idgraph

Provides IPython cell magic for dgraph remote interaction.

Installation

pip install idgraph

Why is this?

While dgraph provides several useful interfaces, I spend most of my time learning, exploring, and developing in Jupyter. Executable documentation is fantastic! This package abstractly provides cell magic for dgraph query, mutation, and alteration execution. You could do the same with curl or requests. But, the sensible defaults help make things less verbose and tedious. (It's inspired by my experience with itikz which proved really beneficial.)

Usage

The easiest way to understand how this works is by following (and cloning) the tutorial notebook. It's a projection dgraph's Tour of Dgraph: A Bigger Dataset.

Cheat Sheet

Load the extension with,

%load_ext idgraph

Then,

By default, %%dgraph assumes a query.
- %%dgraph --alter does an alteration
- %%dgraph --mutate does a mutation
By default, %%dgraph assumes localhost:8080
- %%dgraph --addr=remote-host:8080 overrides the default
- The DGRAPH_ADDR environmental variable overrides default if the --addr flag isn't set
By default, only the value associated with the data key in the response is shown.
- %%dgraph --full-resp shows the full response including metadata.
- %%dgraph --jmespath="query" allows you to extract part of the response with a jmespath query.
By default, each cell execution binds the extracted response to _dgraph and the full response to _dgraph_full
- %%dgraph --into=name will bind the extracted response to name and the full response to {name}_full
By default, the cell contents are executed.
- %%dgraph --skip skips execution.
  - Useful for mutations that are possibly dangerous on someone else's computer.
Jinja
- technically you can use templates in a directory. you almost certainly shouldnt though.

Credits