lei_y601/sqlbot_agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat：修改ddql，增加qa问答

Browse Source
This commit is contained in:
雷雨
2025-09-28 16:44:58 +08:00
parent dc16bfaa8e
commit 6064710f4f
7 changed files with 396 additions and 641 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
.env
View File

@@ -2,11 +2,11 @@ IS_FIRST_LOAD=True
CHAT_MODEL_BASE_URL=https://api.siliconflow.cn
CHAT_MODEL_API_KEY=sk-iyhiltycmrfnhrnbljsgqjrinhbztwdplyvuhfihcdlepole
CHAT_MODEL_NAME=Qwen/Qwen3-Next-80B-A3B-Instruct
CHAT_MODEL_NAME=zai-org/GLM-4.5
EMBEDDING_MODEL_BASE_URL=https://api.siliconflow.cn
EMBEDDING_MODEL_API_KEY=sk-iyhiltycmrfnhrnbljsgqjrinhbztwdplyvuhfihcdlepole
EMBEDDING_MODEL_NAME=Qwen/Qwen3-Embedding-8B
EMBEDDING_MODEL_NAME=BAAI/bge-m3
#向量数据库
#type:memory/remote,如果设置为remote，将IS_FIRST_LOAD 设置成false
@@ -30,7 +30,7 @@ MYSQL_DATABASE_DBNAME=test
#达梦数据库
DAMENG_DATABASE_HOST=10.254.192.191
DAMENG_DATABASE_HOST=10.254.193.63
DAMENG_DATABASE_PORT=5236
DAMENG_DATABASE_PASSWORD=SYSDBA
DAMENG_DATABASE_USER=SYSDBA
DAMENG_DATABASE_PASSWORD=
DAMENG_DATABASE_USER=ai_view

5
main_service.py
View File

@@ -50,6 +50,7 @@ def create_vana():
"model": config('CHAT_MODEL_NAME', default=''),
},
)
return vn
@@ -111,7 +112,7 @@ def generate_sql_2():
data["type"]="success"
return jsonify(data)
except Exception as e:
logger.error("generate sql failed:{e}")
logger.error(f"generate sql failed:{e}")
return jsonify({"type": "error", "error": str(e)})
@@ -170,7 +171,7 @@ def run_sql_2(id: str, sql: str):
)
except Exception as e:
logger.error("run sql failed:{e}")
logger.error(f"run sql failed:{e}")
return jsonify({"type": "sql_error", "error": str(e)})
if __name__ == '__main__':

11
service/cus_vanna_srevice.py
View File

@@ -17,7 +17,7 @@ from datetime import datetime
import logging
from util import train_ddl
logger = logging.getLogger(__name__)
import traceback
class OpenAICompatibleLLM(VannaBase):
def __init__(self, client=None, config_file=None):
VannaBase.__init__(self, config=config_file)
@@ -186,6 +186,9 @@ class OpenAICompatibleLLM(VannaBase):
try:
logger.info("Start to generate_sql_2 in cus_vanna_srevice")
question_sql_list = self.get_similar_question_sql(question, **kwargs)
if question_sql_list and len(question_sql_list)>2:
question_sql_list=question_sql_list[:1]
ddl_list = self.get_related_ddl(question, **kwargs)
#doc_list = self.get_related_documentation(question, **kwargs)
template = get_base_template()
@@ -194,7 +197,8 @@ class OpenAICompatibleLLM(VannaBase):
# --------基于提示词，生成sql以及图表类型
sys_temp = sql_temp['system'].format(engine=config("DB_ENGINE", default='mysql'), lang='中文',
schema=ddl_list, documentation=[train_ddl.train_document],
data_training=question_sql_list)
retrieved_examples_data=question_sql_list,
data_training=question_sql_list,)
logger.info(f"sys_temp:{sys_temp}")
user_temp = sql_temp['user'].format(question=question,
current_time=datetime.now().strftime('%Y-%m-%d %H:%M:%S'))
@@ -223,7 +227,8 @@ class OpenAICompatibleLLM(VannaBase):
logger.info("Finish to generate_sql_2 in cus_vanna_srevice")
return result
except Exception as e:
logger.info("cus_vanna_srevice failed-------------------")
logger.info("cus_vanna_srevice failed-------------------: ")
traceback.print_exc()
raise e
def generate_rewritten_question(self, last_question: str, new_question: str, **kwargs) -> str:

319
template.yaml
View File

@@ -8,151 +8,99 @@ template:
sql:
system: |
<Instruction>
你是"SQLBOT"，智能问数小助手，可以根据用户提问，专业生成SQL与可视化图表。
你当前的任务是根据给定的表结构和用户问题生成SQL语句、可能适合展示的图表类型以及该SQL中所用到的表名。
我们会在<Infos>块内提供给你信息，帮助你生成SQL：
<Infos>内有<db-engine><m-schema><terminologies>等信息；
其中，<db-engine>：提供数据库引擎及版本信息；
<m-schema>：以 M-Schema 格式提供数据库表结构信息；
<terminologies>：提供一组术语，块内每一个<terminology>就是术语，其中同一个<words>内的多个<word>代表术语的多种叫法，也就是术语与它的同义词，<description>即该术语对应的描述，其中也可能是能够用来参考的计算公式，或者是一些其他的查询条件
<sql-examples>：提供一组SQL示例，你可以参考这些示例来生成你的回答，其中<question>内是提问，<suggestion-answer>内是对于该<question>提问的解释或者对应应该回答的SQL示例
用户的提问在<user-question>内，<error-msg>内则会提供上次执行你提供的SQL时会出现的错误信息，<background-infos>内的<current-time>会告诉你用户当前提问的时间
你是"SQLBOT"，一个专业的智能问数助手。你的核心能力是根据用户的问题、数据库表结构以及相关背景信息，精准地生成SQL查询语句，并推荐合适的可视化图表类型。
为此，你需要仔细分析在 <Infos> 块内提供给你的信息，它们包括：
<db-engine>：数据库引擎及其版本，决定了SQL的语法规范。
<m-schema>：以M-Schema格式定义的数据库表结构，包括表名、字段名、字段类型、主键和注释以及values下该字段的枚举值。
<terminologies>：业务术语库。每个<terminology>包含一组同义词<words>和对应的描述<description>，它们是连接用户问题和数据字段的桥梁，请务必利用。
<retrieved-examples>：[RAG核心区] 通过检索与当前问题最相关的历史问答对。**这是最高优先级的s参考**，优先从中寻找与用户问题意图或表述最相似的案例来指导你生成SQL。
<sql-examples>：通用SQL示例库。当<retrieved-examples>中没有足够参考时，可在此处寻找相似的用法、函数模板或Join思路作为补充参考。
<documentation>：数据库或业务相关的补充文档。
<error-msg>：[可选] 上一次生成的SQL执行失败时的错误信息，用于修正和优化你的输出。
<background-infos>：[可选] 背景信息，如当前提问时间<current-time>等。
用户的提问位于 <user-question> 块内。
</Instruction>
你必须遵守以下规则:
<Principles>
<principle>优先级遵循：<retrieved-examples> > <sql-examples> > <terminologies> > <m-schema>。历史成功经验是你的第一指南。</principle>
<principle>理解意图：仔细分析用户问题，结合背景信息，准确识别查询的指标、维度、筛选条件和时间范围。</principle>
<principle>安全第一：严格限制为只读查询（SELECT）。绝不允许生成任何修改、删除或危害数据库数据的SQL（如 INSERT, UPDATE, DELETE, DROP, TRUNCATE 等）。</principle>
<principle>忠于事实：严禁编造 <m-schema> 中未提供的表、字段或关系。</principle>
</Principles>
<Rules>
---
--- A. 核心格式与结构规则
---
<rule>
请使用语言：{lang} 回答，若有深度思考过程，则思考过程也需要使用 {lang} 输出
<rule-title>返回格式</rule-title>
<rule-detail>输出必须是严格的JSON格式。</rule-detail>
<rule-detail>若成功生成SQL，格式为：{{"success": true, "sql": "生成的SQL语句", "tables": ["表名1", "表名2", ...], "chart-type": "图表类型"}}</rule-detail>
<rule-detail>若因任何原因无法生成SQL，格式为：{{"success": false, "message": "清晰说明无法生成的原因 (例如: 问题与数据库不相关 / 缺少必要的表或字段 / 问题意图不明确)"}}</rule-detail>
</rule>
<rule>
你只能生成查询用的SQL语句，不得生成增删改相关或操作数据库以及操作数据库数据的SQL
<rule-title>语言要求</rule-title>
<rule-detail>使用 {lang} 语言进行所有输出，包括思考过程（如果有的话）。</rule-detail>
</rule>
---
--- B. SQL生成规范规则
---
<rule>
<rule-title>表与字段引用</rule-title>
<rule-detail>必须为每个表生成一个英文别名（不带 AS 关键字），例如：`FROM user u`。</rule-detail>
<rule-detail>查询字段禁止使用星号(*)，必须显式写出所有需要的字段名。</rule-detail>
<rule-detail>字段名和别名不能自动翻译，必须使用英文字符。</rule-detail>
<rule-detail>若数据库引擎是 PostgreSQL, Oracle, ClickHouse, 达梦数据库, AWS Redshift, Elasticsearch，则schema、表名、字段名、别名使用双引号，如 "schema_name"."table_name"。</rule-detail>
<rule-detail>若数据库引擎是 MySQL, Doris，则表名、字段名、别名使用反引号，如 `table_name`。</rule-detail>
<rule-detail>生成的SQL必须避免与数据库关键字冲突。</rule-detail>
</rule>
<rule>
如果因为客观原因无法生成sql，请合理分析无法生成的原因并反馈给用户
<rule-title>数据查询与排序</rule-title>
<rule-detail>若未明确指定查询字段，涉及人员信息时，默认返回相关性最强的前10个字段。</rule-detail>
<rule-detail>若查询字段为 VARCHAR 或 TEXT 类型但需要计算，必须先进行合理的类型转换（如 CAST(... AS NUMERIC)）。</rule-detail>
<rule-detail>若查询包含日期/时间字段：
- **默认行为**：若提问未指定排序，**默认按时间字段降序排序**（即最新数据在前）。
- **格式化**：若提问要求时间/日期/年月/年，且未指定格式，则分别格式化为 'yyyy-MM-dd HH:mm:ss' / 'yyyy-MM-dd' / 'yyyy-MM' / 'yyyy'，语法需适配当前数据库引擎。(达梦数据库如果时间字段是varchar类型也可以)
</rule-detail>
</rule>
<rule>
涉及查询人员信息时，如果用户没明确指出要查询哪些字段，主要查询相关性较强的10个字段即可，如果指定要查询所有信息，请返回所有字段信息
<rule-title>聚合与计算</rule-title>
<rule-detail>使用了聚合函数（如 COUNT(), SUM(), AVG()）的SQL，必须配置相应的 GROUP BY 子句。</rule-detail>
<rule-detail>使用了函数（如 COUNT(), CAST(), SUM()）的字段，必须为其指定一个英文别名。</rule-detail>
<rule-detail>计算占比或百分比时，结果保留两位小数，并以 '%' 符号结尾。示例：ROUND(COUNT(*) * 100.0 / (SELECT COUNT(*) FROM table), 2) || '%' (PostgreSQL语法)</rule-detail>
<rule-detail>若查询结果包含枚举字段（如 gender=1,2），必须使用 CASE WHEN 语句将其转换为可读的标签。示例: SELECT CASE WHEN "gender" = '1' THEN '男' WHEN "gender" = '2' THEN '女' END AS "gender"</rule-detail>
</rule>
<rule>
不要编造<m-schema>内没有提供给你的表结构
<rule-title>关联与限制</rule-title>
<rule-detail>多表关联时，优先使用 <m-schema> 中标记为 "Primary key"/"ID"/"主键" 的字段作为关联条件。</rule-detail>
<rule-detail>若用户未指定数据条数，**查询SQL必须包含1000条的限制**。若用户指定的限制大于1000，也按1000处理。
- PostgreSQL: ... LIMIT 1000
</rule-detail>
</rule>
---
--- C. 图表与业务理解规则
---
<rule>
<rule-title>图表类型选择</rule-title>
<rule-detail>若问题与图表展示无关，chart-type 一律使用 "table"。</rule-detail>
<rule-detail>若问题与图表展示相关，根据查询意图推荐最合适的图表类型，参考以下原则：
- **折线图**：展示数据随时间（或其他连续维度）的**趋势**。
- **柱状图/条形图**：展示不同**分类**之间的**数值对比**。柱状图常用于分类较少，条形图常用于分类较多或分类名较长。
- **饼图**：展示单一维度各部分占**整体的比例**，且分类不宜过多（建议少于7个）。
- **表格**：用于展示**详细的原始数据**，或用户明确要求查表的场景。
</rule-detail>
<rule-detail>返回的chart-type值必须是 table, column, bar, line, pie 中的一个。</rule-detail>
</rule>
<rule>
当需要计算的字段类型为varchar或者text时，请根据需求转换为合理的类型格式进行计算
</rule>
<rule>
生成的SQL必须符合<db-engine>内提供数据库引擎的规范
</rule>
<rule>
若用户提问中提供了参考SQL，你需要判断该SQL是否是查询语句
</rule>
<rule>
请注意区分'哪些'和'多少'的区别，哪些是指具体信息，多少是指数量，请注意甄别
</rule>
<rule>
如遇字符串类型的日期要计算时，务必转化为合理的日期格式进行计算
</rule>
<rule>
请使用JSON格式返回你的回答:
若能生成，则返回格式如：{{"success":true,"sql":"你生成的SQL语句","tables":["该SQL用到的表名1","该SQL用到的表名2",...],"chart-type":"table"}}
若不能生成，则返回格式如：{{"success":false,"message":"说明无法生成SQL的原因"}}
</rule>
<rule>
如果问题是图表展示相关，可参考的图表类型为表格(table)、柱状图(column)、条形图(bar)、折线图(line)或饼图(pie), 返回的JSON内chart-type值则为 table/column/bar/line/pie 中的一个
图表类型选择原则推荐：趋势 over time 用 line，分类对比用 column/bar，占比用 pie，原始数据查看用 table
</rule>
<rule>
如果问题是图表展示相关且与生成SQL查询无关时，请参考上一次回答的SQL来生成SQL
</rule>
<rule>
返回的JSON字段中，tables字段为你回答的SQL中所用到的表名，不要包含schema和database，用数组返回
</rule>
<rule>
提问中如果有涉及数据源名称或数据源描述的内容，则忽略数据源的信息，直接根据剩余内容生成SQL
</rule>
<rule>
根据表结构生成SQL语句，需给每个表名生成一个别名（不要加AS）
</rule>
<rule>
SQL查询中不能使用星号(*)，必须明确指定字段名
</rule>
<rule>
SQL查询的字段名不要自动翻译，别名必须为英文
</rule>
<rule>
生成sql时，如果返回字段中有枚举字段,请根据枚举字段选项值生成对应的case when语句
<example>
SELECT
CASE
WHEN gender = 1 THEN '男'
WHEN gender = 2 THEN '女'
ELSE gender
END AS gender,
COUNT(*) AS count
FROM person GROUP BY gender;
</example>
</rule>
<rule>
SQL查询的字段若是函数字段，如 COUNT(),CAST() 等，必须加上别名
</rule>
<rule>
SQL查询的如果用了聚合函数，如SUM(),COUNT()等，必须配合GROUP BY使用
</rule>
<rule>
计算占比，百分比类型字段，保留两位小数，以%结尾
</rule>
<rule>
生成SQL时，必须避免与数据库关键字冲突
</rule>
<rule>
如数据库引擎是 PostgreSQL、Oracle、ClickHouse、达梦（DM）、AWS Redshift、Elasticsearch，则在schema、表名、字段名、别名外层加双引号；
如数据库引擎是 MySQL、Doris，则在表名、字段名、别名外层加反引号；
如数据库引擎是 Microsoft SQL Server，则在schema、表名、字段名、别名外层加方括号。
<example>
以PostgreSQL为例，查询Schema为TEST表TABLE下前1000条id字段，则生成的SQL为：
SELECT "id" FROM "TEST"."TABLE" LIMIT 1000
- 注意在表名外双引号的位置，千万不要生成为:
SELECT "id" FROM "TEST.TABLE" LIMIT 1000
以Microsoft SQL Server为例，查询Schema为TEST表TABLE下前1000条id字段，则生成的SQL为：
SELECT TOP 1000 [id] FROM [TEST].[TABLE]
- 注意在表名外方括号的位置，千万不要生成为:
SELECT TOP 1000 [id] FROM [TEST.TABLE]
</example>
</rule>
<rule>
如果生成SQL的字段内有时间格式的字段:
- 若提问中没有指定查询顺序，则默认按时间升序排序
- 若提问是时间，且没有指定具体格式，则格式化为yyyy-MM-dd HH:mm:ss的格式
- 若提问是日期，且没有指定具体格式，则格式化为yyyy-MM-dd的格式
- 若提问是年月，且没有指定具体格式，则格式化为yyyy-MM的格式
- 若提问是年，且没有指定具体格式，则格式化为yyyy的格式
- 生成的格式化语法需要适配对应的数据库引擎。
</rule>
<rule>
生成的SQL查询结果可以用来进行图表展示，需要注意排序字段的排序优先级，例如：
- 柱状图或折线图：适合展示在横轴的字段优先排序，若SQL包含分类字段，则分类字段次一级排序
</rule>
<rule>
如果用户没有指定数据条数的限制，输出的查询SQL必须加上1000条的数据条数限制
如果用户指定的限制大于1000，则按1000处理
<example>
以PostgreSQL为例，查询Schema为TEST表TABLE下id字段，则生成的SQL为：
SELECT "id" FROM "TEST"."TABLE" LIMIT 1000
以Microsoft SQL Server为例，查询Schema为TEST表TABLE下id字段，则生成的SQL为：
SELECT TOP 1000 [id] FROM [TEST].[TABLE]
</example>
</rule>
<rule>
若需关联多表，优先使用<m-schema>中标记为"Primary key"/"ID"/"主键"的字段作为关联条件。
</rule>
<rule>
我们目前的情况适用于单指标、多分类的场景（展示table除外）
<rule-title>术语与问题解析</rule-title>
<rule-detail>充分利用 <terminologies>。<words> 中的词是用户可能使用的提问方式，<description> 中可能包含计算公式或精确的查询条件，是理解问题并将其翻译为SQL的关键。</rule-detail>
<rule-detail>注意区分“哪些”（具体信息）和“多少”（数量）的区别。</rule-detail>
<rule-detail>若用户提问中提及参考SQL，需先判断该SQL是否为一个合法的、只读的查询语句。</rule-detail>
<rule-detail>忽略问题中提到的“数据源名称”或“数据源描述”等无关信息，聚焦于核心的业务需求。</rule-detail>
</rule>
</Rules>
### 以下<example>帮助你理解问题及返回格式的例子，不要将<example>内的表结构用来回答用户的问题，<example>内的<input>为后续用户提问传入的内容，<output>为根据模版与输入的输出回答
### 以下 <example> 块帮助你理解问题及返回格式，**请勿将此块内的任何表结构用于回答用户的问题**。
<example>
<Info>
<db-engine> PostgreSQL17.6 (Debian 17.6-1.pgdg12+1) </db-engine>
<db-engine>达梦数据库</db-engine>
<m-schema>
【DB_ID】 Sample_Database, 样例数据库
【Schema】
@@ -167,109 +115,63 @@ template:
</m-schema>
<terminologies>
<terminology>
<words>
<word>GDP</word>
<word>国内生产总值</word>
</words>
<words><word>GDP</word><word>国内生产总值</word></words>
<description>指在一个季度或一年，一个国家或地区的经济中所生产出的全部最终产品和劳务的价值。</description>
</terminology>
<terminology>
<words>
<word>中国</word>
<word>中国大陆</word>
</words>
<description>查询SQL时若作为查询条件，将"中国"作为查询用的值</description>
<words><word>中国</word><word>中国大陆</word></words>
<description>查询SQL时若作为查询条件，将"中国"作为查询用的值。</description>
</terminology>
</terminologies>
</Info>
<chat-examples>
<example>
<input>
<user-question>今天天气如何？</user-question>
</input>
<output>
{{"success":false,"message":"我是智能问数小助手，我无法回答您的问题，该问题与当前数据库问数不相关，数据库中无天气等信息。","status":0}}
</output>
<input><user-question>今天天气如何？</user-question></input>
<output>{{"success":false,"message":"我是智能问数小助手，我无法回答您的问题。'天气'与当前数据库中的信息（如国家GDP）不相关，数据库中无天气相关数据表或字段。"}}</output>
</example>
<example>
<input>
<user-question>张三的年龄是多大</user-question>
</input>
<output>
{{"success":true,"sql":"SELECT name, FLOOR(MONTHS_BETWEEN(SYSDATE, birthday) / 12) AS age FROM YJOA_APPSERVICE_DB.t_pr3rl2oj_yj_person_database","tables":["t_pr3rl2oj_yj_person_database"],"chart-type":"columns"}}
</output>
<input><user-question>请清空数据库</user-question></input>
<output>{{"success":false,"message":"我的职责是进行数据查询。'清空数据库'属于破坏性操作，我无法生成此类SQL语句。"}}</output>
</example>
<example>
<input>
<user-question>请清空数据库</user-question>
</input>
<output>
{{"success":false,"message":"我是智能问数小助手，我只能查询数据，不能操作数据库来修改数据或者修改表结构。"}}
</output>
<background-infos><current-time>2025-08-08 11:23:00</current-time></background-infos>
<input><user-question>查询各个国家每年的GDP</user-question></input>
<output>{{"success":true,"sql":"SELECT \"country\" AS \"country_name\", \"continent\" AS \"continent_name\", \"year\" AS \"year\", \"gdp\" AS \"gdp\" FROM \"Sample_Database\".\"sample_country_gdp\" ORDER BY \"year\" DESC, \"country\" ASC LIMIT 1000","tables":["sample_country_gdp"],"chart-type":"line"}}</output>
</example>
<example>
<input>
<user-question>查询所有用户</user-question>
</input>
<output>
{{"success":false,"message":"抱歉，提供的表结构无法生成您需要的SQL"}}
</output>
<background-infos><current-time>2025-08-08 11:23:00</current-time></background-infos>
<input><user-question>使用饼图展示去年各个国家的GDP</user-question></input>
<output>{{"success":true,"sql":"SELECT \"country\" AS \"country_name\", \"gdp\" AS \"gdp\" FROM \"Sample_Database\".\"sample_country_gdp\" WHERE \"year\" = '2024' ORDER BY \"gdp\" DESC LIMIT 1000","tables":["sample_country_gdp"],"chart-type":"pie"}}</output>
</example>
<example>
<input>
<background-infos>
<current-time>
2025-08-08 11:23:00
</current-time>
</background-infos>
<user-question>查询各个国家每年的GDP</user-question>
</input>
<output>
{{"success":true,"sql":"SELECT \"country\" AS \"country_name\", \"continent\" AS \"continent_name\", \"year\" AS \"year\", \"gdp\" AS \"gdp\" FROM \"Sample_Database\".\"sample_country_gdp\" ORDER BY \"country\", \"year\" LIMIT 1000","tables":["sample_country_gdp"],"chart-type":"line"}}
</output>
</example>
<example>
<input>
<background-infos>
<current-time>
2025-08-08 11:23:00
</current-time>
</background-infos>
<user-question>使用饼图展示去年各个国家的GDP</user-question>
</input>
{{"success":true,"sql":"SELECT \"country\" AS \"country_name\", \"gdp\" AS \"gdp\" FROM \"Sample_Database\".\"sample_country_gdp\" WHERE \"year\" = '2024' ORDER BY \"gdp\" DESC LIMIT 1000","tables":["sample_country_gdp"],"chart-type":"pie"}}
<output>
</output>
</example>
<example>
<input>
<background-infos>
<current-time>
2025-08-08 11:24:00
</current-time>
</background-infos>
<user-question>查询今年中国大陆的GDP</user-question>
</input>
{{"success":true,"sql":"SELECT \"country\" AS \"country_name\", \"gdp\" AS \"gdp\" FROM \"Sample_Database\".\"sample_country_gdp\" WHERE \"year\" = '2025' AND \"country\" = '中国' LIMIT 1000","tables":["sample_country_gdp"],"chart-type":"table"}}
<output>
</output>
<background-infos><current-time>2025-08-08 11:24:00</current-time></background-infos>
<input><user-question>查询今年中国大陆的GDP</user-question></input>
<output>{{"success":true,"sql":"SELECT \"country\" AS \"country_name\", \"gdp\" AS \"gdp\" FROM \"Sample_Database\".\"sample_country_gdp\" WHERE \"year\" = '2025' AND \"country\" = '中国' LIMIT 1000","tables":["sample_country_gdp"],"chart-type":"table"}}</output>
</example>
</chat-examples>
</example>
### 下面是提供的信息
### --- 真实任务开始 ---
### 下面是为你提供的完整信息
<Info>
<db-engine>{engine}</db-engine>
<m-schema>
{schema}
</m-schema>
<documentation>
{documentation}
</documentation>
{data_training}
<m-schema>{schema}</m-schema>
<documentation>{documentation}</documentation>
<terminologies>
<terminology>
<words><word国网</word><word>电网</word><word>雅江</word><word>联通</word></words>
<description>这些都可能是内部或者外部单位的名称</description>
</terminology>
</terminologies>
<!-- [RAG 集成区] -->
<!-- 将从向量数据库/知识库中检索到的最相关的N个问答对放在这里 -->
<retrieved-examples>
{retrieved_examples_data}
</retrieved-examples>
</Info>
### 响应, 请根据上述要求直接返回JSON结果:
```json
@@ -282,6 +184,7 @@ template:
<user-question>
{question}
注意查询结果枚举值转换
</user-question>
chart:

7
util/load_ddl_doc.py
View File

@@ -1,8 +1,8 @@
from service.cus_vanna_srevice import CustomVanna
from util import train_ddl
from util import q_and_a_dict
table_ddls = [
train_ddl.ddl_sql,train_ddl.attendance_ddl
train_ddl.person_ddl_sql,train_ddl.rule_ddl,train_ddl.user_status_ddl
]
list_documentions = [
train_ddl.train_document,
@@ -17,3 +17,6 @@ def add_ddl(vn: CustomVanna):
def add_documentation(vn: CustomVanna):
for doc in list_documentions:
vn.add_documentation(doc)
for d in q_and_a_dict.question_and_answer:
vn.add_question_sql(question=d['question'], sql=d['answer'])

206
util/q_and_a_dict.py Normal file
View File

@@ -0,0 +1,206 @@
question_and_answer = [
{"question": "考勤地点有哪些",
"answer": '''
SELECT DISTINCT CASE "region"
WHEN '1' THEN '北京'
WHEN '2' THEN '成都'
WHEN '3' THEN '秭归'
WHEN '4' THEN '拉萨'
WHEN '5' THEN '林芝' END AS "考勤地点"
FROM "YJOA_APPSERVICE_DB"."t_yj_person_attendance_rules"
WHERE "region" IS NOT NULL LIMIT 1000
'''
},
{"question": "成都的考勤规则是什么",
"answer": '''
SELECT "region" AS "region_code",
CASE
WHEN "region" = '1' THEN '北京'
WHEN "region" = '2' THEN '成都'
WHEN "region" = '3' THEN '秭归'
WHEN "region" = '4' THEN '拉萨'
WHEN "region" = '5' THEN '林芝' END AS "region_name",
"morning_check_time" AS "morning_check_time",
"afternoon_check_time" AS "afternoon_check_time",
"before_lunch_time" AS "before_lunch_time",
"after_lunch_time" AS "after_lunch_time"
FROM "YJOA_APPSERVICE_DB"."t_yj_person_attendance_rules"
WHERE "region" = '2' LIMIT 1000
'''
},
{"question": "所有员工男女各有多少人",
"answer": '''
SELECT CASE WHEN "gender" = '1' THEN '男' WHEN "gender" = '2' THEN '女' END AS "gender",
COUNT(*) AS "person_count"
FROM "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database"
WHERE "dr" = 0
GROUP BY "gender"
ORDER BY "gender" ASC LIMIT 1000
'''
},
{
"question": "联通下面有哪些员工",
"answer": '''
SELECT "id" AS "id",
"code" AS "工号",
"name" AS "姓名",
"internal_unit" AS "内部单位",
"external_unit" AS "外部单位"
FROM "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database"
WHERE "internal_unit" LIKE '%联通%'
OR "external_unit" LIKE '%联通%' LIMIT 1000
'''
},
{
"question": "9月有多少哪些员工在休假",
"answer": '''
SELECT DISTINCT p."id" AS "id",
p."code" AS "工号",
p."name" AS "姓名",
p."internal_unit" AS "内部单位",
p."external_unit" AS "外部单位"
FROM "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database" p
INNER JOIN "YJOA_APPSERVICE_DB"."t_yj_person_status" ps ON p."code" = ps."person_id"
WHERE ps."status" = '1003'
AND ps."date_value" LIKE '2025-09%' LIMIT 1000
'''
},
{
"question": "联通的员工，8月份有哪些迟到，旷工的",
"answer": '''
SELECT DISTINCT p."code" AS "工号",
p."name" AS "姓名",
p."internal_unit" AS "内部单位",
p."external_unit" AS "外部单位",
CASE
WHEN ps."status" = '1006' THEN '迟到,早退'
WHEN ps."status" = '1005' THEN '旷工' END AS "人员状态",
ps."date_value" AS "日期"
FROM "YJOA_APPSERVICE_DB"."t_yj_person_status" ps
JOIN "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database" p ON ps."person_id" = p."code"
WHERE (p."internal_unit" LIKE '%联通%' OR p."external_unit" LIKE '%联通%')
AND ps."status" IN ('1005', '1006')
AND ps."date_value" LIKE '2025-08%'
AND ps."dr" = 0
ORDER BY ps."date_value" DESC LIMIT 1000
'''
},
{
"question": "博士和硕士分别有哪些员工",
"answer": '''
SELECT DISTINCT "name" AS "姓名",
"code" AS "工号",
"internal_unit" AS "内部单位",
"external_unit" AS "外部单位",
CASE
WHEN "highest_degree" = '1' THEN '学士学位'
WHEN "highest_degree" = '2' THEN '硕士学位'
WHEN "highest_degree" = '3' THEN '博士学位'
WHEN "highest_degree" = '4' THEN '无' END AS "最高学位"
FROM "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database"
WHERE "highest_degree" IN ('2', '3')
AND "dr" = '0'
ORDER BY "最高学位", "姓名" LIMIT 1000
'''
}, {
"question": "8月份在藏超过10天的有哪些员工",
"answer": '''
SELECT DISTINCT p."code" AS "工号",
p."name" AS "姓名",
CASE
WHEN p."internal_unit" IS NOT NULL AND p."internal_unit" != '' THEN p."internal_unit"
ELSE p."external_unit" END AS "单位",
COUNT(ps."id") AS "在藏天数"
FROM YJOA_APPSERVICE_DB."t_yj_person_status" ps
JOIN YJOA_APPSERVICE_DB."t_pr3rl2oj_yj_person_database" p ON ps."person_id" = p."code"
WHERE ps."is_in_tibet" = 1
AND ps."dr" = 0
AND p."dr" = 0
AND ps."date_value" LIKE '2025-08%'
GROUP BY p."code", p."name",
CASE
WHEN p."internal_unit" IS NOT NULL AND p."internal_unit" != '' THEN p."internal_unit"
ELSE p."external_unit" END
HAVING COUNT(ps."id") > 10
ORDER BY COUNT(ps."id") DESC LIMIT 1000
'''
},
{
"question": "张三 8月的考勤查询",
"answer": '''
SELECT DISTINCT p."code" AS "工号",
p."name" AS "姓名",
p."internal_unit" AS "内部单位",
p."external_unit" AS "外部单位",
CASE
WHEN ps."status" = '1001' THEN '在岗'
WHEN ps."status" = '1002' THEN '出差'
WHEN ps."status" = '1003' THEN '休假,请假'
WHEN ps."status" = '1005' THEN '旷工'
WHEN ps."status" = '1006' THEN '迟到,早退'
WHEN ps."status" = '1007' THEN '休息日'
WHEN ps."status" = '4001' THEN 'am在岗pm缺勤'
WHEN ps."status" = '4002' THEN 'am缺勤pm在岗'
WHEN ps."status" = '6001' THEN 'am在岗pm早退'
WHEN ps."status" = '6002' THEN 'am迟到pm在岗'
WHEN ps."status" = '6004' THEN 'am迟到pm缺勤'
WHEN ps."status" = '4006' THEN 'am缺勤pm早退' END AS "人员状态",
ps."date_value" AS "日期"
FROM "YJOA_APPSERVICE_DB"."t_yj_person_status" ps
JOIN "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database" p ON ps."person_id" = p."code"
WHERE p."name" = '张三'
AND ps."date_value" LIKE '2025-08%'
AND ps."dr" = 0
ORDER BY ps."date_value" DESC LIMIT 1000
'''
},
{
"question": "8月份有多人迟到",
"answer": '''
SELECT count(distinct person_id)
FROM "YJOA_APPSERVICE_DB"."t_yj_person_status" ps
WHERE ps."status" = '1006'
AND ps."date_value" LIKE '2025-08%'
AND ps."dr" = 0 LIMIT 1000
'''
},
{
"question": "负责智能体相关工作的是哪些员工",
"answer": '''
SELECT "id" AS "id",
"code" AS "工号",
"name" AS "姓名",
"internal_unit" AS "内部单位",
"external_unit" AS "外部单位",
"work_content" AS "工作内容"
FROM "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database"
WHERE "work_content" LIKE '%智能体%' LIMIT 1000
'''
},{
"question": "9月，旷工，迟到分别有多少人",
"answer": '''
SELECT CASE WHEN "status" = '1006' THEN '迟到' WHEN "status" = '1005' THEN '旷工' END AS "status_name",
COUNT(DISTINCT "person_id") AS "person_count"
FROM "YJOA_APPSERVICE_DB"."t_yj_person_status"
WHERE "status" IN ('1006', '1005')
AND "date_value" LIKE '2025-09%'
AND "dr" = 0
GROUP BY "status"
ORDER BY "status" LIMIT 1000
'''
},{
"question": "在研发基地工作的有哪些员工",
"answer": '''
SELECT "id" AS "id",
"code" AS "工号",
"name" AS "姓名",
"internal_unit" AS "内部单位",
"external_unit" AS "外部单位",
"office_address" AS "办公地点"
FROM "YJOA_APPSERVICE_DB"."t_pr3rl2oj_yj_person_database"
WHERE ("office_address" LIKE '%研发基地%' OR "office_city" LIKE '%研发基地%')
AND "dr" = '0' LIMIT 1000
'''
}
]

443
util/train_ddl.py
View File

@@ -1,5 +1,5 @@
ddl_sql = """
[
person_ddl_sql = """
{
"db_name":"YJOA_APPSERVICE_DB",
"table_name": "t_pr3rl2oj_yj_person_database",
@@ -17,7 +17,7 @@ ddl_sql = """
"5":"停用"
},
"role": "dimension",
"tags": ["状态信息"]
"tags": ["状态信息","枚举"]
},
{
"name": "gender",
@@ -28,7 +28,7 @@ ddl_sql = """
"2": "女"
},
"role": "dimension",
"tags": ["基本信息", "人口属性"]
"tags": ["基本信息", "人口属性","枚举"]
},
{
"name": "id_card",
@@ -74,7 +74,7 @@ ddl_sql = """
"0": "否"
},
"role": "dimension",
"tags": ["员工类型", "身份标识"]
"tags": ["员工类型", "身份标识","枚举"]
},
{
"name": "phone_number",
@@ -104,7 +104,7 @@ ddl_sql = """
"7":"来访人员",
},
"role": "dimension",
"tags": ["证件信息", "通行权限"]
"tags": ["证件信息", "通行权限","枚举"]
},
{
"name": "expire_time",
@@ -154,7 +154,7 @@ ddl_sql = """
"WQT":"其他外部人员"
},
"role": "dimension",
"tags": ["分类信息", "人员分类"]
"tags": ["分类信息", "人员分类","枚举"]
},
{
"name": "id",
@@ -163,13 +163,7 @@ ddl_sql = """
"role": "dimension",
"tags": ["主键", "ID标识"]
},
{
"name": "pubts",
"type": "DATETIME(39)",
"comment": "发布时间",
"role": "dimension",
"tags": ["时间信息", "系统字段"]
},
{
"name": "dr",
"type": "INT",
@@ -186,7 +180,7 @@ ddl_sql = """
"type": "VARCHAR(600)",
"comment": "编码",
"role": "dimension",
"tags": ["编码信息", "业务编码"]
"tags": ["编码信息", "工号"]
},
{
"name": "ytenant_id",
@@ -195,132 +189,7 @@ ddl_sql = """
"role": "dimension",
"tags": ["租户信息", "系统隔离"]
},
{
"name": "sourcegrand_id",
"type": "VARCHAR(108)",
"comment": "来源表id",
"role": "dimension",
"tags": ["数据来源", "关联信息"]
},
{
"name": "first_id",
"type": "VARCHAR(108)",
"comment": "来源单据主表id",
"role": "dimension",
"tags": ["数据来源", "关联信息"]
},
{
"name": "firstchild_id",
"type": "VARCHAR(108)",
"comment": "来源单据子表id",
"role": "dimension",
"tags": ["数据来源", "关联信息"]
},
{
"name": "firstbusiobj",
"type": "VARCHAR(108)",
"comment": "来源业务对象",
"role": "dimension",
"tags": ["业务对象", "数据来源"]
},
{
"name": "firstcode",
"type": "VARCHAR(600)",
"comment": "来源单据号",
"role": "dimension",
"tags": ["单据信息", "业务编码"]
},
{
"name": "verifystate",
"type": "INT",
"comment": "单据状态",
"role": "dimension",
"tags": ["审批状态", "业务流程"]
},
{
"name": "auditor",
"type": "VARCHAR(180)",
"comment": "终审审批人",
"role": "dimension",
"tags": ["审批信息", "操作人"]
},
{
"name": "audit_time",
"type": "DATETIME(39)",
"comment": "审批日期",
"role": "dimension",
"tags": ["时间信息", "审批流程"]
},
{
"name": "auditnote",
"type": "VARCHAR(600)",
"comment": "当前审批人",
"role": "dimension",
"tags": ["审批信息", "操作人"]
},
{
"name": "procinst_id",
"type": "VARCHAR(108)",
"comment": "流程实例ID",
"role": "dimension",
"tags": ["流程信息", "实例标识"]
},
{
"name": "bizflow_id",
"type": "VARCHAR(108)",
"comment": "业务流id",
"role": "dimension",
"tags": ["业务流程", "流标识"]
},
{
"name": "bizflowname",
"type": "VARCHAR(600)",
"comment": "流程名称",
"role": "dimension",
"tags": ["业务流程", "名称描述"]
},
{
"name": "source_id",
"type": "VARCHAR(108)",
"comment": "上游单据主表id",
"role": "dimension",
"tags": ["数据关联", "上游信息"]
},
{
"name": "sourcechild_id",
"type": "VARCHAR(108)",
"comment": "上游单据子表id",
"role": "dimension",
"tags": ["数据关联", "上游信息"]
},
{
"name": "bizflowinstance_id",
"type": "VARCHAR(108)",
"comment": "业务流实例id",
"role": "dimension",
"tags": ["业务流程", "实例标识"]
},
{
"name": "sourcebusiobj",
"type": "VARCHAR(108)",
"comment": "上游业务对象",
"role": "dimension",
"tags": ["业务对象", "上游信息"]
},
{
"name": "sourcecode",
"type": "VARCHAR(600)",
"comment": "上游单据号",
"role": "dimension",
"tags": ["单据信息", "业务编码"]
},
{
"name": "bizflow_makebillcode",
"type": "VARCHAR(600)",
"comment": "单据转换规则编码",
"role": "dimension",
"tags": ["业务流程", "规则编码"]
},
{
"name": "create_time",
"type": "DATETIME(39)",
@@ -335,27 +204,8 @@ ddl_sql = """
"role": "dimension",
"tags": ["时间信息", "系统记录"]
},
{
"name": "creator",
"type": "VARCHAR(180)",
"comment": "创建人",
"role": "dimension",
"tags": ["操作人信息", "系统记录"]
},
{
"name": "modifier",
"type": "VARCHAR(180)",
"comment": "修改人",
"role": "dimension",
"tags": ["操作人信息", "系统记录"]
},
{
"name": "worker_id",
"type": "VARCHAR(200)",
"comment": "工号",
"role": "dimension",
"tags": ["员工信息", "身份标识"]
},
{
"name": "to_dept",
"type": "VARCHAR(600)",
@@ -419,7 +269,7 @@ ddl_sql = """
"9":"博士 "
},
"role": "dimension",
"tags": ["教育信息", "学历背景"]
"tags": ["教育信息", "学历背景","枚举"]
},
{
"name": "highest_degree",
@@ -432,7 +282,7 @@ ddl_sql = """
"4":"无"
},
"role": "dimension",
"tags": ["教育信息", "学位背景"]
"tags": ["教育信息", "学位背景","枚举"]
},
{
"name": "graduate_school",
@@ -600,196 +450,26 @@ ddl_sql = """
"comment": "关联租户信息"
}
],
"examples": [
{
"question": "查询内部员工的数量",
"sql": "SELECT COUNT(*) as person_count FROM YJOA_APPSERVICE_DB.t_pr3rl2oj_yj_person_database WHERE is_internal = '1' AND dr = 0"
},
{
"question": "按性别统计人员分布情况",
"sql": "SELECT gender, COUNT(*) as person_count FROM YJOA_APPSERVICE_DB.t_pr3rl2oj_yj_person_database WHERE dr = 0 GROUP BY gender"
},
{
"question": "查询最近一个月新增的人员信息",
"sql": "SELECT * FROM YJOA_APPSERVICE_DB.t_pr3rl2oj_yj_person_database WHERE create_time >= ADD_MONTHS(TRUNC(SYSDATE), -1) AND dr = 0"
}
],
"tags": ["人员管理", "人力资源", "审批流程", "基本信息", "工作信息"],
},
{
"db_name":"YJOA_APPSERVICE_DB",
"table_name": "t_yj_person_attendance",
"table_comment": "人员考勤记录表，记录人员的考勤打卡信息",
"columns": [
{
"name": "person_name",
"type": "VARCHAR(50)",
"comment": "人员姓名",
"role": "dimension",
"tags": ["基本信息", "身份标识"]
},
{
"name": "id",
"type": "VARCHAR(200)",
"comment": "主键ID",
"role": "dimension",
"tags": ["主键", "ID标识"]
},
{
"name": "person_id",
"type": "VARCHAR(200)",
"comment": "人员ID",
"role": "dimension",
"tags": ["人员标识", "关联信息"]
},
{
"name": "phone_number",
"type": "VARCHAR(50)",
"comment": "手机号码",
"role": "dimension",
"tags": ["联系方式", "通讯信息"]
},
{
"name": "attendance_time",
"type": "DATETIME(39)",
"comment": "考勤时间",
"role": "dimension",
"tags": ["时间信息", "考勤记录"]
},
{
"name": "attendance_address",
"type": "VARCHAR(200)",
"comment": "考勤地点",
"role": "dimension",
"tags": ["位置信息", "考勤记录"]
},
{
"name": "status",
"type": "INT",
"comment": "状态",
"value":{
"1":"在岗",
"2":"出差",
"3":"休假"
},
"role": "dimension",
"tags": ["状态信息", "考勤状态"]
},
{
"name": "original_id",
"type": "VARCHAR(200)",
"comment": "原数据ID",
"role": "dimension",
"tags": ["数据来源", "原始标识"]
},
{
"name": "source",
"type": "VARCHAR(50)",
"comment": "数据来源",
"role": "dimension",
"tags": ["数据来源", "系统标识"]
},
{
"name": "dr",
"type": "INT",
"comment": "逻辑删除标志",
"role": "dimension",
"tags": ["系统状态", "数据状态"]
},
{
"name": "create_time",
"type": "DATETIME(39)",
"comment": "创建时间",
"role": "dimension",
"tags": ["时间信息", "系统记录"]
},
{
"name": "enter_or_exit",
"type": "INT",
"comment": "进出类型",
"value":{
"1":"进",
"2":"出"
},
"role": "dimension",
"tags": ["考勤类型", "进出标识"]
},
{
"name": "access_control_point",
"type": "VARCHAR(50)",
"comment": "门禁点",
"role": "dimension",
"tags": ["位置信息", "门禁设备"]
},
{
"name": "bv_st",
"type": "VARCHAR(20)",
"comment": "上午打卡时间",
"role": "dimension",
"tags": ["时间信息", "业务时间"]
},
{
"name": "bv_et",
"type": "VARCHAR(20)",
"comment": "下午打卡时间",
"role": "dimension",
"tags": ["时间信息", "业务时间"]
},
{
"name": "bv_st_field",
"type": "VARCHAR(50)",
"comment": "午休前打卡时间",
"role": "dimension",
"tags": ["时间信息", "业务字段"]
},
{
"name": "bv_et_field",
"type": "VARCHAR(50)",
"comment": "午休后打卡时间",
"role": "dimension",
"tags": ["时间信息", "业务字段"]
},
{
"name": "bv_go_type",
"type": "VARCHAR(8)",
"comment": "打卡类型",
"role": "dimension",
"tags": ["业务类型", "分类信息"]
},
],
"relationships": [
{
"from": "person_id",
"to_table": "t_pr3rl2oj_yj_person_database",
"to_field": "id",
"type": "foreign_key",
"comment": "关联人员基本信息表"
}
],
"examples": [
{
"question": "查询今日考勤总人次",
"sql": "SELECT COUNT(*) as attendance_count FROM YJOA_APPSERVICE_DB.t_yj_person_attendance WHERE TO_CHAR(attendance_time, 'YYYY-MM-DD') = TO_CHAR(SYSDATE, 'YYYY-MM-DD') AND dr = '0'"
},
{
"question": "按人员统计本月考勤次数",
"sql": "SELECT person_name, COUNT(*) as attendance_count FROM YJOA_APPSERVICE_DB.t_yj_person_attendance WHERE TO_CHAR(attendance_time, 'YYYY-MM') = TO_CHAR(SYSDATE, 'YYYY-MM') AND dr = '0' GROUP BY person_name, person_id"
},
{
"question": "查询最近一周的考勤记录",
"sql": "SELECT * FROM YJOA_APPSERVICE_DB.t_yj_person_attendance WHERE attendance_time >= SYSDATE - 7 AND dr = '0' ORDER BY attendance_time DESC"
},
{
"question": "统计各门禁点的考勤分布",
"sql": "SELECT access_control_point, COUNT(*) as attendance_count FROM YJOA_APPSERVICE_DB.t_yj_person_attendance WHERE dr = '0' GROUP BY access_control_point"
}
],
"tags": ["考勤管理", "人员考勤", "门禁记录", "时间统计", "考勤分析"]
},
"""
train_document='''
语法为达梦数据库语法;
查询地址，籍贯，公司，单位时，尽量使用like查询;
查询人员信息时,由于数据表字段过多。只需要展示人员关键信息字段，id，工号，姓名，单位以及用户问题中需要查询的字段;
表字段信息以及字段枚举信息在values下，注意相关字段枚举值的转换;
查询单位信息时，内部单位和外部单位都需要查询，用OR条件查询;
'''
rule_ddl='''
{
"db_name":"YJOA_APPSERVICE_DB",
"table_name": "t_yj_person_attendance_rules",
"table_comment": "人员考勤规则表，定义考勤时间规则和区域设置",
"table_comment": "人员考勤规则表，定义考勤时间规则，考勤地点设置",
"columns": [
{
"name": "id",
@@ -838,7 +518,7 @@ ddl_sql = """
"5": "林芝"
},
"role": "dimension",
"tags": ["区域设置", "地理信息"]
"tags": [ "考勤的位置","非办公区域不要混淆","枚举"]
},
],
"relationships": [
@@ -850,26 +530,12 @@ ddl_sql = """
"comment": "关联区域配置信息"
}
],
"examples": [
{
"question": "查询所有考勤规则列表",
"sql": "SELECT * FROM t_yj_person_attendance_rules"
},
{
"question": "统计各区域的考勤规则数量",
"sql": "SELECT region, COUNT(*) as rule_count FROM YJOA_APPSERVICE_DB.t_yj_person_attendance_rules GROUP BY region"
},
{
"question": "查询特定区域的考勤时间设置",
"sql": "SELECT morning_check_time, afternoon_check_time, before_lunch_time, after_lunch_time FROM YJOA_APPSERVICE_DB.t_yj_person_attendance_rules WHERE region = '北京'"
},
{
"question": "检查是否存在重复的考勤规则",
"sql": "SELECT region, morning_check_time, afternoon_check_time, COUNT(*) as rule_count FROM YJOA_APPSERVICE_DB.t_yj_person_attendance_rules GROUP BY region, morning_check_time, afternoon_check_time HAVING COUNT(*) > 1"
}
],
"tags": ["考勤规则", "时间设置", "区域配置", "考勤管理", "规则定义"]
},
}
'''
user_status_ddl='''
{
"db_name":"YJOA_APPSERVICE_DB",
"table_name": "t_yj_person_status",
@@ -896,9 +562,9 @@ ddl_sql = """
"value":{
"1001":"在岗",
"1002":"出差",
"1003":"休假",
"1003":"休假,请假",
"1005":"旷工",
"1006":"迟到早退",
"1006":"迟到,早退",
"1007":"休息日",
"4001":"am在岗pm缺勤",
"4002":"am缺勤pm在岗",
@@ -908,7 +574,7 @@ ddl_sql = """
"4006":"am缺勤pm早退"
},
"role": "dimension",
"tags": ["状态信息", "人员状态"]
"tags": ["状态信息", "人员状态","枚举","迟到早退都是:1006"]
},
{
"name": "date_value",
@@ -943,41 +609,12 @@ ddl_sql = """
{
"from": "person_id",
"to_table": "t_pr3rl2oj_yj_person_database",
"to_field": "id",
"to_field": "code",
"type": "foreign_key",
"comment": "关联人员基本信息表"
}
],
"examples": [
{
"question": "查询今日人员状态记录数量",
"sql": "SELECT COUNT(*) as status_count FROM YJOA_APPSERVICE_DB.t_yj_person_status WHERE date_value = DATE_FORMAT(CURRENT_DATE(), '%Y-%m-%d') AND dr = 0"
},
{
"question": "统计各状态的人员分布",
"sql": "SELECT status, COUNT(*) as status_count FROM YJOA_APPSERVICE_DB.t_yj_person_status WHERE dr = '0' GROUP BY status"
},
{
"question": "查询在西藏地区的人员数量",
"sql": "SELECT COUNT(DISTINCT person_id) as tibet_person_count FROM YJOA_APPSERVICE_DB.t_yj_person_status WHERE is_in_tibet = 1 AND dr = 0"
},
{
"question": "按日期统计人员状态记录",
"sql": "SELECT date_value, status, COUNT(*) as status_count FROM YJOA_APPSERVICE_DB.t_yj_person_status WHERE dr = '0' GROUP BY date_value, status ORDER BY date_value DESC"
},
{
"question": "查询特定人员的状态历史记录",
"sql": "SELECT date_value, status, is_in_tibet FROM YJOA_APPSERVICE_DB.t_yj_person_status WHERE person_id = '123' AND dr = 0 ORDER BY date_value DESC"
}
],
"tags": ["人员状态", "状态记录", "地区管理", "西藏标识", "每日状态"]
}
]
"""
train_document='''
语法为达梦数据库语法;
查询地址，籍贯，公司，单位时，尽量使用like查询;
查询人员信息时,由于数据表字段过多。只需要展示人员关键信息字段，id，工号，姓名，单位以及用户问题中需要查询的字段;
表字段信息以及字段枚举信息:
'''