Pranav Thulasiram Bhat

# RDMBS The use of [JDBC](https://github.com/JuliaDB/JDBC.jl) or [ODBC](https://github.com/JuliaDB/ODBC.jl) should make this fairly straightforward. The current graph definition is: ``` julia type Graph nv::Int ne::Int indxs::SparseMatrixCSC{Int,Int} vdata::AbstractDataFrame edata::AbstractDataFrame lmap::LabelMap end ```...

# GraphDB There seems to be a wrapper for Neo4j already : https://github.com/glesica/Neo4j.jl. `Neo4j.jl` uses HTTP requests to interface with the `Neo4j` server. This is quite slow, and returns text...

I've been looking for this functionality too! @alexhultman I can pass in a PR for this if you'd like

@1st1 Your thoughts on pycapsule vs int?

The build failure looks unrelated.

Maybe we can pretend that the row doesn't exist? That is, if the data column is set to default, then avoid returning the row in `getindex` and in queries. (SparseMatrixCSC...

What will `compact` do?

Ah okay. But I think removing the entries may be a problem. Because, for each entry, we'd have one `deleteat!` operation per column in `Indexes` and one `deleteat!` in the...

Yeah that sounds good. I'll try implementing this soon.

:+1: