chifisource / ipycells.jl

cells, pluto, ipython, and olive readers and writers

IPyCells

IPyCells provides parametric cell-based functionality, as well as readers and writers for different cell formats (Ipynb, JL), as well as offering the option to extend the cells via parametric typing. This module provides

cells

• AbstractCell
• Cell(n::Int64, type::String, content::String, outputs::Any = ""; id::String = "")
• string(Cell{<:Any})
• string(cell::Cell{:md})
• string(cell::Cell{:doc})
• getindex(v::Vector{Cell{<:Any}}, s::String)

read/write

• read_plto(uri::String)
• read_jlcells(uri::String)
• read_jl(uri::String)
• save(cells::Vector{<:AbstractCell}, path::String)
• save_ipynb(cells::Vector{<:AbstractCell}, path::String) (this does not work just right yet) cells are readable by Olive, not jupyter post-save.
• read_ipynb(f::String)
• ipyjl(ipynb_path::String, output_path::String)

(internal)

• plto_cell_lines(uri::String)
• sep(content::Any)

Adding

julia> ]
pkg> add IPyCells

Usage

There are many ways to use IPyCells -- This package could be used to convert Pluto notebooks into Olive notebooks, IPython notebooks into Julia notebooks. Currently, the ipynb save method will break your .ipynb files where IJulia cannot read them, Olive eventually this is planned to be fixed. Anyway, this package could be used to read any package and save it into (currently) ipynb or julia.

ipynbjl("ipynbtestbook.ipynb", "example.jl")

cells = read_ipynb
save_jl(cells)

This preserves both the output and markdown. Alternatively, you could write functions around cells enabling for different cell types to be read by this reader.

ipycells.jl's Issues

Export to Pluto?

Doc revamp

IPy.jl 0.1.0 Release

.ipynb saving breaks notebook for IJulia

The printed JSON seems to be pretty similar to the two. It seems IJulia is trying to call get on something, we have provided a String but evidently this is not the type... It also says str, so it might mean Python String... But that could be because it's a str PyCall object. Either way, yes this is causing problems.

Reading files will need do statement

Reading files should be done better.

need Pluto file reader

We need a Pluto file reader, and for all of the types to be synonymous with the cells, also with all outputs available

Performance update

Need reader for new Julia files

Cells from IPy .jl needs to be written

Cells read using ends for read_jl

It would be nice to not only seperate by /ns, but concatenate In the case that there is still an open end. It would have to be length of find all for everything with an end, and then length findall for end. Then if the two aren't equal we concat the next line (using enumerate) if we cannot find a final end in the next cell, we check the next cell, totalling the ends the whole time. It's certainly a little annoying, but it will make the cell reader work a whole lot better on non-cell formatted Julia files.

md Does not save number

Make tests

Better README