topcat

topological concatenation of files

描述

topcat 是一个简单的工具，用于按拓扑顺序连接文件。当您有一组相互依赖的文件，并且希望按正确顺序连接它们时，它非常有用。

对于我的用例，这是 SQL 文件。

我喜欢将我的 SQL 文件视为相互依赖的函数和视图集合。我喜欢将它们保存在单独的文件中，并按正确顺序连接它们，以创建一个可以在我的数据库中运行的单一文件。

用法

假设您有一个包含以下文件的目录

sql
├── my_other_schema
│   ├── functions
│   │   ├── a.sql
│   │   ├── b.sql
│   │   └── c.sql
│   └── schema.sql
└── my_schema
    ├── functions
    │   └── a.sql
    └── schema.sql

文件内容如下

sql/my_schema/schema.sql:

-- name: my_schema

DROP SCHEMA IF EXISTS my_schema CASCADE;
CREATE SCHEMA IF NOT EXISTS my_schema;

sql/my_schema/functions/a.sql:

-- name: my_schema.a
-- dropped_by: my_schema

CREATE FUNCTION my_schema.a() RETURNS INT AS
$$
SELECT 1;
$$ LANGUAGE SQL IMMUTABLE
                PARALLEL SAFE;

sql/my_schema/functions/b.sql:

-- name: my_schema.b
-- dropped_by: my_schema
-- requires: my_schema.a

CREATE FUNCTION my_schema.b() RETURNS INT AS
$$
SELECT my_schema.a() + 1
$$ LANGUAGE SQL;

sql/my_schema/functions/c.sql:

-- name: my_schema.c
-- dropped_by: my_schema
-- requires: my_schema.b

CREATE FUNCTION my_schema.c() RETURNS INT AS
$$
SELECT my_schema.b() + 1
$$ LANGUAGE SQL IMMUTABLE
                PARALLEL SAFE;

sql/my_other_schema/schema.sql:

-- name: my_other_schema

DROP SCHEMA IF EXISTS my_schema CASCADE;
CREATE SCHEMA IF NOT EXISTS my_schema;

sql/my_other_schema/functions/a.sql:

-- name: my_other_schema.a
-- dropped_by: my_other_schema
-- requires: my_schema.b

CREATE FUNCTION my_other_schema.a() RETURNS INT AS
$$
SELECT my_schema.b() + 1
$$ LANGUAGE SQL IMMUTABLE
                PARALLEL SAFE;

因此，依赖图如下：

现在您可以运行 topcat 以按正确顺序连接文件

topcat -i tests/input/sql -o tests/output/sql/output.sql

output.sql 的内容将是

-- This file was generated by topcat. To regenerate run:
--
-- topcat -i tests/input/sql -o tests/output/sql/output.sql -v

------------------------------------------------------------------------------------------------------------------------
-- tests/input/sql/my_other_schema/schema.sql
-- name: my_schema

DROP SCHEMA IF EXISTS my_schema CASCADE;
CREATE SCHEMA IF NOT EXISTS my_schema;

------------------------------------------------------------------------------------------------------------------------
-- tests/input/sql/my_other_schema/functions/a.sql
-- name: my_schema.a
-- dropped_by: my_schema

CREATE FUNCTION my_schema.a() RETURNS INT AS
$$
SELECT 1;
$$ LANGUAGE SQL;

------------------------------------------------------------------------------------------------------------------------
-- tests/input/sql/my_other_schema/functions/b.sql
-- name: my_schema.b
-- dropped_by: my_schema
-- requires: my_schema.a

CREATE FUNCTION my_schema.b() RETURNS INT AS
$$
SELECT my_schema.a() + 1
$$ LANGUAGE SQL;

------------------------------------------------------------------------------------------------------------------------
-- tests/input/sql/my_schema/schema.sql
-- name: my_other_schema

DROP SCHEMA IF EXISTS my_other_schema CASCADE;
CREATE SCHEMA IF NOT EXISTS my_other_schema;

------------------------------------------------------------------------------------------------------------------------
-- tests/input/sql/my_schema/functions/a.sql
-- name: my_other_schema.a
-- dropped_by: my_other_schema
-- requires: my_schema.b

CREATE FUNCTION my_other_schema.a() RETURNS INT AS
$$
SELECT my_schema.b() + 1
$$ LANGUAGE SQL IMMUTABLE
                PARALLEL SAFE;

------------------------------------------------------------------------------------------------------------------------
-- tests/input/sql/my_other_schema/functions/c.sql
-- name: my_schema.c
-- dropped_by: my_schema
-- requires: my_schema.b
-- requires: my_other_schema.a

CREATE FUNCTION my_schema.c() RETURNS INT AS
$$
SELECT my_schema.b() + my_other_schema.a() + 1
$$ LANGUAGE SQL;