Hard to tell what is the best solution. Telling upstream that this usage breaks XEmacs is always a good idea. (And I think you already did that right?) What upstream then will do about it is another deal. Like many other packages calendar could probably benefit from being upgraded in the XEmacs Packages Tree. In that case, when the upstream version is not compatible with XEmacs, I would suggest just fixing it by using the XEmacs version of functions there. Having said that I also think we should add the optional parameter "when" to 21.5 in order to be compatible with GNU. So this problem will grow away. A first implementation could just accept it and ignore it. If not useless information it seems at least not that critical, but I'm probably ignorant. Yours -- %% Mats _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-beta

You mean that there are other issues as well, as you say in another post, so this on its own will no make it work with 21.4? If those few (?) missing functions are implemented from scratch or synced with GPLv2 versions into 21.4 core or even as party of a compatibility package it could be feasible. Having said that my position is that we should focus om making things work with 21.5 and make 21.5 stable so that it could replace 21.4 out there. Yours -- %% Mats _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-beta

I did a quick check and it does not look good, the new icalendar version has two problem. - One concerns the UFT8 coding (so it would not run for 21.4 no Mule), which most likely could be fixed. - More serious: the following functions are missing in 21.4 replace-regexp-in-string, with-no-warnings It is a pity since the new icalendar will have nice import and export functionality. However I have not seen anybody (besides me) interested in the package so at the moment I don't feel to investigate the 21.4 specific problem in more detail. Uwe Brauer _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-beta

Ok, first a little rant. Although MS Outlook is a terrible product which deliberately breaks almost any standard it can lie hand on, it is very popular in the business world. Why? Among other reasons, because of its included calendar facility which allows to set meetings, send invitations based on these meetings to other (outlook) users, so that their outlook program automatically includes this invitation into the calendar of the recipient. Needless to say that these invitations do not follow any open standard. However there is an open standard called icalendar (ics files) since a couple of years. It gained popularity in the last years, mostly thanks to the lighting[1] calendar project for thunderbird (evolution and kmail have also that facility I think, but they are not cross platform applications.) Anyhow, given that, it is natural to ask, can (X)emacs have some similar functionality compatible with the icalendar standard? Thanks to Ulf the answer is yes (his package goes back to 2002, but since I never received many invitations up to now I scarcely used it). Icalendar allows you to import invitations into your diary (or into any other file you want) (this works more or less ok in 21.4, save the UTF8 coding). And icalendar allows you also to export an invitation from your diary into a ics file, which you then can send to other users. That works in 21.5 but not in 21.4 because of the lack of replace-regexp-in-string, with-no-warnings Now replace-regexp-in-string is in 21.5/lisp/subr.el So maybe it could be put in fsf-compat for 21.4 with-no-warnings however I don't know where it is.... However there is still a small bug in icalendar: invitations whose descriptions are longer than one line are not exported. Ulf wanted to look into this. Once this issue is solved we might discuss again how to make icalendar work under 21.4 Uwe _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-beta

Mats Lidell writes: Sure, but I at least am committed to maintaining backward compatibility for 21.4 for 6 months after the next version is declared stable. It's usually not that much extra work, and if it is, we can (rarely) make exceptions. For `replace-regexp-in-string', we can move it out of 21.5 to xemacs-base (or fsf-compat, if it's too yucky, but it seems like it might be useful). `with-no-warnings' is one of those piles of GNU poop we should avoid if possible. (C-h f with-byte-compiler-warnings-suppressed for how to do it in XEmacs if you really must.) As far as I know there is no warning in XEmacs Lisp that is sufficiently stupid that it makes sense to ignore all warnings just to suppress that one. This one we should definitely put in fsf-compat (my 21.5 seems to have it :-( ). Regards, _______________________________________________ XEmacs-Beta mailing list XEmacs-Beta(a)xemacs.org http://lists.xemacs.org/mailman/listinfo/xemacs-beta