From: shuhei Date: Thu, 19 Apr 2001 07:47:45 +0000 (+0000) Subject: (mime-decode-parameter-value): Removed comments. X-Git-Url: http://git.chise.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=ff2ce11dc4a8e6cec57ad5c23f795a64cbc6bd41;p=elisp%2Fflim.git (mime-decode-parameter-value): Removed comments. (mime-decode-parameter-encode-segment): New function. (mime-decode-parameter-plist): New implementation. Switched from decode-then-concat to concat-then-decode model. (mime-parse-parameters): Strip quoted-pair in quoted-string. (mime-parse-Content-Type): Use `make-mime-content-type'. (mime-parse-Content-Disposition): Use `make-mime-content-disposition'. --- diff --git a/mime-parse.el b/mime-parse.el index 727c0ac..c60c9b7 100644 --- a/mime-parse.el +++ b/mime-parse.el @@ -150,16 +150,6 @@ be the result." (defun mime-decode-parameter-value (text charset language) (let ((start 0)) - ;; RFC 2231 is ambiguous about case-sensitivity. - ;; - ;; ext-octet := "%" 2(DIGIT / "A" / "B" / "C" / "D" / "E" / "F") - ;; - ;; If RFC 2234 is employed, this rule will match "%ab" as well as - ;; "%AB" because ABNF strings are case-insensitive. - ;; But it is not clear whether RFC 2231 employs RFC 2234 or not:-< - ;; - ;; Anyway, we choose to recognize lowercase letters here. - ;; (while (string-match "%[0-9A-F][0-9A-F]" text start) (while (string-match "%[0-9A-Fa-f][0-9A-Fa-f]" text start) (setq text (replace-match (char-to-string @@ -181,6 +171,21 @@ be the result." (put-text-property 0 (length text) 'mime-language language text)) text)) +(defun mime-decode-parameter-encode-segment (segment) + (if (string-match (eval-when-compile + (concat "^" mime-attribute-char-regexp "+$")) + segment) + ;; shortcut + segment + ;; XXX: make too many temporary strings. + (mapconcat + (function + (lambda (chr) + (if (string-match mime-attribute-char-regexp (char-to-string chr)) + (char-to-string chr) + (format "%%%02X" chr)))) + segment ""))) + (defun mime-decode-parameter-plist (params) "Decode PARAMS as a property list of MIME parameter values. @@ -201,30 +206,28 @@ property of the decoded-value." (while params (if (string-match (eval-when-compile (concat "^\\(" mime-attribute-char-regexp "+\\)" - "\\(\\*\\([0-9]+\\)\\)?\\(\\*\\)?$")) + "\\(\\*\\([0-9]+\\)\\)?" ; continuation + "\\(\\*\\)?$")) ; charset/language (car params)) - (let* ((attribute (downcase (substring (car params) 0 (match-end 1)))) - (section (if (match-beginning 3) + (let* ((attribute (downcase + (substring (car params) 0 (match-end 1)))) + (section (if (match-beginning 2) (string-to-int (substring (car params) (match-beginning 3)(match-end 3))) 0)) - ;; EPARAM := (ATTRIBUTE . VALUES) - ;; VALUES := [1*V-ELT] ; vector of (length params) elements. - ;; V-ELT := (VALUE CHARSET LANGUAGE) ; extended-initial-value - ;; | (VALUE t) ; extended-other-values - ;; | (VALUE) ; regular-parameter-values + ;; EPARAM := (ATTRIBUTE CHARSET LANGUAGE VALUES) + ;; VALUES := [1*VALUE] ; vector of (length params) elements. (eparam (assoc attribute eparams))) - ;; should signal an error? - ;; (when (> section len) ...) (unless eparam - (setq eparam (cons attribute (make-vector len nil)) + (setq eparam (cons attribute + (list nil nil (make-vector len nil))) eparams (cons eparam eparams))) (setq params (cdr params)) ;; if parameter-name ends with "*", it is an extended-parameter. (if (match-beginning 4) (if (zerop section) - ;; extended-initial-value contains charset/language info. + ;; extended-initial-parameter. (if (string-match (eval-when-compile (concat "^\\(" @@ -235,71 +238,63 @@ property of the decoded-value." "\\)?'\\)" "\\(" mime-attribute-char-regexp - ;; allow lowercase letters. - ;; "\\|%[0-9A-F][0-9A-F]\\)+$" "\\|%[0-9A-Fa-f][0-9A-Fa-f]\\)+$")) (car params)) - (aset (cdr eparam) - 0 ; section == 0. - (list - ;; text - (substring (car params) - (match-end 2)) - ;; charset - (substring (car params) - 0 (match-beginning 2)) - ;; language - (substring (car params) - (1+ (match-beginning 2)) - (1- (match-end 2))))) - ;; invalid encoding. - (aset (cdr eparam) section - (list (std11-strip-quoted-string - (car params))))) - ;; extended-other-values + (progn + ;; charset + (setcar (cdr eparam) ; (nthcdr 1 eparam) + (downcase + (substring (car params) + 0 (match-beginning 2)))) + ;; language + (setcar (nthcdr 2 eparam) + (downcase + (substring (car params) + (1+ (match-beginning 2)) + (1- (match-end 2))))) + ;; text + (aset (nth 3 eparam) 0 + (substring (car params) + (match-end 2)))) + ;; invalid parameter-value. + (aset (nth 3 eparam) 0 + (mime-decode-parameter-encode-segment + (car params)))) + ;; extended-other-parameter. (if (string-match (eval-when-compile (concat "^\\(" mime-attribute-char-regexp - ;; allow lowercase letters. - ;; "\\|%[0-9A-F][0-9A-F]\\)+$" "\\|%[0-9A-Fa-f][0-9A-Fa-f]\\)+$")) (car params)) - (aset (cdr eparam) section - (list (car params) t)) - ;; invalid encoding. - (aset (cdr eparam) section - (list (std11-strip-quoted-string - (car params)))))) - ;; regular-parameter-name - (aset (cdr eparam) section - (list (std11-strip-quoted-string - (car params)))))) - ;; no parameter value extensions used, or invalid attribute-name. + (aset (nth 3 eparam) section + (car params)) + ;; invalid parameter-value. + (aset (nth 3 eparam) section + (mime-decode-parameter-encode-segment + (car params))))) + ;; regular-parameter. + (aset (nth 3 eparam) section + (mime-decode-parameter-encode-segment + (car params))))) + ;; invalid attribute-name. (setq dest (cons (cons (downcase (car params)) - (std11-strip-quoted-string - (car (cdr params)))) + (car (cdr params))) dest) params (cdr params))) (setq params (cdr params))) ;; decode and concat parameters. (while eparams - (let* ((attribute (car (car eparams))) - (values (cdr (car eparams))) - (charset (nth 1 (aref values 0))) - (language (nth 2 (aref values 0)))) - (setq dest (cons (cons attribute - (mapconcat - (lambda (elt) - (if (car (cdr elt)) - (mime-decode-parameter-value - (car elt) charset language) - ;; this value is not encoded. - ;; should we decode encoded-words here? - (car elt))) - values "")) - dest) - eparams (cdr eparams)))) + (setq dest (cons (cons (car (car eparams)) ; attribute + (mime-decode-parameter-value + (mapconcat (function identity) + (nth 3 (car eparams)) ; values + "") + (nth 1 (car eparams)) ; charset + (nth 2 (car eparams)) ; language + )) + dest) + eparams (cdr eparams))) dest)) (defun mime-parse-alist-to-plist (alist) @@ -346,7 +341,9 @@ Return a property list, which is a list of the form (equal (car tokens) '(tspecials . "=")) (setq tokens (cdr tokens)) (memq (car (car tokens)) '(mime-token quoted-string))) - (setq params (cons (cdr (car tokens)) + (setq params (cons (if (eq (car (car tokens)) 'quoted-string) + (std11-strip-quoted-pair (cdr (car tokens))) + (cdr (car tokens))) (cons attribute params)) tokens (cdr tokens))) (nreverse params))) @@ -375,11 +372,11 @@ PRIMARY-TYPE and SUBTYPE are symbols, and other elements are strings." (when (and (equal (car tokens) '(tspecials . "/")) (setq tokens (cdr tokens)) (eq (car (car tokens)) 'mime-token)) - (cons (cons 'type (intern (downcase primary-type))) - (cons (cons 'subtype - (intern (downcase (cdr (car tokens))))) - (mime-decode-parameters - (mime-parse-parameters (cdr tokens)))))))))) + (make-mime-content-type + (intern (downcase primary-type)) + (intern (downcase (cdr (car tokens)))) + (mime-decode-parameters + (mime-parse-parameters (cdr tokens))))))))) ;;;###autoload (defun mime-read-Content-Type () @@ -399,9 +396,10 @@ Format of return value is same as that of `mime-parse-Content-Type'." "Parse FIELD-BODY as Content-Disposition field. FIELD-BODY is a string." (let ((tokens (mime-lexical-analyze field-body))) (when (eq (car (car tokens)) 'mime-token) - (cons (cons 'type (intern (downcase (cdr (car tokens))))) - (mime-decode-parameters - (mime-parse-parameters (cdr tokens))))))) + (make-mime-content-disposition + (intern (downcase (cdr (car tokens)))) + (mime-decode-parameters + (mime-parse-parameters (cdr tokens))))))) ;;;###autoload (defun mime-read-Content-Disposition ()