These scripts are written for quite specific use, and it is unclear how easy it is to customize them into new uses, but in principle they deal with very ordinary use cases and could be useful. The logic should be such that reimplementing it in some other programmic languages should be easy as well.

All R packages needed to run this package can be installed with command:

install.packages('tidyverse')

The script can be run from Terminal with command:

Rscript get_annotation_lengths.R path/to/the/elan_file.eaf

It should return something like this:

[1] "Segments with annotations: 16M 49.467S"
[1] "Segments with English translations: 12M 48.074S"
[1] "Segments with no Cyrillic characters: 16M 40.542S"
[1] "Segments which contain Cyrillic characters: 8.925S"

This script is more complex exploration into FCPXML file format, with the goal of having tools to parse the files, transform the information from them into different outputs, and then also to plot and visualize it somehow. Eventually also the content from the ELAN files could be merged into the plots.

As one can see, there are some problems left in the time values, of course the video camera files should never overlap etc. But the direction is good anyway.