Open vonbraunbates opened 5 months ago
@vonbraunbates Do you think your ask is covered by this feature request? https://github.com/ministryofjustice/data-platform/issues/3137
@julialawrence I think they complement each other. An experienced programmer could happily do without VS code as long as there's a good shell on the AP.
This request asks for a low-level way to interact with AP data and tools -- the sort of terminal environment you'd experience via SSH: there's a modern shell, some essential command-line tools, everything is open source and free to use.
Whereas #3137 (if I read it correctly) is asking for a high-level program to write the scripts you run with this ask -- an enterprise IDE with lots of bells and whistles and all the package monitoring and licencing complexity attached.
Edwin to Contact Francesca Von Braun-Bates regarding using VS codse.
This issue is being marked as stale because it has been open for 60 days with no activity. Remove stale label or comment to keep the issue open.
Describe the feature request.
Allow users to have direct access to a POSIX-compliant shell on the list of AP tooling. Helpful criteria:
bash
orsh
overzsh
,fish
etc.bash v4.4+
nano
as an interim solution to issue 3137 (doubtless somene will wantvim
too...)yq
for YAML,jq
for JSON,csvkit
for CSVsThis is far more intuitive than starting R Studio just to use a terminal.
Describe the context.
Currently the shell is unintuitively accessed via one panel in R Studio. This is confusing for new users (why do we need a sledgehammer to crack a nut?). It is limiting for advanced users (there are no robust shell syntax highlighting nor
sh
code snippets in RStudio) which slows development.Standard data engineering tools are branching out from python and R, so RStudio is no longer an efficient environment in which to use these tools. Most notably Create A Derived Table -- using SQL, Jija, YAML and shell scripts -- is far easier to use in an SSH-like interface rather than one corner of the RStudio screen.
Value / Purpose
This would enable users to:
This would also streamline the UX experience between data engineers (using the terminal in macbooks) and data scientists to facilitate teaching and debugging. Presumably this would be easier to implement than issue 3137 and provide an interim improvement.
User Types
data scientists, data engineers, AP power users