On Sun, 18 Jun 2023, 17:51 Ralf Hemmecke, <r...@hemmecke.org> wrote:
> On 16.06.23 02:14, Waldek Hebisch wrote:
> > Well, I wrote "binary". The question is if sbcl version I use for
> > creation of binaries has all features needed by jfricas. It should,
> > but testing is better than bling faith.
>
> OK, but building SBCL 1.1.1 from the git repo does not work on my
> computer. Certainly someprerequisite missing. I am building with the
> default SBCL 2.1.11.debian on Ubuntu 22.04.2 LTS.
>
> Is there an easier way to get the sbcl version you wil be using?
>
>
> > As I wrote my current thinking is that hunchentoot is dependence
> > like other dependencies: it is responsibility of person doing build
> > to make sure that it is available. For convenience we can add
> > hsbcl tarball to the release area. And of course instruction
> > in INSTALL.
>
> If we go that way, then it seems as simple as asking the user to provide
> an sbcl image with hunchentoot included such that
>
> (require :asdf)(require :hunchentoot)
>
> can be called and we do not have to change anything in FriCAS. Oh wait,
> there was this thing with stdout.
>
>
> https://github.com/fricas/fricas/commit/7c1d3e8ea7ced544ecb57156415e7b8af64e098b#diff-953a91d70c9e61cba8f0e8f2f97d7141601a66d5d12985e181c9abbb4b8da7d2
>
> (see into the vmlisp.lisp part of the attached patch).
>
> When I jfricas starts webspad.lisp, then *OBEY-STDOUT* will be set to T
> and output goes the way it should go for jfricas in order to capture
> output that would normally go to stdout and otherwise be invisible in
> the notebook.
>
> >> In fact, may I ask you a favour? If you do changes that affects quite a
> big
> >> piece of code and could potentially break things, like the "$ --> %"
> change,
> >> please open a pull request at github so that others have the chance to
> check
> >> things they care about, before the patches become officially committed
> to
> >> the repo.
> >
> > Well, long time ago I tried to provide patches so that people can
> > test various changes. Feedback I received was almost empty,
> > so I rarely do this.
>
> Well, at least for me it feels troublesome to extract the patch from the
> mail, call patch and record it in my local git tree so that I know where
> I am. It would be much easier, if I could simply "git fetch hebisch" and
> look at your branches. Also for you it would be a simple "git push" to
> your github repository.
>
It need not be GitHub. Any way to fetch git branches (from a git server) or
patches which can be applied with "git apply" is infinity less error-prone
that mailing patches without any metadata which specifies a place in the
git tree.
>
> But true, for some things I do not have to say much, but the
> FriCAS-Aldor interface is something I care about.
>
> Ralf
>
> =========================================================
> ; wrote
> /home/hemmecke/v/git/sbcl/obj/from-host/src/code/early-type.fasl-tmp
> ; compilation finished in 0:00:00.236
> Unhandled SIMPLE-ERROR in thread #<SB-THREAD:THREAD "main thread" RUNNING
> {1001834363}>:
> FAILURE-P was set when creating
> "obj/from-host/src/code/early-type.fasl".
>
> Backtrace for: #<SB-THREAD:THREAD "main thread" RUNNING {1001834363}>
> 0: (SB-DEBUG::DEBUGGER-DISABLED-HOOK #<SIMPLE-ERROR "FAILURE-P was set
> when creating ~S." {10040CEC63}> #<unused argument> :QUIT T)
> 1: (SB-DEBUG::RUN-HOOK SB-EXT:*INVOKE-DEBUGGER-HOOK* #<SIMPLE-ERROR
> "FAILURE-P was set when creating ~S." {10040CEC63}>)
> 2: (INVOKE-DEBUGGER #<SIMPLE-ERROR "FAILURE-P was set when creating ~S."
> {10040CEC63}>)
> 3: (ERROR #<SIMPLE-ERROR "FAILURE-P was set when creating ~S."
> {10040CEC63}>)
> 4: (SB-KERNEL:WITH-SIMPLE-CONDITION-RESTARTS ERROR NIL "FAILURE-P was
> set when creating ~S." "obj/from-host/src/code/early-type.fasl")
> 5: (COMPILE-STEM "src/code/early-type" NIL :HOST-COMPILE)
> 6: (IN-HOST-COMPILATION-MODE #<FUNCTION (LAMBDA NIL :IN HOST-CLOAD-STEM)
> {10054D5F2B}>)
> 7: (HOST-CLOAD-STEM "src/code/early-type" NIL)
> 8: (LOAD-OR-CLOAD-XCOMPILER #<FUNCTION HOST-CLOAD-STEM>)
> 9: (SB-INT:SIMPLE-EVAL-IN-LEXENV (LOAD-OR-CLOAD-XCOMPILER (FUNCTION
> HOST-CLOAD-STEM)) #<NULL-LEXENV>)
> 10: (EVAL (LOAD-OR-CLOAD-XCOMPILER (FUNCTION HOST-CLOAD-STEM)))
> 11: (SB-EXT:INTERACTIVE-EVAL (LOAD-OR-CLOAD-XCOMPILER (FUNCTION
> HOST-CLOAD-STEM)) :EVAL NIL)
> 12: (SB-IMPL::REPL-FUN NIL)
> 13: ((LAMBDA NIL :IN SB-IMPL::TOPLEVEL-REPL))
> 14: (SB-IMPL::%WITH-REBOUND-IO-SYNTAX #<FUNCTION (LAMBDA NIL :IN
> SB-IMPL::TOPLEVEL-REPL) {10009BC53B}>)
> 15: (SB-IMPL::TOPLEVEL-REPL NIL)
> 16: (SB-IMPL::TOPLEVEL-INIT)
> 17: ((FLET SB-UNIX::BODY :IN SB-IMPL::START-LISP))
> 18: ((FLET "WITHOUT-INTERRUPTS-BODY-3" :IN SB-IMPL::START-LISP))
> 19: (SB-IMPL::START-LISP)
>
> unhandled condition in --disable-debugger mode, quitting
> deleted
> #P"/home/hemmecke/v/git/sbcl/obj/from-host/src/code/early-type.fasl-tmp"
> Command exited with non-zero status 1
> 4.02user 0.24system 0:06.83elapsed 62%CPU (0avgtext+0avgdata
> 118296maxresident)k
> 67240inputs+4336outputs (370major+69633minor)pagefaults 0swaps
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "FriCAS - computer algebra system" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to fricas-devel+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/fricas-devel/44993d81-51e9-db00-3063-186e8996a79a%40hemmecke.org
> .
>
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to fricas-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/fricas-devel/CAAWYfq1YUGCRWdiAG3tiLYSF%2BFJQUpUruYu3pgMLb22nJ%2BfeqA%40mail.gmail.com.
