/lexy

📙 Lexy enables you to easily build and share data dictionaries to explain and document your data terminology using code.

Primary LanguagePythonMIT LicenseMIT

Lexy

test codecov PyPI Downloads Docs GitHub Discord

📙 Lexy enables you to easily build and share data dictionaries to explain and document your data terminology using code. The name "Lexy" is inspired from lexicographer (/ˌlɛksɪˈkɒɡrəfə/), the person who compiles dictionaries.


Easily document your data objects and generate beautiful data dictionaries:

import lexy as xy
glossary = xy.Glossary()

#Defining glossary terms
glossary("name", "name of the student")
glossary("lastname", "lastname of the student")
glossary("age", "age of the student", sensitivity="private")

#Using the glossary to define pandas dataframe
import pandas as pd
data = [['tom', 'bird', 10], ['nick', 'star', 15], ['juli', 'aston', 14]] 
df = pd.DataFrame(data, columns = [glossary('name'), glossary('lastname'), glossary('age')]) 

xy.display_docs(glossary)

Displayed docs

Beloved Features

Lexy will be soon ready for your use-case:

  • ✔ Clear standard way to define data dictionaries using code.
  • ✔ Tracking of glossary usage throughout the code.
  • ✔ Display / Generate of documentation pages for your data glossaries.
  • ✔ Detection of similarity between the terms and warning about possible data dictionary issues.
  • Validation of data dictionaries using defined templates and rules.
  • ✔ Import / export data dictionary from different formats (csv, excel, etc)
  • AI Suggesting of metadata based on name and definition (personal data, data types, ...)
  • Support for multiple backends (Memory, File, Redis, CloudFile...)
  • Integration with Apache Atlas and Azure Purview.
  • Publish data dictionary to lexyHub using the cli.