Mercurial > hg > xemacs-beta
comparison lisp/mule/misc-lang.el @ 5662:bed39edf91ba
Be better about word boundaries with JIT unicode characters, mule-category.el
lisp/ChangeLog addition:
2012-05-10 Aidan Kehoe <kehoea@parhasard.net>
* mule/mule-category.el (word-combining-categories):
Be better about default word boundaries when text contains
just-in-time-allocated Unicode code points. Document what we
should do instead once we have Unicode internally.
* mule/misc-lang.el: IPA characters are Latin.
| author | Aidan Kehoe <kehoea@parhasard.net> |
|---|---|
| date | Thu, 10 May 2012 13:53:06 +0100 |
| parents | 308d34e9f07d |
| children |
comparison
equal
deleted
inserted
replaced
| 5661:6e5a7278f9bf | 5662:bed39edf91ba |
|---|---|
| 39 final ?0 | 39 final ?0 |
| 40 graphic 1 | 40 graphic 1 |
| 41 short-name "IPA" | 41 short-name "IPA" |
| 42 long-name "IPA")) | 42 long-name "IPA")) |
| 43 | 43 |
| 44 ;; XEmacs; these are Latin, it's not useful to put word boundaries between | |
| 45 ;; them and ASCII. | |
| 46 (modify-category-entry 'ipa ?l nil t) | |
| 47 | |
| 48 ;; XEmacs; why are these Latin? See the following: | |
| 49 ;; | |
| 50 ;; (let ((scripts | |
| 51 ;; (mapcar #'(lambda (character) | |
| 52 ;; (car | |
| 53 ;; (split-string | |
| 54 ;; (cadr (assoc "Name" (describe-char-unicode-data | |
| 55 ;; character)))))) | |
| 56 ;; (loop | |
| 57 ;; for i from 33 to 127 | |
| 58 ;; if (not (eql -1 (char-to-unicode (make-char 'ipa i)))) | |
| 59 ;; nconc (list (make-char 'ipa i)))))) | |
| 60 ;; (mapcar #'(lambda (script) | |
| 61 ;; (cons script (count script scripts :test #'equal))) | |
| 62 ;; (remove-duplicates scripts :test #'equal))) | |
| 63 ;; => (("GREEK" . 1) ("LATIN" . 55) ("MODIFIER" . 3)) | |
| 64 | |
| 65 | |
| 44 ;;; misc-lang.el ends here | 66 ;;; misc-lang.el ends here |