On May 14, 2020, at 11:11 AM, Tom Tromey <t...@tromey.com> wrote: > >>>>>> "Rob" == Rob Savoye <r...@welcomehome.org> writes: > > Rob> Not that team, the folks I talked to thought I was crazy for wanting > Rob> to refactor it. :-) > > I don't think refactoring dejagnu is crazy, but I think it's pretty hard > to imagine rewriting the gdb test suite in Python. It's 260 KLOC.
So, TCL is subject to being easy to parse, and if you can reliable move each feature to a new system with a re-engineering style system that is complete enough to handle converting code from TCL to python, for example; one merely needs to complete the work for a few of the odd corner cases one might use. At some point, I do think as an industry, we do need tools to migrate code from system to system, updating the language used. C++ may well fall outside of the possibility for the next 30-90 years, but TCL, lisp and python might not be so unreasonable in a shorter timeframe. I one saw someone convert TCL into lisp I think it was, which I thought was neat. One day, would be nice if language implementors and designers implemented conversions into and out of their language from _the_ re-engineering toolkit as they did their language. 10 or 30 years after they decide, oh, no more support for you, you're dead, you can then migrate to the next new wiz bang language. Yes, I say this all, even knowing that people can't even do the python 2.7 -> 3.x conversion program yet. Anyway, love to have software that can move code wholesale. Love to move the testsuite into a new language.