Update ui/ui_app.py

ui/ui_app.py

@@ -1,231 +1,260 @@
 # ui/ui_app.py
 from __future__ import annotations
-import json, math
-from typing import Dict, Any, List, Tuple
+import json
+from typing import Dict, Any, List
 
 import gradio as gr
 import plotly.graph_objects as go
 import pandas as pd
 
-from core.extract import parse_pdf                      # 既存：PDF→(fin, df, meta, log)
+from core.extract import parse_pdf
+from core.market_infer import infer_market_metrics
 from core.external_scoring import (
     get_external_template_df,
     fill_missing_with_external,
+    merge_market_into_external_df,
     score_external_from_df,
 )
 from core.ai_judgement import suggest_external_with_llm, ai_evaluate
 
-# ---------- 可視化ヘルパ ----------
-def _radar_from_categories(title: str, cat_scores: Dict[str, float]) -> go.Figure:
+# ---------- chart helpers ----------
+def _radar(title: str, cat_scores: Dict[str, float]) -> go.Figure:
     if not cat_scores:
         cat_scores = {"N/A": 0.0}
     labels = list(cat_scores.keys())
-    values = [cat_scores[k] for k in labels]
+    vals = [cat_scores[k] for k in labels]
     fig = go.Figure()
-    fig.add_trace(go.Scatterpolar(r=values+[values[0]], theta=labels+[labels[0]],
-                                  fill="toself", name=title))
-    fig.update_layout(
-        polar=dict(radialaxis=dict(visible=True, range=[0,100])),
-        showlegend=False, margin=dict(l=30,r=30,t=40,b=30), height=380,
-        title=title
-    )
+    fig.add_trace(go.Scatterpolar(r=vals+[vals[0]], theta=labels+[labels[0]], fill="toself", name=title))
+    fig.update_layout(polar=dict(radialaxis=dict(visible=True, range=[0,100])),
+                      showlegend=False, height=360, margin=dict(l=30,r=30,t=40,b=30), title=title)
     return fig
 
 def _diff_bar(ext: Dict[str,float], ai: Dict[str,float]) -> go.Figure:
-    keys = sorted(set(ext.keys()) | set(ai.keys()))
-    diffs = [(ai.get(k,0)-ext.get(k,0)) for k in keys]
-    fig = go.Figure(data=[go.Bar(x=keys, y=diffs)])
-    fig.update_layout(title="AI評点 - 外部評価（カテゴリ差分）", height=360,
-                      margin=dict(l=30,r=30,t=40,b=30))
+    ks = sorted(set(ext.keys()) | set(ai.keys()))
+    diffs = [(ai.get(k,0)-ext.get(k,0)) for k in ks]
+    fig = go.Figure(data=[go.Bar(x=ks, y=diffs)])
+    fig.update_layout(title="AI評点 - 外部評価（カテゴリ差分）", height=320, margin=dict(l=30,r=30,t=40,b=30))
     return fig
 
-def _format_num(x):
-    if x is None: return "—"
+def _fmt(x):
     try:
-        x = float(x)
-        if abs(x)>=1e12: return f"{x/1e12:.2f}兆"
-        if abs(x)>=1e8:  return f"{x/1e8:.2f}億"
-        if abs(x)>=1e6:  return f"{x/1e6:.2f}百万円"
-        if abs(x)>=1e3:  return f"{x/1e3:.2f}千円"
-        return f"{x:.0f}"
+        f = float(x)
+        if abs(f) >= 1e8: return f"{f/1e8:.2f}億"
+        if abs(f) >= 1e6: return f"{f/1e6:.2f}百万円"
+        if abs(f) >= 1e3: return f"{f/1e3:.1f}千"
+        return f"{f:.0f}"
     except Exception:
-        return str(x)
+        return str(x) if x not in (None,"") else "—"
 
-def _cards_html(company: str, meta: Dict[str,Any], fin: Dict[str,Any], ext_total: float, ai_total: float) -> str:
-    bs = fin.get("balance_sheet", {}) or {}
-    is_ = fin.get("income_statement", {}) or {}
-    equity_ratio = None
+def _cards(company, meta, fin, ext_total, ai_total) -> str:
+    bs = fin.get("balance_sheet", {}) or {}; is_ = fin.get("income_statement", {}) or {}
+    ta = bs.get("total_assets") or 0; te = bs.get("total_equity") or 0
+    er = ""
     try:
-        ta = float(bs.get("total_assets") or 0)
-        te = float(bs.get("total_equity") or 0)
-        if ta>0: equity_ratio = te/ta*100
+        ta = float(ta); te = float(te); er = f"{(te/ta*100):.1f}%" if ta>0 else "—"
     except Exception:
-        equity_ratio = None
-
+        er = "—"
     period = ""
-    if meta and meta.get("period"):
-        p = meta["period"]
-        if isinstance(p, dict):
-            period = f"{p.get('start_date','')} ~ {p.get('end_date','')}"
-        else:
-            period = str(p)
+    if meta and isinstance(meta.get("period"), dict):
+        period = f"{meta['period'].get('start_date','')} ~ {meta['period'].get('end_date','')}"
     unit = (meta.get("unit") or "円").replace("JPY","円")
-
-    html = f"""
+    return f"""
 <div style="display:grid;grid-template-columns:repeat(4,1fr);gap:12px;">
   <div style="background:#F8FAFF;border:1px solid #E5E7EB;border-radius:12px;padding:12px;">
     <div style="font-size:12px;color:#6B7280;">企業名</div>
     <div style="font-size:18px;font-weight:700;">{company or fin.get('company',{}).get('name','—')}</div>
     <div style="font-size:12px;color:#6B7280;margin-top:4px;">期間: {period or '—'} / 単位: {unit}</div>
   </div>
-
   <div style="background:#F8FAFF;border:1px solid #E5E7EB;border-radius:12px;padding:12px;">
     <div style="font-size:12px;color:#6B7280;">売上高</div>
-    <div style="font-size:18px;font-weight:700;">{_format_num(is_.get('sales'))}</div>
+    <div style="font-size:18px;font-weight:700;">{_fmt(is_.get('sales'))}</div>
   </div>
-
   <div style="background:#F8FAFF;border:1px solid #E5E7EB;border-radius:12px;padding:12px;">
     <div style="font-size:12px;color:#6B7280;">営業利益</div>
-    <div style="font-size:18px;font-weight:700;">{_format_num(is_.get('operating_income'))}</div>
+    <div style="font-size:18px;font-weight:700;">{_fmt(is_.get('operating_income'))}</div>
   </div>
-
   <div style="background:#F8FAFF;border:1px solid #E5E7EB;border-radius:12px;padding:12px;">
     <div style="font-size:12px;color:#6B7280;">自己資本比率</div>
-    <div style="font-size:18px;font-weight:700;">{ '—' if equity_ratio is None else f'{equity_ratio:.1f}%' }</div>
+    <div style="font-size:18px;font-weight:700;">{er}</div>
   </div>
-
   <div style="background:#F0FDF4;border:1px solid #DCFCE7;border-radius:12px;padding:12px;">
     <div style="font-size:12px;color:#047857;">外部評価（定量）</div>
     <div style="font-size:22px;font-weight:800;color:#065F46;">{ext_total:.1f}</div>
   </div>
-
   <div style="background:#EFF6FF;border:1px solid #DBEAFE;border-radius:12px;padding:12px;">
     <div style="font-size:12px;color:#1D4ED8;">AI評点</div>
     <div style="font-size:22px;font-weight:800;color:#1E40AF;">{ai_total:.1f}</div>
   </div>
 </div>
 """
-    return html
 
-# ---------- 解析フロー ----------
+# ---------- core flows ----------
+def _market_df_from_dict(d: Dict[str, Any]) -> pd.DataFrame:
+    rows = []
+    order = [
+        "市場の年成長率(%)","市場成熟度(0-1)","競争強度(0-10)","参入障壁(0-10)","価格決定力(0-10)",
+        "サイクル感応度(0-10)","規制リスク(0-10)","技術破壊リスク(0-10)","TAM_億円","SAM_億円","SOM_億円"
+    ]
+    for k in order:
+        rows.append([k, d.get(k,"")])
+    return pd.DataFrame(rows, columns=["指標","値"])
+
+def _dict_from_market_df(df: pd.DataFrame) -> Dict[str, Any]:
+    out = {}
+    for _, r in df.iterrows():
+        k = str(r["指標"]); v = r["値"]
+        try:
+            out[k] = float(v)
+        except Exception:
+            out[k] = None
+    return out
+
 def on_analyze(company: str, use_vision: bool, files: List[str]):
     if not files:
         raise gr.Error("PDF をアップロードしてください。")
-
-    # PDF→財務JSON/抽出DF/メタ/ログ
     fin, df_fin, meta, log = parse_pdf(files, company, use_vision)
 
-    # 外部入力テンプレ & LLMで不足を提案
+    # 外部入力テンプレ＋不足項目のLLMサジェスト（従来どおり）
     ext_df = get_external_template_df()
-    suggestions = suggest_external_with_llm(fin, company)
-    ext_df = fill_missing_with_external(ext_df, suggestions)
+    ext_df = fill_missing_with_external(ext_df, suggest_external_with_llm(fin, company))
 
-    # 外部スコア
+    # 初期（空の市場DF）
+    market_df = _market_df_from_dict({})
+    # 外部/AIは市場未反映のまま仮計算（UI起動のため）
     ext_res = score_external_from_df(ext_df)
-
-    # AI評点（基準分離）
-    # ext_dfから必要要素だけピック（市場成長率/商品系）
-    ext_like = {}
-    for k in ["市場の年成長率(%)", "主力商品数", "成長中主力商品数"]:
-        try:
-            v = float(ext_df.loc[ext_df["入力項目"].eq(k), "値"].values[0])
-            ext_like[k] = v
-        except Exception:
-            pass
+    ai_res  = ai_evaluate(fin, {})
+
+    cards = _cards(company, meta, fin, ext_res["external_total"], ai_res["ai_total"])
+    ext_fig = _radar("外部評価（カテゴリ）", ext_res.get("category_scores", {}))
+    ai_fig  = _radar("AI評点（カテゴリ）", ai_res.get("category_scores", {}))
+    diff    = _diff_bar(ext_res.get("category_scores", {}), ai_res.get("category_scores", {}))
+
+    return (cards, df_fin, ext_df, market_df,
+            ext_res["external_total"], ai_res["ai_total"],
+            ext_fig, ai_fig, diff,
+            json.dumps(fin, ensure_ascii=False, indent=2),
+            json.dumps(ext_res, ensure_ascii=False, indent=2),
+            json.dumps(ai_res, ensure_ascii=False, indent=2),
+            "\n".join([str(x) for x in (log if isinstance(log,list) else [log])]))
+
+def on_market_infer(industry: str, products_text: str, country: str, horizon: int,
+                    ext_df: pd.DataFrame, fin_json: str):
+    prods = [p.strip() for p in (products_text or "").splitlines() if p.strip()]
+    market = infer_market_metrics(industry, prods, country, horizon)
+    market_df = _market_df_from_dict(market)
+
+    # ext_dfに市場推定を統合
+    ext_df2 = merge_market_into_external_df(ext_df, market, prods)
+
+    # スコア更新
+    fin = json.loads(fin_json)
+    ext_res = score_external_from_df(ext_df2)
+    # AIは市場の“別軸”影響を軽めに（成長余地構成で反映）
+    ext_like = {"市場の年成長率(%)": market.get("市場の年成長率(%)"),
+                "主力商品数": len(prods),
+                "成長中主力商品数": sum(1 for p in prods if (market.get("製品別年成長率(%)",{}).get(p,0) or 0)>10)}
     ai_res = ai_evaluate(fin, ext_like)
 
-    # 表示物
-    cards_html = _cards_html(company, meta, fin, ext_res["external_total"], ai_res["ai_total"])
-    ext_fig = _radar_from_categories("外部評価（カテゴリ）", ext_res.get("category_scores", {}))
-    ai_fig = _radar_from_categories("AI評点（カテゴリ）", ai_res.get("category_scores", {}))
-    diff_fig = _diff_bar(ext_res.get("category_scores", {}), ai_res.get("category_scores", {}))
+    ext_fig = _radar("外部評価（カテゴリ）", ext_res.get("category_scores", {}))
+    ai_fig  = _radar("AI評点（カテゴリ）", ai_res.get("category_scores", {}))
+    diff    = _diff_bar(ext_res.get("category_scores", {}), ai_res.get("category_scores", {}))
 
-    # JSON（詳細で見る用）
-    fin_json = json.dumps(fin, ensure_ascii=False, indent=2)
-    ext_json = json.dumps(ext_res, ensure_ascii=False, indent=2)
-    ai_json = json.dumps(ai_res, ensure_ascii=False, indent=2)
+    return (market_df, ext_df2,
+            ext_res["external_total"], ai_res["ai_total"],
+            ext_fig, ai_fig, diff,
+            json.dumps(ext_res, ensure_ascii=False, indent=2),
+            json.dumps(ai_res, ensure_ascii=False, indent=2),
+            "市場推定OK: " + "; ".join(market.get("注記", [])[:3]))
 
-    return cards_html, df_fin, ext_df, ext_fig, ai_fig, diff_fig, ai_res.get("memo",""), fin_json, ext_json, ai_json, log
-
-def on_rescore_external(ext_df: pd.DataFrame, fin_json: str, company: str):
+def on_rescore_all(ext_df: pd.DataFrame, market_df: pd.DataFrame, fin_json: str, products_text: str):
     fin = json.loads(fin_json)
-    ext_res = score_external_from_df(ext_df)
-    ext_fig = _radar_from_categories("外部評価（カテゴリ）", ext_res.get("category_scores", {}))
-    return ext_res["external_total"], ext_fig, json.dumps(ext_res, ensure_ascii=False, indent=2)
-
-def on_rescore_ai(ext_df: pd.DataFrame, fin_json: str):
-    fin = json.loads(fin_json)
-    # ext_dfから市場成長率/商品系を再取得
-    ext_like = {}
-    for k in ["市場の年成長率(%)", "主力商品数", "成長中主力商品数"]:
-        try:
-            v = float(ext_df.loc[ext_df["入力項目"].eq(k), "値"].values[0])
-            ext_like[k] = v
-        except Exception:
-            pass
+    prods = [p.strip() for p in (products_text or "").splitlines() if p.strip()]
+    market = _dict_from_market_df(market_df)
+    # ext_dfへ反映（手で編集されたmarket_dfも取り込む）
+    ext_df2 = merge_market_into_external_df(ext_df, market, prods)
+
+    ext_res = score_external_from_df(ext_df2)
+    ext_like = {"市場の年成長率(%)": market.get("市場の年成長率(%)"),
+                "主力商品数": len(prods),
+                "成長中主力商品数": sum(1 for p in prods if (market.get("製品別年成長率(%)",{}).get(p,0) or 0)>10)}
     ai_res = ai_evaluate(fin, ext_like)
-    ai_fig = _radar_from_categories("AI評点（カテゴリ）", ai_res.get("category_scores", {}))
-    return ai_res["ai_total"], ai_fig, json.dumps(ai_res, ensure_ascii=False, indent=2), ai_res.get("memo","")
 
-# ---------- UI ----------
+    ext_fig = _radar("外部評価（カテゴリ）", ext_res.get("category_scores", {}))
+    ai_fig  = _radar("AI評点（カテゴリ）", ai_res.get("category_scores", {}))
+    diff    = _diff_bar(ext_res.get("category_scores", {}), ai_res.get("category_scores", {}))
+
+    return (ext_df2,
+            ext_res["external_total"], ai_res["ai_total"],
+            ext_fig, ai_fig, diff,
+            json.dumps(ext_res, ensure_ascii=False, indent=2),
+            json.dumps(ai_res, ensure_ascii=False, indent=2),
+            "再計算完了")
+
 def build_ui():
     with gr.Blocks(theme=gr.themes.Soft(primary_hue="indigo"), analytics_enabled=False) as demo:
-        gr.Markdown("## 🧮 企業スコアリング（PDF解析 × 外部定量 × AI評点）")
+        gr.Markdown("## 🧮 企業スコアリング：PDF抽出 × 市場推定（LLM）× 外部定量 × AI評点")
 
         with gr.Row():
             with gr.Column(scale=1):
-                company = gr.Textbox(label="企業名（任意）", placeholder="例：株式会社OO")
-                use_vision = gr.Checkbox(value=True, label="OpenAI Visionで表・数値を補完")
+                company = gr.Textbox(label="企業名（任意）")
+                use_vision = gr.Checkbox(value=True, label="OpenAI VisionでPDF表を補完")
                 files = gr.File(label="決算書PDF（複数可）", file_count="multiple", type="filepath")
-                run_btn = gr.Button("📄 解析する", variant="primary")
-                gr.Markdown("※ JSONは下部の「詳細」に折りたたみ表示します。")
-
+                run_btn = gr.Button("📄 PDFを解析", variant="primary")
             with gr.Column(scale=2):
                 cards = gr.HTML(label="サマリー")
 
-        with gr.Tabs():
-            with gr.Tab("財務・外部入力"):
-                with gr.Row():
-                    df_fin = gr.Dataframe(label="抽出テーブル（編集可）", interactive=True, wrap=True)
-                with gr.Row():
-                    ext_df = gr.Dataframe(label="外部入力（編集可）", interactive=True, wrap=True)
-
-            with gr.Tab("スコア"):
-                with gr.Row():
-                    ext_plot = gr.Plot(label="外部評価（レーダー）")
-                    ai_plot = gr.Plot(label="AI評点（レーダー）")
-                diff_plot = gr.Plot(label="差分（棒）")
-                with gr.Row():
-                    ext_total = gr.Number(label="外部評価 合計(0-100)", value=0, precision=1, interactive=False)
-                    ai_total = gr.Number(label="AI評点 合計(0-100)", value=0, precision=1, interactive=False)
-                ai_memo = gr.Markdown(label="AI短評")
-
-            with gr.Tab("詳細"):
-                fin_json = gr.Code(label="抽出JSON", language="json")
-                ext_json = gr.Code(label="外部評価JSON", language="json")
-                ai_json  = gr.Code(label="AI評点JSON", language="json")
-                debug_out = gr.Textbox(label="ログ", lines=8)
-
-        # 動作
+        with gr.Tab("入力/市場推定"):
+            with gr.Row():
+                industry = gr.Textbox(label="事業領域（業界・カテゴリ）", placeholder="例）ヘルスケアIT / 産業ロボット 等")
+                products = gr.Textbox(label="主力商品（1行1件）", lines=4, placeholder="製品A\n製品B\n…")
+            with gr.Row():
+                country = gr.Dropdown(choices=["JP","US","EU","APAC","GLOBAL"], value="JP", label="対象地域")
+                horizon = gr.Slider(1, 7, value=3, step=1, label="予測年数")
+                infer_btn = gr.Button("🔎 市場を推定（LLM）", variant="secondary")
+
+            market_df = gr.Dataframe(label="市場メトリクス（編集可）", interactive=True, wrap=True)
+
+        with gr.Tab("外部入力/財務"):
+            df_fin  = gr.Dataframe(label="抽出テーブル（編集可）", interactive=True, wrap=True)
+            ext_df  = gr.Dataframe(label="外部入力（編集可）", interactive=True, wrap=True)
+
+        with gr.Tab("スコア"):
+            with gr.Row():
+                ext_total = gr.Number(label="外部評価 合計(0-100)", value=0, precision=1, interactive=False)
+                ai_total  = gr.Number(label="AI評点 合計(0-100)", value=0, precision=1, interactive=False)
+            with gr.Row():
+                ext_plot = gr.Plot(label="外部評価（レーダー）")
+                ai_plot  = gr.Plot(label="AI評点（レーダー）")
+            diff_plot = gr.Plot(label="差分（棒）")
+            rescore_btn = gr.Button("🔁 すべて再計算", variant="secondary")
+
+        with gr.Tab("詳細"):
+            fin_json = gr.Code(label="抽出JSON", language="json")
+            ext_json = gr.Code(label="外部評価JSON", language="json")
+            ai_json  = gr.Code(label="AI評点JSON", language="json")
+            debug    = gr.Textbox(label="ログ", lines=8)
+
+        # state
+        fin_state = gr.State("")
+
+        # wire
         run_btn.click(
             on_analyze,
             inputs=[company, use_vision, files],
-            outputs=[cards, df_fin, ext_df, ext_plot, ai_plot, diff_plot, ai_memo, fin_json, ext_json, ai_json, debug_out],
+            outputs=[cards, df_fin, ext_df, market_df, ext_total, ai_total, ext_plot, ai_plot, diff_plot,
+                     fin_json, ext_json, ai_json, debug],
+        ).then(lambda x: x, inputs=[fin_json], outputs=[fin_state])
+
+        infer_btn.click(
+            on_market_infer,
+            inputs=[industry, products, country, horizon, ext_df, fin_state],
+            outputs=[market_df, ext_df, ext_total, ai_total, ext_plot, ai_plot, diff_plot, ext_json, ai_json, debug],
         )
 
-        # 外部再スコア / AI再スコア（外部入力編集後）
-        rescore_ext = gr.Button("🔁 外部評価を再計算")
-        rescore_ai  = gr.Button("🔁 AI評点を再計算")
-
-        rescore_ext.click(
-            on_rescore_external,
-            inputs=[ext_df, fin_json, company],
-            outputs=[ext_total, ext_plot, ext_json],
-        )
-        rescore_ai.click(
-            on_rescore_ai,
-            inputs=[ext_df, fin_json],
-            outputs=[ai_total, ai_plot, ai_json, ai_memo],
+        rescore_btn.click(
+            on_rescore_all,
+            inputs=[ext_df, market_df, fin_state, products],
+            outputs=[ext_df, ext_total, ai_total, ext_plot, ai_plot, diff_plot, ext_json, ai_json, debug],
         )
 
     return demo