简单易用的SQL执行引擎。
- 可在本地运行MySQL语法结构的SQL
- 支持查询常用mysql、mongodb、postgresql、sqlserver、elasticsearch、influxdb、clickhouse、sqlite数据库及execl、csv、json和普通文本文件
- 支持本地临时数据表逻辑做中间结果保存
- 数据库数据加载使用简单条件过滤及IN条件查询
- 因由本地完成Join匹配所以支持不同库表、不同主机及不同类型数据库间Join关联查询
- Group By分组聚合计算及Order By排序也由本地执行,保证数据库安全性
- 数据写Insert Into支持 ”仅插入 I“、”存在更新否则插入 UI“、”存在更新否则插入其余删除 UDI“、”删除后插入 DI“四种模式
- 大数据量支持批次执行,有Group By或Having条件过滤自动执行Reduce合并结果
- 支持流式执行
pip3 install syncanysql
docker pull sujin190/syncany-sql
-- 查询访问量最高的三个IP
SELECT seg0 AS ip, COUNT(*) AS cnt FROM `file://data/access.log?sep= ` GROUP BY seg0 ORDER BY cnt DESC LIMIT 3;
SELECT
a.site_id,
b.name AS site_name,
IF(c.site_amount > 0, c.site_amount, 0) AS site_amount,
MAX(a.timeout_at) AS timeout_at,
MAX(a.vip_timeout_at) AS vip_timeout_at,
now() as `created_at?`
FROM
(SELECT
YIELD_ARRAY(sites) AS site_id,
IF(vip_type = '2', GET_VALUE(rules, 0, 'timeout_time'), '') AS timeout_at,
IF(vip_type = '1', GET_VALUE(rules, 0, 'timeout_time'), '') AS vip_timeout_at
FROM
`data/demo.json`
WHERE
start_date >= '2021-01-01') a
JOIN
`data/sites.json` b ON a.site_id = b.site_id
JOIN
(SELECT
site_id, SUM(amount) AS site_amount
FROM
`data/orders.json`
WHERE
status <= 0
GROUP BY site_id) c ON a.site_id = c.site_id
GROUP BY a.site_id;
from syncanysql import ScriptEngine
with ScriptEngine() as engine:
engine.execute('''
INSERT INTO `top_ips` SELECT
ip, cnt
FROM
(SELECT
seg0 AS ip, COUNT(*) AS cnt
FROM
`file:///var/log/nginx/access.log?sep= `
GROUP BY seg0) a
ORDER BY cnt DESC
LIMIT 3;
''')
print(engine.pop_memory_datas("top_ips"))
Syncany-SQL uses the MIT license, see LICENSE file for the details.