nullの許容範囲を見直し

app/repositories/query.sql

@@ -69,4 +69,5 @@ LEFT JOIN (
 ) AS sns ON sns.circleProjectId = cp.id
 
 -- 除外条件：学部祭や実行委員会企画など
-WHERE cp.`projectType` NOT IN ('FACULTY', 'EXECUTIVE_COMMITTEE', 'OTHER', 'DEFAULT');
+WHERE cp.`projectType` NOT IN ('FACULTY', 'EXECUTIVE_COMMITTEE', 'OTHER', 'DEFAULT')
+  AND cp.`isArchived` = false;

schemas/projects.py

@@ -1,4 +1,4 @@
-from typing import Optional, Literal
+from typing import Literal
 from pydantic import BaseModel
 
 
@@ -47,7 +47,7 @@ class Project(BaseModel):
 
     projectId: str
     circleName: str
-    circleNameKana: Optional[str]
+    circleNameKana: str
     name: str
     category: Literal[
         "パフォーマンス", "飲食", "物販", "展示", "参加型", "音楽", "その他"
@@ -55,12 +55,12 @@ class Project(BaseModel):
     firstDay: bool
     secondDay: bool
     thirdDay: bool
-    location: str
-    description: str
-    prSummary: str
-    prDetail: Optional[str]
-    notes: Optional[list[str]]
-    images: Optional[list[Image]]
+    location: str | None
+    description: str | None
+    prSummary: str | None
+    prDetail: str | None
+    notes: list[str] | None
+    images: list[Image]
     urls: list[Sns]
 
 
@@ -89,11 +89,11 @@ class ProjectSummary(BaseModel):
     firstDay: bool
     secondDay: bool
     thirdDay: bool
-    location: str
-    description: str
-    prSummary: str
-    notes: Optional[list[str]]
-    image: Optional[Image]
+    location: str | None
+    description: str | None
+    prSummary: str | None
+    notes: list[str] | None
+    image: Image | None
 
 
 class ProjectDetail(BaseModel):
@@ -123,12 +123,12 @@ class ProjectDetail(BaseModel):
     firstDay: bool
     secondDay: bool
     thirdDay: bool
-    location: str
-    description: str
-    prSummary: str
-    prDetail: Optional[str]
-    notes: Optional[list[str]]
-    images: Optional[list[Image]]
+    location: str | None
+    description: str | None
+    prSummary: str | None
+    prDetail: str | None
+    notes: list[str] | None
+    images: list[Image]
     urls: list[Sns]
 
 

scripts/4_prepare_bm25f_meta.py

@@ -93,6 +93,8 @@ def main():
         seen_in_doc = set()
         for field in target_fields:
             text = p.get(field) or ""
+            if not text:
+                continue
             toks = tokenize(str(text), tok, mode, target_pos_l1, ban_list, stopwords)
             field_token_lens_sum[field] += len(toks)
             for t in set(toks):