>>>>> "Andreas" == Andreas Franke <[EMAIL PROTECTED]> writes:
Andreas> Hello,
Andreas> in 19c, ext:*command-line-switches* is still nil when the first
function
Andreas> from ext:*after-save-initializations* is run; by the time the
Andreas> read-eval-print loop is running, it works as advertised. Is there
an
Andreas> alternative way to access the path of the current lisp image (the
value
Andreas> of the "core" switch) at that early point, or possibly from a
later
Andreas> stage during the *after-save-initializations*?
>From looking at the save-lisp in save.lisp, the command line is
processed after all of the *after-save-initializations* have been
run. I suppose it's possible to reverse this order, but I don't know
what impact that has. Perhaps you can do a save-lisp with
:process-command-line nil, and have an *after-save-initializations*
function that is run early enough to call
(ext:process-command-strings).
Also, perhaps looking at lisp::*cmucl-core-path* might help. It's the
(string) path to lisp.core that cmucl used to find the core file. I
think if you specify a -core, it's exactly what you specified.
Ray
