🍌 Banana Manager
Welcome to Banana Manager! Banana Manager is a Python package designed to connect to a database and create a simple web app that displays and allows updates to selected tables. This tool is perfect for non-technical end-users who need to interact with database tables without using complex DBA tools.
Powered by
- Dash and AG Grid: User-friendly, intuitive, and interactive web interface with powerful table displays and editing capabilities.
- Pydantic and YAML: Fast and accurate data handling and configuration.
- SQLAlchemy: Secure, efficient, and flexible database operations for multiple database backends.
Installation
To install Banana Manager, simply use pip:
pip install banana-manager
Also, remember to install the appropriate database connector for your project, like pyodbc
or psycopg2
. If you’re unsure, SQLAlchemy will inform you when you load your application.
Additionally, consider installing a production server like waitress
:
pip install waitress
Getting started
At the end of this tutorial, you’ll have a folder structure similar to the following:
my_manager
└─ app.py
└─ config.yaml
└─ my_tables
└─ my_group_of_tables.yaml
└─ another_group_of_tables.yaml
Configuring the Manager
Create a config.yaml
file in the root folder of your project with the following structure:
connection:
drivername: <optional str>
username: <optional str>
password: <optional str>
host: <optional str>
port: <optional str>
database: <optional str>
dataPath: "data"
tablePaths: ["tables"]
title: "Banana Database Manager"
theme: <optional str>
defaultColDef: <optional dict>
defaultGridOptions: <optional dict>
- connection (dict) : This will create a SQLAlchemy URL object. See the section Dialets for more information about the appropriate driver.
- dataPath (str, default="data") : The folder where the app data files will be stored.
- tablePaths (list[str], default=["tables"]) : List of folder where the table models YAML files are stored.
- title (str, default="Banana Database Manager") : HTML header title attribute.
- theme (str, default="cyan") : One of the default Mantine colors.
- defaultColDef (dict[str, Any]) : AG Grid column properties.
- defaultGridOptions (dict[str, Any]) : AG Grid options.
Defining the tables
The tables can be defined using YAML files located in the folders specified in the config.yaml
. If no folder is specified, create a new folder named "tables" in the root folder.
Each YAML file represents a group containing a list of tables. Here’s the structure:
groupName: <optional string>
displayOrder: <optional integer>
tables:
- name: <string>
schemaName: <optional string>
displayName: <optional string>
columns:
- name: <string>
displayName: <optional string>
dataType: (optional)
data: <optional dict>
type: <string>
columnDef: <optional dict>
- <other columns>
orderBy: (optional)
- column: <string>
desc: <optional bool>
- <other columns>
limit: <optional int>
defaultColDef: <optional dict>
gridOptions: <optional dict>
- <other tables>
Group configuration
- groupName (str, optional) : Name of the group that will be displayed in the side menu.
- displayOrder (int, optional) : Order which the groups will be stacked in the side menu.
- tables (list) : List of table configurations.
Table configuration
- name (str) : Name of the table in the database.
- schemaName (str, optional) : Schema where the table is located in the database.
- displayName : (str, optional) : Name that will be displayed at the side menu.
- columns (list) : List of column configurations.
- orderBy (list[dict]) : Default ordering of the table rows.
- limit (int) : Maximum of rows returned after order by.
- defaultColDef (dict[str, Any]) : AG Grid column properties.
- gridOptions (dict[str, Any]) : AG Grid options.
Order by configuration
- column (str) : Name of the column
- desc (bool, optional) : If True, order table by column in descending order.
Column configuration
- name (str) : Name of the column in the database.
- displayName (str, optional) : Name that will be displayed in the table.
- dataType (dict, optional) : Set the column to one of the special data types.
- columnDef (dict[str, Any]) : AG Grid column properties.
Data type configuration
The dataType
argument customizes how column data is interpreted and displayed. It consists of two keys:
-
type
(str): Specifies the data type. Available options are:
default
: Standard data type, no special handling.color
: Interprets the data as a color, rendering a visual color representation.enumerator
: Maps raw database values to user-friendly labels.foreign
: Creates a relationship with another table for enhanced data display.
-
data
(dict, optional): Contains additional information required for specific type
values. Each type requires a different data
structure:
default
: No data
required (ignored if provided).color
: No data
required (ignored if provided).enumerator
: A dictionary mapping raw database values (keys) to display labels (values). For example:
data: {"active": "Active", "inactive": "Inactive"}
Here, a database value of "active"
is displayed as "Active"
in the frontend.foreign
: Specifies the relationship details for joining with another table. Structure:
data:
tableName: <string>
schemaName: <string>
columnName: <string>
columnDisplay: <string>
Example:
data:
tableName: "users"
schemaName: "public"
columnName: "id"
columnDisplay: "username"
This configuration joins the current table with the users
table on the id
column, displaying the username
column instead of the raw id
.
About column definitions
You can set the columnDef
property in the Column
and PrimaryKey
models and set the defaultColDef
property in the Config
and Table
models.
The column definitions set in the Config
model are applied to every table and column in the application, but they will be overwritten by the definitions set in the Table
, and lastly overwritten by the definitions set in Column
and PrimaryKey
.
BananaManager does not offer support to every single definition of the original AG Grid, specially the Enterprise-only properties, but are some highlights that were fully tested:
- editable (bool) : Set to
true
if this column is editable, otherwise false
. - hide (bool) : Set to
true
for this column to be hidden. - sortable (bool) : Set to
false
to disable sorting which is enabled by default. - filter (bool) : Set to
true
to use the default filter component.
Load the application
Create an app.py file in the root folder:
from banana import Banana
app = Banana()
MY_PORT = 4000
if __name__ == "__main__":
app.run_server(port=MY_PORT)
This will load a development server in the selected port. Consider running a production server with waitress
:
from banana import Banana
from waitress import serve
app = Banana()
MY_PORT = 4000
if __name__ == "__main__":
serve(app.server, port=MY_PORT)
Roadmap
Version | Description | Release date |
---|
v0.1 | Load table and update cells | First half of July 2024 |
v0.2 | Table groups | Second half of July 2024 |
v0.3 | Logging and configurations | First half of August 2024 |
v0.4 | Insert rows and color themes | Second half of August 2024 |
v0.5 | Change history | September 2024 |
v0.6 | Color and enumerator data types | December 2024 |
v0.7 | Undo changes | First quarter of 2025 |
v0.8 | Delete rows | Second quarter of 2025 |
v0.9 | User authentication | 2025 |
v0.10 | Advanced user authorization | 2025 |
License
Banana Manager is released under the MIT License. See the LICENSE file for more details.