zhaoqingliang/data-ge
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

数据知识回调入库

Browse Source
This commit is contained in:
zhaoawd
2025-11-04 20:28:50 +08:00
parent 0b765e6719
commit 7eb3c059a1
7 changed files with 346 additions and 111 deletions
Download Patch File Download Diff File Expand all files Collapse all files

91
table_snippet.sql
View File

@ -1,54 +1,37 @@
CREATE TABLE IF NOT EXISTS action_results (
id BIGINT NOT NULL AUTO_INCREMENT COMMENT '主键',
table_id BIGINT NOT NULL COMMENT '表ID',
version_ts BIGINT NOT NULL COMMENT '版本时间戳（版本号）',
action_type ENUM('ge_profiling','ge_result_desc','snippet','snippet_alias') NOT NULL COMMENT '动作类型',
status ENUM('pending','running','success','failed','partial') NOT NULL DEFAULT 'pending' COMMENT '执行状态',
error_code VARCHAR(128) NULL,
error_message TEXT NULL,
-- 回调 & 观测
callback_url VARCHAR(1024) NOT NULL,
started_at DATETIME NULL,
finished_at DATETIME NULL,
duration_ms INT NULL,
-- 本次schema信息
table_schema_version_id BIGINT NOT NULL,
table_schema JSON NOT NULL,
-- ===== 动作1：GE Profiling =====
ge_profiling_full JSON NULL COMMENT 'Profiling完整结果JSON',
ge_profiling_full_size_bytes BIGINT NULL,
ge_profiling_summary JSON NULL COMMENT 'Profiling摘要（剔除大value_set等）',
ge_profiling_summary_size_bytes BIGINT NULL,
ge_profiling_total_size_bytes BIGINT NULL COMMENT '上两者合计',
ge_profiling_html_report_url VARCHAR(1024) NULL COMMENT 'GE报告HTML路径/URL',
-- ===== 动作2：GE Result Desc =====
ge_result_desc_full JSON NULL COMMENT '表描述结果JSON',
ge_result_desc_full_size_bytes BIGINT NULL,
-- ===== 动作3：Snippet 生成 =====
snippet_full JSON NULL COMMENT 'SQL知识片段结果JSON',
snippet_full_size_bytes BIGINT NULL,
-- ===== 动作4：Snippet Alias 改写 =====
snippet_alias_full JSON NULL COMMENT 'SQL片段改写/丰富结果JSON',
snippet_alias_full_size_bytes BIGINT NULL,
-- 通用可选指标
result_checksum VARBINARY(32) NULL COMMENT '对当前action有效载荷计算的MD5/xxhash',
created_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,
updated_at TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
PRIMARY KEY (id),
UNIQUE KEY uq_table_ver_action (table_id, version_ts, action_type),
KEY idx_status (status),
KEY idx_table (table_id, updated_at),
KEY idx_action_time (action_type, version_ts),
KEY idx_schema_version (table_schema_version_id)
) ENGINE=InnoDB
ROW_FORMAT=DYNAMIC
COMMENT='数据分析知识片段表';
CREATE TABLE `action_results` (
`id` bigint NOT NULL AUTO_INCREMENT COMMENT '主键',
`table_id` bigint NOT NULL COMMENT '表ID',
`version_ts` bigint NOT NULL COMMENT '版本时间戳（版本号）',
`action_type` enum('ge_profiling','ge_result_desc','snippet','snippet_alias') COLLATE utf8mb4_bin NOT NULL COMMENT '动作类型',
`status` enum('pending','running','success','failed','partial') COLLATE utf8mb4_bin NOT NULL DEFAULT 'pending' COMMENT '执行状态',
`llm_usage` json DEFAULT NULL COMMENT 'LLM token usage统计',
`error_code` varchar(128) COLLATE utf8mb4_bin DEFAULT NULL,
`error_message` text COLLATE utf8mb4_bin,
`started_at` datetime DEFAULT NULL,
`finished_at` datetime DEFAULT NULL,
`duration_ms` int DEFAULT NULL,
`table_schema_version_id` varchar(19) COLLATE utf8mb4_bin NOT NULL,
`table_schema` json NOT NULL,
`ge_profiling_json` json DEFAULT NULL COMMENT 'Profiling完整结果JSON',
`ge_profiling_json_size_bytes` bigint DEFAULT NULL,
`ge_profiling_summary` json DEFAULT NULL COMMENT 'Profiling摘要（剔除大value_set等）',
`ge_profiling_summary_size_bytes` bigint DEFAULT NULL,
`ge_profiling_total_size_bytes` bigint DEFAULT NULL COMMENT '上两者合计',
`ge_profiling_html_report_url` varchar(1024) COLLATE utf8mb4_bin DEFAULT NULL COMMENT 'GE报告HTML路径/URL',
`ge_result_desc_json` json DEFAULT NULL COMMENT '表描述结果JSON',
`ge_result_desc_json_size_bytes` bigint DEFAULT NULL,
`snippet_json` json DEFAULT NULL COMMENT 'SQL知识片段结果JSON',
`snippet_json_size_bytes` bigint DEFAULT NULL,
`snippet_alias_json` json DEFAULT NULL COMMENT 'SQL片段改写/丰富结果JSON',
`snippet_alias_json_size_bytes` bigint DEFAULT NULL,
`callback_url` varchar(1024) CHARACTER SET utf8mb4 COLLATE utf8mb4_bin NOT NULL,
`result_checksum` varbinary(32) DEFAULT NULL COMMENT '对当前action有效载荷计算的MD5/xxhash',
`created_at` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP,
`updated_at` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
PRIMARY KEY (`id`),
UNIQUE KEY `uq_table_ver_action` (`table_id`,`version_ts`,`action_type`),
KEY `idx_status` (`status`),
KEY `idx_table` (`table_id`,`updated_at`),
KEY `idx_action_time` (`action_type`,`version_ts`),
KEY `idx_schema_version` (`table_schema_version_id`)
) ENGINE=InnoDB AUTO_INCREMENT=53 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin ROW_FORMAT=DYNAMIC COMMENT='数据分析知识片段表';