From: Alan Mackenzie Date: Fri, 29 Sep 2023 12:07:32 +0000 (+0000) Subject: CC Mode: Fix bug in whitespace scanning functions X-Git-Url: http://git.eshelyaron.com/gitweb/?a=commitdiff_plain;h=688c893b18e757249e8b2f6a61964a76417a75bb;p=emacs.git CC Mode: Fix bug in whitespace scanning functions c-forward-sws and c-backward-sws were scanning over spaces and linefeeds marked with the string-fence syntax-table text property. Fix this by (i) removing the WS text properties c-in-sws and c-is-sws from characters when setting the string-fence on them; (ii) checking the syntax of "space" characters when scanning over them. * lisp/progmodes/cc-defs.el (c-skip-ws-chars-forward) c-skip-ws-chars-backward, c-put-string-fence): New macros. * lisp/progmodes/cc-awk.el (c-awk-set-string-regexp-syntax-table-properties): Use c-put-string-fence. * lisp/progmodes/cc-engine.el (c-beginning-of-statement-1): Correct the determination of macro-start. (c-forward-sws, c-backward-sws): Replace skip-chars-forward by c-skip-ws-chars-forward and skip-chars-backward by c-skip-ws-chars-backward. (c-unmark-<>-around-region, c-after-change-unmark-ml-strings) (c-propertize-ml-string-opener): Use c-put-string-fence. * lisp/progmodes/cc-mode.el (c-put-syn-tab): Use c-put-string-fence when appropriate. --- diff --git a/lisp/progmodes/cc-awk.el b/lisp/progmodes/cc-awk.el index c367341345d..22f63bb5be7 100644 --- a/lisp/progmodes/cc-awk.el +++ b/lisp/progmodes/cc-awk.el @@ -754,14 +754,14 @@ (if (eq (char-after beg) ?_) (setq beg (1+ beg))) ;; First put the properties on the delimiters. - (cond ((eq end (point-max)) ; string/regexp terminated by EOB - (c-put-char-property beg 'syntax-table '(15))) ; (15) = "string fence" - ((/= (char-after beg) (char-after end)) ; missing end delimiter - (c-put-char-property beg 'syntax-table '(15)) - (c-put-char-property end 'syntax-table '(15))) - ((eq (char-after beg) ?/) ; Properly bracketed regexp - (c-put-char-property beg 'syntax-table '(7)) ; (7) = "string" - (c-put-char-property end 'syntax-table '(7))) + (cond ((eq end (point-max)) ; string/regexp terminated by EOB + (c-put-string-fence beg)) + ((/= (char-after beg) (char-after end)) ; missing end delimiter + (c-put-string-fence beg) + (c-put-string-fence end)) + ((eq (char-after beg) ?/) ; Properly bracketed regexp + (c-put-char-property beg 'syntax-table '(7)) ; (7) = "string" + (c-put-char-property end 'syntax-table '(7))) (t)) ; Properly bracketed string: Nothing to do. ;; Now change the properties of any escaped "s in the string to punctuation. (save-excursion diff --git a/lisp/progmodes/cc-defs.el b/lisp/progmodes/cc-defs.el index 6e4b570c2e8..8662e0cade6 100644 --- a/lisp/progmodes/cc-defs.el +++ b/lisp/progmodes/cc-defs.el @@ -733,9 +733,10 @@ various buffer change hooks." (defmacro c-forward-syntactic-ws (&optional limit) "Forward skip over syntactic whitespace. -Syntactic whitespace is defined as whitespace characters, comments, -and preprocessor directives. However if point starts inside a comment -or preprocessor directive, the content of it is not treated as +Syntactic whitespace is defined as whitespace characters with +whitespace (or comment-end) syntax, comments, and preprocessor +directives. However if point starts inside a comment or +preprocessor directive, the content of it is not treated as whitespace. LIMIT sets an upper limit of the forward movement, if specified. If @@ -755,9 +756,10 @@ comment at the start of cc-engine.el for more info." (defmacro c-backward-syntactic-ws (&optional limit) "Backward skip over syntactic whitespace. -Syntactic whitespace is defined as whitespace characters, comments, -and preprocessor directives. However if point starts inside a comment -or preprocessor directive, the content of it is not treated as +Syntactic whitespace is defined as whitespace characters with +whitespace (or comment-end) syntax, comments, and preprocessor +directives. However if point starts inside a comment or +preprocessor directive, the content of it is not treated as whitespace. LIMIT sets a lower limit of the backward movement, if specified. If @@ -1102,6 +1104,38 @@ continuations." (eq (char-before) ?\\))) (backward-char)))) +(defmacro c-skip-ws-chars-forward (string &optional lim) + ;; Move point forward, stopping before a char which isn't in STRING, or a + ;; char whose syntax isn't whitespace or comment-end, or at pos LIM. + ;; Note that \n usually has comment-end syntax. + ;; + ;; Returns the distance traveled, either zero or positive. + (declare (debug t)) + `(let ((-lim- ,lim) + (here (point)) + count) + (setq count (skip-chars-forward ,string -lim-)) + (when (> count 0) + (goto-char here) + (setq count (skip-syntax-forward " >" (+ here count)))) + count)) + +(defmacro c-skip-ws-chars-backward (string &optional lim) + ;; Move point backward, stopping after a char which isn't in STRING, or a + ;; char whose syntax isn't whitespace or comment-end, or at pos LIM. Note + ;; that \n usually has comment-end syntax. + ;; + ;; Returns the distance traveled, either zero or negative. + (declare (debug t)) + `(let ((-lim- ,lim) + (here (point)) + count) + (setq count (skip-chars-backward ,string -lim-)) + (when (< count 0) + (goto-char here) + (setq count (skip-syntax-backward " >" (+ here count)))) + count)) + (eval-and-compile (defvar c-langs-are-parametric nil)) @@ -1208,6 +1242,17 @@ MODE is either a mode symbol or a list of mode symbols." `((setq c-syntax-table-hwm (min c-syntax-table-hwm -pos-)))) (put-text-property -pos- (1+ -pos-) ',property ,value)))) +(defmacro c-put-string-fence (pos) + ;; Put the string-fence syntax-table text property at POS. + ;; Since the character there cannot then count as syntactic whitespace, + ;; clear the properties `c-is-sws' and `c-in-sws' (see functions + ;; `c-forward-sws' and `c-backward-sws' in cc-engine.el for details). + (declare (debug t)) + `(let ((-pos- ,pos)) + (c-put-char-property -pos- 'syntax-table '(15)) + (c-clear-char-property -pos- 'c-is-sws) + (c-clear-char-property -pos- 'c-in-sws))) + (eval-and-compile ;; Constant to decide at compilation time whether to use category ;; properties. Currently (2010-03) they're available only on GNU diff --git a/lisp/progmodes/cc-engine.el b/lisp/progmodes/cc-engine.el index abcc20fcb82..e687f44d657 100644 --- a/lisp/progmodes/cc-engine.el +++ b/lisp/progmodes/cc-engine.el @@ -976,10 +976,10 @@ comment at the start of cc-engine.el for more info." (point-min))) (widen) - (if (save-excursion - (and (c-beginning-of-macro) - (/= (point) start))) - (setq macro-start (point))) + (save-excursion + (if (and (c-beginning-of-macro) + (/= (point) start)) + (setq macro-start (point)))) ;; Try to skip back over unary operator characters, to register ;; that we've moved. @@ -2130,7 +2130,7 @@ comment at the start of cc-engine.el for more info." ;; Skip simple ws and do a quick check on the following character to see ;; if it's anything that can't start syntactic ws, so we can bail out ;; early in the majority of cases when there just are a few ws chars. - (skip-chars-forward " \t\n\r\f\v") + (c-skip-ws-chars-forward " \t\n\r\f\v") (when (or (looking-at c-syntactic-ws-start) (and c-opt-cpp-prefix (looking-at c-noise-macro-name-re)) @@ -2180,7 +2180,7 @@ comment at the start of cc-engine.el for more info." rung-pos (point) (point-max)) (setq rung-pos (point)) - (and (> (skip-chars-forward " \t\n\r\f\v") 0) + (and (> (c-skip-ws-chars-forward " \t\n\r\f\v") 0) (not (eobp)))) ;; We'll loop here if there is simple ws after the last rung. @@ -2246,7 +2246,7 @@ comment at the start of cc-engine.el for more info." (and c-opt-cpp-prefix (looking-at c-opt-cpp-start) (setq macro-start (point)) - (progn (skip-chars-backward " \t") + (progn (c-skip-ws-chars-backward " \t") (bolp)) (or (bobp) (progn (backward-char) @@ -2286,7 +2286,7 @@ comment at the start of cc-engine.el for more info." ;; We've searched over a piece of non-white syntactic ws. See if this ;; can be cached. (setq next-rung-pos (point)) - (skip-chars-forward " \t\n\r\f\v") + (c-skip-ws-chars-forward " \t\n\r\f\v") (setq rung-end-pos (min (1+ (point)) (point-max))) (if (or @@ -2383,7 +2383,7 @@ comment at the start of cc-engine.el for more info." ;; bail out early in the majority of cases when there just are a few ws ;; chars. Newlines are complicated in the backward direction, so we can't ;; skip over them. - (skip-chars-backward " \t\f") + (c-skip-ws-chars-backward " \t\f") (when (and (not (bobp)) (save-excursion (or (and @@ -2411,7 +2411,7 @@ comment at the start of cc-engine.el for more info." (setq simple-ws-beg (or attr-end ; After attribute. (match-end 1) ; Noise macro, etc. (match-end 0))) ; c-syntactic-ws-end - (skip-chars-backward " \t\n\r\f\v") + (c-skip-ws-chars-backward " \t\n\r\f\v") (if (setq rung-is-marked (text-property-any (point) (min (1+ rung-pos) (point-max)) 'c-is-sws t)) @@ -2448,10 +2448,10 @@ comment at the start of cc-engine.el for more info." (point) rung-pos (point-min)) (setq rung-pos (point)) - (if (and (< (min (skip-chars-backward " \t\f\v") + (if (and (< (min (c-skip-ws-chars-backward " \t\f\v") (progn (setq simple-ws-beg (point)) - (skip-chars-backward " \t\n\r\f\v"))) + (c-skip-ws-chars-backward " \t\n\r\f\v"))) 0) (setq rung-is-marked (text-property-any (point) rung-pos @@ -2531,7 +2531,7 @@ comment at the start of cc-engine.el for more info." ;; the macro, and then `simple-ws-beg' must be kept on the ;; same side of those comments. (goto-char simple-ws-beg) - (skip-chars-backward " \t\n\r\f\v") + (c-skip-ws-chars-backward " \t\n\r\f\v") (if (eq (char-before) ?\\) (forward-char)) (forward-line 1) @@ -2544,7 +2544,7 @@ comment at the start of cc-engine.el for more info." t))) ((/= (save-excursion - (skip-chars-forward " \t\n\r\f\v" simple-ws-beg) + (c-skip-ws-chars-forward " \t\n\r\f\v" simple-ws-beg) (setq next-rung-pos (point))) simple-ws-beg) ;; Skipped over comments. Must put point at the end of @@ -2581,7 +2581,7 @@ comment at the start of cc-engine.el for more info." ;; We've searched over a piece of non-white syntactic ws. See if this ;; can be cached. (setq next-rung-pos (point)) - (skip-chars-backward " \t\f\v") + (c-skip-ws-chars-backward " \t\f\v") (if (or ;; Cache if we started either from a marked rung or from a @@ -2591,7 +2591,7 @@ comment at the start of cc-engine.el for more info." ;; Cache if there's a marked rung in the encountered simple ws. (save-excursion - (skip-chars-backward " \t\n\r\f\v") + (c-skip-ws-chars-backward " \t\n\r\f\v") (text-property-any (point) (min (1+ next-rung-pos) (point-max)) 'c-is-sws t))) @@ -7202,10 +7202,8 @@ comment at the start of cc-engine.el for more info." (progn (c-clear-char-property (1- beg-literal-end) 'syntax-table) - (c-put-char-property (1- end-literal-end) - 'syntax-table '(15))) - (c-put-char-property (1- beg-literal-end) - 'syntax-table '(15)) + (c-put-string-fence (1- end-literal-end))) + (c-put-string-fence (1- beg-literal-end)) (c-clear-char-property (1- end-literal-end) 'syntax-table))) @@ -7284,10 +7282,8 @@ comment at the start of cc-engine.el for more info." (progn (c-clear-char-property (1- beg-literal-end) 'syntax-table) - (c-put-char-property (1- end-literal-end) - 'syntax-table '(15))) - (c-put-char-property (1- beg-literal-end) - 'syntax-table '(15)) + (c-put-string-fence (1- end-literal-end))) + (c-put-string-fence (1- beg-literal-end)) (c-clear-char-property (1- end-literal-end) 'syntax-table))))) ;; Extend the fontification region, if needed. @@ -7946,7 +7942,7 @@ multi-line strings (but not C++, for example)." (insert (nth 3 (car state)))) ((eq (nth 3 (car state)) t) (insert ?\") - (c-put-char-property end 'syntax-table '(15)))) + (c-put-string-fence end))) (c-truncate-lit-pos-cache end) ;; ....ensure c-new-END extends right to the end of the about ;; to be un-stringed raw string.... @@ -8191,7 +8187,7 @@ multi-line strings (but not C++, for example)." (goto-char (cadr end-delim)) t) (c-put-char-property (cddr delim) 'syntax-table '(1)) - (c-put-char-property (1- (cadr delim)) 'syntax-table '(15)) + (c-put-string-fence (1- (cadr delim))) (c-truncate-lit-pos-cache (1- (cddr delim))) (when bound ;; In a CPP construct, we try to apply a generic-string @@ -8221,10 +8217,10 @@ multi-line strings (but not C++, for example)." (cadr delim) t)) (if (match-beginning 10) (progn - (c-put-char-property (match-beginning 10) 'syntax-table '(15)) + (c-put-string-fence (match-beginning 10)) (c-truncate-lit-pos-cache (match-beginning 10))) (c-put-char-property (match-beginning 5) 'syntax-table '(1)) - (c-put-char-property (1+ (match-beginning 5)) 'syntax-table '(15)) + (c-put-string-fence (1+ (match-beginning 5))) (c-truncate-lit-pos-cache (match-beginning 5)))) (goto-char bound)) nil)) diff --git a/lisp/progmodes/cc-mode.el b/lisp/progmodes/cc-mode.el index 1dbe91b5633..8dea599ed98 100644 --- a/lisp/progmodes/cc-mode.el +++ b/lisp/progmodes/cc-mode.el @@ -1279,7 +1279,9 @@ Note that the style variables are always made local to the buffer." ;; VALUE (which should not be nil). ;; `(let ((-pos- ,pos) ;; (-value- ,value)) - (c-put-char-property pos 'syntax-table value) + (if (equal value '(15)) + (c-put-string-fence pos) + (c-put-char-property pos 'syntax-table value)) (c-put-char-property pos 'c-fl-syn-tab value) (cond ((null c-min-syn-tab-mkr)