codex: DB連携・スキーマ変更に係る各所の変更

app/main.py

@@ -17,7 +17,7 @@ sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
 sys.path.append(os.path.dirname(__file__))
 
 from utils.logger import setup_logger
-from schemas.projects import ProjectSummary, ProjectDetail, ProjectIds
+from schemas.projects import Image, ProjectSummary, ProjectDetail, ProjectIds
 
 from .repositories.projects_repository import ProjectsRepository
 from .search.engine import SearchEngine
@@ -51,6 +51,18 @@ projects_service = ProjectsService(
 )
 
 
+def _select_primary_image(images: list[Image]) -> Image | None:
+    primary: Image | None = None
+    for image in images:
+        if image.order == 0:
+            primary = image
+    if primary is not None:
+        return primary
+    if images:
+        return images[0]
+    return None
+
+
 @asynccontextmanager
 async def lifespan(app: FastAPI):
     """アプリケーションの起動時と終了時に実行されるコード"""
@@ -93,13 +105,30 @@ def health_check():
 )
 def get_summary_data():
     # ここで必要なフィールドのみ抽出して返す
-    summaries = [
-        ProjectSummary(
-            **item.model_dump(include=set(ProjectSummary.model_fields.keys()))
+    summaries_payload: list[dict[str, object]] = []
+    for project in projects_service.list_projects():
+        # 画像はprimaryの1枚だけ返す
+        image = _select_primary_image(project.images)
+        if not image:
+            log.error("企画ID %s の画像選択に失敗しました", project.projectId)
+
+        summary = ProjectSummary(
+            projectId=project.projectId,
+            circleName=project.circleName,
+            name=project.name,
+            category=project.category,
+            firstDay=project.firstDay,
+            secondDay=project.secondDay,
+            thirdDay=project.thirdDay,
+            location=project.location,
+            description=project.description,
+            prSummary=project.prSummary,
+            notes=project.notes,
+            image=image,
         )
-        for item in projects_service.list_projects()
-    ]
-    content = json.dumps(summaries, ensure_ascii=False).encode("utf-8")
+        summaries_payload.append(summary.model_dump(mode="json"))
+
+    content = json.dumps(summaries_payload, ensure_ascii=False).encode("utf-8")
     return Response(
         content=gzip.compress(content),
         headers={"Content-Encoding": "gzip", "Content-Type": "application/json"},

app/repositories/_data_example.py

@@ -12,10 +12,10 @@
         "description": "自律走行ロボットの展示とデモを行います。",
         "prSummary": "ロボの魅力をぎゅっと！",
         "prDetail": "部員が制作した各種ロボットを分かりやすく紹介します。体験コーナーもあり。",
-        "note": ["写真撮影OK", "混雑時は入場制限あり"],
+        "notes": ["写真撮影OK", "混雑時は入場制限あり"],
         "images": [
-            {"filename": "rb001", "extension": "webp", "order": 1},
-            {"filename": "rb002", "extension": "webp", "order": 2},
+            {"filename": "rb001", "extension": "webp", "order": 0},
+            {"filename": "rb002", "extension": "webp", "order": 1},
         ],
         "urls": [
             {"service": "X", "url": "https://x.com/roboken"},
@@ -34,10 +34,10 @@
         "location": "野外ステージ",
         # texts.* は該当APPROVEDが無い場合 None になり得る
         "description": None,
-        "prSummary": None,
+        "prSummary": "",
         "prDetail": None,
         # COALESCE により空集合は [] になる
-        "note": [],
+        "notes": [],
         "images": [],
         "urls": [],
     },

app/repositories/projects_repository.py

@@ -1,3 +1,4 @@
+import json
 from typing import Any, Dict, List
 
 from pymysql import MySQLError
@@ -47,6 +48,29 @@ class ProjectsRepository:
             "MUSIC": "音楽",
             "OTHER": "その他",
         }
+        def _coerce_list(value: Any) -> list[Any]:
+            if value is None:
+                return []
+            if isinstance(value, list):
+                return value
+            if isinstance(value, str):
+                try:
+                    parsed = json.loads(value)
+                except json.JSONDecodeError:
+                    return []
+                return parsed if isinstance(parsed, list) else []
+            return []
+
+        def _coerce_dict_list(value: Any) -> list[Dict[str, Any]]:
+            return [item for item in _coerce_list(value) if isinstance(item, dict)]
+
+        def _coerce_str_list(value: Any) -> list[str]:
+            items = []
+            for item in _coerce_list(value):
+                if isinstance(item, str):
+                    items.append(item)
+            return items
+
         return Project(
             projectId=row["projectId"],
             circleName=row["circleName"],
@@ -60,7 +84,7 @@ class ProjectsRepository:
             description=row["description"],
             prSummary=row["prSummary"],
             prDetail=row["prDetail"],
-            notes=row["notes"],
-            images=[Image(**img) for img in row["images"]],
-            urls=[Sns(**url) for url in row["urls"]],
+            notes=_coerce_str_list(row.get("notes")) or None,
+            images=[Image(**img) for img in _coerce_dict_list(row.get("images"))],
+            urls=[Sns(**url) for url in _coerce_dict_list(row.get("urls"))],
         )

app/repositories/query.sql

@@ -11,7 +11,7 @@ SELECT
   texts.description AS "description",
   texts.prSummary AS "prSummary",
   texts.prDetail AS "prDetail",
-  COALESCE(visitor_note.note, '[]'::jsonb) AS "notes",
+  COALESCE(visitor_note.notes, '[]'::jsonb) AS "notes",
   COALESCE(images."images", '[]'::jsonb) AS "images",
   COALESCE(sns.urls, '[]'::jsonb) AS "urls"
 
@@ -56,7 +56,7 @@ LEFT JOIN LATERAL (
 
 -- 来場者向けタグ情報（注意事項）を結合
 LEFT JOIN LATERAL (
-  SELECT array_agg(t.text ORDER BY cpt.id ASC) AS note
+  SELECT array_agg(t.text ORDER BY cpt.id ASC) AS notes
   FROM public."CPTag" AS cpt
   JOIN public."Tag" AS t
     ON t.id = cpt."tagId"

app/search/engine.py

@@ -169,7 +169,7 @@ class SearchEngine:
             p["projectId"]: normalize_text_for_org(p.get("circleName") or "")
             for p in self.projects
         }
-        self.circleNameKana_norms = {
+        self.reading_norms = {
             p["projectId"]: normalize_text_for_org(p.get("circleNameKana") or "")
             for p in self.projects
         }
@@ -641,8 +641,8 @@ class SearchEngine:
             details.append(
                 {
                     "projectId": ids[idx],
-                    "organization": project.get("organization"),
-                    "title": project.get("title"),
+                    "circleName": project.get("circleName"),
+                    "name": project.get("name"),
                     "bm25": float(bm25[idx]),
                     "ws_filter_topk": float(ws_filter[idx]),
                     "ws_rerank_pairavg": float(ws_rerank[idx])

config/files.json

@@ -8,8 +8,8 @@
         "stopwords": "resources/stopwords.json"
     },
     "projects": {
-        "original_csv": "resources/projects_original.csv",
-        "projects_json": "data/generated/projects.json"
+        "projects_json": "data/generated/projects.json",
+        "output_json": "data/generated/projects.json"
     },
     "bm25": {
         "bm25_meta": "data/generated/bm25_meta.json",

config/search_model.json

@@ -10,8 +10,8 @@
     "circleName",
     "circleNameKana",
     "description",
-    "prComment",
-    "prCommentLong"
+    "prSummary",
+    "prDetail"
   ],
   "synonyms": {
     "enable": true,
@@ -38,8 +38,8 @@
       "circleName": 1.5,
       "circleNameKana": 0.6,
       "description": 1.0,
-      "prComment": 1.0,
-      "prCommentLong": 0.8
+      "prSummary": 1.0,
+      "prDetail": 0.8
     }
   },
   "word_sim": {

docs/project_schema_change_points.md

@@ -6,8 +6,7 @@
 - The API layer (`app/main.py`) serves the DB-derived models directly, while the search engine (`app/search/engine.py`) consumes the statically generated artefacts.
 
 ## Config (`config/files.json`)
-- `config/files.json:11-12` still references `projects.original_csv`; this key can be retired once the CSV path is unused.
-- Add an explicit output mapping (e.g. `projects.output_json`) so scripts resolve the generated JSON path without relying on hard-coded fallbacks. Update all callers to reference the new key.
+- Remove the unused `projects.original_csv` entry and expose an explicit `projects.output_json` mapping so scripts resolve the generated JSON path without relying on hard-coded fallbacks.
 
 ## Batch Scripts (`scripts/`)
 ### `scripts/2_create_projects_data.py`
@@ -33,7 +32,7 @@
 - Ensure the summary payload exposes `notes` (renamed from `note`) and the PR fields `prSummary` / `prDetail` as returned by the repository.
 
 ## Search Engine (`app/search/engine.py`)
-- In `SearchEngine.initialize` (`app/search/engine.py:164-175`) we populate kana norms into `self.circleNameKana_norms`, but the boosting logic reads `self.reading_norms` (`app/search/engine.py:545`). Assign the kana norms into `self.reading_norms` so substring boosts still consider kana.
+- In `SearchEngine.initialize` (`app/search/engine.py:164-175`) ensure kana norms are stored directly in `self.reading_norms` so substring boosts continue to consider kana values during scoring.
 - Debug responses at `app/search/engine.py:643-645` reference `organization` / `title`; switch these to `circleName` / `name` so diagnostics stay meaningful with the new schema.
 - Rebuild all search assets (`projects.json`, `tf_token.json`, `bm25_meta.json`, `substring_index.json`) after applying the script updates above.
 

schemas/projects.py

@@ -60,7 +60,7 @@ class Project(BaseModel):
     prSummary: str
     prDetail: Optional[str]
     notes: Optional[list[str]]
-    images: list[Image]
+    images: Optional[list[Image]]
     urls: list[Sns]
 
 
@@ -93,7 +93,7 @@ class ProjectSummary(BaseModel):
     description: str
     prSummary: str
     notes: Optional[list[str]]
-    image: Image
+    image: Optional[Image]
 
 
 class ProjectDetail(BaseModel):
@@ -128,7 +128,7 @@ class ProjectDetail(BaseModel):
     prSummary: str
     prDetail: Optional[str]
     notes: Optional[list[str]]
-    images: list[Image]
+    images: Optional[list[Image]]
     urls: list[Sns]
 
 

schemas/tf_token.py

@@ -11,8 +11,8 @@ class Fields(BaseModel):
     circleName: TfOfField
     circleNameKana: TfOfField
     description: TfOfField
-    prComment: TfOfField
-    prCommentLong: TfOfField
+    prSummary: TfOfField
+    prDetail: TfOfField
 
 
 class Project(BaseModel):

scripts/2_create_projects_data.py

@@ -28,8 +28,10 @@ def main():
         log.error(f"企画データの取得に失敗しました: {exc}")
         sys.exit(1)
 
+    project_dicts = [project.model_dump(mode="json") for project in projects]
+
     os.makedirs(os.path.dirname(output_json_path), exist_ok=True)
-    json_dumps(projects, output_json_path)
+    json_dumps(project_dicts, output_json_path)
 
     log.info(f"企画データの生成が完了しました: {output_json_path}")
     sys.exit(0)

scripts/5_prepare_tf_token.py

@@ -106,12 +106,12 @@ def main():
             return field_objs.get(name, tf_token.TfOfField(len=0, tf={}))
 
         fields_obj = tf_token.Fields(
-            title=get_field("title"),
+            name=get_field("name"),
             circleName=get_field("circleName"),
             circleNameKana=get_field("circleNameKana"),
             description=get_field("description"),
-            prComment=get_field("prComment"),
-            prCommentLong=get_field("prCommentLong"),
+            prSummary=get_field("prSummary"),
+            prDetail=get_field("prDetail"),
         )
 
         project_entry = tf_token.Project(

scripts/7_prepare_circle_names.py

@@ -68,9 +68,9 @@ def main():
     circle_names = [
         {
             "projectId": p.projectId,
-            "circle": p.organization,
-            "circleNormalized": normalized_circle_name(p.organization),
-            "circleKana": p.reading,
+            "circle": p.circleName,
+            "circleNormalized": normalized_circle_name(p.circleName),
+            "circleKana": p.circleNameKana or "",
         }
         for p in projects
     ]

tests/repositories/test_projects_repository.py

@@ -1,3 +1,4 @@
+import json
 import os
 import sys
 import types
@@ -74,43 +75,42 @@ class ProjectsRepositoryTestCase(unittest.TestCase):
     def test_transform_row_maps_database_columns_to_api_keys(self):
         repo = ProjectsRepository()
         db_row = {
-            "project_id": "100",
-            "circle_name": "Circle A",
-            "circle_name_kana": "サークルエー",
+            "projectId": "100",
+            "circleName": "Circle A",
+            "circleNameKana": "サークルエー",
             "name": "Project A",
-            "category": "展示",
-            "first_day": True,
-            "second_day": False,
-            "third_day": True,
+            "category": "PERFORMANCE",
+            "firstDay": True,
+            "secondDay": False,
+            "thirdDay": True,
             "location": "Room 1",
             "description": "Description",
-            "pr_comment": "PR",
-            "pr_comment_long": "Long PR",
-            "note": None,
-            "image_name": "img.png",
-            "optional_image_name1": None,
-            "optional_image_name2": None,
-            "optional_image_name3": None,
-            "optional_image_name4": None,
-            "optional_image_name5": None,
-            "url_x": "https://x.com",
-            "url_instagram": None,
-            "url_official_website": None,
-            "url_youtube": None,
-            "url_others": None,
+            "prSummary": "PR",
+            "prDetail": "Long PR",
+            "notes": json.dumps(["注意事項"]),
+            "images": [
+                {"filename": "primary", "extension": "png", "order": 0},
+                {"filename": "sub", "extension": "png", "order": 1},
+            ],
+            "urls": json.dumps([
+                {"service": "X", "url": "https://x.com"},
+                {"service": "Website", "url": "https://example.com"},
+            ]),
         }
 
         api_row = repo._transform_row(db_row)
 
         self.assertEqual(api_row.projectId, "100")
         self.assertEqual(api_row.circleName, "Circle A")
-        self.assertEqual(api_row.category, "展示")
+        self.assertEqual(api_row.category, "パフォーマンス")
         self.assertTrue(api_row.firstDay)
-        self.assertIsNone(api_row.optionalImageName1)
+        self.assertEqual(api_row.notes, ["注意事項"])
+        self.assertEqual(api_row.images[0].filename, "primary")
+        self.assertEqual(api_row.urls[0].service, "X")
 
     def test_transform_row_handles_missing_columns(self):
         repo = ProjectsRepository()
-        db_row = {"project_id": "200"}
+        db_row = {"projectId": "200"}
 
         # 必須フィールドが欠けているため、ValidationError が発生するはず
         self.assertRaises(ValidationError, lambda: repo._transform_row(db_row))

tests/services/test_projects_service.py

@@ -82,20 +82,17 @@ class FakeProjectsRepository:
             "thirdDay": True,
             "location": "Room 1",
             "description": "Description",
-            "prComment": "PR",
-            "prCommentLong": "Long PR",
-            "note": None,
-            "imageName": "img.png",
-            "optionalImageName1": None,
-            "optionalImageName2": None,
-            "optionalImageName3": None,
-            "optionalImageName4": None,
-            "optionalImageName5": None,
-            "urlX": "https://x.com",
-            "urlInstagram": None,
-            "urlOfficialWebsite": None,
-            "urlYoutube": None,
-            "urlOthers": None,
+            "prSummary": "PR",
+            "prDetail": "Long PR",
+            "notes": ["注意事項"],
+            "images": [
+                {"filename": "img", "extension": "png", "order": 0},
+                {"filename": "img2", "extension": "png", "order": 1},
+            ],
+            "urls": [
+                {"service": "X", "url": "https://x.com"},
+                {"service": "Website", "url": "https://example.com"},
+            ],
         }
         projects = [Project(**p)]
         self._projects = projects