diff options
Diffstat (limited to 'lisp/progmodes/perl-mode.el')
| -rw-r--r-- | lisp/progmodes/perl-mode.el | 43 |
1 files changed, 19 insertions, 24 deletions
diff --git a/lisp/progmodes/perl-mode.el b/lisp/progmodes/perl-mode.el index ef372a34fdb..092aa2b2fac 100644 --- a/lisp/progmodes/perl-mode.el +++ b/lisp/progmodes/perl-mode.el @@ -66,22 +66,7 @@ ;; a rich language; writing a more suitable parser would be a big job): ;; 2) The globbing syntax <pattern> is not recognized, so special ;; characters in the pattern string must be backslashed. -;; 3) The << quoting operators are not recognized; see below. -;; 5) To make '$' work correctly, $' is not recognized as a variable. -;; Use "$'" or $POSTMATCH instead. ;; -;; If you don't use font-lock, additional problems will appear: -;; 1) Regular expression delimiters do not act as quotes, so special -;; characters such as `'"#:;[](){} may need to be backslashed -;; in regular expressions and in both parts of s/// and tr///. -;; 4) The q and qq quoting operators are not recognized; see below. -;; 5) To make variables such a $' and $#array work, perl-mode treats -;; $ just like backslash, so '$' is not treated correctly. -;; 6) Unfortunately, treating $ like \ makes ${var} be treated as an -;; unmatched }. See below. -;; 7) When ' (quote) is used as a package name separator, perl-mode -;; doesn't understand, and thinks it is seeing a quoted string. - ;; Here are some ugly tricks to bypass some of these problems: the perl ;; expression /`/ (that's a back-tick) usually evaluates harmlessly, ;; but will trick perl-mode into starting a quoted string, which @@ -218,6 +203,13 @@ (defvar perl-quote-like-pairs '((?\( . ?\)) (?\[ . ?\]) (?\{ . ?\}) (?\< . ?\>))) +(eval-and-compile + (defconst perl--syntax-exp-intro-regexp + (concat "\\(?:\\(?:^\\|[^$@&%[:word:]]\\)" + (regexp-opt '("split" "if" "unless" "until" "while" "print" + "grep" "map" "not" "or" "and" "for" "foreach")) + "\\|[-?:.,;|&+*=!~({[]\\|\\(^\\)\\)[ \t\n]*"))) + ;; FIXME: handle here-docs and regexps. ;; <<EOF <<"EOF" <<'EOF' (no space) ;; see `man perlop' @@ -278,10 +270,7 @@ ;; *opening* slash. We can afford to mis-match the closing ones ;; here, because they will be re-treated separately later in ;; perl-font-lock-special-syntactic-constructs. - ((concat "\\(?:\\(?:^\\|[^$@&%[:word:]]\\)" - (regexp-opt '("split" "if" "unless" "until" "while" "split" - "grep" "map" "not" "or" "and" "for" "foreach")) - "\\|[-?:.,;|&+*=!~({[]\\|\\(^\\)\\)[ \t\n]*\\(/\\)") + ((concat perl--syntax-exp-intro-regexp "\\(/\\)") (2 (ignore (if (and (match-end 1) ; / at BOL. (save-excursion @@ -316,10 +305,15 @@ (string-to-syntax "\""))) (perl-syntax-propertize-special-constructs end))))) ;; Here documents. - ;; TODO: Handle <<WORD. These are trickier because you need to - ;; disambiguate with the shift operator. - ("<<[ \t]*\\('[^'\n]*'\\|\"[^\"\n]*\"\\|\\\\[[:alpha:]][[:alnum:]]*\\).*\\(\n\\)" - (2 (let* ((st (get-text-property (match-beginning 2) 'syntax-table)) + ((concat + "\\(?:" + ;; << "EOF", << 'EOF', or << \EOF + "<<[ \t]*\\('[^'\n]*'\\|\"[^\"\n]*\"\\|\\\\[[:alpha:]][[:alnum:]]*\\)" + ;; The <<EOF case which needs perl--syntax-exp-intro-regexp, to + ;; disambiguate with the left-bitshift operator. + "\\|" perl--syntax-exp-intro-regexp "<<\\(?1:\\sw+\\)\\)" + ".*\\(\n\\)") + (3 (let* ((st (get-text-property (match-beginning 3) 'syntax-table)) (name (match-string 1))) (goto-char (match-end 1)) (if (save-excursion (nth 8 (syntax-ppss (match-beginning 0)))) @@ -329,7 +323,8 @@ ;; Remember the names of heredocs found on this line. (cons (pcase (aref name 0) (`?\\ (substring name 1)) - (_ (substring name 1 -1))) + ((or `?\" `?\' `?\`) (substring name 1 -1)) + (_ name)) (cdr st))))))) ;; We don't call perl-syntax-propertize-special-constructs directly ;; from the << rule, because there might be other elements (between |