Uwe Brauer writes: The problem latin-unity faces is that it would require a lot of effort to write a detection system that can distinguish a disk file in ISO 8859/15 from one in ISO 8859/1, especially since many (perhaps the vast majority) are byte-for-byte identical in the two encodings. So I didn't try. If you are willing to give ISO 8859/15 precedence over ISO 8859/1 everywhere in your use of XEmacs, then you can do ;; Prefer Latin-9 when reading files. (prefer-coding-system 'iso-8859-15) ;; Automatically use Latin-9 if feasible when writing files. ;; The default value of this variable would also work but if you ;; later prefer-coding-system another coding system, it would stop ;; working suddenly. (setq latin-unity-preapproved-coding-system-list '(buffer-default iso-8859-15)) The problem is you may not want to do that, if you have a lot of files that use the 1/2 character of ISO 8859/1 and only a few in French requiring œ. Check the values of the variable latin-unity-preapproved-coding-system-list and take note of the warnings in its docstring: "*List of coding systems used without querying the user if feasible. The first feasible coding system in this list is used. The special values 'preferred and 'buffer-default may be present: buffer-default Use the coding system used by `write-region', if feasible. preferred Use the coding system specified by `prefer-coding-system' if feasible. \"Feasible\" means that all characters in the buffer can be represented by the coding system. Coding systems in `latin-unity-ucs-list' are always considered feasible. Other feasible coding systems are computed by `latin-unity-representations-feasible-region'. Most users will want at least one ISO 8859 coding system in this list, as otherwise pure ASCII files will not be preapproved. (This is a bug, due to the limitation of applicability of this package to Latin and universal. The condition that an ISO 8859 coding system be included will be satisfied implicitly by 'buffer-default or 'preferred for most users, but it can be annoying for users of ISO 2022 or EUC coding systems.) Note that the first universal coding system in this list shadows all other coding systems. In particular, if your preferred coding system is a universal coding system, and 'preferred is a member of this list, latin-unity will blithely convert all your files to that coding system. This is considered a feature, but it may surprise most users. Users who don't like this behavior should move 'preferred to `latin-unity-preferred-coding-system-list'." HTH Steve _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta

Uwe Brauer <oub(a)mat.ucm.es&gt; writes: It is not possible to distinguish Latin-1 from Latin-9 without an appropriate coding cookie or other hint separately from just the encoded file. XEmacs can't do magic, neither can GNU Emacs. And I never stated anything else. -- David Kastrup, Kriemhildstr. 15, 44793 Bochum _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta

David Kastrup wrote: Hi David, interesting point. What about to teach them? Seems worthwhile the effort. Is magic still the tool of choice in that kind of matters? Do you know alternatives? Thanks Andreas Röhler _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta