This cuts it: cd d:\tmp\21.2\xemacs\ patch --strip=3 --reverse < d:\tmp\21.2\select.2.diff Compilation started at Sun Jul 16 00:15:27 2000 +0200 (W. Europe Daylight Time) patching file `lisp/select.el' patching file `src/select.c' patching file `src/select-x.c' patching file `src/select-msw.c' patching file `src/select.h' Compilation finished at Sun Jul 16 00:15:31 cd d:\tmp\21.2\xemacs\ patch --strip=3 < d:\tmp\select.3.diff Compilation started at Sun Jul 16 00:16:09 2000 +0200 (W. Europe Daylight Time) patching file `src/select.h' patching file `src/select.c' patching file `src/select-msw.c' patching file `src/select-x.c' patching file `src/buffer.c' patching file `lisp/select.el' patching file `lisp/x-select.el' Compilation finished at Sun Jul 16 00:16:14 I'm building now while I wait for the mswindows-get-clipboard patch. Should mswindows-paste-clipboard just call get-clipboard instead? Best regards, Adrian APA> Applying this patch on top of your last one fails badly: APA> cd d:\tmp\21.2\xemacs\ APA> patch --dry-run --strip=3 < d:\tmp\select.3.diff APA> Compilation started at Sun Jul 16 00:08:34 2000 +0200 (W. Europe Daylight Time) APA> patching file `src/select.h' APA> Hunk #1 FAILED at 38. APA> 1 out of 1 hunk FAILED -- saving rejects to src/select.h.rej APA> patching file `src/select.c' APA> Hunk #3 FAILED at 80. APA> Hunk #4 FAILED at 94. APA> Hunk #5 FAILED at 128. APA> Hunk #6 FAILED at 350. APA> Hunk #7 succeeded at 604 (offset 140 lines). APA> Hunk #8 FAILED at 664. APA> Hunk #9 succeeded at 714 (offset -3 lines). APA> Hunk #10 succeeded at 869 (offset 140 lines). APA> Hunk #11 succeeded at 777 (offset -3 lines). APA> Hunk #12 succeeded at 932 (offset 140 lines). APA> 5 out of 12 hunks FAILED -- saving rejects to src/select.c.rej APA> patching file `src/select-msw.c' APA> Reversed (or previously applied) patch detected! Assume -R? [n] APA> Apply anyway? [n] APA> Skipping patch. APA> 4 out of 4 hunks ignored -- saving rejects to src/select-msw.c.rej APA> patching file `src/select-x.c' APA> Hunk #2 FAILED at 374. APA> Hunk #3 succeeded at 568 (offset -2 lines). APA> Hunk #5 FAILED at 611. APA> Hunk #6 succeeded at 674 (offset -2 lines). APA> Hunk #8 succeeded at 1146 (offset -2 lines). APA> 2 out of 9 hunks FAILED -- saving rejects to src/select-x.c.rej APA> patching file `src/buffer.c' APA> patching file `lisp/select.el' APA> Hunk #5 succeeded at 418 with fuzz 2 (offset 4 lines). APA> Hunk #6 succeeded at 451 with fuzz 2 (offset 3 lines). APA> Hunk #7 succeeded at 462 (offset 4 lines). APA> Hunk #8 succeeded at 488 with fuzz 2 (offset 8 lines). APA> Hunk #9 FAILED at 609. APA> Hunk #10 FAILED at 702. APA> Hunk #11 FAILED at 748. APA> Hunk #12 succeeded at 844 (offset 74 lines). APA> 3 out of 12 hunks FAILED -- saving rejects to lisp/select.el.rej APA> patching file `lisp/x-select.el' APA> Compilation exited abnormally with code 1 at Sun Jul 16 00:08:41 APA> Please send clear instructions with your patches how they should be APA> applied. APA> This is easy to do for the author and there is no point in others APA> trying to figure this out. It's a waste of everybody's time. APA> My best guess: APA> d:\tmp\select.3.diff APA> should be applied INSTEAD of APA> d:\tmp\select.2.diff APA> , yes? APA> Best regards, APA> Adrian Alastair> OK. There were some *very* silly problems, especially in the Alastair> X code, which was attempting to hack with selection-alist Alastair> (an internal variable of select.c). I *think* this should Alastair> solve most of the problems I've been hearing about. In Alastair> particular, the _EMACS_INTERNAL selection type has gone, Alastair> for good I hope. I haven't tested it on X yet (will do, as Alastair> soon as I download a new set of sources from the Alastair> repository) - apologies for this, but I know it works (and Alastair> fixes a few things) on Windows, so I'll send it out *now* Alastair> and then have another go at building on X. Alastair> The new selection code is still a work in progress - there Alastair> are several things still to do as far as I can see; next on Alastair> the list is a set of functions for encoding and decoding Alastair> binary data, so that the existing code in select-x.c that Alastair> decodes the binary data can be moved into Lisp, where it Alastair> belongs. Alastair> Regards, Alastair> Alastair. Alastair> lisp/ChangeLog: Alastair> 2000-07-15 Alastair J. Houghton <ajhoughton(a)lineone.net&gt; Alastair> * select.el (select-convert-from-text): New. Alastair> * select.el (select-convert-from-length): New. Alastair> * select.el (select-convert-from-integer): New. Alastair> * select.el (select-convert-from-identity): New. Alastair> * select.el (select-convert-from-filename): New. Alastair> * select.el (selection-converter-in-alist): Use them. Alastair> New converter functions for X. Alastair> * select.el (get-selection): Removed comment. Alastair> * select.el (own-selection): Removed append code - I thought Alastair> I'd already done this, but apparently not... Alastair> * select.el: Changed comment about TIMESTAMP. Alastair> * select.el (select-convert-in): New. Alastair> * select.el (select-convert-out): New. Alastair> New functions that get used by select.c. Alastair> * select.el (select-buffer-killed-default): New. Alastair> * select.el (select-buffer-killed-text): New. Alastair> * select.el (selection-buffer-killed-alist): New. Alastair> New list and functions that get called if a relevant selection Alastair> exists and a buffer gets killed. Alastair> * select.el (select-convert-to-targets): Alastair> * select.el (select-convert-to-identity): Removed. Alastair> * select.el (select-convert-from-identity): Removed. Alastair> * select.el (select-converter-out-alist): Alastair> Removed _EMACS_INTERNAL selection type. Alastair> * x-select.el (xselect-kill-buffer-hook): Removed. Alastair> * x-select.el (xselect-kill-buffer-hook-1): Removed. Alastair> Removed this X-specific nonesense. Use the generic support Alastair> in select.el instead. Alastair> src/ChangeLog: Alastair> 2000-07-15 Alastair J. Houghton <ajhoughton(a)lineone.net&gt; Alastair> * select.h (selection-alist): Removed declaration. Alastair> * select.h (get_local_selection): Alastair> * select.c (get_local_selection): Alastair> Made static. Alastair> * select.h (convert_selection): Removed declaration. Alastair> * select.c (convert_selection): Removed. Alastair> This function belongs in Lisp. Alastair> * select.h (select_convert_in): Declare. Alastair> * select.h (select_convert_out): Declare. Alastair> * select.c (select_convert_in): New. Alastair> * select.c (select_convert_out): New. Alastair> New conversion functions for other files to call. Alastair> * select.h (select_notify_buffer_kill): Declare. Alastair> * select.c (select_notify_buffer_kill): New. Alastair> New functions that get called from kill-buffer. Alastair> * buffer.c (kill-buffer): Call select_notify_buffer_kill, rather than Alastair> X-specific lisp code. Alastair> * select.h: Declare some of the lisp-visible functions for Alastair> external use. Alastair> * select.c (clean_local_selection_data): Removed. This was Alastair> a disgusting function, and previously should have been in Alastair> select-x.c in any case. The functionality is now provided Alastair> in select-convert-from-integer (select.el). Alastair> * select.c (available-selection-types): Fixed stupidity where Alastair> INTEGER and ATOM got added twice. Also add STRING when we see an Alastair> extent. Alastair> * select.c (get-selection-internal): Removed symbol stripping. No Alastair> longer causes conversion when data comes from the internal cache. Alastair> * select.c (syms_of_select): Added new functions. Alastair> * select-x.c (motif_clipboard_cb): Use select_convert_out. Rewrote Alastair> error checking - previously this called abort! Alastair> * select-x.c (x_own_selection): Changed comment. Alastair> * select-x.c (x_handle_selection_request): Use select_convert_out. Alastair> Don't mess with selection-alist; it's an internal variable of select.c. Alastair> * select-x.c (x_get_foreign_selection): Use select_convert_in. Alastair> * select-x.c (x_handle_selection_clear): Use get-selection-timestamp, Alastair> rather than messing with selection-alist. Alastair> * select-msw.c (mswindows_get_foreign_selection): Alastair> Use TO_INTERNAL_FORMAT rather than hacking. Alastair> -- Alastair> ____________________________________________________________ Alastair> Alastair Houghton ajhoughton(a)lineone.net