[Ecls-list] Project status and changes (directions? help?)

juanjose.garciaripoll at gmail.com juanjose.garciaripoll at gmail.com
Fri Oct 25 07:14:19 UTC 2013


Just FYI, Sage (www.sagemath.org) project
tests a distribution of ECL as a part of Sage, mainly used to build/run
Maxima (as a library callable from Python, as well as a standalone
Maxima version). 


I am afraid we are confused about what “testing” here means. ECL has not broken any of its own tests for the last couple of years, except for minor mistakes on my commits. Thus ANSI tests and regression tests become quickly useless.


The problem is Common Lisp library testing, which is the only thing that hindrances adoption. This is a major issue and it is only solved by Anton’s testing platform, that checks for regressions among different versions of a lisp implementation using Quicklisp to test for all libraries.


Given that it typically takes about 12 hours to run such tests vs. the bare 5 minutes of running “make check” on ECL (something any developer can do after making a couple of changes), I would say it is not a task for regular users to do. OTOH it becomes essential to keep the lisp implementation up and running and useful for the general public.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.common-lisp.net/pipermail/ecl-devel/attachments/20131025/d21f3c10/attachment.html>


More information about the ecl-devel mailing list