>>>>> "Martin" == Martin Buchholz <martin(a)xemacs.org> writes:
Martin> Somewhere along the line we renamed the 21.2 tarballs from
Martin> xemacs-21.2-b3.tar.gz
Martin> to
Martin> xemacs-21.2.16.tar.gz
Martin> Is the current naming convention documented?
Martin> Is it a bug that xemacs is still *installed* into directories with
Martin> names like xemacs-21.2-b23 ??
Martin> ObPatch (completely untested):
Martin> --- configure.in 1999/12/08 07:43:35 1.111.2.53
Martin> +++ configure.in 1999/12/09 12:17:43
Martin> @@ -954,12 +954,11 @@
Martin> AC_DEFINE_UNQUOTED(EMACS_MAJOR_VERSION, $emacs_major_version)
Martin> AC_DEFINE_UNQUOTED(EMACS_MINOR_VERSION, $emacs_minor_version)
Martin> if test -n "$emacs_beta_version" ; then
Martin> + version="${version}.${emacs_beta_version}"
Martin> if test "$beta" = "yes"; then
Martin> - version="${version}-b${emacs_beta_version}"
Martin> - AC_DEFINE_UNQUOTED(EMACS_BETA_VERSION, $emacs_beta_version)
Martin> + AC_DEFINE_UNQUOTED(EMACS_BETA_VERSION, $emacs_beta_version)
Martin> else
Martin> - version="${version}.${emacs_beta_version}"
Martin> - AC_DEFINE_UNQUOTED(EMACS_PATCH_LEVEL, $emacs_beta_version)
Martin> + AC_DEFINE_UNQUOTED(EMACS_PATCH_LEVEL, $emacs_beta_version)
Martin> fi
Martin> fi
Martin> AC_DEFINE_UNQUOTED(XEMACS_CODENAME, "$xemacs_codename")
Does this mean "-b" will vanish from the installation directory name
and the version string?
I, for one, would definitely dislike this.
Here's what 21.2-b23 xemacs.mak has:
!if "$(emacs_beta_version)" != ""
XEMACS_VERSION_STRING=$(emacs_major_version).$(emacs_minor_version)-b$(emacs_beta_version)
!else
XEMACS_VERSION_STRING=$(emacs_major_version).$(emacs_minor_version)
!endif
Best regards,
Adrian