Google 드라이브를 데이터 소스로 사용하는 Vertex AI Search 앱에서 Python의 검색 결과가 0이 되는 경우

from google.cloud.discoveryengine_v1alpha import SearchServiceClient, SearchRequest
from google.protobuf.json_format import MessageToDict
 
PROJECT_ID = "xxx"  # Google Cloud 프로젝트 ID
VERTEX_AI_APP_ID = "xxx"  # Vertex AI Search 앱 ID
 
client = SearchServiceClient(credentials=credentials)
 
serving_config = f "projects/{PROJECT_ID}/locations/global/collections/default_collection/engines/{VERTEX_AI_APP_ID}/servingConfigs/default_serving_config"
 
content_search_spec = SearchRequest.ContentSearchSpec (
    # 스 니펫을 출력하지 않음
    snippet_spec=SearchRequest.ContentSearchSpec().SnippetSpec(
        return_snippet= False
    ),
    # 요약문을 출력한다
    summary_spec=SearchRequest.ContentSearchSpec().SummarySpec(
        summary_result_count= 3 ,
        include_citations = False ,
        # Gemini Pro를 사용하도록 지정
        model_spec=SearchRequest.ContentSearchSpec().SummarySpec().ModelSpec(
            version= "gemini-1.5-flash-001/answer_gen/v1"
        )
    )
)
 
# Vertex AI Search에 검색어 던지기
response = client.search(
    SearchRequest (
        serving_config=serving_config,
        query= "G-gen이란?" ,
        page_size= 3 ,
        content_search_spec=content_search_spec
    )
)
 
# 요약 문을 표준 출력
print (response.summary.summary_text)
 
# 검색 결과를 표준 출력
for r in response.results:
    r_dct = MessageToDict(r._pb)
    print (r_dct)

from google.cloud.discoveryengine_v1alpha import SearchServiceClient, SearchRequest

# 지정되지 않음
from google.cloud.discoveryengine import SearchServiceClient, SearchRequest
 
# v1 지정
from google.cloud.discoveryengine_v1 import SearchServiceClient, SearchRequest
 
# v1beta 지정
from google.cloud.discoveryengine_v1beta import SearchServiceClient, SearchRequest

import requests
 
PROJECT_NUMBER = "xxx"  # Google Cloud 프로젝트 번호
VERTEX_AI_APP_ID = "xxx"  # Vertex AI Search 앱 ID
 
# API URL
url = f "https://discoveryengine.googleapis.com/v1alpha/projects/{PROJECT_NUMBER}/locations/global/collections/default_collection/engines/{VERTEX_AI_APP_ID}/servingConfigs/default_search"
 
# 요청 헤더
headers={
    "Authorization" : "Bearer " + credentials.token,
    "Content-Type" : "application/json" ,
}
 
# 요청 본문
session = f "projects/{PROJECT_NUMBER}/locations/global/collections/default_collection/engines/{VERTEX_AI_APP_ID}/sessions/-"
data={
    "query" : "G-gen이란?" ,
    "pageSize" : 3 ,
    "contentSearchSpec" : {
        "snippetSpec" : {
            "returnSnippet" : False
        },
        "extractiveContentSpec" : {
            "maxExtractiveAnswerCount" : 1
        }
    },
    "session" : session
}
 
# 검색 요청 제출
response = requests.post(f "{url}:search" , headers=headers, json=data)
 
# 검색 결과를 표준 출력
for r in response.json().get( "results" ):
    print (r)
 
data={
    "query" : {
        "text" : "G-gen이란?" ,
        "queryId" : response.json().get( "sessionInfo" ).get( "queryId" )
    },
    "session" : response.json().get( "sessionInfo" ).get( "name" ),
    "answerGenerationSpec" : {
        "modelSpec" : {
            "modelVersion" : "gemini-1.5-flash-001/answer_gen/v1"
        }
    }
}
 
# 요약 요청 제출
response = requests.post(f "{url}:answer" , headers=headers, json=data)
 
# 요약 문을 표준 출력
print (response.json().get( "answer" ).get( "answerText" ))