_bytes,
-            parquet_path=f"candidates/{file_hash}.parquet",
-            json_path=f"candidates/{file_hash}.json",
-            pdf_path=f"candidates/{file_hash}.anon.pdf",
+            parquet_path=f"candidates/{cid}.parquet",
+            json_path=f"candidates/{cid}.json",
+            pdf_path=f"candidates/{cid}.anon.pdf",
         )
 
-    anon_pdf = (result_json["candidate_id"] + ".anon.pdf", anon_pdf_bytes)
+    anon_pdf = (f"{cid}.anon.pdf", anon_pdf_bytes)
 
     return (
-        json.dumps(result_json, ensure_ascii=False, indent=2),      # JSON全体
-        json.dumps(skills, ensure_ascii=False, indent=2),           # ★ skillsは文字列で返す
-        json.dumps(score, ensure_ascii=False, indent=2),            # スコアも文字列
-        summaries.get("300chars", ""),
-        summaries.get("100chars", ""),
-        summaries.get("onesent", ""),
+        json.dumps(result_json, ensure_ascii=False, indent=2),
+        json.dumps(skills, ensure_ascii=False, indent=2),   # ← out_skillsをCodeにしたのでJSON文字列を返す
+        json.dumps(score, ensure_ascii=False, indent=2),
+        summaries["300chars"],
+        summaries["100chars"],
+        summaries["onesent"],
         anon_pdf,
         json.dumps(commit_info or {"status": "skipped (DATASET_REPO not set)"}, ensure_ascii=False, indent=2),
     )
@@ -126,21 +159,21 @@ 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="file"   # fileオブジェクトを受ける（process_resumesの実装に合致）
+            type="filepath"  # ← v4では 'filepath' or 'binary'
         )
         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)
 
-    run_btn = gr.Button("実行")
+    run_btn = gr.Button("実行", variant="primary")
 
     with gr.Tab("構造化JSON"):
-        out_json = gr.Code(label="統合出力 (JSON文字列)")
+        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文字列)")  # ★ gr.JSON を使わない
+        out_skills = gr.Code(label="スキル一覧 (JSON)")  # ← JSON Schema問題回避
 
     with gr.Tab("品質スコア"):
-        out_score = gr.Code(label="品質評価 (JSON文字列)")
+        out_score = gr.Code(label="品質評価 (JSON)")
 
     with gr.Tab("要約 (300/100/1文)"):
         out_sum_300 = gr.Textbox(label="300字要約")
@@ -151,7 +184,7 @@ with gr.Blocks(title=APP_TITLE) as demo:
         out_pdf = gr.File(label="匿名PDFダウンロード")
 
     with gr.Tab("Datasets 保存ログ"):
-        out_commit = gr.Code(label="コミット情報 (JSON文字列)")
+        out_commit = gr.Code(label="コミット情報")
 
     run_btn.click(
         process_resumes,