Update app.py

app.py

@@ -2,9 +2,9 @@ import os
 import io
 import json
 import hashlib
-import gradio  # 前方参照の解決用（必要なら）
+import gradio  # 一部の前方参照バグ回避用
 import gradio as gr
-from typing import Any, List
+from typing import List
 
 from pipelines.openai_ingest import (
     extract_text_with_openai,
@@ -15,7 +15,7 @@ from pipelines.parsing import normalize_resume
 from pipelines.merge import merge_normalized_records
 from pipelines.skills import extract_skills
 
-# ---- anonymize フォールバック（pipelines/anonymize.py が未実装でも動く） ----
+# --- 匿名化のフォールバック（pipelines/anonymize.py が空/未実装でも動く） ---
 try:
     from pipelines.anonymize import anonymize_text, render_anonymized_pdf  # type: ignore
 except Exception:
@@ -28,7 +28,6 @@ except Exception:
         A4 = None
 
     def anonymize_text(text: str):
-        # 超簡易：メール/電話っぽい所をマスク。氏名は見出し候補を雑にマスク。
         masked = re.sub(r"([A-Za-z0-9._%+-]+)@([A-Za-z0-9.-]+\.[A-Za-z]{2,})", r"***@\2", text)
         masked = re.sub(r"(?:\+?\d{1,3}[ -]?)?(?:\(\d{2,4}\)[ -]?)?\d{2,4}[ -]?\d{2,4}[ -]?\d{3,4}", "***-****-****", masked)
         masked = re.sub(r"(氏名[:：]?\s*)(\S+)", r"\1＊＊＊", masked)
@@ -36,22 +35,21 @@ except Exception:
 
     def render_anonymized_pdf(text: str) -> bytes:
         if canvas is None:
-            # reportlab が無ければテキストファイルで代替（UIは .pdf 名で返す）
             return text.encode("utf-8")
         buf = io.BytesIO()
         c = canvas.Canvas(buf, pagesize=A4)
         width, height = A4
-        margin = 40
-        y = height - margin
+        m = 40
+        y = height - m
         for line in text.splitlines() or ["(no content)"]:
-            if y < margin:
+            if y < m:
                 c.showPage()
-                y = height - margin
-            c.drawString(margin, y, line[:95])
+                y = height - m
+            c.drawString(m, y, line[:95])
             y -= 14
         c.save()
         return buf.getvalue()
-# ---------------------------------------------------------------------------
+# ----------------------------------------------------------------------
 
 from pipelines.scoring import compute_quality_score
 from pipelines.storage import persist_to_hf
@@ -60,7 +58,7 @@ 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 _read_bytes_from_path(path: str) -> bytes:
+def _read_bytes(path: str) -> bytes:
     with open(path, "rb") as f:
         return f.read()
 
@@ -77,10 +75,10 @@ def process_resumes(files: List[str], candidate_id: str = "", additional_notes:
 
     for path in files:
         filename = os.path.basename(path)
-        raw_bytes = _read_bytes_from_path(path)
+        raw_bytes = _read_bytes(path)
         filetype = detect_filetype(filename, 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
+        # 1) 抽出
         if filetype in {"pdf", "image"}:
             text = extract_text_with_openai(raw_bytes, filename=filename, filetype=filetype)
         else:
@@ -89,7 +87,7 @@ def process_resumes(files: List[str], candidate_id: str = "", additional_notes:
 
         raw_texts.append({"filename": filename, "text": text})
 
-        # 2) OpenAIでセクション構造化 → ルール正規化
+        # 2) 構造化→正規化
         structured = structure_with_openai(text)
         normalized = normalize_resume({
             "work_experience": structured.get("work_experience_raw", ""),
@@ -104,10 +102,10 @@ def process_resumes(files: List[str], candidate_id: str = "", additional_notes:
             "normalized": normalized,
         })
 
-    # 3) 統合（複数ファイル→1候補者）
+    # 3) 統合
     merged = merge_normalized_records([r["normalized"] for r in partial_records])
 
-    # 4) スキル抽出（辞書/正規表現）
+    # 4) スキル抽出
     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", ""),
@@ -123,10 +121,10 @@ def process_resumes(files: List[str], candidate_id: str = "", additional_notes:
     # 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
+    # 8) 出力まとめ
     cid = candidate_id or hashlib.sha256(merged_text.encode("utf-8")).hexdigest()[:16]
     result_json = {
         "candidate_id": cid,
@@ -152,10 +150,11 @@ def process_resumes(files: List[str], candidate_id: str = "", additional_notes:
             pdf_path=f"candidates/{cid}.anon.pdf",
         )
 
-    # UI 向け出力を整形
     anon_pdf = (f"{cid}.anon.pdf", anon_pdf_bytes)
+
+    # ★ UI用：すべて文字列化して返す（gr.JSON を使わない）
     out_json_str = json.dumps(result_json, ensure_ascii=False, indent=2)
-    out_skills_str = json.dumps(skills, ensure_ascii=False, indent=2)  # gr.Code で表示
+    out_skills_str = json.dumps(skills, ensure_ascii=False, indent=2)
     out_score_str = json.dumps(score, ensure_ascii=False, indent=2)
     out_commit_str = json.dumps(commit_info or {"status": "skipped (DATASET_REPO not set)"}, ensure_ascii=False, indent=2)
 
@@ -179,7 +178,7 @@ 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="filepath",  # ← 重要：'file' ではなく 'filepath'
         )
         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)
@@ -190,7 +189,7 @@ 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)")  # ★ JSON -> Code に変更
+        out_skills = gr.Code(label="スキル一覧 (JSON)")  # ← gr.JSON を避ける
 
     with gr.Tab("品質スコア"):
         out_score = gr.Code(label="品質評価 (JSON)")
@@ -214,5 +213,5 @@ with gr.Blocks(title=APP_TITLE) as demo:
     )
 
 if __name__ == "__main__":
-    # 環境により localhost へ到達できない場合があるため share=True を既定に
+    # 到達性のため share=True 推奨
     demo.launch(server_name="0.0.0.0", server_port=7860, share=True)