Dmitry Semikin wrote:
>
> --20cf303bffd208055f04ae58d0f5
> Content-Type: text/plain; charset=KOI8-R
> Content-Transfer-Encoding: quoted-printable
>
> Once more about building trunk version under cygwin.
>
> I've tried to it one more time and finished with following situation:
>
> Build failed with the error:
>
> >> System error:
> LOAD: A file with name
> /home/dmitry_semikin/builds/fricas-trunk/fricas-build/src/algebra/IEVALAB.N=
> RLIB/IEVALAB
> does not exist
>
> directory
> make[3]: *** [stamp-db] =EF=DB=C9=C2=CB=C1 1
> make[3]: Leaving directory
> `/home/dmitry_semikin/builds/fricas-trunk/fricas-build/src/algebra'
>
> As it can be seen, the system cannot find the file IEVALAB.NRLIB/IEVALAB
>
> I looked at that directory and found, that there is IEVALAB.erlib directory
> and IEVALAB-.NRLIB (with dash at the end). So, for some reason compiler
> added dash at the end of IEVALAB, while creating .NRLIB directory and all
> files within it.
>
IEVALAB-.NRLIB is OK. Spad compiler should create both IEVALAB-.NRLIB
and IEVALAB.NRLIB.
> I tried to look into the source spad files, but did not find there any
> differences from other spad files (the only thing I've noticed is that afte=
> r
> )abbrev line there is '--' comment (second line in file)... but I do not
> think, it can play so (although I did not try to build without this line to
> make sure, that it is not important)
>
> So, I wonder if sombody has any idea about why this can happen (dash '-'
> sign is appended to the name of .erlib, when converted to .NRLIB)? Or may b=
> e
> somebody can give some hints about how this can be debugged... I'll try to
> think about it myself, but I'm not quite familiar with fricas building
> process and compiler internals yet.
>
One possibility is to capture all messages to a file (for sucessful
full build expect about 18 Mb of messages). After failure search
for messages about compilation of IEVALAB.spad. If there is an
error message it may explain what is happening. If this does not
help there is possibility to run compiler by hand. The failing stage
can be re-run by going to 'src/algebra' subdirectory, running
(note setting environment variables, otherwise interpsys will
not start)
AXIOM=./axp DAASE=./r1 ../../build/arch/bin/interpsys
where 'arch' is your build architecture (probably i386-pc-cygwin)
and typing
)set break break
)read "boo_db.input"
to the interpsys prompt. Note: ')set break break' is so that
on error you will get into Lisp debugger. Unfortunately,
the failing stage is doing a lot of work that has to be done
sequentlialy in single run.
--
Waldek Hebisch
[email protected]
--
You received this message because you are subscribed to the Google Groups
"FriCAS - computer algebra system" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/fricas-devel?hl=en.
