In nbstats, add size statistics for sources and outputs
It can be useful to know something about the distribution of the source size of markdown, code (and its outputs), and raw cells in a Jupyter notebook.
The appropriate tool to report this is
Keep in mind that the
source attribute in a Jupyter notebook can be either a single string (possibly multi-line through embedded
'\n'), or a list of (typically, one-line) strings.
outputs attribute is a list of output elements, classified as
error(with error name)
Minimum, maximum, total, average, median, possibly also 25% and 75% quartiles; overall and per cell type.
Also report about empty (code/markdown) cells.