python.md

author: Martin Weise

{ tabindex=-1 }

tl;dr

:fontawesome-solid-cube:  View Docs{ .md-button .md-button--primary tabindex=-1 }

Installing

:octicons-tag-16:{ title="Minimum version" } 1.4.2

$ python -m pip install dbrepo

To use DBRepo in your Jupyter notebook, install the dbrepo library` directly in a code cell and type:

!pip install dbrepo

This package supports Python 3.11+.

Quickstart

Create a table and import a .csv file from your computer.

from dbrepo.RestClient import RestClient
from dbrepo.api.dto import CreateTableColumn, ColumnType, CreateTableConstraints

client = RestClient(endpoint='https://test.dbrepo.tuwien.ac.at', username="foo",
                    password="bar")

# analyse csv
analysis = client.analyse_datatypes(file_path="sensor.csv", separator=",")
print(f"Analysis result: {analysis}")
# -> columns=(date=date, precipitation=decimal, lat=decimal, lng=decimal), separator=,
#    line_termination=\n

# create table
table = client.create_table(database_id=1,
                            name="Sensor Data",
                            constraints=CreateTableConstraints(checks=['precipitation >= 0'],
                                                               uniques=[['precipitation']]),
                            columns=[CreateTableColumn(name="date",
                                                       type=ColumnType.DATE,
                                                       dfid=3,  # YYYY-MM-dd
                                                       primary_key=True,
                                                       null_allowed=False),
                                     CreateTableColumn(name="precipitation",
                                                       type=ColumnType.DECIMAL,
                                                       size=10,
                                                       d=4,
                                                       primary_key=False,
                                                       null_allowed=True),
                                     CreateTableColumn(name="lat",
                                                       type=ColumnType.DECIMAL,
                                                       size=10,
                                                       d=4,
                                                       primary_key=False,
                                                       null_allowed=True),
                                     CreateTableColumn(name="lng",
                                                       type=ColumnType.DECIMAL,
                                                       size=10,
                                                       d=4,
                                                       primary_key=False,
                                                       null_allowed=True)])
print(f"Create table result {table}")
# -> (id=1, internal_name=sensor_data, ...)

client.import_table_data(database_id=1, table_id=1, file_path="sensor.csv", separator=",",
                         skip_lines=1, line_encoding="\n")
print(f"Finished.")

The library is well-documented, please see the full documentation or the PyPI page.

Supported Features & Best-Practices

• Manage user account (docs)
• Manage databases (docs)
• Manage database access & visibility (docs)
• Import dataset (docs)
• Create persistent identifiers (docs)
• Execute queries (docs)
• Get data from tables/views/subsets

Secrets

It is not recommended to store credentials directly in the notebook as they will be versioned with git, etc. Use environment variables instead:

DBREPO_ENDPOINT=https://test.dbrepo.tuwien.ac.at
DBREPO_USERNAME=foo
DBREPO_PASSWORD=bar
DBREPO_SECURE=True

Then use the default constructor of the RestClient to e.g. analyse a CSV. Your secrets are automatically passed:

from dbrepo.RestClient import RestClient

client = RestClient()
analysis = client.analyse_datatypes(file_path="sensor.csv", separator=",")

Future

• Searching

Links

This information is also mirrored on PyPI.