kedro-cmd much faster than kedro cmd, should plugins adopt the former?

[astrojuanlu]

Recently @ankatiyar has been working on improving the startup performance of Kedro commands, see discussion at #1476

We discarded one of the solutions, which was lazy loading commands from plugins #3901 because we would lose the complete list of commands in kedro --help.

On the other hand, @noklam's personal project, kedro-inspect, shows really good startup times, possibly because it's launched with kedro-inspect rather than kedro inspect. I can confirm from anecdotal evidence that it's the same for my kedro-init.

It's well known that CLIs in Python are slow in general:

Consider the example of a Python command line program (CLI) with a number of subcommands. Each subcommand may perform different tasks, requiring the import of different dependencies. But a given invocation of the program will only execute a single subcommand, or possibly none (i.e. if just --help usage info is requested). Top-level eager imports in such a program will result in the import of many modules that will never be used at all; the time spent (possibly compiling and) executing these modules is pure waste.

To improve startup time, some large Python CLIs make imports lazy by manually placing imports inline into functions to delay imports of expensive subsystems. This manual approach is labor-intensive and fragile; one misplaced import or refactor can easily undo painstaking optimization work.

https://peps.python.org/pep-0690/#motivation (rejected)

And maybe the part of Kedro that collects commands from entry points can't really be optimized any further.

As a first step, should we encourage plugins to support the kedro-cmd version on top of the currently available kedro cmd version?

[noklam]

First thing affected will be telemetry data