Skip to content

strolid/strolid_vcon

 
 

Repository files navigation

The Home Repo for vCons and the Conserver

Introduction

vCons are PDFs for human conversations, defining them so they can be shared, analyzed and secured. The Conserver is a domain specific data platform based on vCons, converting the raw materials of recorded conversations into self-serve data sources for any team. The Conserver represents the most modern set of tools for data engineers to responsibly and scalably use customer conversations in data pipelines.

The Vcon library consists of two primary components:

  • The Python Vcon package for constructing and operating on Vcon objects
  • The Conserver for storing, managing and manipulating Vcon objects and operation streams on Vcon objects

Table of Contents

Presentations, Whitepapers and Tutorials

See the presentation at TADSummit

See the presentation at IETF

See the presentation at IIT

Read the IETF draft proposal

Read the white paper

See the key note proposal for vCons.

Testing the Vcon Package

A suite of pytest unit tests exist for the Vcon package in: tests

These can be run using the following command in the current directory:

pytest -v -rP tests

Testing the conserver

A suite of pytest unit tests exist for the conserver in: server/tests

Running and testing the conserver requires a running instance of Redis. Be sure to create and edit your server/.env file to reflect your redis server address and port. It can be generated like the following command line:

cat <<EOF>.env
#!/usr/bin/sh
export AWS_BUCKET=vcon-storage
export AWS_KEY_ID=aaaaaaaaaaaaaaa
export AWS_SECRET_KEY=bbbbbbbbbbb
export DEEPGRAM_KEY=ccccccccccccc
export ENV=dev
export HOSTNAME=http://0.0.0.0:8000
export REDIS_URL=redis://172.17.0.4:6379
#export MONOREPO_DATABASE_URL=postgresql://password:userid@your.postgres.domain.com:5432/postgres
EOF

The unit tests for the conserver can be run using the following command in the server directory:

source .env
pytest -v -rP tests

Releases

No releases published

Packages

No packages published

Languages

  • Python 88.6%
  • HTML 8.7%
  • Shell 1.6%
  • Other 1.1%