d0dc94e36d
新功能: - 创建AI聊天助手组件（AIChatWidget） - 支持文字+图片多模态对话 - 集成Gemini AI API - 支持图片编辑和内容生成 - 实时对话历史管理 - 优雅的UI设计，支持黑夜模式 技术实现: - 后端API代理Gemini API（/api/ai-chat） - 支持最多5张图片上传 - 流式响应处理 - 对话历史管理 - Base64图片处理 配置: - 在.env中添加Gemini API配置 - GEMINI_API_KEY - GEMINI_API_BASE_URL - GEMINI_MODEL 替换原弹幕功能，提供更强大的AI助手体验！
106 lines
6.0 KiB
Python
Executable File
106 lines
6.0 KiB
Python
Executable File
# coding=utf-8
|
||
from __future__ import absolute_import
|
||
from __future__ import division
|
||
from __future__ import print_function
|
||
|
||
import os
|
||
|
||
from volcengine.tls.TLSService import TLSService
|
||
from volcengine.tls.tls_requests import *
|
||
|
||
|
||
if __name__ == "__main__":
|
||
# 初始化客户端，推荐通过环境变量动态获取火山引擎密钥等身份认证信息，以免AccessKey硬编码引发数据安全风险。详细说明请参考 https://www.volcengine.com/docs/6470/1166455
|
||
# 使用STS时，ak和sk均使用临时密钥，且设置VOLCENGINE_TOKEN；不使用STS时，VOLCENGINE_TOKEN部分传空
|
||
endpoint = os.environ["VOLCENGINE_ENDPOINT"]
|
||
region = os.environ["VOLCENGINE_REGION"]
|
||
access_key_id = os.environ["VOLCENGINE_ACCESS_KEY_ID"]
|
||
access_key_secret = os.environ["VOLCENGINE_ACCESS_KEY_SECRET"]
|
||
|
||
# 实例化TLS客户端
|
||
tls_service = TLSService(endpoint, access_key_id, access_key_secret, region)
|
||
now = str(int(time.time()))
|
||
|
||
# 创建日志项目
|
||
create_project_request = CreateProjectRequest(project_name="project-name-" + now, region=region,
|
||
description="project-description")
|
||
create_project_response = tls_service.create_project(create_project_request)
|
||
project_id = create_project_response.project_id
|
||
|
||
# 创建日志主题
|
||
create_topic_request = CreateTopicRequest(topic_name="topic-name-" + now, project_id=project_id,
|
||
ttl=3650, description="topic-description", shard_count=2)
|
||
create_topic_response = tls_service.create_topic(create_topic_request)
|
||
topic_id = create_topic_response.get_topic_id()
|
||
|
||
# 创建索引
|
||
full_text = FullTextInfo(case_sensitive=False, delimiter=",-;", include_chinese=False)
|
||
value_info_a = ValueInfo(value_type="text", delimiter="", case_sensitive=True,
|
||
include_chinese=False, sql_flag=True)
|
||
value_info_b = ValueInfo(value_type="long", delimiter="", case_sensitive=False,
|
||
include_chinese=False, sql_flag=True)
|
||
key_value_info_a = KeyValueInfo(key="key1", value=value_info_a)
|
||
key_value_info_b = KeyValueInfo(key="key2", value=value_info_b)
|
||
key_value = [key_value_info_a, key_value_info_b]
|
||
create_index_request = CreateIndexRequest(topic_id, full_text, key_value)
|
||
create_index_response = tls_service.create_index(create_index_request)
|
||
|
||
# 写入日志数据
|
||
# 建议您一次性聚合多条日志后调用一次put_logs_v2接口，以提高日志上传吞吐率
|
||
# 请根据您的需要，填写topic_id、source、filename和logs列表，建议您使用lz4压缩
|
||
# PutLogs API的请求参数规范和限制请参阅 https://www.volcengine.com/docs/6470/112191
|
||
logs = PutLogsV2Logs(source="192.168.1.1", filename="sys.log")
|
||
for i in range(100):
|
||
logs.add_log(contents={"key1": "value1-" + str(i + 1), "key2": "value2-" + str(i + 1)},
|
||
log_time=int(round(time.time())))
|
||
tls_service.put_logs_v2(PutLogsV2Request(topic_id, logs))
|
||
time.sleep(30)
|
||
|
||
# 查询日志直方图情况
|
||
# 请根据您的需要，填写topic_id、query、start_time、end_time、interval
|
||
# DescribeHistogramV1 API的请求参数规范请参阅 https://www.volcengine.com/docs/6470/1347899
|
||
start_time = int(round(time.time())) - 60
|
||
end_time = start_time + 60 * 2
|
||
# Deprecated, use DescribeHistogramV1 instead.
|
||
describe_histogram_request = DescribeHistogramRequest(topic_id, "", start_time, end_time, interval=None)
|
||
describe_histogram_response = tls_service.describe_histogram(describe_histogram_request)
|
||
describe_histogram_v1_request = DescribeHistogramV1Request(topic_id, "", start_time, end_time, interval=None)
|
||
describe_histogram_v1_response = tls_service.describe_histogram_v1(describe_histogram_v1_request)
|
||
|
||
# 查询消费游标
|
||
# 请根据您的需要，填写topic_id、shard_id和from_time
|
||
# DescribeCursor API的请求参数规范请参阅 https://www.volcengine.com/docs/6470/112193
|
||
describe_cursor_request = DescribeCursorRequest(topic_id, shard_id=0, from_time="begin")
|
||
describe_cursor_response = tls_service.describe_cursor(describe_cursor_request)
|
||
|
||
# 消费日志数据
|
||
# 请根据您的需要，填写topic_id、shard_id、cursor等参数
|
||
# ConsumeLogs API的请求参数规范请参阅 https://www.volcengine.com/docs/6470/112194
|
||
consume_logs_request = ConsumeLogsRequest(topic_id, shard_id=0, cursor=describe_cursor_response.cursor)
|
||
consume_logs_response = tls_service.consume_logs(consume_logs_request)
|
||
|
||
# 查询分析日志数据
|
||
# 请根据您的需要，填写topic_id、query、start_time、end_time、limit等参数值
|
||
# SearchLogs API的请求参数规范和限制请参阅 https://www.volcengine.com/docs/6470/112195
|
||
|
||
# 当您需要检索和分析日志时，推荐您使用Python SDK提供的search_logs_v2方法，下面的代码提供了具体的调用示例
|
||
# 查询日志数据（全文检索）
|
||
search_logs_request = SearchLogsRequest(topic_id, query="error", limit=10,
|
||
start_time=1672502400000, end_time=1688140800000)
|
||
search_logs_response = tls_service.search_logs_v2(search_logs_request)
|
||
|
||
# 查询日志数据（键值检索）
|
||
search_logs_request = SearchLogsRequest(topic_id, query="key1:error", limit=10,
|
||
start_time=1672502400000, end_time=1688140800000)
|
||
search_logs_response = tls_service.search_logs_v2(search_logs_request)
|
||
|
||
# 查询日志数据（SQL分析）
|
||
search_logs_request = SearchLogsRequest(topic_id, query="* | select key1, key2", limit=10,
|
||
start_time=1672502400000, end_time=1688140800000)
|
||
search_logs_response = tls_service.search_logs_v2(search_logs_request)
|
||
|
||
# 查询日志数据（SQL分析）
|
||
search_logs_request = SearchLogsRequest(topic_id, query="* | select key1, key2", limit=10,
|
||
start_time=1672502400000, end_time=1688140800000)
|
||
search_logs_response = tls_service.search_logs(search_logs_request)
|