Skip to content

FeatureBaseDB/getting-started

Repository files navigation

Getting Started

This repository contains the dataset and sample code for the Getting Started section of Pilosa documentation.

The Dataset

The sample dataset contains stargazer and language data for Github projects which were retrieved for the search keyword "Go". See the Generating the Dataset section below to create other datasets.

  • languages.txt: Language name to languageID mapping. The line number corresponds to the languageID.
  • language.csv: languageID, projectID
  • stargazer.csv: stargazerID, projectID, timestamp(starred)

Usage

Docker

Run the Pilosa Docker image with Getting Started data using:

docker run -it --rm -p 10101:10101 pilosa/getting-started:latest

Continue with Getting Started: Make Some Queries.

Without Docker

  1. Pilosa server should be running: Starting Pilosa
  2. The appropriate schema should be initialized: Create the Schema
  3. Finally, the data can be imported: Import Some Data

Continue with Getting Started: Make Some Queries.

Sample Projects

Generating the Dataset

Using a Github token is strongly recommended for avoiding throttling. If you don't already have a token for the GitHub API, see Creating a personal access token for the command line.

A recent version of Python is required. We test the script with 2.7 and 3.5.

Below are the steps to run commands:

  1. Create a virtual env:
    • Using Python 2.7: virtualenv getting-started
    • Using Python 3.5: python3 -m venv getting-started
  2. Activate the virtual env:
    • On Linux, MacOS, other UNIX: source getting-started/bin/activate
    • On Windows: getting-started\Scripts\activate
  3. Install requirements: pip install -r requirements.txt
  4. If you have a Github token, save it as token in the root directory of the project.

To generate csv files:

The fetch.py script searches Github for a given keyword and creates the dataset explained in The Dataset section.

Run the script: python fetch.py KEYWORD. KEYWORD is the search term to use for searching repository names.

Creating the Docker Image

make docker VERSION=some-version