1 ;;; cc-langs.el --- language specific settings for CC Mode -*- coding: utf-8 -*-
3 ;; Copyright (C) 1985, 1987, 1992-2014 Free Software Foundation, Inc.
5 ;; Authors: 2002- Alan Mackenzie
6 ;; 1998- Martin Stjernholm
7 ;; 1992-1999 Barry A. Warsaw
10 ;; 1985 Richard M. Stallman
11 ;; Maintainer: bug-cc-mode@gnu.org
12 ;; Created: 22-Apr-1997 (split from cc-mode.el)
13 ;; Keywords: c languages
16 ;; This file is part of GNU Emacs.
18 ;; GNU Emacs is free software: you can redistribute it and/or modify
19 ;; it under the terms of the GNU General Public License as published by
20 ;; the Free Software Foundation, either version 3 of the License, or
21 ;; (at your option) any later version.
23 ;; GNU Emacs is distributed in the hope that it will be useful,
24 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
25 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
26 ;; GNU General Public License for more details.
28 ;; You should have received a copy of the GNU General Public License
29 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
33 ;; HACKERS NOTE: There's heavy macro magic here. If you need to make
34 ;; changes in this or other files containing `c-lang-defconst' but
35 ;; don't want to read through the longer discussion below then read
38 ;; o A change in a `c-lang-defconst' or `c-lang-defvar' will not take
39 ;; effect if the file containing the mode init function (typically
40 ;; cc-mode.el) is byte compiled.
41 ;; o To make changes show in font locking you need to reevaluate the
42 ;; `*-font-lock-keywords-*' constants, which normally is easiest to
43 ;; do with M-x eval-buffer in cc-fonts.el.
44 ;; o In either case it's necessary to reinitialize the mode to make
45 ;; the changes show in an existing buffer.
47 ;;; Introduction to the language dependent variable system:
49 ;; This file contains all the language dependent variables, except
50 ;; those specific for font locking which reside in cc-fonts.el. As
51 ;; far as possible, all the differences between the languages that CC
52 ;; Mode supports are described with these variables only, so that the
53 ;; code can be shared.
55 ;; The language constant system (see cc-defs.el) is used to specify
56 ;; various language dependent info at a high level, such as lists of
57 ;; keywords, and then from them generate - at compile time - the
58 ;; various regexps and other low-level structures actually employed in
59 ;; the code at runtime.
61 ;; This system is also designed to make it easy for developers of
62 ;; derived modes to customize the source constants for new language
63 ;; variants, without having to keep up with the exact regexps etc that
64 ;; are used in each CC Mode version. It's possible from an external
65 ;; package to add a new language by inheriting an existing one, and
66 ;; then change specific constants as necessary for the new language.
67 ;; The old values for those constants (and the values of all the other
68 ;; high-level constants) may be used to build the new ones, and those
69 ;; new values will in turn be used by the low-level definitions here
70 ;; to build the runtime constants appropriately for the new language
71 ;; in the current version of CC Mode.
73 ;; Like elsewhere in CC Mode, the existence of a doc string signifies
74 ;; that a language constant is part of the external API, and that it
75 ;; therefore can be used with a high confidence that it will continue
76 ;; to work with future versions of CC Mode. Even so, it's not
77 ;; unlikely that such constants will change meaning slightly as this
78 ;; system is refined further; a certain degree of dependence on the CC
79 ;; Mode version is unavoidable when hooking in at this level. Also
80 ;; note that there's still work to be done to actually use these
81 ;; constants everywhere inside CC Mode; there are still hardcoded
82 ;; values in many places in the code.
84 ;; Separate packages will also benefit from the compile time
85 ;; evaluation; the byte compiled file(s) for them will contain the
86 ;; compiled runtime constants ready for use by (the byte compiled) CC
87 ;; Mode, and the source definitions in this file don't have to be
88 ;; loaded then. However, if a byte compiled package is loaded that
89 ;; has been compiled with a different version of CC Mode than the one
90 ;; currently loaded, then the compiled-in values will be discarded and
91 ;; new ones will be built when the mode is initialized. That will
92 ;; automatically trig a load of the file(s) containing the source
93 ;; definitions (i.e. this file and/or cc-fonts.el) if necessary.
95 ;; A small example of a derived mode is available at
96 ;; <http://cc-mode.sourceforge.net/derived-mode-ex.el>. It also
97 ;; contains some useful hints for derived mode developers.
99 ;;; Using language variables:
101 ;; The `c-lang-defvar' forms in this file comprise the language
102 ;; variables that CC Mode uses. It does not work to use
103 ;; `c-lang-defvar' anywhere else (which isn't much of a limitation
104 ;; since these variables sole purpose is to interface with the CC Mode
105 ;; core functions). The values in these `c-lang-defvar's are not
106 ;; evaluated right away but instead collected to a single large `setq'
107 ;; that can be inserted for a particular language with the
108 ;; `c-init-language-vars' macro.
110 ;; This file is only required at compile time, or when not running
111 ;; from byte compiled files, or when the source definitions for the
112 ;; language constants are requested.
118 (unless (fboundp 'declare-function) (defmacro declare-function (&rest r))))
122 (if (and (boundp 'byte-compile-dest-file)
123 (stringp byte-compile-dest-file))
124 (cons (file-name-directory byte-compile-dest-file) load-path)
126 (load "cc-bytecomp" nil t)))
128 (cc-require 'cc-defs)
129 (cc-require 'cc-vars)
132 ;; This file is not always loaded. See note above.
133 ;; Except it is always loaded - see bug#17463.
134 ;;;(cc-external-require 'cl)
138 ;;; Setup for the `c-lang-defvar' system.
141 ;; These are used to collect the init forms from the subsequent
142 ;; `c-lang-defvar' and `c-lang-setvar'. They are used to build the
143 ;; lambda in `c-make-init-lang-vars-fun' below, and to build `defvar's
144 ;; and `make-variable-buffer-local's in cc-engine and
145 ;; `make-local-variable's in `c-init-language-vars-for'.
146 (defvar c-lang-variable-inits nil)
147 (defvar c-lang-variable-inits-tail nil)
148 (setq c-lang-variable-inits (list nil)
149 c-lang-variable-inits-tail c-lang-variable-inits)
150 (defvar c-emacs-variable-inits nil)
151 (defvar c-emacs-variable-inits-tail nil)
152 (setq c-emacs-variable-inits (list nil)
153 c-emacs-variable-inits-tail c-emacs-variable-inits))
155 (defmacro c-lang-defvar (var val &optional doc)
156 "Declares the buffer local variable VAR to get the value VAL. VAL is
157 evaluated and assigned at mode initialization. More precisely, VAL is
158 evaluated and bound to VAR when the result from the macro
159 `c-init-language-vars' is evaluated.
161 `c-lang-const' is typically used in VAL to get the right value for the
162 language being initialized, and such calls will be macro expanded to
163 the evaluated constant value at compile time."
166 (eq (car-safe val) 'c-lang-const)
169 ;; Special case: If there's no docstring and the value is a
170 ;; simple (c-lang-const foo) where foo is the same name as VAR
171 ;; then take the docstring from the language constant foo.
172 (setq doc (get (intern (symbol-name (nth 1 val)) c-lang-constants)
173 'variable-documentation)))
177 (let ((elem (assq var (cdr c-lang-variable-inits))))
179 (setcdr elem (list val doc))
180 (setcdr c-lang-variable-inits-tail (list (list var val doc)))
181 (setq c-lang-variable-inits-tail (cdr c-lang-variable-inits-tail))))
183 ;; Return the symbol, like the other def* forms.
186 (defmacro c-lang-setvar (var val)
187 "Causes the variable VAR to be made buffer local and to get set to the
188 value VAL. VAL is evaluated and assigned at mode initialization. More
189 precisely, VAL is evaluated and bound to VAR when the result from the
190 macro `c-init-language-vars' is evaluated. VAR is typically a standard
191 Emacs variable like `comment-start'.
193 `c-lang-const' is typically used in VAL to get the right value for the
194 language being initialized, and such calls will be macro expanded to
195 the evaluated constant value at compile time."
196 (let ((elem (assq var (cdr c-emacs-variable-inits))))
198 (setcdr elem (list val)) ; Maybe remove "list", sometime. 2006-07-19
199 (setcdr c-emacs-variable-inits-tail (list (list var val)))
200 (setq c-emacs-variable-inits-tail (cdr c-emacs-variable-inits-tail))))
202 ;; Return the symbol, like the other def* forms.
205 (put 'c-lang-defvar 'lisp-indent-function 'defun)
206 ; (eval-after-load "edebug" ; 2006-07-09: def-edebug-spec is now in subr.el.
208 (def-edebug-spec c-lang-defvar
209 (&define name def-form &optional stringp)) ;)
211 ;; Suppress "might not be defined at runtime" warning.
212 ;; This file is only used when compiling other cc files.
213 ;; These are defined in cl as aliases to the cl- versions.
214 ;(declare-function delete-duplicates "cl-seq" (cl-seq &rest cl-keys) t)
215 ;(declare-function mapcan "cl-extra" (cl-func cl-seq &rest cl-rest) t)
218 ;; Some helper functions used when building the language constants.
220 (defun c-filter-ops (ops opgroup-filter op-filter &optional xlate)
221 ;; Extract a subset of the operators in the list OPS in a DWIM:ey
222 ;; way. The return value is a plain list of operators:
224 ;; OPS either has the structure of `c-operators', is a single
225 ;; group in `c-operators', or is a plain list of operators.
227 ;; OPGROUP-FILTER specifies how to select the operator groups. It
228 ;; can be t to choose all groups, a list of group type symbols
229 ;; (such as 'prefix) to accept, or a function which will be called
230 ;; with the group symbol for each group and should return non-nil
231 ;; if that group is to be included.
233 ;; If XLATE is given, it's a function which is called for each
234 ;; matching operator and its return value is collected instead.
235 ;; If it returns a list, the elements are spliced directly into
236 ;; the final result, which is returned as a list with duplicates
237 ;; removed using `equal'.
239 ;; `c-mode-syntax-table' for the current mode is in effect during
240 ;; the whole procedure.
241 (unless (listp (car-safe ops))
242 (setq ops (list ops)))
243 (cond ((eq opgroup-filter t)
244 (setq opgroup-filter (lambda (opgroup) t)))
245 ((not (functionp opgroup-filter))
246 (setq opgroup-filter `(lambda (opgroup)
247 (memq opgroup ',opgroup-filter)))))
248 (cond ((eq op-filter t)
249 (setq op-filter (lambda (op) t)))
251 (setq op-filter `(lambda (op)
252 (string-match ,op-filter op)))))
254 (setq xlate 'identity))
255 (c-with-syntax-table (c-lang-const c-mode-syntax-table)
256 (cl-delete-duplicates
257 (cl-mapcan (lambda (opgroup)
258 (when (if (symbolp (car opgroup))
259 (when (funcall opgroup-filter (car opgroup))
260 (setq opgroup (cdr opgroup))
263 (cl-mapcan (lambda (op)
264 (when (funcall op-filter op)
265 (let ((res (funcall xlate op)))
266 (if (listp res) res (list res)))))
272 ;;; Various mode specific values that aren't language related.
274 (c-lang-defconst c-mode-menu
275 ;; The definition for the mode menu. The menu title is prepended to
276 ;; this before it's fed to `easy-menu-define'.
277 t `(["Comment Out Region" comment-region
278 (c-fn-region-is-active-p)]
279 ["Uncomment Region" (comment-region (region-beginning)
281 (c-fn-region-is-active-p)]
282 ["Indent Expression" c-indent-exp
283 (memq (char-after) '(?\( ?\[ ?\{))]
284 ["Indent Line or Region" c-indent-line-or-region t]
285 ["Fill Comment Paragraph" c-fill-paragraph t]
287 ["Backward Statement" c-beginning-of-statement t]
288 ["Forward Statement" c-end-of-statement t]
289 ,@(when (c-lang-const c-opt-cpp-prefix)
290 ;; Only applicable if there's a cpp preprocessor.
291 `(["Up Conditional" c-up-conditional t]
292 ["Backward Conditional" c-backward-conditional t]
293 ["Forward Conditional" c-forward-conditional t]
295 ["Macro Expand Region" c-macro-expand
296 (c-fn-region-is-active-p)]
297 ["Backslashify" c-backslash-region
298 (c-fn-region-is-active-p)]))
301 ["Set Style..." c-set-style t]
302 ["Show Current Style Name" (message
304 c-indentation-style) t]
305 ["Guess Style from this Buffer" c-guess-buffer-no-install t]
306 ["Install the Last Guessed Style..." c-guess-install
307 (and c-guess-guessed-offsets-alist
308 c-guess-guessed-basic-offset) ]
309 ["View the Last Guessed Style" c-guess-view
310 (and c-guess-guessed-offsets-alist
311 c-guess-guessed-basic-offset) ])
314 ["Syntactic indentation" c-toggle-syntactic-indentation
315 :style toggle :selected c-syntactic-indentation]
316 ["Electric mode" c-toggle-electric-state
317 :style toggle :selected c-electric-flag]
318 ["Auto newline" c-toggle-auto-newline
319 :style toggle :selected c-auto-newline]
320 ["Hungry delete" c-toggle-hungry-state
321 :style toggle :selected c-hungry-delete-key]
322 ["Subword mode" subword-mode
323 :style toggle :selected (and (boundp 'subword-mode)
329 (defun c-populate-syntax-table (table)
330 "Populate the given syntax table as necessary for a C-like language.
331 This includes setting ' and \" as string delimiters, and setting up
332 the comment syntax to handle both line style \"//\" and block style
333 \"/*\" \"*/\" comments."
335 (modify-syntax-entry ?_ "_" table)
336 (modify-syntax-entry ?\\ "\\" table)
337 (modify-syntax-entry ?+ "." table)
338 (modify-syntax-entry ?- "." table)
339 (modify-syntax-entry ?= "." table)
340 (modify-syntax-entry ?% "." table)
341 (modify-syntax-entry ?< "." table)
342 (modify-syntax-entry ?> "." table)
343 (modify-syntax-entry ?& "." table)
344 (modify-syntax-entry ?| "." table)
345 (modify-syntax-entry ?\' "\"" table)
346 (modify-syntax-entry ?\240 "." table)
348 ;; Set up block and line oriented comments. The new C
349 ;; standard mandates both comment styles even in C, so since
350 ;; all languages now require dual comments, we make this the
354 ((memq '8-bit c-emacs-features)
355 (modify-syntax-entry ?/ ". 1456" table)
356 (modify-syntax-entry ?* ". 23" table))
358 ((memq '1-bit c-emacs-features)
359 (modify-syntax-entry ?/ ". 124b" table)
360 (modify-syntax-entry ?* ". 23" table))
362 (t (error "CC Mode is incompatible with this version of Emacs")))
364 (modify-syntax-entry ?\n "> b" table)
365 ;; Give CR the same syntax as newline, for selective-display
366 (modify-syntax-entry ?\^m "> b" table))
368 (c-lang-defconst c-make-mode-syntax-table
369 "Functions that generates the mode specific syntax tables.
370 The syntax tables aren't stored directly since they're quite large."
372 (let ((table (make-syntax-table)))
373 (c-populate-syntax-table table)
374 ;; Mode specific syntaxes.
375 ,(cond ((or (c-major-mode-is 'objc-mode) (c-major-mode-is 'java-mode))
376 ;; Let '@' be part of symbols in ObjC to cope with
377 ;; its compiler directives as single keyword tokens.
378 ;; This is then necessary since it's assumed that
379 ;; every keyword is a single symbol.
380 `(modify-syntax-entry ?@ "_" table))
381 ((c-major-mode-is 'pike-mode)
382 `(modify-syntax-entry ?@ "." table)))
385 (c-lang-defconst c-mode-syntax-table
386 ;; The syntax tables in evaluated form. Only used temporarily when
387 ;; the constants in this file are evaluated.
388 t (funcall (c-lang-const c-make-mode-syntax-table)))
390 (c-lang-defconst c++-make-template-syntax-table
391 ;; A variant of `c++-mode-syntax-table' that defines `<' and `>' as
392 ;; parenthesis characters. Used temporarily when template argument
393 ;; lists are parsed. Note that this encourages incorrect parsing of
394 ;; templates since they might contain normal operators that uses the
395 ;; '<' and '>' characters. Therefore this syntax table might go
396 ;; away when CC Mode handles templates correctly everywhere. WHILE
397 ;; THIS SYNTAX TABLE IS CURRENT, `c-parse-state' MUST _NOT_ BE
400 (java c++) `(lambda ()
401 (let ((table (funcall ,(c-lang-const c-make-mode-syntax-table))))
402 (modify-syntax-entry ?< "(>" table)
403 (modify-syntax-entry ?> ")<" table)
405 (c-lang-defvar c++-template-syntax-table
406 (and (c-lang-const c++-make-template-syntax-table)
407 (funcall (c-lang-const c++-make-template-syntax-table))))
409 (c-lang-defconst c-make-no-parens-syntax-table
410 ;; A variant of the standard syntax table which is used to find matching
411 ;; "<"s and ">"s which have been marked as parens using syntax table
412 ;; properties. The other paren characters (e.g. "{", ")" "]") are given a
413 ;; non-paren syntax here. so that the list commands will work on "< ... >"
414 ;; even when there's unbalanced other parens inside them.
416 ;; This variable is nil for languages which don't have template stuff.
417 t (if (c-lang-const c-recognize-<>-arglists)
419 ;(if (c-lang-const c-recognize-<>-arglists)
420 (let ((table (funcall ,(c-lang-const c-make-mode-syntax-table))))
421 (modify-syntax-entry ?\( "." table)
422 (modify-syntax-entry ?\) "." table)
423 (modify-syntax-entry ?\[ "." table)
424 (modify-syntax-entry ?\] "." table)
425 (modify-syntax-entry ?\{ "." table)
426 (modify-syntax-entry ?\} "." table)
428 (c-lang-defvar c-no-parens-syntax-table
429 (and (c-lang-const c-make-no-parens-syntax-table)
430 (funcall (c-lang-const c-make-no-parens-syntax-table))))
432 (c-lang-defconst c-identifier-syntax-modifications
433 "A list that describes the modifications that should be done to the
434 mode syntax table to get a syntax table that matches all identifiers
435 and keywords as words.
437 The list is just like the one used in `font-lock-defaults': Each
438 element is a cons where the car is the character to modify and the cdr
439 the new syntax, as accepted by `modify-syntax-entry'."
440 ;; The $ character is not allowed in most languages (one exception
441 ;; is Java which allows it for legacy reasons) but we still classify
442 ;; it as an identifier character since it's often used in various
443 ;; machine generated identifiers.
444 t '((?_ . "w") (?$ . "w"))
445 (objc java) (append '((?@ . "w"))
446 (c-lang-const c-identifier-syntax-modifications))
448 (c-lang-defvar c-identifier-syntax-modifications
449 (c-lang-const c-identifier-syntax-modifications))
451 (c-lang-defvar c-identifier-syntax-table
452 (let ((table (copy-syntax-table (c-mode-var "mode-syntax-table")))
453 (mods c-identifier-syntax-modifications)
458 (modify-syntax-entry (car mod) (cdr mod) table))
460 "Syntax table built on the mode syntax table but additionally
461 classifies symbol constituents like '_' and '$' as word constituents,
462 so that all identifiers are recognized as words.")
464 (c-lang-defconst c-get-state-before-change-functions
465 ;; For documentation see the following c-lang-defvar of the same name.
466 ;; The value here may be a list of functions or a single function.
468 c++ '(c-extend-region-for-CPP
469 c-before-change-check-<>-operators
470 c-invalidate-macro-cache)
471 (c objc) '(c-extend-region-for-CPP c-invalidate-macro-cache)
472 ;; java 'c-before-change-check-<>-operators
473 awk 'c-awk-record-region-clear-NL)
474 (c-lang-defvar c-get-state-before-change-functions
475 (let ((fs (c-lang-const c-get-state-before-change-functions)))
479 "If non-nil, a list of functions called from c-before-change-hook.
480 Typically these will record enough state to allow
481 `c-before-font-lock-function' to extend the region to fontify,
482 and may do such things as removing text-properties which must be
485 These functions will be run in the order given. Each of them
486 takes 2 parameters, the BEG and END supplied to every
487 before-change function; on entry, the buffer will have been
488 widened and match-data will have been saved; point is undefined
489 on both entry and exit; the return value is ignored.
491 The functions are called even when font locking isn't enabled.
493 When the mode is initialized, the functions are called with
494 parameters \(point-min) and \(point-max).")
496 (c-lang-defconst c-before-font-lock-functions
497 ;; For documentation see the following c-lang-defvar of the same name.
498 ;; The value here may be a list of functions or a single function.
499 t 'c-change-set-fl-decl-start
500 (c c++ objc) '(c-neutralize-syntax-in-and-mark-CPP
501 c-change-set-fl-decl-start)
502 awk 'c-awk-extend-and-syntax-tablify-region)
503 (c-lang-defvar c-before-font-lock-functions
504 (let ((fs (c-lang-const c-before-font-lock-functions)))
508 "If non-nil, a list of functions called just before font locking.
509 Typically they will extend the region about to be fontified \(see
510 below) and will set `syntax-table' text properties on the region.
512 These functions will be run in the order given. Each of them
513 takes 3 parameters, the BEG, END, and OLD-LEN supplied to every
514 after-change function; point is undefined on both entry and exit;
515 on entry, the buffer will have been widened and match-data will
516 have been saved; the return value is ignored.
518 The functions may extend the region to be fontified by setting the
519 buffer local variables c-new-BEG and c-new-END.
521 The functions are called even when font locking is disabled.
523 When the mode is initialized, these functions are called with
524 parameters \(point-min), \(point-max) and <buffer size>.")
526 (c-lang-defconst c-before-context-fontification-functions
528 t 'c-context-set-fl-decl-start)
529 ;; For documentation see the following c-lang-defvar of the same name.
530 ;; The value here may be a list of functions or a single function.
531 (c-lang-defvar c-before-context-fontification-functions
532 (let ((fs (c-lang-const c-before-context-fontification-functions)))
536 "If non-nil, a list of functions called just before context (or
537 other non-change) fontification is done. Typically they will
540 These functions will be run in the order given. Each of them
541 takes 2 parameters, the BEG and END of the region to be
542 fontified. Point is undefined on both entry and exit. On entry,
543 the buffer will have been widened and match-data will have been
544 saved; the return value is a cons of the adjusted
545 region, (NEW-BEG . NEW-END).")
548 ;;; Syntactic analysis ("virtual semicolons") for line-oriented languages (AWK).
549 (c-lang-defconst c-at-vsemi-p-fn
550 "Contains a function \"Is there a virtual semicolon at POS or point?\".
551 Such a function takes one optional parameter, a buffer position (defaults to
552 point), and returns nil or t. This variable contains nil for languages which
553 don't have EOL terminated statements. "
555 (c c++ objc) 'c-at-macro-vsemi-p
556 awk 'c-awk-at-vsemi-p)
557 (c-lang-defvar c-at-vsemi-p-fn (c-lang-const c-at-vsemi-p-fn))
559 (c-lang-defconst c-vsemi-status-unknown-p-fn
560 "Contains a function \"are we unsure whether there is a virtual semicolon on this line?\".
561 The (admittedly kludgy) purpose of such a function is to prevent an infinite
562 recursion in c-beginning-of-statement-1 when point starts at a `while' token.
563 The function MUST NOT UNDER ANY CIRCUMSTANCES call c-beginning-of-statement-1,
564 even indirectly. This variable contains nil for languages which don't have
565 EOL terminated statements."
567 (c c++ objc) 'c-macro-vsemi-status-unknown-p
568 awk 'c-awk-vsemi-status-unknown-p)
569 (c-lang-defvar c-vsemi-status-unknown-p-fn
570 (c-lang-const c-vsemi-status-unknown-p-fn))
573 ;;; Lexer-level syntax (identifiers, tokens etc).
575 (c-lang-defconst c-has-bitfields
576 "Whether the language has bitfield declarations."
579 (c-lang-defvar c-has-bitfields (c-lang-const c-has-bitfields))
581 (c-lang-defconst c-symbol-start
582 "Regexp that matches the start of a symbol, i.e. any identifier or
583 keyword. It's unspecified how far it matches. Does not contain a \\|
584 operator at the top level."
585 t (concat "[" c-alpha "_]")
586 java (concat "[" c-alpha "_@]")
587 objc (concat "[" c-alpha "_@]")
588 pike (concat "[" c-alpha "_`]"))
589 (c-lang-defvar c-symbol-start (c-lang-const c-symbol-start))
591 (c-lang-defconst c-symbol-chars
592 "Set of characters that can be part of a symbol.
593 This is of the form that fits inside [ ] in a regexp."
594 ;; Pike note: With the backquote identifiers this would include most
595 ;; operator chars too, but they are handled with other means instead.
596 t (concat c-alnum "_$")
597 objc (concat c-alnum "_$@"))
598 (c-lang-defvar c-symbol-chars (c-lang-const c-symbol-chars))
600 (c-lang-defconst c-symbol-key
601 "Regexp matching identifiers and keywords (with submatch 0). Assumed
602 to match if `c-symbol-start' matches on the same position."
603 t (concat (c-lang-const c-symbol-start)
604 "[" (c-lang-const c-symbol-chars) "]*")
606 ;; Use the value from C here since the operator backquote is
607 ;; covered by the other alternative.
608 (c-lang-const c-symbol-key c)
610 (c-make-keywords-re nil
611 (c-lang-const c-overloadable-operators))))
612 (c-lang-defvar c-symbol-key (c-lang-const c-symbol-key))
614 (c-lang-defconst c-symbol-key-depth
615 ;; Number of regexp grouping parens in `c-symbol-key'.
616 t (regexp-opt-depth (c-lang-const c-symbol-key)))
618 (c-lang-defconst c-nonsymbol-chars
619 "This is the set of chars that can't be part of a symbol, i.e. the
620 negation of `c-symbol-chars'."
621 t (concat "^" (c-lang-const c-symbol-chars)))
622 (c-lang-defvar c-nonsymbol-chars (c-lang-const c-nonsymbol-chars))
624 (c-lang-defconst c-nonsymbol-key
625 "Regexp that matches any character that can't be part of a symbol.
626 It's usually appended to other regexps to avoid matching a prefix.
627 It's assumed to not contain any submatchers."
628 ;; The same thing regarding Unicode identifiers applies here as to
630 t (concat "[" (c-lang-const c-nonsymbol-chars) "]"))
632 (c-lang-defconst c-identifier-ops
633 "The operators that make up fully qualified identifiers. nil in
634 languages that don't have such things. See `c-operators' for a
635 description of the format. Binary operators can concatenate symbols,
636 e.g. \"::\" in \"A::B::C\". Prefix operators can precede identifiers,
637 e.g. \"~\" in \"~A::B\". Other types of operators aren't supported.
639 This value is by default merged into `c-operators'."
641 c++ '((prefix "~" "??-" "compl")
644 ;; Java has "." to concatenate identifiers but it's also used for
645 ;; normal indexing. There's special code in the Java font lock
646 ;; rules to fontify qualified identifiers based on the standard
647 ;; naming conventions. We still define "." here to make
648 ;; `c-forward-name' move over as long names as possible which is
649 ;; necessary to e.g. handle throws clauses correctly.
650 java '((left-assoc "."))
651 idl '((left-assoc "::")
653 pike '((left-assoc "::")
657 (c-lang-defconst c-opt-identifier-concat-key
658 ;; Appendable adorned regexp matching the operators that join
659 ;; symbols to fully qualified identifiers, or nil in languages that
660 ;; don't have such things.
662 ;; This was a docstring constant in 5.30. It still works but is now
663 ;; considered internal - change `c-identifier-ops' instead.
664 t (let ((ops (c-filter-ops (c-lang-const c-identifier-ops)
665 '(left-assoc right-assoc)
668 (c-make-keywords-re 'appendable ops))))
669 (c-lang-defvar c-opt-identifier-concat-key
670 (c-lang-const c-opt-identifier-concat-key)
673 (c-lang-defconst c-opt-identifier-concat-key-depth
674 ;; Number of regexp grouping parens in `c-opt-identifier-concat-key'.
675 t (regexp-opt-depth (c-lang-const c-opt-identifier-concat-key)))
677 (c-lang-defconst c-opt-identifier-prefix-key
678 ;; Appendable adorned regexp matching operators that might precede
679 ;; an identifier and that are part of the identifier in that case.
680 ;; nil in languages without such things.
681 t (let ((ops (c-filter-ops (c-lang-const c-identifier-ops)
685 (c-make-keywords-re 'appendable ops))))
687 (c-lang-defconst c-after-id-concat-ops
688 "Operators that can occur after a binary operator on `c-identifier-ops'
689 in identifiers. nil in languages that don't have such things.
691 Operators here should also have appropriate entries in `c-operators' -
692 it's not taken care of by default."
694 ;; '~' for destructors in C++, '*' for member pointers.
696 ;; In Java we recognize '*' to deal with "foo.bar.*" that can occur
697 ;; in import declarations. (This will also match bogus things like
698 ;; "foo.*bar" but we don't bother.)
701 (c-lang-defconst c-opt-after-id-concat-key
702 ;; Regexp that must match the token after
703 ;; `c-opt-identifier-concat-key' for it to be considered an
704 ;; identifier concatenation operator (which e.g. causes the
705 ;; preceding identifier to be fontified as a reference). Assumed to
706 ;; be a string if `c-opt-identifier-concat-key' is.
708 ;; This was a docstring constant in 5.30. It still works but is now
709 ;; considered internal - change `c-after-id-concat-ops' instead.
710 t (concat (c-lang-const c-symbol-start)
711 (if (c-lang-const c-after-id-concat-ops)
712 (concat "\\|" (c-make-keywords-re 'appendable
713 (c-lang-const c-after-id-concat-ops)))
716 (c-lang-defconst c-identifier-start
717 "Regexp that matches the start of an (optionally qualified) identifier.
718 It should also match all keywords. It's unspecified how far it
720 t (concat (c-lang-const c-symbol-start)
721 (if (c-lang-const c-opt-identifier-prefix-key)
723 (c-lang-const c-opt-identifier-prefix-key))
725 (c-lang-defvar c-identifier-start (c-lang-const c-identifier-start))
727 (c-lang-defconst c-identifier-key
728 "Regexp matching a fully qualified identifier, like \"A::B::c\" in
729 C++. It does not recognize the full range of syntactic whitespace
730 between the tokens; `c-forward-name' has to be used for that. It
731 should also not match identifiers containing parenthesis groupings,
732 e.g. identifiers with template arguments such as \"A<X,Y>\" in C++."
733 ;; This regexp is more complex than strictly necessary to ensure
734 ;; that it can be matched with a minimum of backtracking.
735 t (concat (if (c-lang-const c-opt-identifier-prefix-key)
738 (c-lang-const c-opt-identifier-prefix-key)
739 (c-lang-const c-simple-ws) "*"
742 "\\(" (c-lang-const c-symbol-key) "\\)"
743 (if (c-lang-const c-opt-identifier-concat-key)
746 (c-lang-const c-simple-ws) "*"
747 (c-lang-const c-opt-identifier-concat-key)
748 (c-lang-const c-simple-ws) "*"
749 (if (c-lang-const c-after-id-concat-ops)
752 (c-make-keywords-re 'appendable
753 (c-lang-const c-after-id-concat-ops))
755 ;; For flexibility, consider the symbol match
756 ;; optional if we've hit a
757 ;; `c-after-id-concat-ops' operator. This is
758 ;; also necessary to handle the "*" that can
759 ;; end import declaration identifiers in Java.
761 (c-lang-const c-simple-ws) "*"
762 "\\(" (c-lang-const c-symbol-key) "\\)"
765 "\\(" (c-lang-const c-symbol-key) "\\)"
767 (concat "\\(" (c-lang-const c-symbol-key) "\\)"))
770 (c-lang-defvar c-identifier-key (c-lang-const c-identifier-key))
772 (c-lang-defconst c-identifier-last-sym-match
773 ;; This was a docstring constant in 5.30 but it's no longer used.
774 ;; It's only kept to avoid breaking third party code.
776 ;; Used to identify the submatch in `c-identifier-key' that
777 ;; surrounds the last symbol in the qualified identifier. It's a
778 ;; list of submatch numbers, of which the first that has a match is
779 ;; taken. It's assumed that at least one does when the regexp has
783 (c-lang-defconst c-string-escaped-newlines
784 "Set if the language support backslash escaped newlines inside string
788 (c-lang-defvar c-string-escaped-newlines
789 (c-lang-const c-string-escaped-newlines))
791 (c-lang-defconst c-multiline-string-start-char
792 "Set if the language supports multiline string literals without escaped
793 newlines. If t, all string literals are multiline. If a character,
794 only literals where the open quote is immediately preceded by that
795 literal are multiline."
798 (c-lang-defvar c-multiline-string-start-char
799 (c-lang-const c-multiline-string-start-char))
801 (c-lang-defconst c-opt-cpp-symbol
802 "The symbol which starts preprocessor constructs when in the margin."
805 (c-lang-defvar c-opt-cpp-symbol (c-lang-const c-opt-cpp-symbol))
807 (c-lang-defconst c-opt-cpp-prefix
808 "Regexp matching the prefix of a cpp directive in the languages that
809 normally use that macro preprocessor. Tested at bol or at boi.
810 Assumed to not contain any submatches or \\| operators."
811 ;; TODO (ACM, 2005-04-01). Amend the following to recognize escaped NLs;
812 ;; amend all uses of c-opt-cpp-prefix which count regexp-depth.
815 (c-lang-defvar c-opt-cpp-prefix (c-lang-const c-opt-cpp-prefix))
817 (c-lang-defconst c-anchored-cpp-prefix
818 "Regexp matching the prefix of a cpp directive anchored to BOL,
819 in the languages that have a macro preprocessor."
820 t "^\\s *\\(#\\)\\s *"
822 (c-lang-defvar c-anchored-cpp-prefix (c-lang-const c-anchored-cpp-prefix))
824 (c-lang-defconst c-opt-cpp-start
825 "Regexp matching the prefix of a cpp directive including the directive
826 name, or nil in languages without preprocessor support. The first
827 submatch surrounds the directive name."
828 t (if (c-lang-const c-opt-cpp-prefix)
829 (concat (c-lang-const c-opt-cpp-prefix)
830 "\\([" c-alnum "]+\\)"))
831 ;; Pike, being a scripting language, recognizes hash-bangs too.
832 pike (concat (c-lang-const c-opt-cpp-prefix)
833 "\\([" c-alnum "]+\\|!\\)"))
834 (c-lang-defvar c-opt-cpp-start (c-lang-const c-opt-cpp-start))
836 (c-lang-defconst c-cpp-message-directives
837 "List of cpp directives (without the prefix) that are followed by a
839 t (if (c-lang-const c-opt-cpp-prefix)
841 (c c++ objc pike) '("error" "warning"))
843 (c-lang-defconst c-cpp-include-directives
844 "List of cpp directives (without the prefix) that are followed by a
845 file name in angle brackets or quotes."
846 t (if (c-lang-const c-opt-cpp-prefix)
848 objc '("include" "import"))
850 (c-lang-defconst c-opt-cpp-macro-define
851 "Cpp directive (without the prefix) that is followed by a macro
852 definition, or nil if the language doesn't have any."
853 t (if (c-lang-const c-opt-cpp-prefix)
855 (c-lang-defvar c-opt-cpp-macro-define
856 (c-lang-const c-opt-cpp-macro-define))
858 (c-lang-defconst c-opt-cpp-macro-define-start
859 ;; Regexp matching everything up to the macro body of a cpp define, or the
860 ;; end of the logical line if there is none. Submatch 1 is the name of the
861 ;; macro. Set if c-opt-cpp-macro-define is.
862 t (if (c-lang-const c-opt-cpp-macro-define)
863 (concat (c-lang-const c-opt-cpp-prefix)
864 (c-lang-const c-opt-cpp-macro-define)
865 "[ \t]+\\(\\(\\sw\\|_\\)+\\)\\(\([^\)]*\)\\)?"
867 "\\([ \t]\\|\\\\\n\\)*")))
868 (c-lang-defvar c-opt-cpp-macro-define-start
869 (c-lang-const c-opt-cpp-macro-define-start))
871 (c-lang-defconst c-opt-cpp-macro-define-id
872 ;; Regexp matching everything up to the end of the identifier defined
874 t (if (c-lang-const c-opt-cpp-macro-define)
875 (concat (c-lang-const c-opt-cpp-prefix) ; #
876 (c-lang-const c-opt-cpp-macro-define) ; define
877 "[ \t]+\\(\\sw\\|_\\)+")))
878 (c-lang-defvar c-opt-cpp-macro-define-id
879 (c-lang-const c-opt-cpp-macro-define-id))
881 (c-lang-defconst c-cpp-expr-directives
882 "List of cpp directives (without the prefix) that are followed by an
884 t (if (c-lang-const c-opt-cpp-prefix)
887 (c-lang-defconst c-cpp-expr-intro-re
888 "Regexp which matches the start of a CPP directive which contains an
889 expression, or nil if there aren't any in the language."
890 t (if (c-lang-const c-cpp-expr-directives)
892 (c-lang-const c-opt-cpp-prefix)
893 (c-make-keywords-re t (c-lang-const c-cpp-expr-directives)))))
894 (c-lang-defvar c-cpp-expr-intro-re
895 (c-lang-const c-cpp-expr-intro-re))
897 (c-lang-defconst c-cpp-expr-functions
898 "List of functions in cpp expressions."
899 t (if (c-lang-const c-opt-cpp-prefix)
901 pike '("defined" "efun" "constant"))
903 (c-lang-defconst c-assignment-operators
904 "List of all assignment operators."
905 t '("=" "*=" "/=" "%=" "+=" "-=" ">>=" "<<=" "&=" "^=" "|=")
906 java (append (c-lang-const c-assignment-operators)
908 c++ (append (c-lang-const c-assignment-operators)
909 '("and_eq" "or_eq" "xor_eq" "??!=" "??'="))
912 (c-lang-defconst c-operators
913 "List describing all operators, along with their precedence and
914 associativity. The order in the list corresponds to the precedence of
915 the operators: The operators in each element are a group with the same
916 precedence, and the group has higher precedence than the groups in all
917 following elements. The car of each element describes the type of the
918 operator group, and the cdr is a list of the operator tokens in it.
919 The operator group types are:
921 'prefix Unary prefix operators.
922 'postfix Unary postfix operators.
924 Unary postfix operators if and only if the chars have
926 'left-assoc Binary left associative operators (i.e. a+b+c means (a+b)+c).
927 'right-assoc Binary right associative operators (i.e. a=b=c means a=(b=c)).
928 'right-assoc-sequence
929 Right associative operator that constitutes of a
930 sequence of tokens that separate expressions. All the
931 tokens in the group are in this case taken as
932 describing the sequence in one such operator, and the
933 order between them is therefore significant.
935 Operators containing a character with paren syntax are taken to match
936 with a corresponding open/close paren somewhere else. A postfix
937 operator with close paren syntax is taken to end a postfix expression
938 started somewhere earlier, rather than start a new one at point. Vice
939 versa for prefix operators with open paren syntax.
941 Note that operators like \".\" and \"->\" which in language references
942 often are described as postfix operators are considered binary here,
943 since CC Mode treats every identifier as an expression."
945 ;; There's currently no code in CC Mode that exploits all the info
946 ;; in this variable; precedence, associativity etc are present as a
947 ;; preparation for future work.
949 ;; FIXME!!! C++11's "auto" operators "=" and "->" need to go in here
950 ;; somewhere. 2012-03-24.
953 ,@(when (c-lang-const c-opt-cpp-prefix)
955 ,@(when (c-major-mode-is '(c-mode c++-mode))
958 ,@(when (c-major-mode-is '(c-mode c++-mode))
959 '("%:%:" "??=??=")))))
962 ,@(c-lang-const c-identifier-ops)
963 ,@(cond ((or (c-major-mode-is 'c++-mode) (c-major-mode-is 'java-mode))
964 `((postfix-if-paren "<" ">"))) ; Templates.
965 ((c-major-mode-is 'pike-mode)
966 `((prefix "global" "predef")))
967 ((c-major-mode-is 'java-mode)
968 `((prefix "super"))))
971 ,@(when (c-major-mode-is 'c++-mode)
972 ;; The following need special treatment.
973 `((prefix "dynamic_cast" "static_cast"
974 "reinterpret_cast" "const_cast" "typeid")))
976 ,@(unless (c-major-mode-is 'java-mode)
978 (postfix "++" "--" "[" "]" "(" ")"
979 ,@(when (c-major-mode-is '(c-mode c++-mode))
980 '("<:" ":>" "??(" "??)")))
983 (prefix "++" "--" "+" "-" "!" "~"
984 ,@(when (c-major-mode-is 'c++-mode) '("not" "compl"))
985 ,@(when (c-major-mode-is '(c-mode c++-mode))
986 '("*" "&" "sizeof" "??-"))
987 ,@(when (c-major-mode-is 'objc-mode)
988 '("@selector" "@protocol" "@encode"))
989 ;; The following need special treatment.
990 ,@(cond ((c-major-mode-is 'c++-mode)
992 ((c-major-mode-is 'java-mode)
994 ((c-major-mode-is 'pike-mode)
995 '("class" "lambda" "catch" "throw" "gauge")))
997 ,@(when (c-major-mode-is 'pike-mode)
998 '("[" "]"))) ; Type cast.
1000 ;; Member selection.
1001 ,@(when (c-major-mode-is 'c++-mode)
1002 `((left-assoc ".*" "->*")))
1005 (left-assoc "*" "/" "%")
1008 (left-assoc "+" "-")
1011 (left-assoc "<<" ">>"
1012 ,@(when (c-major-mode-is 'java-mode)
1016 (left-assoc "<" ">" "<=" ">="
1017 ,@(when (c-major-mode-is 'java-mode)
1021 (left-assoc "==" "!="
1022 ,@(when (c-major-mode-is 'c++-mode) '("not_eq")))
1026 ,@(when (c-major-mode-is 'c++-mode) '("bitand")))
1028 ;; Bitwise exclusive or.
1030 ,@(when (c-major-mode-is '(c-mode c++-mode))
1032 ,@(when (c-major-mode-is 'c++-mode) '("xor")))
1036 ,@(when (c-major-mode-is '(c-mode c++-mode))
1038 ,@(when (c-major-mode-is 'c++-mode) '("bitor")))
1042 ,@(when (c-major-mode-is 'c++-mode) '("and")))
1046 ,@(when (c-major-mode-is '(c-mode c++-mode))
1048 ,@(when (c-major-mode-is 'c++-mode) '("or")))
1051 (right-assoc-sequence "?" ":")
1054 (right-assoc ,@(c-lang-const c-assignment-operators))
1057 ,@(when (c-major-mode-is 'c++-mode)
1058 '((prefix "throw")))
1063 ;; IDL got its own definition since it has a much smaller operator
1064 ;; set than the other languages.
1065 idl `(;; Preprocessor.
1069 ,@(c-lang-const c-identifier-ops)
1071 (prefix "+" "-" "~")
1073 (left-assoc "*" "/" "%")
1075 (left-assoc "+" "-")
1077 (left-assoc "<<" ">>")
1085 (c-lang-defconst c-operator-list
1086 ;; The operators as a flat list (without duplicates).
1087 t (c-filter-ops (c-lang-const c-operators) t t))
1089 (c-lang-defconst c-overloadable-operators
1090 "List of the operators that are overloadable, in their \"identifier
1091 form\". See also `c-op-identifier-prefix'."
1093 c++ '("new" "delete" ;; Can be followed by "[]" but we ignore that.
1095 "^" "??'" "xor" "&" "bitand" "|" "??!" "bitor" "~" "??-" "compl"
1096 "!" "=" "<" ">" "+=" "-=" "*=" "/=" "%=" "^="
1097 "??'=" "xor_eq" "&=" "and_eq" "|=" "??!=" "or_eq"
1098 "<<" ">>" ">>=" "<<=" "==" "!=" "not_eq" "<=" ">="
1099 "&&" "and" "||" "??!??!" "or" "++" "--" "," "->*" "->"
1100 "()" "[]" "<::>" "??(??)")
1101 ;; These work like identifiers in Pike.
1102 pike '("`+" "`-" "`&" "`|" "`^" "`<<" "`>>" "`*" "`/" "`%" "`~"
1103 "`==" "`<" "`>" "`!" "`[]" "`[]=" "`->" "`->=" "`()" "``+"
1104 "``-" "``&" "``|" "``^" "``<<" "``>>" "``*" "``/" "``%"
1107 (c-lang-defconst c-overloadable-operators-regexp
1108 ;; Regexp tested after an "operator" token in C++.
1110 c++ (c-make-keywords-re nil (c-lang-const c-overloadable-operators)))
1111 (c-lang-defvar c-overloadable-operators-regexp
1112 (c-lang-const c-overloadable-operators-regexp))
1114 (c-lang-defconst c-opt-op-identifier-prefix
1115 "Regexp matching the token before the ones in
1116 `c-overloadable-operators' when operators are specified in their
1117 \"identifier form\". This typically matches \"operator\" in C++ where
1118 operator functions are specified as e.g. \"operator +\". It's nil in
1119 languages without operator functions or where the complete operator
1120 identifier is listed in `c-overloadable-operators'.
1122 This regexp is assumed to not match any non-operator identifier."
1124 c++ (c-make-keywords-re t '("operator")))
1125 (c-lang-defvar c-opt-op-identifier-prefix
1126 (c-lang-const c-opt-op-identifier-prefix))
1128 ;; Note: the following alias is an old name which was a mis-spelling. It has
1129 ;; been corrected above and throughout cc-engine.el. It will be removed at
1130 ;; some release very shortly in the future. ACM, 2006-04-14.
1131 (defvaralias 'c-opt-op-identitier-prefix 'c-opt-op-identifier-prefix)
1132 (make-obsolete-variable 'c-opt-op-identitier-prefix 'c-opt-op-identifier-prefix
1133 "CC Mode 5.31.4, 2006-04-14")
1135 (c-lang-defconst c-other-op-syntax-tokens
1136 "List of the tokens made up of characters in the punctuation or
1137 parenthesis syntax classes that have uses other than as expression
1139 t '("{" "}" "(" ")" "[" "]" ";" ":" "," "=" "/*" "*/" "//")
1140 (c c++ pike) (append '("#" "##" ; Used by cpp.
1142 (c-lang-const c-other-op-syntax-tokens))
1143 (c c++) (append '("*") (c-lang-const c-other-op-syntax-tokens))
1144 c++ (append '("&" "<%" "%>" "<:" ":>" "%:" "%:%:")
1145 (c-lang-const c-other-op-syntax-tokens))
1146 objc (append '("#" "##" ; Used by cpp.
1147 "+" "-") (c-lang-const c-other-op-syntax-tokens))
1148 idl (append '("#" "##") ; Used by cpp.
1149 (c-lang-const c-other-op-syntax-tokens))
1150 pike (append '("..")
1151 (c-lang-const c-other-op-syntax-tokens)
1152 (c-lang-const c-overloadable-operators))
1153 awk '("{" "}" "(" ")" "[" "]" ";" "," "=" "/"))
1155 (c-lang-defconst c-all-op-syntax-tokens
1156 ;; List of all tokens in the punctuation and parenthesis syntax
1158 t (cl-delete-duplicates (append (c-lang-const c-other-op-syntax-tokens)
1159 (c-lang-const c-operator-list))
1160 :test 'string-equal))
1162 (c-lang-defconst c-nonsymbol-token-char-list
1163 ;; List containing all chars not in the word, symbol or
1164 ;; syntactically irrelevant syntax classes, i.e. all punctuation,
1165 ;; parenthesis and string delimiter chars.
1166 t (c-with-syntax-table (c-lang-const c-mode-syntax-table)
1167 ;; Only go through the chars in the printable ASCII range. No
1168 ;; language so far has 8-bit or widestring operators.
1169 (let (list (char 32))
1171 (or (memq (char-syntax char) '(?w ?_ ?< ?> ?\ ))
1172 (setq list (cons (c-int-to-char char) list)))
1173 (setq char (1+ char)))
1176 (c-lang-defconst c-nonsymbol-token-regexp
1177 ;; Regexp matching all tokens in the punctuation and parenthesis
1178 ;; syntax classes. Note that this also matches ".", which can start
1180 t (c-make-keywords-re nil
1181 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1183 "\\`\\(\\s.\\|\\s\(\\|\\s\)\\)+\\'")))
1184 (c-lang-defvar c-nonsymbol-token-regexp
1185 (c-lang-const c-nonsymbol-token-regexp))
1187 (c-lang-defconst c-assignment-op-regexp
1188 ;; Regexp matching all assignment operators and only them. The
1189 ;; beginning of the first submatch is used to detect the end of the
1190 ;; token, along with the end of the whole match.
1191 t (if (c-lang-const c-assignment-operators)
1193 ;; Need special case for "=" since it's a prefix of "==".
1196 (c-make-keywords-re nil
1197 (set-difference (c-lang-const c-assignment-operators)
1199 :test 'string-equal)))
1201 (c-lang-defvar c-assignment-op-regexp
1202 (c-lang-const c-assignment-op-regexp))
1204 (c-lang-defconst c-<>-multichar-token-regexp
1205 ;; Regexp matching all tokens containing "<" or ">" which are longer
1207 t (c-make-keywords-re nil
1208 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1211 (c-lang-defvar c-<>-multichar-token-regexp
1212 (c-lang-const c-<>-multichar-token-regexp))
1214 (c-lang-defconst c-<-op-cont-regexp
1215 ;; Regexp matching the second and subsequent characters of all
1216 ;; multicharacter tokens that begin with "<".
1217 t (c-make-keywords-re nil
1218 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1221 (lambda (op) (substring op 1)))))
1223 (c-lang-defvar c-<-op-cont-regexp (c-lang-const c-<-op-cont-regexp))
1225 (c-lang-defconst c->-op-cont-tokens
1226 ;; A list of second and subsequent characters of all multicharacter tokens
1227 ;; that begin with ">".
1228 t (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1231 (lambda (op) (substring op 1)))
1232 java (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1234 "\\`>[^>]\\|\\`>>[^>]"
1235 (lambda (op) (substring op 1))))
1237 (c-lang-defconst c->-op-cont-regexp
1238 ;; Regexp matching the second and subsequent characters of all
1239 ;; multicharacter tokens that begin with ">".
1240 t (c-make-keywords-re nil (c-lang-const c->-op-cont-tokens)))
1242 (c-lang-defvar c->-op-cont-regexp (c-lang-const c->-op-cont-regexp))
1244 (c-lang-defconst c->-op-without->-cont-regexp
1245 ;; Regexp matching the second and subsequent characters of all
1246 ;; multicharacter tokens that begin with ">" except for those beginning with
1248 t (c-make-keywords-re nil
1250 (c-lang-const c->-op-cont-tokens)
1251 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1254 (lambda (op) (substring op 1)))
1255 :test 'string-equal)))
1257 (c-lang-defvar c->-op-without->-cont-regexp
1258 (c-lang-const c->-op-without->-cont-regexp))
1260 (c-lang-defconst c-stmt-delim-chars
1261 ;; The characters that should be considered to bound statements. To
1262 ;; optimize `c-crosses-statement-barrier-p' somewhat, it's assumed to
1263 ;; begin with "^" to negate the set. If ? : operators should be
1264 ;; detected then the string must end with "?:".
1266 (c-lang-defvar c-stmt-delim-chars (c-lang-const c-stmt-delim-chars))
1268 (c-lang-defconst c-stmt-delim-chars-with-comma
1269 ;; Variant of `c-stmt-delim-chars' that additionally contains ','.
1271 (c-lang-defvar c-stmt-delim-chars-with-comma
1272 (c-lang-const c-stmt-delim-chars-with-comma))
1274 (c-lang-defconst c-auto-ops
1275 ;; Ops which signal C++11's new auto uses.
1278 (c-lang-defconst c-auto-ops-re
1279 t (c-make-keywords-re nil (c-lang-const c-auto-ops)))
1280 (c-lang-defvar c-auto-ops-re (c-lang-const c-auto-ops-re))
1282 (c-lang-defconst c-haskell-op
1283 ;; Op used in the new C++11 auto function definition, indicating type.
1286 (c-lang-defconst c-haskell-op-re
1287 t (c-make-keywords-re nil (c-lang-const c-haskell-op)))
1288 (c-lang-defvar c-haskell-op-re (c-lang-const c-haskell-op-re))
1290 ;;; Syntactic whitespace.
1292 (c-lang-defconst c-simple-ws
1293 "Regexp matching an ordinary whitespace character.
1294 Does not contain a \\| operator at the top level."
1295 ;; "\\s " is not enough since it doesn't match line breaks.
1296 t "\\(\\s \\|[\n\r]\\)")
1298 (c-lang-defconst c-simple-ws-depth
1299 ;; Number of regexp grouping parens in `c-simple-ws'.
1300 t (regexp-opt-depth (c-lang-const c-simple-ws)))
1302 (c-lang-defconst c-line-comment-starter
1303 "String that starts line comments, or nil if such don't exist.
1304 Line comments are always terminated by newlines. At least one of
1305 `c-block-comment-starter' and this one is assumed to be set.
1307 Note that it's currently not enough to set this to support a new
1308 comment style. Other stuff like the syntax table must also be set up
1312 (c-lang-defvar c-line-comment-starter (c-lang-const c-line-comment-starter))
1314 (c-lang-defconst c-block-comment-starter
1315 "String that starts block comments, or nil if such don't exist.
1316 Block comments are ended by `c-block-comment-ender', which is assumed
1317 to be set if this is. At least one of `c-line-comment-starter' and
1318 this one is assumed to be set.
1320 Note that it's currently not enough to set this to support a new
1321 comment style. Other stuff like the syntax table must also be set up
1326 (c-lang-defconst c-block-comment-ender
1327 "String that ends block comments, or nil if such don't exist.
1329 Note that it's currently not enough to set this to support a new
1330 comment style. Other stuff like the syntax table must also be set up
1335 (c-lang-defconst c-comment-start-regexp
1336 ;; Regexp to match the start of any type of comment.
1337 t (let ((re (c-make-keywords-re nil
1338 (list (c-lang-const c-line-comment-starter)
1339 (c-lang-const c-block-comment-starter)))))
1340 (if (memq 'gen-comment-delim c-emacs-features)
1341 (concat re "\\|\\s!")
1343 (c-lang-defvar c-comment-start-regexp (c-lang-const c-comment-start-regexp))
1345 (c-lang-defconst c-block-comment-start-regexp
1346 ;; Regexp which matches the start of a block comment (if such exists in the
1348 t (if (c-lang-const c-block-comment-starter)
1349 (regexp-quote (c-lang-const c-block-comment-starter))
1351 (c-lang-defvar c-block-comment-start-regexp
1352 (c-lang-const c-block-comment-start-regexp))
1354 (c-lang-defconst c-line-comment-start-regexp
1355 ;; Regexp which matches the start of a line comment (if such exists in the
1356 ;; language; it does in all 7 CC Mode languages).
1357 t (if (c-lang-const c-line-comment-starter)
1358 (regexp-quote (c-lang-const c-line-comment-starter))
1360 (c-lang-defvar c-line-comment-start-regexp
1361 (c-lang-const c-line-comment-start-regexp))
1363 (c-lang-defconst c-literal-start-regexp
1364 ;; Regexp to match the start of comments and string literals.
1365 t (concat (c-lang-const c-comment-start-regexp)
1367 (if (memq 'gen-string-delim c-emacs-features)
1370 (c-lang-defvar c-literal-start-regexp (c-lang-const c-literal-start-regexp))
1372 (c-lang-defconst c-doc-comment-start-regexp
1373 "Regexp to match the start of documentation comments."
1375 ;; From font-lock.el: `doxygen' uses /*! while others use /**.
1376 (c c++ objc) "/\\*[*!]"
1379 (c-lang-defvar c-doc-comment-start-regexp
1380 (c-lang-const c-doc-comment-start-regexp))
1382 (c-lang-defconst comment-start
1383 "String that starts comments inserted with M-; etc.
1384 `comment-start' is initialized from this."
1385 ;; Default: Prefer line comments to block comments, and pad with a space.
1386 t (concat (or (c-lang-const c-line-comment-starter)
1387 (c-lang-const c-block-comment-starter))
1389 ;; In C we still default to the block comment style since line
1390 ;; comments aren't entirely portable.
1392 (c-lang-setvar comment-start (c-lang-const comment-start))
1394 (c-lang-defconst comment-end
1395 "String that ends comments inserted with M-; etc.
1396 `comment-end' is initialized from this."
1397 ;; Default: Use block comment style if comment-start uses block
1398 ;; comments, and pad with a space in that case.
1399 t (if (string-match (concat "\\`\\("
1400 (c-lang-const c-block-comment-start-regexp)
1402 (c-lang-const comment-start))
1403 (concat " " (c-lang-const c-block-comment-ender))
1405 (c-lang-setvar comment-end (c-lang-const comment-end))
1407 (c-lang-defconst comment-start-skip
1408 "Regexp to match the start of a comment plus everything up to its body.
1409 `comment-start-skip' is initialized from this."
1410 ;; Default: Allow the last char of the comment starter(s) to be
1411 ;; repeated, then allow any amount of horizontal whitespace.
1414 (mapcar (lambda (cs)
1416 (concat (regexp-quote cs) "+")))
1417 (list (c-lang-const c-line-comment-starter)
1418 (c-lang-const c-block-comment-starter)))
1421 (c-lang-setvar comment-start-skip (c-lang-const comment-start-skip))
1423 (c-lang-defconst c-syntactic-ws-start
1424 ;; Regexp matching any sequence that can start syntactic whitespace.
1425 ;; The only uncertain case is '#' when there are cpp directives.
1427 (c-make-keywords-re nil
1428 (append (list (c-lang-const c-line-comment-starter)
1429 (c-lang-const c-block-comment-starter)
1430 (when (c-lang-const c-opt-cpp-prefix)
1434 (when (memq 'gen-comment-delim c-emacs-features)
1436 (c-lang-defvar c-syntactic-ws-start (c-lang-const c-syntactic-ws-start))
1438 (c-lang-defconst c-syntactic-ws-end
1439 ;; Regexp matching any single character that might end syntactic whitespace.
1441 (c-make-keywords-re nil
1442 (append (when (c-lang-const c-block-comment-ender)
1445 (elt (c-lang-const c-block-comment-ender)
1447 (c-lang-const c-block-comment-ender)))))))
1449 (when (memq 'gen-comment-delim c-emacs-features)
1451 (c-lang-defvar c-syntactic-ws-end (c-lang-const c-syntactic-ws-end))
1453 (c-lang-defconst c-unterminated-block-comment-regexp
1454 ;; Regexp matching an unterminated block comment that doesn't
1455 ;; contain line breaks, or nil in languages without block comments.
1456 ;; Does not contain a \| operator at the top level.
1457 t (when (c-lang-const c-block-comment-starter)
1459 (regexp-quote (c-lang-const c-block-comment-starter))
1460 ;; It's messy to cook together a regexp that matches anything
1461 ;; but c-block-comment-ender.
1462 (let ((end (c-lang-const c-block-comment-ender)))
1463 (cond ((= (length end) 1)
1464 (concat "[^" end "\n\r]*"))
1466 (concat "[^" (substring end 0 1) "\n\r]*"
1468 (regexp-quote (substring end 0 1)) "+"
1470 ;; The quoting rules inside char classes are silly. :P
1471 (cond ((= (elt end 0) (elt end 1))
1472 (concat (substring end 0 1) "\n\r"))
1473 ((= (elt end 1) ?\])
1474 (concat (substring end 1 2) "\n\r"
1475 (substring end 0 1)))
1477 (concat (substring end 0 1) "\n\r"
1478 (substring end 1 2))))
1480 "[^" (substring end 0 1) "\n\r]*"
1483 (error "Can't handle a block comment ender of length %s"
1486 (c-lang-defconst c-block-comment-regexp
1487 ;; Regexp matching a block comment that doesn't contain line breaks,
1488 ;; or nil in languages without block comments. The reason we don't
1489 ;; allow line breaks is to avoid going very far and risk running out
1490 ;; of regexp stack; this regexp is intended to handle only short
1491 ;; comments that might be put in the middle of limited constructs
1492 ;; like declarations. Does not contain a \| operator at the top
1494 t (when (c-lang-const c-unterminated-block-comment-regexp)
1496 (c-lang-const c-unterminated-block-comment-regexp)
1497 (let ((end (c-lang-const c-block-comment-ender)))
1498 (cond ((= (length end) 1)
1501 (concat (regexp-quote (substring end 0 1)) "+"
1502 (regexp-quote (substring end 1 2))))
1504 (error "Can't handle a block comment ender of length %s"
1507 (c-lang-defconst c-nonwhite-syntactic-ws
1508 ;; Regexp matching a piece of syntactic whitespace that isn't a
1509 ;; sequence of simple whitespace characters. As opposed to
1510 ;; `c-(forward|backward)-syntactic-ws', this doesn't regard cpp
1511 ;; directives as syntactic whitespace.
1512 t (c-concat-separated
1513 (list (when (c-lang-const c-line-comment-starter)
1514 (concat (regexp-quote (c-lang-const c-line-comment-starter))
1516 (c-lang-const c-block-comment-regexp)
1518 (when (memq 'gen-comment-delim c-emacs-features)
1522 (c-lang-defconst c-syntactic-ws
1523 ;; Regexp matching syntactic whitespace, including possibly the
1524 ;; empty string. As opposed to `c-(forward|backward)-syntactic-ws',
1525 ;; this doesn't regard cpp directives as syntactic whitespace. Does
1526 ;; not contain a \| operator at the top level.
1527 t (concat (c-lang-const c-simple-ws) "*"
1529 (concat "\\(" (c-lang-const c-nonwhite-syntactic-ws) "\\)"
1530 (c-lang-const c-simple-ws) "*")
1533 (c-lang-defconst c-syntactic-ws-depth
1534 ;; Number of regexp grouping parens in `c-syntactic-ws'.
1535 t (regexp-opt-depth (c-lang-const c-syntactic-ws)))
1537 (c-lang-defconst c-nonempty-syntactic-ws
1538 ;; Regexp matching syntactic whitespace, which is at least one
1539 ;; character long. As opposed to `c-(forward|backward)-syntactic-ws',
1540 ;; this doesn't regard cpp directives as syntactic whitespace. Does
1541 ;; not contain a \| operator at the top level.
1543 (c-lang-const c-simple-ws)
1545 (c-lang-const c-nonwhite-syntactic-ws)
1548 (c-lang-defconst c-nonempty-syntactic-ws-depth
1549 ;; Number of regexp grouping parens in `c-nonempty-syntactic-ws'.
1550 t (regexp-opt-depth (c-lang-const c-nonempty-syntactic-ws)))
1552 (c-lang-defconst c-single-line-syntactic-ws
1553 ;; Regexp matching syntactic whitespace without any line breaks. As
1554 ;; opposed to `c-(forward|backward)-syntactic-ws', this doesn't
1555 ;; regard cpp directives as syntactic whitespace. Does not contain
1556 ;; a \| operator at the top level.
1557 t (if (c-lang-const c-block-comment-regexp)
1559 (c-lang-const c-block-comment-regexp)
1563 (c-lang-defconst c-single-line-syntactic-ws-depth
1564 ;; Number of regexp grouping parens in `c-single-line-syntactic-ws'.
1565 t (regexp-opt-depth (c-lang-const c-single-line-syntactic-ws)))
1567 (c-lang-defconst c-syntactic-eol
1568 ;; Regexp that matches when there is no syntactically significant
1569 ;; text before eol. Macros are regarded as syntactically
1570 ;; significant text here.
1571 t (concat (c-lang-const c-single-line-syntactic-ws)
1572 ;; Match eol (possibly inside a block comment or preceded
1573 ;; by a line continuation backslash), or the beginning of a
1574 ;; line comment. Note: This has to be modified for awk
1575 ;; where line comments start with '#'.
1578 (list (when (c-lang-const c-line-comment-starter)
1579 (regexp-quote (c-lang-const c-line-comment-starter)))
1580 (when (c-lang-const c-unterminated-block-comment-regexp)
1581 (concat (c-lang-const c-unterminated-block-comment-regexp)
1587 (c-lang-defvar c-syntactic-eol (c-lang-const c-syntactic-eol))
1592 ;; The Emacs variables beginning-of-defun-function and
1593 ;; end-of-defun-function will be set so that commands like
1594 ;; `mark-defun' and `narrow-to-defun' work right. The key sequences
1595 ;; C-M-a and C-M-e are, however, bound directly to the CC Mode
1596 ;; functions, allowing optimization for large n.
1597 (c-lang-defconst beginning-of-defun-function
1598 "Function to which beginning-of-defun-function will be set."
1599 t 'c-beginning-of-defun
1600 awk 'c-awk-beginning-of-defun)
1601 (c-lang-setvar beginning-of-defun-function
1602 (c-lang-const beginning-of-defun-function))
1604 (c-lang-defconst end-of-defun-function
1605 "Function to which end-of-defun-function will be set."
1607 awk 'c-awk-end-of-defun)
1608 (c-lang-setvar end-of-defun-function (c-lang-const end-of-defun-function))
1610 ;;; In-comment text handling.
1612 (c-lang-defconst c-paragraph-start
1613 "Regexp to append to `paragraph-start'."
1615 java "\\(@[a-zA-Z]+\\>\\|$\\)" ; For Javadoc.
1616 pike "\\(@[a-zA-Z_-]+\\>\\([^{]\\|$\\)\\|$\\)") ; For Pike refdoc.
1617 (c-lang-defvar c-paragraph-start (c-lang-const c-paragraph-start))
1619 (c-lang-defconst c-paragraph-separate
1620 "Regexp to append to `paragraph-separate'."
1622 pike (c-lang-const c-paragraph-start))
1623 (c-lang-defvar c-paragraph-separate (c-lang-const c-paragraph-separate))
1628 ;; Note: All and only all language constants containing keyword lists
1629 ;; should end with "-kwds"; they're automatically collected into the
1630 ;; `c-kwds-lang-consts' list below and used to build `c-keywords' etc.
1632 (c-lang-defconst c-primitive-type-kwds
1633 "Primitive type keywords. As opposed to the other keyword lists, the
1634 keywords listed here are fontified with the type face instead of the
1637 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1638 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1639 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1642 Do not try to modify this list for end user customizations; the
1643 `*-font-lock-extra-types' variable, where `*' is the mode prefix, is
1644 the appropriate place for that."
1645 t '("char" "double" "float" "int" "long" "short" "signed"
1648 '("_Bool" "_Complex" "_Imaginary") ; Conditionally defined in C99.
1649 (c-lang-const c-primitive-type-kwds))
1652 (c-lang-const c-primitive-type-kwds))
1653 ;; Objective-C extends C, but probably not the new stuff in C99.
1655 '("id" "Class" "SEL" "IMP" "BOOL")
1656 (c-lang-const c-primitive-type-kwds))
1657 java '("boolean" "byte" "char" "double" "float" "int" "long" "short" "void")
1658 idl '("Object" "ValueBase" "any" "boolean" "char" "double" "fixed" "float"
1659 "long" "octet" "sequence" "short" "string" "void" "wchar" "wstring"
1662 ;; The following can't really end a type, but we have to specify them
1663 ;; here due to the assumption in `c-primitive-type-prefix-kwds'. It
1664 ;; doesn't matter that much.
1665 "unsigned" "strong")
1666 pike '(;; this_program isn't really a keyword, but it's practically
1667 ;; used as a builtin type.
1668 "array" "float" "function" "int" "mapping" "mixed" "multiset"
1669 "object" "program" "string" "this_program" "void"))
1671 (c-lang-defconst c-primitive-type-key
1672 ;; An adorned regexp that matches `c-primitive-type-kwds'.
1673 t (c-make-keywords-re t (c-lang-const c-primitive-type-kwds)))
1674 (c-lang-defvar c-primitive-type-key (c-lang-const c-primitive-type-key))
1676 (c-lang-defconst c-primitive-type-prefix-kwds
1677 "Keywords that might act as prefixes for primitive types. Assumed to
1678 be a subset of `c-primitive-type-kwds'."
1680 (c c++) '("long" "short" "signed" "unsigned")
1681 idl '("long" "unsigned"
1685 (c-lang-defconst c-typedef-kwds
1686 "Prefix keyword\(s\) like \"typedef\" which make a type declaration out
1687 of a variable declaration."
1691 (c-lang-defconst c-typedef-key
1692 ;; Adorned regexp matching `c-typedef-kwds'.
1693 t (c-make-keywords-re t (c-lang-const c-typedef-kwds)))
1694 (c-lang-defvar c-typedef-key (c-lang-const c-typedef-key))
1696 (c-lang-defconst c-typeof-kwds
1697 "Keywords followed by a parenthesized expression, which stands for
1698 the type of that expression."
1700 c '("typeof") ; longstanding GNU C(++) extension.
1701 c++ '("decltype" "typeof"))
1703 (c-lang-defconst c-typeof-key
1704 ;; Adorned regexp matching `c-typeof-kwds'.
1705 t (c-make-keywords-re t (c-lang-const c-typeof-kwds)))
1706 (c-lang-defvar c-typeof-key (c-lang-const c-typeof-key))
1708 (c-lang-defconst c-type-prefix-kwds
1709 "Keywords where the following name - if any - is a type name, and
1710 where the keyword together with the symbol works as a type in
1713 Note that an alternative if the second part doesn't hold is
1714 `c-type-list-kwds'. Keywords on this list are typically also present
1715 on one of the `*-decl-kwds' lists."
1717 c '("struct" "union" "enum")
1718 c++ (append '("class" "typename")
1719 (c-lang-const c-type-prefix-kwds c)))
1721 (c-lang-defconst c-type-prefix-key
1722 ;; Adorned regexp matching `c-type-prefix-kwds'.
1723 t (c-make-keywords-re t (c-lang-const c-type-prefix-kwds)))
1724 (c-lang-defvar c-type-prefix-key (c-lang-const c-type-prefix-key))
1726 (c-lang-defconst c-type-modifier-kwds
1727 "Type modifier keywords. These can occur almost anywhere in types
1728 but they don't build a type of themselves. Unlike the keywords on
1729 `c-primitive-type-kwds', they are fontified with the keyword face and
1732 c '("const" "restrict" "volatile")
1733 c++ '("const" "volatile" "throw")
1734 objc '("const" "volatile"))
1736 (c-lang-defconst c-opt-type-modifier-key
1737 ;; Adorned regexp matching `c-type-modifier-kwds', or nil in
1738 ;; languages without such keywords.
1739 t (and (c-lang-const c-type-modifier-kwds)
1740 (c-make-keywords-re t (c-lang-const c-type-modifier-kwds))))
1741 (c-lang-defvar c-opt-type-modifier-key (c-lang-const c-opt-type-modifier-key))
1743 (c-lang-defconst c-opt-type-component-key
1744 ;; An adorned regexp that matches `c-primitive-type-prefix-kwds' and
1745 ;; `c-type-modifier-kwds', or nil in languages without any of them.
1746 t (and (or (c-lang-const c-primitive-type-prefix-kwds)
1747 (c-lang-const c-type-modifier-kwds))
1748 (c-make-keywords-re t
1749 (append (c-lang-const c-primitive-type-prefix-kwds)
1750 (c-lang-const c-type-modifier-kwds)))))
1751 (c-lang-defvar c-opt-type-component-key
1752 (c-lang-const c-opt-type-component-key))
1754 (c-lang-defconst c-type-start-kwds
1755 ;; All keywords that can start a type (i.e. are either a type prefix
1756 ;; or a complete type).
1757 t (cl-delete-duplicates (append (c-lang-const c-primitive-type-kwds)
1758 (c-lang-const c-type-prefix-kwds)
1759 (c-lang-const c-type-modifier-kwds))
1760 :test 'string-equal))
1762 (c-lang-defconst c-class-decl-kwds
1763 "Keywords introducing declarations where the following block (if any)
1764 contains another declaration level that should be considered a class.
1766 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1767 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1768 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1771 Note that presence on this list does not automatically treat the
1772 following identifier as a type; the keyword must also be present on
1773 `c-type-prefix-kwds' or `c-type-list-kwds' to accomplish that."
1775 c '("struct" "union")
1776 c++ '("class" "struct" "union")
1777 objc '("struct" "union"
1778 "@interface" "@implementation" "@protocol")
1779 java '("class" "@interface" "interface")
1780 idl '("component" "eventtype" "exception" "home" "interface" "struct"
1783 "storagehome" "storagetype"
1785 "catalog" "executor" "manages" "segment")
1788 (c-lang-defconst c-class-key
1789 ;; Regexp matching the start of a class.
1790 t (c-make-keywords-re t (c-lang-const c-class-decl-kwds)))
1791 (c-lang-defvar c-class-key (c-lang-const c-class-key))
1793 (c-lang-defconst c-brace-list-decl-kwds
1794 "Keywords introducing declarations where the following block (if
1795 any) is a brace list.
1797 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1798 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1799 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1804 (c-lang-defconst c-brace-list-key
1805 ;; Regexp matching the start of declarations where the following
1806 ;; block is a brace list.
1807 t (c-make-keywords-re t (c-lang-const c-brace-list-decl-kwds)))
1808 (c-lang-defvar c-brace-list-key (c-lang-const c-brace-list-key))
1810 (c-lang-defconst c-other-block-decl-kwds
1811 "Keywords where the following block (if any) contains another
1812 declaration level that should not be considered a class. For every
1813 keyword here, CC Mode will add a set of special syntactic symbols for
1814 those blocks. E.g. if the keyword is \"foo\" then there will be
1815 `foo-open', `foo-close', and `infoo' symbols.
1817 The intention is that this category should be used for block
1818 constructs that aren't related to object orientation concepts like
1819 classes (which thus also include e.g. interfaces, templates,
1820 contracts, structs, etc). The more pragmatic distinction is that
1821 while most want some indentation inside classes, it's fairly common
1822 that they don't want it in some of these constructs, so it should be
1823 simple to configure that differently from classes. See also
1824 `c-class-decl-kwds'.
1826 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1827 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1828 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1831 (c objc) '("extern")
1832 c++ '("namespace" "extern")
1837 (c-lang-defconst c-other-decl-block-key
1838 ;; Regexp matching the start of blocks besides classes that contain
1839 ;; another declaration level.
1840 t (c-make-keywords-re t (c-lang-const c-other-block-decl-kwds)))
1841 (c-lang-defvar c-other-decl-block-key (c-lang-const c-other-decl-block-key))
1843 (c-lang-defvar c-other-decl-block-key-in-symbols-alist
1847 (if (string= elt "extern")
1849 (intern (concat "in" elt)))))
1850 (c-lang-const c-other-block-decl-kwds))
1851 "Alist associating keywords in c-other-decl-block-decl-kwds with
1852 their matching \"in\" syntactic symbols.")
1854 (c-lang-defconst c-typedef-decl-kwds
1855 "Keywords introducing declarations where the identifier(s) being
1858 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1859 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1860 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1862 ;; Default to `c-class-decl-kwds' and `c-brace-list-decl-kwds'
1863 ;; (since e.g. "Foo" is a type that's being defined in "class Foo
1865 t (append (c-lang-const c-class-decl-kwds)
1866 (c-lang-const c-brace-list-decl-kwds))
1867 ;; Languages that have a "typedef" construct.
1868 (c c++ objc idl pike) (append (c-lang-const c-typedef-decl-kwds)
1870 ;; Unlike most other languages, exception names are not handled as
1871 ;; types in IDL since they only can occur in "raises" specs.
1872 idl (delete "exception" (append (c-lang-const c-typedef-decl-kwds) nil)))
1874 (c-lang-defconst c-typedef-decl-key
1875 t (c-make-keywords-re t (c-lang-const c-typedef-decl-kwds)))
1876 (c-lang-defvar c-typedef-decl-key (c-lang-const c-typedef-decl-key))
1878 (c-lang-defconst c-typeless-decl-kwds
1879 "Keywords introducing declarations where the \(first) identifier
1880 \(declarator) follows directly after the keyword, without any type.
1882 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1883 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1884 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1886 ;; Default to `c-class-decl-kwds' and `c-brace-list-decl-kwds'
1887 ;; (since e.g. "Foo" is the identifier being defined in "class Foo
1889 t (append (c-lang-const c-class-decl-kwds)
1890 (c-lang-const c-brace-list-decl-kwds))
1891 c++ (append (c-lang-const c-typeless-decl-kwds) '("auto")) ; C++11.
1892 ;; Note: "manages" for CORBA CIDL clashes with its presence on
1893 ;; `c-type-list-kwds' for IDL.
1894 idl (append (c-lang-const c-typeless-decl-kwds)
1895 '("factory" "finder" "native"
1900 pike (append (c-lang-const c-class-decl-kwds)
1903 (c-lang-defconst c-modifier-kwds
1904 "Keywords that can prefix normal declarations of identifiers
1905 \(and typically act as flags). Things like argument declarations
1906 inside function headers are also considered declarations in this
1909 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1910 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1911 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1914 (c c++) '("auto" "extern" "inline" "register" "static")
1915 c++ (append '("explicit" "friend" "mutable" "template" "using" "virtual")
1916 (c-lang-const c-modifier-kwds))
1917 objc '("auto" "bycopy" "byref" "extern" "in" "inout" "oneway" "out" "static")
1918 ;; FIXME: Some of those below ought to be on `c-other-decl-kwds' instead.
1919 idl '("abstract" "attribute" "const" "consumes" "custom" "emits" "import"
1920 "in" "inout" "local" "multiple" "oneway" "out" "private" "provides"
1921 "public" "publishes" "readonly" "typeid" "typeprefix" "uses"
1925 "bindsTo" "delegatesTo" "implements" "proxy" "storedOn")
1926 ;; Note: "const" is not used in Java, but it's still a reserved keyword.
1927 java '("abstract" "const" "final" "native" "private" "protected" "public"
1928 "static" "strictfp" "synchronized" "transient" "volatile")
1929 pike '("final" "inline" "local" "nomask" "optional" "private" "protected"
1930 "public" "static" "variant"))
1932 (c-lang-defconst c-other-decl-kwds
1933 "Keywords that can start or prefix any declaration level construct,
1934 besides those on `c-class-decl-kwds', `c-brace-list-decl-kwds',
1935 `c-other-block-decl-kwds', `c-typedef-decl-kwds',
1936 `c-typeless-decl-kwds' and `c-modifier-kwds'.
1938 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1939 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1940 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1943 objc '("@class" "@end" "@defs")
1944 java '("import" "package")
1945 pike '("import" "inherit"))
1947 (c-lang-defconst c-decl-start-kwds
1948 "Keywords that always start declarations, wherever they occur.
1949 This can be used for declarations that aren't recognized by the normal
1950 combination of `c-decl-prefix-re' and `c-decl-start-re'."
1952 ;; Classes can be declared anywhere in a Pike expression.
1955 (c-lang-defconst c-decl-hangon-kwds
1956 "Keywords that can occur anywhere in a declaration level construct.
1957 This is used for self-contained things that can be tacked on anywhere
1958 on a declaration and that should be ignored to be able to recognize it
1959 correctly. Typical cases are compiler extensions like
1960 \"__attribute__\" or \"__declspec\":
1962 __declspec(noreturn) void foo();
1963 class __declspec(dllexport) classname {...};
1964 void foo() __attribute__((noreturn));
1966 Note that unrecognized plain symbols are skipped anyway if they occur
1967 before the type, so such things are not necessary to mention here.
1968 Mentioning them here is necessary only if they can occur in other
1969 places, or if they are followed by a construct that must be skipped
1970 over \(like the parens in the \"__attribute__\" and \"__declspec\"
1971 examples above). In the last case, they alse need to be present on
1972 one of `c-type-list-kwds', `c-ref-list-kwds',
1973 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1974 `c-<>-type-kwds', or `c-<>-arglist-kwds'."
1975 ;; NB: These are currently not recognized in all parts of a
1976 ;; declaration. Specifically, they aren't recognized in the middle
1977 ;; of multi-token types, inside declarators, and between the
1978 ;; identifier and the arglist paren of a function declaration.
1980 ;; FIXME: This ought to be user customizable since compiler stuff
1981 ;; like this usually is wrapped in project specific macros. (It'd
1982 ;; of course be even better if we could cope without knowing this.)
1984 (c c++) '(;; GCC extension.
1989 (c-lang-defconst c-decl-hangon-key
1990 ;; Adorned regexp matching `c-decl-hangon-kwds'.
1991 t (c-make-keywords-re t (c-lang-const c-decl-hangon-kwds)))
1992 (c-lang-defvar c-decl-hangon-key (c-lang-const c-decl-hangon-key))
1994 (c-lang-defconst c-prefix-spec-kwds
1995 ;; All keywords that can occur in the preamble of a declaration.
1996 ;; They typically occur before the type, but they are also matched
1997 ;; after presumptive types since we often can't be sure that
1998 ;; something is a type or just some sort of macro in front of the
1999 ;; declaration. They might be ambiguous with types or type
2001 t (cl-delete-duplicates (append (c-lang-const c-class-decl-kwds)
2002 (c-lang-const c-brace-list-decl-kwds)
2003 (c-lang-const c-other-block-decl-kwds)
2004 (c-lang-const c-typedef-decl-kwds)
2005 (c-lang-const c-typeless-decl-kwds)
2006 (c-lang-const c-modifier-kwds)
2007 (c-lang-const c-other-decl-kwds)
2008 (c-lang-const c-decl-start-kwds)
2009 (c-lang-const c-decl-hangon-kwds))
2010 :test 'string-equal))
2012 (c-lang-defconst c-prefix-spec-kwds-re
2013 ;; Adorned regexp of `c-prefix-spec-kwds'.
2014 t (c-make-keywords-re t (c-lang-const c-prefix-spec-kwds)))
2016 (c-lang-defvar c-prefix-spec-kwds-re (c-lang-const c-prefix-spec-kwds-re))
2018 (c-lang-defconst c-specifier-key
2019 ;; Adorned regexp of the keywords in `c-prefix-spec-kwds' that aren't
2020 ;; ambiguous with types or type prefixes. These are the keywords (like
2021 ;; extern, namespace, but NOT template) that can modify a declaration.
2022 t (c-make-keywords-re t
2023 (set-difference (c-lang-const c-prefix-spec-kwds)
2024 (append (c-lang-const c-type-start-kwds)
2025 (c-lang-const c-<>-arglist-kwds))
2026 :test 'string-equal)))
2027 (c-lang-defvar c-specifier-key (c-lang-const c-specifier-key))
2029 (c-lang-defconst c-postfix-spec-kwds
2030 ;; Keywords that can occur after argument list of a function header
2031 ;; declaration, i.e. in the "K&R region".
2032 t (append (c-lang-const c-postfix-decl-spec-kwds)
2033 (c-lang-const c-decl-hangon-kwds)))
2035 (c-lang-defconst c-not-decl-init-keywords
2036 ;; Adorned regexp matching all keywords that can't appear at the
2037 ;; start of a declaration.
2038 t (c-make-keywords-re t
2039 (set-difference (c-lang-const c-keywords)
2040 (append (c-lang-const c-type-start-kwds)
2041 (c-lang-const c-prefix-spec-kwds)
2042 (c-lang-const c-typeof-kwds))
2043 :test 'string-equal)))
2044 (c-lang-defvar c-not-decl-init-keywords
2045 (c-lang-const c-not-decl-init-keywords))
2047 (c-lang-defconst c-not-primitive-type-keywords
2048 "List of all keywords apart from primitive types (like \"int\")."
2049 t (set-difference (c-lang-const c-keywords)
2050 (c-lang-const c-primitive-type-kwds)
2051 :test 'string-equal)
2052 ;; The "more" for C++ is the QT keyword (as in "more slots:").
2053 ;; This variable is intended for use in c-beginning-of-statement-1.
2054 c++ (append (c-lang-const c-not-primitive-type-keywords) '("more")))
2056 (c-lang-defconst c-not-primitive-type-keywords-regexp
2057 t (c-make-keywords-re t
2058 (c-lang-const c-not-primitive-type-keywords)))
2059 (c-lang-defvar c-not-primitive-type-keywords-regexp
2060 (c-lang-const c-not-primitive-type-keywords-regexp))
2062 (c-lang-defconst c-protection-kwds
2063 "Access protection label keywords in classes."
2065 c++ '("private" "protected" "public")
2066 objc '("@private" "@protected" "@public"))
2068 (c-lang-defconst c-block-decls-with-vars
2069 "Keywords introducing declarations that can contain a block which
2070 might be followed by variable declarations, e.g. like \"foo\" in
2071 \"class Foo { ... } foo;\". So if there is a block in a declaration
2072 like that, it ends with the following ';' and not right away.
2074 The keywords on list are assumed to also be present on one of the
2075 `*-decl-kwds' lists."
2077 (c objc) '("struct" "union" "enum" "typedef")
2078 c++ '("class" "struct" "union" "enum" "typedef"))
2080 (c-lang-defconst c-opt-block-decls-with-vars-key
2081 ;; Regexp matching the `c-block-decls-with-vars' keywords, or nil in
2082 ;; languages without such constructs.
2083 t (and (c-lang-const c-block-decls-with-vars)
2084 (c-make-keywords-re t (c-lang-const c-block-decls-with-vars))))
2085 (c-lang-defvar c-opt-block-decls-with-vars-key
2086 (c-lang-const c-opt-block-decls-with-vars-key))
2088 (c-lang-defconst c-postfix-decl-spec-kwds
2089 "Keywords introducing extra declaration specifiers in the region
2090 between the header and the body \(i.e. the \"K&R-region\") in
2093 java '("extends" "implements" "throws")
2094 idl '("context" "getraises" "manages" "primarykey" "raises" "setraises"
2097 "as" "const" "implements" "of" "ref"))
2099 (c-lang-defconst c-postfix-decl-spec-key
2100 ;; Regexp matching the keywords in `c-postfix-decl-spec-kwds'.
2101 t (c-make-keywords-re t (c-lang-const c-postfix-decl-spec-kwds)))
2102 (c-lang-defvar c-postfix-decl-spec-key
2103 (c-lang-const c-postfix-decl-spec-key))
2105 (c-lang-defconst c-nonsymbol-sexp-kwds
2106 "Keywords that may be followed by a nonsymbol sexp before whatever
2107 construct it's part of continues."
2109 (c c++ objc) '("extern"))
2111 (c-lang-defconst c-type-list-kwds
2112 "Keywords that may be followed by a comma separated list of type
2113 identifiers, where each optionally can be prefixed by keywords. (Can
2114 also be used for the special case when the list can contain only one
2117 Assumed to be mutually exclusive with `c-ref-list-kwds'. There's no
2118 reason to put keywords on this list if they are on `c-type-prefix-kwds'.
2119 There's also no reason to add keywords that prefixes a normal
2120 declaration consisting of a type followed by a declarator (list), so
2121 the keywords on `c-modifier-kwds' should normally not be listed here
2124 Note: Use `c-typeless-decl-kwds' for keywords followed by a function
2125 or variable identifier (that's being defined)."
2129 java '("import" "new" "extends" "super" "implements" "throws")
2130 idl '("manages" "native" "primarykey" "supports"
2132 "as" "implements" "of" "scope")
2135 (c-lang-defconst c-ref-list-kwds
2136 "Keywords that may be followed by a comma separated list of
2137 reference (i.e. namespace/scope/module) identifiers, where each
2138 optionally can be prefixed by keywords. (Can also be used for the
2139 special case when the list can contain only one element.) Assumed to
2140 be mutually exclusive with `c-type-list-kwds'.
2142 Note: Use `c-typeless-decl-kwds' for keywords followed by a function
2143 or variable identifier (that's being defined)."
2147 idl '("import" "module"
2152 (c-lang-defconst c-colon-type-list-kwds
2153 "Keywords that may be followed (not necessarily directly) by a colon
2154 and then a comma separated list of type identifiers, where each
2155 optionally can be prefixed by keywords. (Can also be used for the
2156 special case when the list can contain only one element.)"
2158 c++ '("class" "struct")
2159 idl '("component" "eventtype" "home" "interface" "valuetype"
2161 "storagehome" "storagetype"))
2163 (c-lang-defconst c-colon-type-list-re
2164 "Regexp matched after the keywords in `c-colon-type-list-kwds' to skip
2165 forward to the colon. The end of the match is assumed to be directly
2166 after the colon, so the regexp should end with \":\". Must be a
2167 regexp if `c-colon-type-list-kwds' isn't nil."
2168 t (if (c-lang-const c-colon-type-list-kwds)
2169 ;; Disallow various common punctuation chars that can't come
2170 ;; before the ":" that starts the inherit list after "class"
2171 ;; or "struct" in C++. (Also used as default for other
2173 "[^\]\[{}();,/#=:]*:"))
2174 (c-lang-defvar c-colon-type-list-re (c-lang-const c-colon-type-list-re))
2176 (c-lang-defconst c-paren-nontype-kwds
2177 "Keywords that may be followed by a parenthesis expression that doesn't
2178 contain type identifiers."
2180 (c c++) '(;; GCC extension.
2185 (c-lang-defconst c-paren-type-kwds
2186 "Keywords that may be followed by a parenthesis expression containing
2187 type identifiers separated by arbitrary tokens."
2192 pike '("array" "function" "int" "mapping" "multiset" "object" "program"))
2194 (c-lang-defconst c-paren-any-kwds
2195 t (cl-delete-duplicates (append (c-lang-const c-paren-nontype-kwds)
2196 (c-lang-const c-paren-type-kwds))
2197 :test 'string-equal))
2199 (c-lang-defconst c-<>-type-kwds
2200 "Keywords that may be followed by an angle bracket expression
2201 containing type identifiers separated by \",\". The difference from
2202 `c-<>-arglist-kwds' is that unknown names are taken to be types and
2203 not other identifiers. `c-recognize-<>-arglists' is assumed to be set
2211 (c-lang-defconst c-<>-arglist-kwds
2212 "Keywords that can be followed by a C++ style template arglist; see
2213 `c-recognize-<>-arglists' for details. That language constant is
2214 assumed to be set if this isn't nil."
2217 idl '("fixed" "string" "wstring"))
2219 (c-lang-defconst c-<>-sexp-kwds
2220 ;; All keywords that can be followed by an angle bracket sexp.
2221 t (cl-delete-duplicates (append (c-lang-const c-<>-type-kwds)
2222 (c-lang-const c-<>-arglist-kwds))
2223 :test 'string-equal))
2225 (c-lang-defconst c-opt-<>-sexp-key
2226 ;; Adorned regexp matching keywords that can be followed by an angle
2227 ;; bracket sexp. Always set when `c-recognize-<>-arglists' is.
2228 t (c-make-keywords-re t (c-lang-const c-<>-sexp-kwds)))
2229 (c-lang-defvar c-opt-<>-sexp-key (c-lang-const c-opt-<>-sexp-key))
2231 (c-lang-defconst c-brace-id-list-kwds
2232 "Keywords that may be followed by a brace block containing a comma
2233 separated list of identifier definitions, i.e. like the list of
2234 identifiers that follows the type in a normal declaration."
2235 t (c-lang-const c-brace-list-decl-kwds))
2237 (c-lang-defconst c-block-stmt-1-kwds
2238 "Statement keywords followed directly by a substatement."
2240 c++ '("do" "else" "try")
2241 objc '("do" "else" "@finally" "@try")
2242 java '("do" "else" "finally" "try")
2245 (c-lang-defconst c-block-stmt-1-key
2246 ;; Regexp matching the start of any statement followed directly by a
2247 ;; substatement (doesn't match a bare block, however).
2248 t (c-make-keywords-re t (c-lang-const c-block-stmt-1-kwds)))
2249 (c-lang-defvar c-block-stmt-1-key (c-lang-const c-block-stmt-1-key))
2251 (c-lang-defconst c-block-stmt-1-2-kwds
2252 "Statement keywords optionally followed by a paren sexp.
2253 Keywords here should also be in `c-block-stmt-1-kwds'."
2257 (c-lang-defconst c-block-stmt-1-2-key
2258 ;; Regexp matching the start of a statement which may be followed by a
2259 ;; paren sexp and will then be followed by a substatement.
2260 t (c-make-keywords-re t (c-lang-const c-block-stmt-1-2-kwds)))
2261 (c-lang-defvar c-block-stmt-1-2-key (c-lang-const c-block-stmt-1-2-key))
2263 (c-lang-defconst c-block-stmt-2-kwds
2264 "Statement keywords followed by a paren sexp and then by a substatement."
2265 t '("for" "if" "switch" "while")
2266 c++ '("for" "if" "switch" "while" "catch")
2267 objc '("for" "if" "switch" "while" "@catch" "@synchronized")
2268 java '("for" "if" "switch" "while" "catch" "synchronized")
2270 pike '("for" "if" "switch" "while" "foreach")
2271 awk '("for" "if" "while"))
2273 (c-lang-defconst c-block-stmt-2-key
2274 ;; Regexp matching the start of any statement followed by a paren sexp
2275 ;; and then by a substatement.
2276 t (c-make-keywords-re t (c-lang-const c-block-stmt-2-kwds)))
2277 (c-lang-defvar c-block-stmt-2-key (c-lang-const c-block-stmt-2-key))
2279 (c-lang-defconst c-block-stmt-kwds
2280 ;; Union of `c-block-stmt-1-kwds' and `c-block-stmt-2-kwds'.
2281 t (cl-delete-duplicates (append (c-lang-const c-block-stmt-1-kwds)
2282 (c-lang-const c-block-stmt-2-kwds))
2283 :test 'string-equal))
2285 (c-lang-defconst c-opt-block-stmt-key
2286 ;; Regexp matching the start of any statement that has a
2287 ;; substatement (except a bare block). Nil in languages that
2288 ;; don't have such constructs.
2289 t (if (or (c-lang-const c-block-stmt-1-kwds)
2290 (c-lang-const c-block-stmt-2-kwds))
2291 (c-make-keywords-re t
2292 (append (c-lang-const c-block-stmt-1-kwds)
2293 (c-lang-const c-block-stmt-2-kwds)))))
2294 (c-lang-defvar c-opt-block-stmt-key (c-lang-const c-opt-block-stmt-key))
2296 (c-lang-defconst c-simple-stmt-kwds
2297 "Statement keywords followed by an expression or nothing."
2298 t '("break" "continue" "goto" "return")
2299 objc '("break" "continue" "goto" "return" "@throw")
2300 ;; Note: `goto' is not valid in Java, but the keyword is still reserved.
2301 java '("break" "continue" "goto" "return" "throw")
2303 pike '("break" "continue" "return")
2304 awk '(;; Not sure about "delete", "exit", "getline", etc. ; ACM 2002/5/30
2305 "break" "continue" "return" "delete" "exit" "getline" "next"
2306 "nextfile" "print" "printf"))
2308 (c-lang-defconst c-simple-stmt-key
2309 ;; Adorned regexp matching `c-simple-stmt-kwds'.
2310 t (c-make-keywords-re t (c-lang-const c-simple-stmt-kwds)))
2311 (c-lang-defvar c-simple-stmt-key (c-lang-const c-simple-stmt-key))
2313 (c-lang-defconst c-paren-stmt-kwds
2314 "Statement keywords followed by a parenthesis expression that
2315 nevertheless contains a list separated with ';' and not ','."
2319 (c-lang-defconst c-paren-stmt-key
2320 ;; Adorned regexp matching `c-paren-stmt-kwds'.
2321 t (c-make-keywords-re t (c-lang-const c-paren-stmt-kwds)))
2322 (c-lang-defvar c-paren-stmt-key (c-lang-const c-paren-stmt-key))
2324 (c-lang-defconst c-asm-stmt-kwds
2325 "Statement keywords followed by an assembler expression."
2327 (c c++) '("asm" "__asm__")) ;; Not standard, but common.
2329 (c-lang-defconst c-opt-asm-stmt-key
2330 ;; Regexp matching the start of an assembler statement. Nil in
2331 ;; languages that don't support that.
2332 t (if (c-lang-const c-asm-stmt-kwds)
2333 (c-make-keywords-re t (c-lang-const c-asm-stmt-kwds))))
2334 (c-lang-defvar c-opt-asm-stmt-key (c-lang-const c-opt-asm-stmt-key))
2336 (c-lang-defconst c-case-kwds
2337 "The keyword\(s) which introduce a \"case\" like construct.
2338 This construct is \"<keyword> <expression> :\"."
2342 (c-lang-defconst c-case-kwds-regexp
2343 ;; Adorned regexp matching any "case"-like keyword.
2344 t (c-make-keywords-re t (c-lang-const c-case-kwds)))
2345 (c-lang-defvar c-case-kwds-regexp (c-lang-const c-case-kwds-regexp))
2347 (c-lang-defconst c-label-kwds
2348 "Keywords introducing colon terminated labels in blocks."
2349 t '("case" "default"))
2351 (c-lang-defconst c-label-kwds-regexp
2352 ;; Adorned regexp matching any keyword that introduces a label.
2353 t (c-make-keywords-re t (c-lang-const c-label-kwds)))
2354 (c-lang-defvar c-label-kwds-regexp (c-lang-const c-label-kwds-regexp))
2356 (c-lang-defconst c-before-label-kwds
2357 "Keywords that might be followed by a label identifier."
2359 (java pike) (append '("break" "continue")
2360 (c-lang-const c-before-label-kwds))
2364 (c-lang-defconst c-constant-kwds
2365 "Keywords for constants."
2367 (c c++) '("NULL" ;; Not a keyword, but practically works as one.
2368 "false" "true") ; Defined in C99.
2369 objc '("nil" "Nil" "YES" "NO" "NS_DURING" "NS_HANDLER" "NS_ENDHANDLER")
2370 idl '("TRUE" "FALSE")
2371 java '("true" "false" "null") ; technically "literals", not keywords
2372 pike '("UNDEFINED")) ;; Not a keyword, but practically works as one.
2374 (c-lang-defconst c-primary-expr-kwds
2375 "Keywords besides constants and operators that start primary expressions."
2377 c++ '("operator" "this")
2378 objc '("super" "self")
2380 pike '("this")) ;; Not really a keyword, but practically works as one.
2382 (c-lang-defconst c-expr-kwds
2383 ;; Keywords that can occur anywhere in expressions. Built from
2384 ;; `c-primary-expr-kwds' and all keyword operators in `c-operators'.
2385 t (cl-delete-duplicates
2386 (append (c-lang-const c-primary-expr-kwds)
2387 (c-filter-ops (c-lang-const c-operator-list)
2389 "\\`\\(\\w\\|\\s_\\)+\\'"))
2390 :test 'string-equal))
2392 (c-lang-defconst c-lambda-kwds
2393 "Keywords that start lambda constructs, i.e. function definitions in
2398 (c-lang-defconst c-inexpr-block-kwds
2399 "Keywords that start constructs followed by statement blocks which can
2400 be used in expressions \(the gcc extension for this in C and C++ is
2401 handled separately by `c-recognize-paren-inexpr-blocks')."
2403 pike '("catch" "gauge"))
2405 (c-lang-defconst c-inexpr-class-kwds
2406 "Keywords that can start classes inside expressions."
2411 (c-lang-defconst c-inexpr-brace-list-kwds
2412 "Keywords that can start brace list blocks inside expressions.
2413 Note that Java specific rules are currently applied to tell this from
2414 `c-inexpr-class-kwds'."
2418 (c-lang-defconst c-opt-inexpr-brace-list-key
2419 ;; Regexp matching the start of a brace list in an expression, or
2420 ;; nil in languages that don't have such things. This should not
2421 ;; match brace lists recognized through `c-special-brace-lists'.
2422 t (and (c-lang-const c-inexpr-brace-list-kwds)
2423 (c-make-keywords-re t (c-lang-const c-inexpr-brace-list-kwds))))
2424 (c-lang-defvar c-opt-inexpr-brace-list-key
2425 (c-lang-const c-opt-inexpr-brace-list-key))
2427 (c-lang-defconst c-decl-block-key
2428 ;; Regexp matching keywords in any construct that contain another
2429 ;; declaration level, i.e. that isn't followed by a function block
2430 ;; or brace list. When the first submatch matches, it's an
2431 ;; unambiguous construct, otherwise it's an ambiguous match that
2432 ;; might also be the return type of a function declaration.
2433 t (let* ((decl-kwds (append (c-lang-const c-class-decl-kwds)
2434 (c-lang-const c-other-block-decl-kwds)
2435 (c-lang-const c-inexpr-class-kwds)))
2436 (unambiguous (set-difference decl-kwds
2437 (c-lang-const c-type-start-kwds)
2438 :test 'string-equal))
2439 (ambiguous (intersection decl-kwds
2440 (c-lang-const c-type-start-kwds)
2441 :test 'string-equal)))
2443 (concat (c-make-keywords-re t unambiguous)
2445 (c-make-keywords-re t ambiguous))
2446 (c-make-keywords-re t unambiguous))))
2447 (c-lang-defvar c-decl-block-key (c-lang-const c-decl-block-key))
2449 (c-lang-defconst c-bitfield-kwds
2450 "Keywords that can introduce bitfields."
2452 (c c++ objc) '("char" "int" "long" "signed" "unsigned"))
2454 (c-lang-defconst c-opt-bitfield-key
2455 ;; Regexp matching the start of a bitfield (not uniquely), or nil in
2456 ;; languages without bitfield support.
2458 (c c++) (c-make-keywords-re t (c-lang-const c-bitfield-kwds)))
2459 (c-lang-defvar c-opt-bitfield-key (c-lang-const c-opt-bitfield-key))
2461 (c-lang-defconst c-other-kwds
2462 "Keywords not accounted for by any other `*-kwds' language constant."
2465 ;; In CORBA CIDL: (These are declaration keywords that never
2466 ;; can start a declaration.)
2467 "entity" "process" "service" "session" "storage"))
2470 ;;; Constants built from keywords.
2472 ;; Note: No `*-kwds' language constants may be defined below this point.
2475 (defconst c-kwds-lang-consts
2476 ;; List of all the language constants that contain keyword lists.
2478 (mapatoms (lambda (sym)
2479 (when (and (boundp sym)
2480 (string-match "-kwds\\'" (symbol-name sym)))
2481 ;; Make the list of globally interned symbols
2482 ;; instead of ones interned in `c-lang-constants'.
2483 (setq list (cons (intern (symbol-name sym)) list))))
2487 (c-lang-defconst c-keywords
2488 ;; All keywords as a list.
2489 t (cl-delete-duplicates
2490 (c-lang-defconst-eval-immediately
2491 `(append ,@(mapcar (lambda (kwds-lang-const)
2492 `(c-lang-const ,kwds-lang-const))
2495 :test 'string-equal))
2497 (c-lang-defconst c-keywords-regexp
2498 ;; All keywords as an adorned regexp.
2499 t (c-make-keywords-re t (c-lang-const c-keywords)))
2500 (c-lang-defvar c-keywords-regexp (c-lang-const c-keywords-regexp))
2502 (c-lang-defconst c-keyword-member-alist
2503 ;; An alist with all the keywords in the cars. The cdr for each
2504 ;; keyword is a list of the symbols for the `*-kwds' lists that
2506 t (let ((kwd-list-alist
2507 (c-lang-defconst-eval-immediately
2508 `(list ,@(mapcar (lambda (kwds-lang-const)
2509 `(cons ',kwds-lang-const
2510 (c-lang-const ,kwds-lang-const)))
2511 c-kwds-lang-consts))))
2512 lang-const kwd-list kwd
2514 (while kwd-list-alist
2515 (setq lang-const (caar kwd-list-alist)
2516 kwd-list (cdar kwd-list-alist)
2517 kwd-list-alist (cdr kwd-list-alist))
2519 (setq kwd (car kwd-list)
2520 kwd-list (cdr kwd-list))
2521 (unless (setq elem (assoc kwd result-alist))
2522 (setq result-alist (cons (setq elem (list kwd)) result-alist)))
2523 (unless (memq lang-const (cdr elem))
2524 (setcdr elem (cons lang-const (cdr elem))))))
2527 (c-lang-defvar c-keywords-obarray
2528 ;; An obarray containing all keywords as symbols. The property list
2529 ;; of each symbol has a non-nil entry for the specific `*-kwds'
2530 ;; lists it's a member of.
2532 ;; E.g. to see whether the string str contains a keyword on
2533 ;; `c-class-decl-kwds', one can do like this:
2534 ;; (get (intern-soft str c-keyword-obarray) 'c-class-decl-kwds)
2535 ;; Which preferably is written using the associated functions in
2537 ;; (c-keyword-member (c-keyword-sym str) 'c-class-decl-kwds)
2539 ;; The obarray is not stored directly as a language constant since
2540 ;; the printed representation for obarrays used in .elc files isn't
2543 (let* ((alist (c-lang-const c-keyword-member-alist))
2545 (obarray (make-vector (* (length alist) 2) 0)))
2547 (setq kwd (caar alist)
2548 lang-const-list (cdar alist)
2550 (setplist (intern kwd obarray)
2551 ;; Emacs has an odd bug that causes `mapcan' to fail
2552 ;; with unintelligible errors. (XEmacs works.)
2553 ;;(mapcan (lambda (lang-const)
2554 ;; (list lang-const t))
2556 (apply 'nconc (mapcar (lambda (lang-const)
2557 (list lang-const t))
2561 (c-lang-defconst c-regular-keywords-regexp
2562 ;; Adorned regexp matching all keywords that should be fontified
2563 ;; with the keywords face. I.e. that aren't types or constants.
2564 t (c-make-keywords-re t
2565 (set-difference (c-lang-const c-keywords)
2566 (append (c-lang-const c-primitive-type-kwds)
2567 (c-lang-const c-constant-kwds))
2568 :test 'string-equal)))
2569 (c-lang-defvar c-regular-keywords-regexp
2570 (c-lang-const c-regular-keywords-regexp))
2572 (c-lang-defconst c-primary-expr-regexp
2573 ;; Regexp matching the start of any primary expression, i.e. any
2574 ;; literal, symbol, prefix operator, and '('. It doesn't need to
2575 ;; exclude keywords; they are excluded afterwards unless the second
2576 ;; submatch matches. If the first but not the second submatch
2577 ;; matches then it is an ambiguous primary expression; it could also
2578 ;; be a match of e.g. an infix operator. (The case with ambiguous
2579 ;; keyword operators isn't handled.)
2581 t (let* ((prefix-ops
2582 (c-filter-ops (c-lang-const c-operators)
2585 ;; Filter out the special case prefix
2586 ;; operators that are close parens.
2587 (not (string-match "\\s)" op)))))
2589 (nonkeyword-prefix-ops
2590 (c-filter-ops prefix-ops
2592 "\\`\\(\\s.\\|\\s(\\|\\s)\\)+\\'"))
2595 (c-filter-ops (c-lang-const c-operators)
2600 right-assoc-sequence)
2603 (unambiguous-prefix-ops (set-difference nonkeyword-prefix-ops
2605 :test 'string-equal))
2606 (ambiguous-prefix-ops (intersection nonkeyword-prefix-ops
2608 :test 'string-equal)))
2612 ;; Take out all symbol class operators from `prefix-ops' and make the
2613 ;; first submatch from them together with `c-primary-expr-kwds'.
2614 (c-make-keywords-re t
2615 (append (c-lang-const c-primary-expr-kwds)
2616 (set-difference prefix-ops nonkeyword-prefix-ops
2617 :test 'string-equal)))
2620 ;; Match all ambiguous operators.
2621 (c-make-keywords-re nil
2622 (intersection nonkeyword-prefix-ops in-or-postfix-ops
2623 :test 'string-equal))
2627 ;; Now match all other symbols.
2628 (c-lang-const c-symbol-start)
2631 ;; The chars that can start integer and floating point
2636 ;; The unambiguous operators from `prefix-ops'.
2637 (c-make-keywords-re nil
2638 (set-difference nonkeyword-prefix-ops in-or-postfix-ops
2639 :test 'string-equal))
2642 ;; Match string and character literals.
2644 (if (memq 'gen-string-delim c-emacs-features)
2647 (c-lang-defvar c-primary-expr-regexp (c-lang-const c-primary-expr-regexp))
2650 ;;; Additional constants for parser-level constructs.
2652 (c-lang-defconst c-decl-start-colon-kwd-re
2653 "Regexp matching a keyword that is followed by a colon, where
2654 the whole construct can precede a declaration.
2655 E.g. \"public:\" in C++."
2657 c++ (c-make-keywords-re t (c-lang-const c-protection-kwds)))
2658 (c-lang-defvar c-decl-start-colon-kwd-re
2659 (c-lang-const c-decl-start-colon-kwd-re))
2661 (c-lang-defconst c-decl-prefix-re
2662 "Regexp matching something that might precede a declaration, cast or
2663 label, such as the last token of a preceding statement or declaration.
2664 This is used in the common situation where a declaration or cast
2665 doesn't start with any specific token that can be searched for.
2667 The regexp should not match bob; that is done implicitly. It can't
2668 require a match longer than one token. The end of the token is taken
2669 to be at the end of the first submatch, which is assumed to always
2670 match. It's undefined whether identifier syntax (see
2671 `c-identifier-syntax-table') is in effect or not. This regexp is
2672 assumed to be a superset of `c-label-prefix-re' if
2673 `c-recognize-colon-labels' is set.
2675 Besides this, `c-decl-start-kwds' is used to find declarations.
2677 Note: This variable together with `c-decl-start-re' and
2678 `c-decl-start-kwds' is only used to detect \"likely\"
2679 declaration/cast/label starts. I.e. they might produce more matches
2680 but should not miss anything (or else it's necessary to use text
2681 properties - see the next note). Wherever they match, the following
2682 construct is analyzed to see if it indeed is a declaration, cast or
2683 label. That analysis is not cheap, so it's important that not too
2684 many false matches are triggered.
2686 Note: If a declaration/cast/label start can't be detected with this
2687 variable, it's necessary to use the `c-type' text property with the
2688 value `c-decl-end' on the last char of the last token preceding the
2689 declaration. See the comment blurb at the start of cc-engine.el for
2692 ;; We match a sequence of characters to skip over things like \"};\"
2693 ;; more quickly. We match ")" in C for K&R region declarations, and
2694 ;; in all languages except Java for when a cpp macro definition
2695 ;; begins with a declaration.
2696 t "\\([\{\}\(\);,]+\\)"
2697 java "\\([\{\}\(;,<]+\\)"
2698 ;; Match "<" in C++ to get the first argument in a template arglist.
2699 ;; In that case there's an additional check in `c-find-decl-spots'
2700 ;; that it got open paren syntax. Match ":" to aid in picking up
2701 ;; "public:", etc. This involves additional checks in
2702 ;; `c-find-decl-prefix-search' to prevent a match of identifiers
2704 c++ "\\([\{\}\(\);:,<]+\\)"
2705 ;; Additionally match the protection directives in Objective-C.
2706 ;; Note that this doesn't cope with the longer directives, which we
2707 ;; would have to match from start to end since they don't end with
2708 ;; any easily recognized characters.
2709 objc (concat "\\([\{\}\(\);,]+\\|"
2710 (c-make-keywords-re nil (c-lang-const c-protection-kwds))
2712 ;; Pike is like C but we also match "[" for multiple value
2713 ;; assignments and type casts.
2714 pike "\\([\{\}\(\)\[;,]+\\)")
2715 (c-lang-defvar c-decl-prefix-re (c-lang-const c-decl-prefix-re)
2718 (c-lang-defconst c-decl-start-re
2719 "Regexp matching the start of any declaration, cast or label.
2720 It's used on the token after the one `c-decl-prefix-re' matched. This
2721 regexp should not try to match those constructs accurately as it's
2722 only used as a sieve to avoid spending more time checking other
2724 t (c-lang-const c-identifier-start))
2725 (c-lang-defvar c-decl-start-re (c-lang-const c-decl-start-re))
2727 (c-lang-defconst c-decl-prefix-or-start-re
2728 ;; Regexp matching something that might precede or start a
2729 ;; declaration, cast or label.
2731 ;; If the first submatch matches, it's taken to match the end of a
2732 ;; token that might precede such a construct, e.g. ';', '}' or '{'.
2733 ;; It's built from `c-decl-prefix-re'.
2735 ;; If the first submatch did not match, the match of the whole
2736 ;; regexp is taken to be at the first token in the declaration.
2737 ;; `c-decl-start-re' is not checked in this case.
2739 ;; Design note: The reason the same regexp is used to match both
2740 ;; tokens that precede declarations and start them is to avoid an
2741 ;; extra regexp search from the previous declaration spot in
2742 ;; `c-find-decl-spots'. Users of `c-find-decl-spots' also count on
2743 ;; that it finds all declaration/cast/label starts in approximately
2744 ;; linear order, so we can't do the searches in two separate passes.
2745 t (if (c-lang-const c-decl-start-kwds)
2746 (concat (c-lang-const c-decl-prefix-re)
2748 (c-make-keywords-re t (c-lang-const c-decl-start-kwds)))
2749 (c-lang-const c-decl-prefix-re)))
2750 (c-lang-defvar c-decl-prefix-or-start-re
2751 (c-lang-const c-decl-prefix-or-start-re)
2754 (c-lang-defconst c-cast-parens
2755 ;; List containing the paren characters that can open a cast, or nil in
2756 ;; languages without casts.
2757 t (c-filter-ops (c-lang-const c-operators)
2760 (lambda (op) (elt op 0))))
2761 (c-lang-defvar c-cast-parens (c-lang-const c-cast-parens))
2763 (c-lang-defconst c-block-prefix-disallowed-chars
2764 "List of syntactically relevant characters that never can occur before
2765 the open brace in any construct that contains a brace block, e.g. in
2766 the \"class Foo: public Bar\" part of:
2768 class Foo: public Bar {int x();} a, *b;
2770 If parens can occur, the chars inside those aren't filtered with this
2773 '<' and '>' should be disallowed even if angle bracket arglists can
2774 occur. That since the search function needs to stop at them anyway to
2775 ensure they are given paren syntax.
2777 This is used to skip backward from the open brace to find the region
2778 in which to look for a construct like \"class\", \"enum\",
2779 \"namespace\" or whatever. That skipping should be as tight as
2780 possible for good performance."
2782 ;; Default to all chars that only occurs in nonsymbol tokens outside
2785 (c-lang-const c-nonsymbol-token-char-list)
2786 (c-filter-ops (append (c-lang-const c-identifier-ops)
2788 (c-lang-const c-after-id-concat-ops))))
2793 (while (string-match "\\(\\s.\\|\\s(\\|\\s)\\)"
2795 (setq res (cons (aref op (match-beginning 1)) res)
2799 ;; Allow cpp operations (where applicable).
2800 t (if (c-lang-const c-opt-cpp-prefix)
2801 (set-difference (c-lang-const c-block-prefix-disallowed-chars)
2803 (c-lang-const c-block-prefix-disallowed-chars))
2805 ;; Allow ':' for inherit list starters.
2806 (c++ objc idl) (set-difference (c-lang-const c-block-prefix-disallowed-chars)
2809 ;; Allow ',' for multiple inherits.
2810 (c++ java) (set-difference (c-lang-const c-block-prefix-disallowed-chars)
2813 ;; Allow parentheses for anonymous inner classes in Java and class
2814 ;; initializer lists in Pike.
2815 (java pike) (set-difference (c-lang-const c-block-prefix-disallowed-chars)
2818 ;; Allow '"' for extern clauses (e.g. extern "C" {...}).
2819 (c c++ objc) (set-difference (c-lang-const c-block-prefix-disallowed-chars)
2822 (c-lang-defconst c-block-prefix-charset
2823 ;; `c-block-prefix-disallowed-chars' as an inverted charset suitable
2824 ;; for `c-syntactic-skip-backward'.
2825 t (c-make-bare-char-alt (c-lang-const c-block-prefix-disallowed-chars) t))
2826 (c-lang-defvar c-block-prefix-charset (c-lang-const c-block-prefix-charset))
2828 (c-lang-defconst c-type-decl-prefix-key
2829 "Regexp matching the declarator operators that might precede the
2830 identifier in a declaration, e.g. the \"*\" in \"char *argv\". This
2831 regexp should match \"(\" if parentheses are valid in declarators.
2832 The end of the first submatch is taken as the end of the operator.
2833 Identifier syntax is in effect when this is matched \(see
2834 `c-identifier-syntax-table')."
2835 t (if (c-lang-const c-type-modifier-kwds)
2836 (concat (regexp-opt (c-lang-const c-type-modifier-kwds) t) "\\>")
2837 ;; Default to a regexp that never matches.
2839 ;; Check that there's no "=" afterwards to avoid matching tokens
2841 (c objc) (concat "\\("
2844 (c-lang-const c-type-decl-prefix-key)
2850 (c-lang-const c-type-decl-prefix-key)
2853 ;; If this matches there's special treatment in
2854 ;; `c-font-lock-declarators' and
2855 ;; `c-font-lock-declarations' that check for a
2856 ;; complete name followed by ":: *".
2857 (c-lang-const c-identifier-start)
2861 pike "\\(\\*\\)\\([^=]\\|$\\)")
2862 (c-lang-defvar c-type-decl-prefix-key (c-lang-const c-type-decl-prefix-key)
2865 (c-lang-defconst c-type-decl-suffix-key
2866 "Regexp matching the declarator operators that might follow after the
2867 identifier in a declaration, e.g. the \"[\" in \"char argv[]\". This
2868 regexp should match \")\" if parentheses are valid in declarators. If
2869 it matches an open paren of some kind, the type declaration check
2870 continues at the corresponding close paren, otherwise the end of the
2871 first submatch is taken as the end of the operator. Identifier syntax
2872 is in effect when this is matched (see `c-identifier-syntax-table')."
2873 ;; Default to a regexp that matches `c-type-modifier-kwds' and a
2874 ;; function argument list parenthesis.
2875 t (if (c-lang-const c-type-modifier-kwds)
2877 (regexp-opt (c-lang-const c-type-modifier-kwds) t) "\\>"
2880 (c c++ objc) (concat
2883 (if (c-lang-const c-type-modifier-kwds)
2886 ;; "throw" in `c-type-modifier-kwds' is followed
2887 ;; by a parenthesis list, but no extra measures
2888 ;; are necessary to handle that.
2889 (regexp-opt (c-lang-const c-type-modifier-kwds) t)
2893 java "\\([\[\(\)]\\)"
2895 (c-lang-defvar c-type-decl-suffix-key (c-lang-const c-type-decl-suffix-key)
2898 (c-lang-defconst c-after-suffixed-type-decl-key
2899 "This regexp is matched after a declarator expression where
2900 `c-type-decl-suffix-key' has matched. If it matches then the
2901 construct is taken as a declaration. It's typically used to match the
2902 beginning of a function body or whatever might occur after the
2903 function header in a function declaration or definition. It's
2904 undefined whether identifier syntax (see `c-identifier-syntax-table')
2905 is in effect or not.
2907 Note that it's used in cases like after \"foo (bar)\" so it should
2908 only match when it's certain that it's a declaration, e.g., \"{\" but
2909 not \",\" or \";\"."
2911 ;; If K&R style declarations should be recognized then one could
2912 ;; consider to match the start of any symbol since we want to match
2913 ;; the start of the first declaration in the "K&R region". That
2914 ;; could however produce false matches on code like "FOO(bar) x"
2915 ;; where FOO is a cpp macro, so it's better to leave it out and rely
2916 ;; on the other heuristics in that case.
2917 t (if (c-lang-const c-postfix-spec-kwds)
2918 ;; Add on the keywords in `c-postfix-spec-kwds'.
2919 (concat (c-lang-const c-after-suffixed-type-decl-key)
2921 (c-make-keywords-re t (c-lang-const c-postfix-spec-kwds)))
2922 (c-lang-const c-after-suffixed-type-decl-key))
2923 ;; Also match the colon that starts a base class initializer list in
2924 ;; C++. That can be confused with a function call before the colon
2925 ;; in a ? : operator, but we count on that `c-decl-prefix-re' won't
2926 ;; match before such a thing (as a declaration-level construct;
2927 ;; matches inside arglist contexts are already excluded).
2929 (c-lang-defvar c-after-suffixed-type-decl-key
2930 (c-lang-const c-after-suffixed-type-decl-key)
2933 (c-lang-defconst c-after-suffixed-type-maybe-decl-key
2934 ;; Regexp that in addition to `c-after-suffixed-type-decl-key'
2935 ;; matches ";" and ",".
2936 t (concat "\\(" (c-lang-const c-after-suffixed-type-decl-key) "\\)"
2938 (c-lang-defvar c-after-suffixed-type-maybe-decl-key
2939 (c-lang-const c-after-suffixed-type-maybe-decl-key))
2941 (c-lang-defconst c-opt-type-concat-key
2942 "Regexp matching operators that concatenate types, e.g. the \"|\" in
2943 \"int|string\" in Pike. The end of the first submatch is taken as the
2944 end of the operator. nil in languages without such operators. It's
2945 undefined whether identifier syntax (see `c-identifier-syntax-table')
2946 is in effect or not."
2948 pike "\\([|.&]\\)\\($\\|[^|.&]\\)")
2949 (c-lang-defvar c-opt-type-concat-key (c-lang-const c-opt-type-concat-key)
2952 (c-lang-defconst c-opt-type-suffix-key
2953 "Regexp matching operators that might follow after a type, or nil in
2954 languages that don't have such operators. The end of the first
2955 submatch is taken as the end of the operator. This should not match
2956 things like C++ template arglists if `c-recognize-<>-arglists' is set.
2957 It's undefined whether identifier syntax (see `c-identifier-syntax-table')
2958 is in effect or not."
2960 (c c++ objc pike) "\\(\\.\\.\\.\\)"
2961 java (concat "\\(\\[" (c-lang-const c-simple-ws) "*\\]\\|\\.\\.\\.\\)"))
2962 (c-lang-defvar c-opt-type-suffix-key (c-lang-const c-opt-type-suffix-key))
2964 (c-lang-defvar c-known-type-key
2965 ;; Regexp matching the known type identifiers. This is initialized
2966 ;; from the type keywords and `*-font-lock-extra-types'. The first
2967 ;; submatch is the one that matches the type. Note that this regexp
2968 ;; assumes that symbol constituents like '_' and '$' have word
2971 (when (boundp (c-mode-symbol "font-lock-extra-types"))
2972 (c-mode-var "font-lock-extra-types")))
2975 (mapcar (lambda (re)
2976 (when (string-match "[][.*+?^$\\]" re)
2981 (mapcar (lambda (re)
2982 (unless (string-match "[][.*+?^$\\]" re)
2987 (append (list (c-make-keywords-re nil
2988 (append (c-lang-const c-primitive-type-kwds)
2994 (c-lang-defconst c-special-brace-lists
2995 "List of open- and close-chars that makes up a pike-style brace list,
2996 i.e. for a ([Â ]) list there should be a cons (?\\[ . ?\\]) in this
2999 pike '((?{ . ?}) (?\[ . ?\]) (?< . ?>)))
3000 (c-lang-defvar c-special-brace-lists (c-lang-const c-special-brace-lists))
3002 (c-lang-defconst c-recognize-knr-p
3003 "Non-nil means K&R style argument declarations are valid."
3006 (c-lang-defvar c-recognize-knr-p (c-lang-const c-recognize-knr-p))
3008 (c-lang-defconst c-recognize-typeless-decls
3009 "Non-nil means function declarations without return type should be
3010 recognized. That can introduce an ambiguity with parenthesized macro
3011 calls before a brace block. This setting does not affect declarations
3012 that are preceded by a declaration starting keyword, so
3013 e.g. `c-typeless-decl-kwds' may still be used when it's set to nil."
3015 (c c++ objc java) t)
3016 (c-lang-defvar c-recognize-typeless-decls
3017 (c-lang-const c-recognize-typeless-decls))
3019 (c-lang-defconst c-recognize-<>-arglists
3020 "Non-nil means C++ style template arglists should be handled. More
3021 specifically, this means a comma separated list of types or
3022 expressions surrounded by \"<\" and \">\". It's always preceded by an
3023 identifier or one of the keywords on `c-<>-type-kwds' or
3024 `c-<>-arglist-kwds'. If there's an identifier before then the whole
3025 expression is considered to be a type."
3026 t (or (consp (c-lang-const c-<>-type-kwds))
3027 (consp (c-lang-const c-<>-arglist-kwds)))
3029 (c-lang-defvar c-recognize-<>-arglists (c-lang-const c-recognize-<>-arglists))
3031 (c-lang-defconst c-enums-contain-decls
3032 "Non-nil means that an enum structure can contain declarations."
3035 (c-lang-defvar c-enums-contain-decls (c-lang-const c-enums-contain-decls))
3037 (c-lang-defconst c-recognize-paren-inits
3038 "Non-nil means that parenthesis style initializers exist,
3039 i.e. constructs like
3043 in addition to the more classic
3048 (c-lang-defvar c-recognize-paren-inits (c-lang-const c-recognize-paren-inits))
3050 (c-lang-defconst c-recognize-paren-inexpr-blocks
3051 "Non-nil to recognize gcc style in-expression blocks,
3052 i.e. compound statements surrounded by parentheses inside expressions."
3055 (c-lang-defvar c-recognize-paren-inexpr-blocks
3056 (c-lang-const c-recognize-paren-inexpr-blocks))
3058 (c-lang-defconst c-opt-<>-arglist-start
3059 ;; Regexp matching the start of angle bracket arglists in languages
3060 ;; where `c-recognize-<>-arglists' is set. Does not exclude
3061 ;; keywords outside `c-<>-arglist-kwds'. The first submatch is
3062 ;; assumed to surround the preceding symbol. The whole match is
3063 ;; assumed to end directly after the opening "<".
3064 t (if (c-lang-const c-recognize-<>-arglists)
3066 (c-lang-const c-symbol-key)
3068 (c-lang-const c-syntactic-ws)
3070 (c-lang-defvar c-opt-<>-arglist-start (c-lang-const c-opt-<>-arglist-start))
3072 (c-lang-defconst c-opt-<>-arglist-start-in-paren
3073 ;; Regexp that in addition to `c-opt-<>-arglist-start' matches close
3074 ;; parens. The first submatch is assumed to surround
3075 ;; `c-opt-<>-arglist-start'.
3076 t (if (c-lang-const c-opt-<>-arglist-start)
3078 (c-lang-const c-opt-<>-arglist-start)
3080 (c-lang-defvar c-opt-<>-arglist-start-in-paren
3081 (c-lang-const c-opt-<>-arglist-start-in-paren))
3083 (c-lang-defconst c-opt-postfix-decl-spec-key
3084 ;; Regexp matching the beginning of a declaration specifier in the
3085 ;; region between the header and the body of a declaration.
3087 ;; TODO: This is currently not used uniformly; c++-mode and
3088 ;; java-mode each have their own ways of using it.
3091 (c-lang-const c-simple-ws) "*"
3092 "\\(virtual" (c-lang-const c-simple-ws) "+\\)?\\("
3093 (c-make-keywords-re nil (c-lang-const c-protection-kwds))
3094 "\\)" (c-lang-const c-simple-ws) "+"
3095 "\\(" (c-lang-const c-symbol-key) "\\)")
3096 java (c-make-keywords-re t (c-lang-const c-postfix-spec-kwds)))
3097 (c-lang-defvar c-opt-postfix-decl-spec-key
3098 (c-lang-const c-opt-postfix-decl-spec-key))
3100 (c-lang-defconst c-recognize-colon-labels
3101 "Non-nil if generic labels ending with \":\" should be recognized.
3102 That includes labels in code and access keys in classes. This does
3103 not apply to labels recognized by `c-label-kwds' and
3104 `c-opt-extra-label-key'."
3106 (c c++ objc java pike) t)
3107 (c-lang-defvar c-recognize-colon-labels
3108 (c-lang-const c-recognize-colon-labels))
3110 (c-lang-defconst c-label-prefix-re
3111 "Regexp like `c-decl-prefix-re' that matches any token that can precede
3112 a generic colon label. Not used if `c-recognize-colon-labels' is
3115 (c-lang-defvar c-label-prefix-re
3116 (c-lang-const c-label-prefix-re))
3118 (c-lang-defconst c-nonlabel-token-key
3119 "Regexp matching things that can't occur in generic colon labels,
3120 neither in a statement nor in a declaration context. The regexp is
3121 tested at the beginning of every sexp in a suspected label,
3122 i.e. before \":\". Only used if `c-recognize-colon-labels' is set."
3124 ;; All keywords except `c-label-kwds' and `c-protection-kwds'.
3125 (c-make-keywords-re t
3126 (set-difference (c-lang-const c-keywords)
3127 (append (c-lang-const c-label-kwds)
3128 (c-lang-const c-protection-kwds))
3129 :test 'string-equal)))
3130 ;; Don't allow string literals, except in AWK. Character constants are OK.
3131 (c objc java pike idl) (concat "\"\\|"
3132 (c-lang-const c-nonlabel-token-key))
3133 ;; Also check for open parens in C++, to catch member init lists in
3134 ;; constructors. We normally allow it so that macros with arguments
3136 c++ (concat "\\s\(\\|\"\\|" (c-lang-const c-nonlabel-token-key)))
3137 (c-lang-defvar c-nonlabel-token-key (c-lang-const c-nonlabel-token-key))
3139 (c-lang-defconst c-nonlabel-token-2-key
3140 "Regexp matching things that can't occur two symbols before a colon in
3141 a label construct. This catches C++'s inheritance construct \"class foo
3142 : bar\". Only used if `c-recognize-colon-labels' is set."
3143 t "\\<\\>" ; matches nothing
3144 c++ (c-make-keywords-re t '("class")))
3145 (c-lang-defvar c-nonlabel-token-2-key (c-lang-const c-nonlabel-token-2-key))
3147 (c-lang-defconst c-opt-extra-label-key
3148 "Optional regexp matching labels.
3149 Normally, labels are detected according to `c-nonlabel-token-key',
3150 `c-decl-prefix-re' and `c-nonlabel-decl-prefix-re'. This regexp can
3151 be used if there are additional labels that aren't recognized that
3154 objc (c-make-keywords-re t (c-lang-const c-protection-kwds)))
3155 (c-lang-defvar c-opt-extra-label-key (c-lang-const c-opt-extra-label-key))
3157 (c-lang-defconst c-opt-friend-key
3158 ;; Regexp describing friend declarations classes, or nil in
3159 ;; languages that don't have such things.
3161 ;; TODO: Ought to use `c-prefix-spec-kwds-re' or similar, and the
3162 ;; template skipping isn't done properly. This will disappear soon.
3164 c++ (concat "friend" (c-lang-const c-simple-ws) "+"
3167 (c-lang-const c-simple-ws) "*"
3169 (c-lang-const c-simple-ws) "*"
3171 (c-lang-const c-simple-ws) "+")))
3172 (c-lang-defvar c-opt-friend-key (c-lang-const c-opt-friend-key))
3174 (c-lang-defconst c-opt-method-key
3175 ;; Special regexp to match the start of Objective-C methods. The
3176 ;; first submatch is assumed to end after the + or - key.
3179 ;; TODO: Ought to use a better method than anchoring on bol.
3182 (c-lang-const c-simple-ws) "*"
3183 (concat "\\(" ; Return type.
3185 (c-lang-const c-simple-ws) "*"
3187 "\\(" (c-lang-const c-symbol-key) "\\)"))
3188 (c-lang-defvar c-opt-method-key (c-lang-const c-opt-method-key))
3190 (c-lang-defconst c-type-decl-end-used
3191 ;; Must be set in buffers where the `c-type' text property might be
3192 ;; used with the value `c-decl-end'.
3194 ;; `c-decl-end' is used to mark the ends of labels and access keys
3195 ;; to make interactive refontification work better.
3196 t (or (c-lang-const c-recognize-colon-labels)
3197 (and (c-lang-const c-label-kwds) t))
3198 ;; `c-decl-end' is used to mark the end of the @-style directives in
3201 (c-lang-defvar c-type-decl-end-used (c-lang-const c-type-decl-end-used))
3204 ;;; Wrap up the `c-lang-defvar' system.
3206 ;; Compile in the list of language variables that has been collected
3207 ;; with the `c-lang-defvar' and `c-lang-setvar' macros. Note that the
3208 ;; first element of each is nil.
3209 (defconst c-lang-variable-inits (cc-eval-when-compile c-lang-variable-inits))
3210 (defconst c-emacs-variable-inits (cc-eval-when-compile c-emacs-variable-inits))
3212 ;; Make the `c-lang-setvar' variables buffer local in the current buffer.
3213 ;; These are typically standard emacs variables such as `comment-start'.
3214 (defmacro c-make-emacs-variables-local ()
3216 ,@(mapcar (lambda (init)
3217 `(make-local-variable ',(car init)))
3218 (cdr c-emacs-variable-inits))))
3220 (defun c-make-init-lang-vars-fun (mode)
3221 "Create a function that initializes all the language dependent variables
3224 This function should be evaluated at compile time, so that the
3225 function it returns is byte compiled with all the evaluated results
3226 from the language constants. Use the `c-init-language-vars' macro to
3227 accomplish that conveniently."
3229 (if (and (not load-in-progress)
3230 (boundp 'byte-compile-dest-file)
3231 (stringp byte-compile-dest-file))
3233 ;; No need to byte compile this lambda since the byte compiler is
3234 ;; smart enough to detect the `funcall' construct in the
3235 ;; `c-init-language-vars' macro below and compile it all straight
3236 ;; into the function that contains `c-init-language-vars'.
3239 ;; This let sets up the context for `c-mode-var' and similar
3240 ;; that could be in the result from `macroexpand-all'.
3241 (let ((c-buffer-is-cc-mode ',mode)
3242 current-var source-eval)
3243 (c-make-emacs-variables-local)
3246 (if (eq c-version-sym ',c-version-sym)
3247 (setq ,@(let ((c-buffer-is-cc-mode mode)
3248 (c-lang-const-expansion 'immediate))
3249 ;; `c-lang-const' will expand to the evaluated
3250 ;; constant immediately in `macroexpand-all'
3254 `(current-var ',(car init)
3255 ,(car init) ,(macroexpand-all
3257 ;; Note: The following `append' copies the
3258 ;; first argument. That list is small, so
3259 ;; this doesn't matter too much.
3260 (append (cdr c-emacs-variable-inits)
3261 (cdr c-lang-variable-inits)))))
3263 ;; This diagnostic message isn't useful for end
3264 ;; users, so it's disabled.
3265 ;;(unless (get ',mode 'c-has-warned-lang-consts)
3266 ;; (message ,(concat "%s compiled with CC Mode %s "
3267 ;; "but loaded with %s - evaluating "
3268 ;; "language constants from source")
3269 ;; ',mode ,c-version c-version)
3270 ;; (put ',mode 'c-has-warned-lang-consts t))
3272 (setq source-eval t)
3273 (let ((init ',(append (cdr c-emacs-variable-inits)
3274 (cdr c-lang-variable-inits))))
3276 (setq current-var (caar init))
3277 (set (caar init) (eval (cadar init)))
3278 (setq init (cdr init)))))
3282 (message "Eval error in the `c-lang-defvar' or `c-lang-setvar' for `%s'%s: %S"
3286 (fallback source eval - %s compiled with CC Mode %s but loaded with %s)"
3287 ',mode ,c-version c-version)
3290 (signal (car err) (cdr err)))))))
3292 ;; Being evaluated from source. Always use the dynamic method to
3293 ;; work well when `c-lang-defvar's in this file are reevaluated
3297 (let ((c-buffer-is-cc-mode ',mode)
3298 (init (append (cdr c-emacs-variable-inits)
3299 (cdr c-lang-variable-inits)))
3301 (c-make-emacs-variables-local)
3305 (setq current-var (caar init))
3306 (set (caar init) (eval (cadar init)))
3307 (setq init (cdr init)))
3312 "Eval error in the `c-lang-defvar' or `c-lang-setver' for `%s' (source eval): %S"
3314 (signal (car err) (cdr err)))))))
3317 (defmacro c-init-language-vars (mode)
3318 "Initialize all the language dependent variables for the given mode.
3319 This macro is expanded at compile time to a form tailored for the mode
3320 in question, so MODE must be a constant. Therefore MODE is not
3321 evaluated and should not be quoted."
3322 `(funcall ,(c-make-init-lang-vars-fun mode)))
3325 (cc-provide 'cc-langs)
3327 ;;; cc-langs.el ends here