Greg.  Sorry

Le ven. 11 nov. 2022 à 03:45, Grégory Vanuxem <[email protected]> a
écrit :

> Hello,
>
> Can  you  give a diff
>
> Please
> __
> Hreg
>
> Le lun. 7 nov. 2022 à 04:37, Qian Yun <[email protected]> a écrit :
>
>> Hi, I'd like a review for this patch.
>>
>> After this patch, I can move on to "load GMP on Windows".
>>
>> - Qian
>>
>> On 10/22/22 10:15, Qian Yun wrote:
>> > The subset of pr#96 to solve issue#110 is this:
>> >
>> > diff --git a/src/interp/util.lisp b/src/interp/util.lisp
>> > index 63cd20f3..2a6219cd 100644
>> > --- a/src/interp/util.lisp
>> > +++ b/src/interp/util.lisp
>> > @@ -77,7 +77,7 @@ at load time.
>> >   ;;; It is set up in the {\bf reroot} function.
>> >   (defvar $library-directory-list ())
>> >
>> > -;;; Prefix a filename with the {\bf FRICAS} shell variable.
>> > +;;; Prefix a filename with the {\bf $spadroot} variable.
>> >   (defun make-absolute-filename (name)
>> >    (concatenate 'string $spadroot name))
>> >
>> > @@ -88,13 +88,10 @@ the system. In particular, these variables are
>> > sensitive to the
>> >   be {\bf \$spadroot}. The {\bf reroot} function will change the
>> >   system to use a new root directory and will have the same effect
>> >   as changing the {\bf FRICAS} shell variable and rerunning the system
>> > -from scratch.  A correct call looks like:
>> > -\begin{verbatim}
>> > -(in-package "BOOT")
>> > -(reroot "${FRICAS}")
>> > -\end{verbatim}
>> > -where the [[${FRICAS}]] variable points to installed tree.
>> > +from scratch.
>> >   |#
>> > +(defvar $spadroot "")
>> > +
>> >   (defun reroot (dir)
>> >     (setq $spadroot dir)
>> >     (setq $directory-list
>> > @@ -267,10 +264,9 @@ After this function is called the image is clean
>> > and can be saved.
>> >   #+:GCL (system:gbc-time 0)
>> >       #+(or :sbcl :clisp :openmcl :lispworks)
>> >       (if *fricas-load-libspad*
>> > -        (let* ((ax-dir (|getEnv| "FRICAS"))
>> > -               (spad-lib (concatenate 'string ax-dir
>> "/lib/libspad.so")))
>> > +        (let ((spad-lib (make-absolute-filename "/lib/libspad.so")))
>> >               (format t "Checking for foreign routines~%")
>> > -            (format t "FRICAS=~S~%" ax-dir)
>> > +            (format t "FRICAS=~S~%" $spadroot)
>> >               (format t "spad-lib=~S~%" spad-lib)
>> >               (if (|fricas_probe_file| spad-lib)
>> >                   (progn
>> > @@ -280,7 +276,7 @@ After this function is called the image is clean
>> and
>> > can be saved.
>> >                       (|quiet_load_alien| spad-lib)
>> >                       #+(or :sbcl :openmcl)
>> >                       (fricas-lisp::init-gmp
>> > -                        (concatenate 'string ax-dir
>> "/lib/gmp_wrap.so"))
>> > +                        (make-absolute-filename "/lib/gmp_wrap.so"))
>> >                       #+(and :clisp :ffi)
>> >                       (progn
>> >                           (eval `(FFI:DEFAULT-FOREIGN-LIBRARY
>> ,spad-lib))
>> >
>> >
>> > On 10/14/22 20:19, Qian Yun wrote:
>> >> (This is a follow up discussion of
>> >> https://github.com/fricas/fricas/pull/96)
>> >>
>> >> I'd like to discuss this again, to avoid problem like
>> >> https://github.com/fricas/fricas/issues/110,
>> >> aka to load "libspad.so" when "FRICAS" is not set.
>> >>
>> >> First, the binary (instead of shell script) FRICASsys should
>> >> work when "FRICAS" is not set.
>> >>
>> >> Second, during a fricas session, we can safely assume that
>> >> the environment variable "FRICAS" does not change.
>> >>
>> >> Third, if so, the we can use "FRICAS" from C side and use
>> >> "$spadroot" in Lisp side and still be consistent.
>> >>
>> >> Finally, if we are considering the use case of using FRICASsys
>> >> alone, aka using the Lisp image without the C part, then it
>> >> is clear that we should (|getEnv| "FRICAS") once, if it does not
>> >> exist, use some other default value, and store it into an
>> >> variable ($spadroot) and use that variable later.
>> >>
>> >> - Qian
>>
>> --
>> 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 [email protected].
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/fricas-devel/a7abd54b-cc5e-c97c-2045-9dea36cf8d29%40gmail.com
>> .
>>
>

-- 
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 [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/fricas-devel/CAHnU2dZfnVF_Y%3D0haoxPc2T-qrAq8%2BnnSNcpSxkCg2jmGH9KLQ%40mail.gmail.com.

Reply via email to