From octave-maintainers-request at bevo dot che dot wisc dot edu Thu May 16 14:29:29 2002 Subject: RE: rebuilding still From: "Lippert, Ross A." To: "'John W. Eaton'" Cc: octave-maintainers mailing list Date: Thu, 16 May 2002 15:13:37 -0400 OK, problem fixed now. In the make-break-edit-make cycle, somehow the DOCSTRINGS for scripts ended up being created empty, so munge-texi was choking inelegantly on it, and it wasn't clear to me why. I have a new one though. I'll bet this is a gcc issue: /usr/local/gcc/3.0.2/bin/g++ -c -I. -I.. -I../liboctave -I../src -I../libcruft/misc -I../glob -I../glob -DHAVE_CONFIG_H -mieee-with-inexact -g -O2 -Wall data.cc -o data.o In file included from defun-int.h:31, from defun.h:30, from data.cc:35: symtab.h:449: `ostream' was not declared in this scope symtab.h:449: `os' was not declared in this scope symtab.h:449: invalid data member initialization symtab.h:449: (use `=' to initialize static data members) symtab.h:449: variable or field `print_info' declared void gmake-3.75-97[2]: *** [data.o] Error 1 gmake-3.75-97[1]: *** [src] Error 2 gmake-3.75-97: *** [all] Error 2 -r -----Original Message----- From: John W. Eaton [mailto:jwe at bevo dot che dot wisc dot edu] Sent: Thursday, May 16, 2002 2:14 PM To: Lippert, Ross A. Cc: octave-maintainers mailing list Subject: RE: rebuilding still On 16-May-2002, Lippert, Ross A. wrote: | | OK, I think I found the problem. It isn't a compiler problem, it is a | Makefile | problem. | | In the Makefile for doc/interpreter the rule for going from txi to texi has | an | extra -d in it. | | %.texi : %.txi | at echo making $@ from $< | at dot /munge-texi \ | -d $(TOPDIR)/src/DOCSTRINGS \ | !bad! -d $(TOPDIR)/scripts/DOCSTRINGS < $< > $ at dot t | at mv $@.t $@ But if you remove that part of the command, doesn't that omit all the M-file doc strings from the manual? jwe