view lisp/international/utf-7.el @ 67575:d0f7b15488df

(bibtex-expand-strings) (bibtex-autokey-expand-string, bibtex-name-part) (bibtex-entry-type-whitespace, bibtex-entry-type-str) (bibtex-any-entry-maybe-empty-head, bibtex-string-type) (bibtex-preamble-prefix, bibtex-string-empty-key): New variables. (bibtex-entry-type, bibtex-entry-head): Match only valid entries. (bibtex-entry-postfix, bibtex-known-entry-type-re) (bibtex-valid-entry-re, bibtex-any-valid-entry-re) (bibtex-valid-entry-whitespace-re, bibtex-empty-field-re) (bibtex-field-name-for-parsing, bibtex-remove-delimiters-string) (bibtex-beginning-of-last-entry): Remove. (bibtex-parse-field-name): Use bibtex-field-name. Issue error message if comma is missing but buffer is read-only. (bibtex-parse-field-text): Handle whitespaces at the end of field text. Return 3-element list with beginning and end of field text and end of field. (bibtex-end-of-text-in-field, bibtex-end-of-field): Change accordingly. (bibtex-parse-field): Remove arg name. Use bibtex-field-name. (bibtex-search-forward-field, bibtex-search-backward-field): Search always delimited by limits of entry. Use more efficient search algorithms. (bibtex-name-in-field): Use bibtex-start-of-name-in-field and bibtex-end-of-name-in-field. (bibtex-text-in-field-bounds): Handle BibTeX strings when extracting the content of a field. (bibtex-text-in-field): Use search limits. (bibtex-parse-string-prefix): Handle empty string keys based on bibtex-string-empty-key. (bibtex-parse-string): Fix docstring. (bibtex-text-in-string): Use bibtex-text-in-field-bounds. (bibtex-preamble-prefix, bibtex-strings): New functions. (bibtex-skip-to-valid-entry): Include preceding whitespace in BibTeX entries (consistent with other BibTeX functions). (bibtex-map-entries): Use bibtex-skip-to-valid-entry. (bibtex-search-entry): Fix docstring. Simplify. (bibtex-flash-head, bibtex-complete-string-cleanup) (bibtex-count-entries, bibtex-sort-buffer): Simplify. (bibtex-beginning-of-first-entry): Use bibtex-skip-to-valid-entry. (bibtex-parse-entry): New optional arg content. (bibtex-format-entry, bibtex-autofill-entry, bibtex-url): Use it. Use bibtex-text-in-field-bounds. (bibtex-print-help-message): Handle BibTeX strings and preambles. (bibtex-end-of-entry): Use bibtex-preamble-prefix and bibtex-parse-string-postfix. (bibtex-find-text-internal): New function. (bibtex-remove-delimiters): Use it. (bibtex-find-text): Use it. New optional arg help. (bibtex-complete): Handle BibTeX string and preamble entries. (bibtex-Preamble): Fix order of closing delimiters.

;;; utf-7.el --- utf-7 coding system

;; Copyright (C) 2003  Free Software Foundation, Inc.

;; Author: Dave Love <[e-mail]>
;; Keywords: i18n, mail

;; This file is free software; you can redistribute it and/or modify
;; it under the terms of the GNU General Public License as published by
;; the Free Software Foundation; either version 2, or (at your option)
;; any later version.

;; This file is distributed in the hope that it will be useful,
;; but WITHOUT ANY WARRANTY; without even the implied warranty of
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
;; GNU General Public License for more details.

;; You should have received a copy of the GNU General Public License
;; along with GNU Emacs; see the file COPYING.  If not, write to
;; the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
;; Boston, MA 02110-1301, USA.

;;; Commentary:

;; Defines a coding system for UTF-7, defined in RFC 2152.  Non-ASCII
;; segments are encoded as base64-encoded big endian UTF-16.  Also
;; defines a variation required for IMAP (RFC 2060).

;; The encoding and decoding was originally taken from Jon K Hellan's
;; implementation in Gnus, but has been substantially re-done.

;; This probably needs more attention.  In particular, it's not
;; completely consistent with iconv's behaviour.  It's arguable
;; whether the IMAP version should be a coding system since it's
;; apparently only used for IMAP mailbox names, so it's commented out.

;;; Code:

;;;###autoload(autoload-coding-system 'utf-7 '(require 'utf-7))
(make-coding-system
 'utf-7 0 ?U
 "UTF-7 encoding of Unicode (RFC 2152)"
 nil
 `((safe-chars . ,(coding-system-get 'utf-16be 'safe-chars))
   (mime-charset . utf-7)
   (pre-write-conversion . utf-7-pre-write-conversion)
   (post-read-conversion . utf-7-post-read-conversion)))

;; (make-coding-system
;;  'utf-7-imap 0 ?u
;;  "UTF-7 encoding of Unicode, IMAP version (RFC 2060)"
;;  nil
;;  `((safe-chars . ,(coding-system-get 'utf-16be 'safe-chars))
;;    (pre-write-conversion . utf-7-imap-pre-write-conversion)
;;    (post-read-conversion . utf-7-imap-post-read-conversion)))

(defun utf-7-decode (len imap)
  "Decode LEN bytes of UTF-7 at point.
IMAP non-nil means use the IMAP version."
  (save-excursion
    (save-restriction
      (narrow-to-region (point) (+ (point) len))
      (let ((not-esc (if imap "^&" "^+"))
	    (skip-chars (if imap "A-Za-z0-9+," "A-Za-z0-9+/")))
	(while (not (eobp))
	  (skip-chars-forward not-esc)
	  (unless (eobp)
	    (forward-char)
	    (let ((p (point))
		  (run-length (skip-chars-forward skip-chars)))
	      (if (eq ?- (char-after))
		  (delete-char 1))
	      (unless (= run-length 0)	; encoded lone esc-char
		(let ((pl (mod (- run-length) 4)))
		  (insert-char ?= pl)
		  (if imap
		      (subst-char-in-region p (point) ?, ?/))
		  (base64-decode-region p (point)))
		(decode-coding-region p (point) 'utf-16be)
		(save-excursion
		  (goto-char p)
		  (delete-backward-char 1)))))))
      (- (point-max) (point-min)))))

(defun utf-7-post-read-conversion (len)
  (utf-7-decode len nil))

;; (defun utf-7-imap-post-read-conversion (len)
;;   (utf-7-decode len t))

(defun utf-7-encode (from to imap)
  "Encode bytes between FROM and TO to UTF-7.
ESC and SKIP-CHARS are adjusted for the normal and IMAP versions."
  (let* ((old-buf (current-buffer))
	 (esc (if imap ?& ?+))
	 ;; These are characters which can be encoded asis.
	 (skip-chars (if imap
			 "\t\n\r\x20-\x25\x27-\x7e" ; rfc2060 
		       ;; This includes the rfc2152 optional set.
		       ;; Perhaps it shouldn't (like iconv).
		       "\t\n\r -*,-[]-}"))
	 (not-skip-chars (format "^%s%c" skip-chars esc)))
    (set-buffer (generate-new-buffer " *temp*"))
    (if (stringp from)
	(insert from)
      (insert-buffer-substring old-buf from to))
    (goto-char (point-min))
    (while (not (eobp))
      (skip-chars-forward skip-chars)
      (if (eq ?+ (char-after))
	  (progn (forward-char)
		 (insert ?-))
	(unless (eobp)
	  (insert esc)
	  (let ((p (point)))
	    (skip-chars-forward not-skip-chars)
	    (save-restriction
	      ;; encode-coding-region doesn't preserve point
	      (narrow-to-region p (point))
	      (encode-coding-region p (point-max) 'utf-16be)
	      (base64-encode-region p (point-max))
	      (if imap
		  (subst-char-in-region p (point-max) ?/ ?,))
	      (goto-char p)
	      ;; As I read the RFC, this isn't correct, but it's
	      ;; consistent with iconv, at least regarding `='.
	      (skip-chars-forward "^= \t\n")
	      (delete-region (point) (point-max))))
	  (unless (eobp)
	    (insert ?-)))))
    nil))

(defun utf-7-pre-write-conversion (from to)
  (utf-7-encode from to nil))

;; (defun utf-7-imap-pre-write-conversion (from to)
;;   (utf-7-encode from to t))

(provide 'utf-7)

;;; arch-tag: 975ee403-90a4-4286-97d2-4ed1323f4ef9
;;; utf-7.el ends here