On Fri, Dec 24, 2010 at 06:14, Felix <[email protected]> wrote:
[possible tinyclos issues] > I can't reproduce this myself (4.6.4). Trying the latest tinyclos > is a good idea. A sure way to pinpoint the problem is to use > a debug-build of chicken (DEBUGBUILD=1) and do: > > chicken-install -r tinyclos > cd tinyclos > chicken-install -k > gdb csi > b C_dbg_hook > run > (use tinyclos) > bt There is one thing here that bothers me a bit... According to http://code.call-cc.org/, the current Chicken release is 4.6.0. That is also the release I have installed on this machine. So, when I install an egg (without specifying a version), I would expect to get a version of that egg that works for 4.6.0, or maybe an error message if there is no such version. Instead, the behavior seems to be, that a version of said egg is downloaded which does not work under 4.6.0, but which does work for later versions. This is a confusing situation. If I use the stable version 4.6.0, I cannot use tinyclos (and maybe other eggs, I don't know; it is not my intention to single out tinyclos here). On the other hand, I am reluctant to use 4.6.1 or newer, because that seems to be a development version, which is likely to have recent and undocumented (so far) changes, and maybe more egg breakage. Isn't there some sort of mechanism in place that handles issues like this? (E.g. if I use Chicken 4.6.0, download a version of egg 'foo' that works for it. If I use a bleeding edge version of Chicken, download the development version of that egg. Etc.) Maybe there is, but I am not aware of it? Or is there something else I am misunderstanding or doing wrong? Thanks, -- Hans Nowak _______________________________________________ Chicken-users mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/chicken-users
