mariasql 3

# MariaSQL

Nowadays selecting data is easy, but inserting and updating data is annoying.

* https://git.osuv.de/m/MariaSQL
* Mirror, Issues & PRs: https://gitlab.com/markuman/MariaSQL

# install

with `pip3 install mariasql --user`

# Example / Usage

start a temporary MariaDB
`docker run -d -p 3307:3306 --rm --name mariadb -e MYSQL_ROOT_PASSWORD=password mariadb`

start python3 and go

```python
>>> import MariaSQL
>>> db = MariaSQL.MariaSQL(host='127.0.0.1', port=3307)
>>> db.query('create database mariasql;')
()
>>> db.use('mariasql')
>>> # we can create tables based on dict definitions
>>> mytable = dict()
>>> mytable['id'] = int
>>> mytable['name'] = str
>>> mytable['some shitty column name'] = float
>>> db.create_table('test_table', mytable)
()
>>> db.show_tables()
[{'Tables_in_mariasql': 'test_table'}]
>>> # now we inserting a dict into the create tables
>>> data = dict()
>>> data['id'] = 12
>>> data['name'] = 'Alf'
>>> data['some shitty column name'] = 3.1415
>>> db.insert('test_table', data)
()
# let's read from create table
>>> dataset = db.query('select * from test_table')
>>> dataset
[{'id': 12, 'name': 'Alf', 'some shitty column name': 3.1415}]
>>> db.query('show create table test_table')
[{'Table': 'test_table', 'Create Table': 'CREATE TABLE `test_table` (\n `id` int(11) DEFAULT NULL,\n `name` varchar(255) DEFAULT NULL,\n `some shitty column name` double DEFAULT NULL\n) ENGINE=InnoDB DEFAULT CHARSET=utf8'}]
```

# Properties

| variabele | default value |
| --- | --- |
| host | localhost |
| port | 3306 |
| user | root |
| password | password |
| db | mysql |
| charset | utf8mb4 |

# Methods

* `db.use(dbname)`
* to change the database
* `db.show_tables()`
* get list of current tables in selected database
* `db.query(sql)`
* execute raw sql commands
* `db.create_db(dbname)`
* create a new database if not exists
* `db.create_table(name, tabledef = None)`
* When `tabledef` is not given, `name` must be a raw sql string which will be executed
* When `tabledef` is a dict() and its `keys` are just datatype definitions, a table based on this dict will be created
* datatye limitations when using dict() definitions
* `str` -> `VARCHAR(255)`
* `int` -> `INT`
* `float` -> `DOUBLE`
* `dict` -> `JSON`
* `db.insert(table, data, on_duplicate = False)`
* `data` must be a `dict()` with `keys` which exists as `COLUMN_NAME` in the sql table. But it must not include all column names.
* when `on_duplicate` is set to `True`, it will perform an update when the PKs already exists in the table.
* `db.insert_on_duplicate(table, data)`
* just a wrapper for `db.insert()` with `on_duplicate = True`.

# todo

* some tests and setup ci/cd
* update method
* insert/update list of dicts