1 ;;; cc-langs.el --- language specific settings for CC Mode -*- coding: utf-8 -*-
3 ;; Copyright (C) 1985, 1987, 1992-2016 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 (cc-external-require (if (eq c--mapcan-status 'cl-mapcan) 'cl-lib 'cl))
136 ;;; Setup for the `c-lang-defvar' system.
139 ;; These are used to collect the init forms from the subsequent
140 ;; `c-lang-defvar' and `c-lang-setvar'. They are used to build the
141 ;; lambda in `c-make-init-lang-vars-fun' below, and to build `defvar's
142 ;; and `make-variable-buffer-local's in cc-engine and
143 ;; `make-local-variable's in `c-init-language-vars-for'.
144 (defvar c-lang-variable-inits nil)
145 (defvar c-lang-variable-inits-tail nil)
146 (setq c-lang-variable-inits (list nil)
147 c-lang-variable-inits-tail c-lang-variable-inits)
148 (defvar c-emacs-variable-inits nil)
149 (defvar c-emacs-variable-inits-tail nil)
150 (setq c-emacs-variable-inits (list nil)
151 c-emacs-variable-inits-tail c-emacs-variable-inits))
153 (defmacro c-lang-defvar (var val &optional doc)
154 "Declares the buffer local variable VAR to get the value VAL. VAL is
155 evaluated and assigned at mode initialization. More precisely, VAL is
156 evaluated and bound to VAR when the result from the macro
157 `c-init-language-vars' is evaluated.
159 `c-lang-const' is typically used in VAL to get the right value for the
160 language being initialized, and such calls will be macro expanded to
161 the evaluated constant value at compile time."
164 (eq (car-safe val) 'c-lang-const)
167 ;; Special case: If there's no docstring and the value is a
168 ;; simple (c-lang-const foo) where foo is the same name as VAR
169 ;; then take the docstring from the language constant foo.
170 (setq doc (get (intern (symbol-name (nth 1 val)) c-lang-constants)
171 'variable-documentation)))
175 (let ((elem (assq var (cdr c-lang-variable-inits))))
177 (setcdr elem (list val doc))
178 (setcdr c-lang-variable-inits-tail (list (list var val doc)))
179 (setq c-lang-variable-inits-tail (cdr c-lang-variable-inits-tail))))
181 ;; Return the symbol, like the other def* forms.
184 (defmacro c-lang-setvar (var val)
185 "Causes the variable VAR to be made buffer local and to get set to the
186 value VAL. VAL is evaluated and assigned at mode initialization. More
187 precisely, VAL is evaluated and bound to VAR when the result from the
188 macro `c-init-language-vars' is evaluated. VAR is typically a standard
189 Emacs variable like `comment-start'.
191 `c-lang-const' is typically used in VAL to get the right value for the
192 language being initialized, and such calls will be macro expanded to
193 the evaluated constant value at compile time."
194 (let ((elem (assq var (cdr c-emacs-variable-inits))))
196 (setcdr elem (list val)) ; Maybe remove "list", sometime. 2006-07-19
197 (setcdr c-emacs-variable-inits-tail (list (list var val)))
198 (setq c-emacs-variable-inits-tail (cdr c-emacs-variable-inits-tail))))
200 ;; Return the symbol, like the other def* forms.
203 (put 'c-lang-defvar 'lisp-indent-function 'defun)
204 ; (eval-after-load "edebug" ; 2006-07-09: def-edebug-spec is now in subr.el.
206 (def-edebug-spec c-lang-defvar
207 (&define name def-form &optional stringp)) ;)
209 ;; Suppress "might not be defined at runtime" warning.
210 ;; This file is only used when compiling other cc files.
211 ;; These are defined in cl as aliases to the cl- versions.
212 ;(declare-function delete-duplicates "cl-seq" (cl-seq &rest cl-keys) t)
213 ;(declare-function mapcan "cl-extra" (cl-func cl-seq &rest cl-rest) t)
216 ;; Some helper functions used when building the language constants.
218 (defun c-filter-ops (ops opgroup-filter op-filter &optional xlate)
219 ;; Extract a subset of the operators in the list OPS in a DWIM:ey
220 ;; way. The return value is a plain list of operators:
222 ;; OPS either has the structure of `c-operators', is a single
223 ;; group in `c-operators', or is a plain list of operators.
225 ;; OPGROUP-FILTER specifies how to select the operator groups. It
226 ;; can be t to choose all groups, a list of group type symbols
227 ;; (such as 'prefix) to accept, or a function which will be called
228 ;; with the group symbol for each group and should return non-nil
229 ;; if that group is to be included.
231 ;; OP-FILTER selects the operators. It is either t to select all
232 ;; operators, a string to select all operators for which `string-match'
233 ;; matches the operator with the string, or a function which will be
234 ;; called with the operator and should return non-nil when the operator
235 ;; is to be selected.
237 ;; If XLATE is given, it's a function which is called for each
238 ;; matching operator and its return value is collected instead.
239 ;; If it returns a list, the elements are spliced directly into
240 ;; the final result, which is returned as a list with duplicates
241 ;; removed using `equal'.
243 ;; `c-mode-syntax-table' for the current mode is in effect during
244 ;; the whole procedure.
245 (unless (listp (car-safe ops))
246 (setq ops (list ops)))
247 (cond ((eq opgroup-filter t)
248 (setq opgroup-filter (lambda (opgroup) t)))
249 ((not (functionp opgroup-filter))
250 (setq opgroup-filter `(lambda (opgroup)
251 (memq opgroup ',opgroup-filter)))))
252 (cond ((eq op-filter t)
253 (setq op-filter (lambda (op) t)))
255 (setq op-filter `(lambda (op)
256 (string-match ,op-filter op)))))
258 (setq xlate 'identity))
259 (c-with-syntax-table (c-lang-const c-mode-syntax-table)
260 (c--delete-duplicates
261 (c--mapcan (lambda (opgroup)
262 (when (if (symbolp (car opgroup))
263 (when (funcall opgroup-filter (car opgroup))
264 (setq opgroup (cdr opgroup))
267 (c--mapcan (lambda (op)
268 (when (funcall op-filter op)
269 (let ((res (funcall xlate op)))
270 (if (listp res) res (list res)))))
276 ;;; Various mode specific values that aren't language related.
278 (c-lang-defconst c-mode-menu
279 ;; The definition for the mode menu. The menu title is prepended to
280 ;; this before it's fed to `easy-menu-define'.
281 t `(["Comment Out Region" comment-region
282 (c-fn-region-is-active-p)]
283 ["Uncomment Region" (comment-region (region-beginning)
285 (c-fn-region-is-active-p)]
286 ["Indent Expression" c-indent-exp
287 (memq (char-after) '(?\( ?\[ ?\{))]
288 ["Indent Line or Region" c-indent-line-or-region t]
289 ["Fill Comment Paragraph" c-fill-paragraph t]
291 ["Backward Statement" c-beginning-of-statement t]
292 ["Forward Statement" c-end-of-statement t]
293 ,@(when (c-lang-const c-opt-cpp-prefix)
294 ;; Only applicable if there's a cpp preprocessor.
295 `(["Up Conditional" c-up-conditional t]
296 ["Backward Conditional" c-backward-conditional t]
297 ["Forward Conditional" c-forward-conditional t]
299 ["Macro Expand Region" c-macro-expand
300 (c-fn-region-is-active-p)]
301 ["Backslashify" c-backslash-region
302 (c-fn-region-is-active-p)]))
305 ["Set Style..." c-set-style t]
306 ["Show Current Style Name" (message
310 ["Guess Style from this Buffer" c-guess-buffer-no-install t]
311 ["Install the Last Guessed Style..." c-guess-install
312 (and c-guess-guessed-offsets-alist
313 c-guess-guessed-basic-offset) ]
314 ["View the Last Guessed Style" c-guess-view
315 (and c-guess-guessed-offsets-alist
316 c-guess-guessed-basic-offset) ])
319 ["Syntactic indentation" c-toggle-syntactic-indentation
320 :style toggle :selected c-syntactic-indentation]
321 ["Electric mode" c-toggle-electric-state
322 :style toggle :selected c-electric-flag]
323 ["Auto newline" c-toggle-auto-newline
324 :style toggle :selected c-auto-newline]
325 ["Hungry delete" c-toggle-hungry-state
326 :style toggle :selected c-hungry-delete-key]
327 ["Subword mode" c-subword-mode
328 :style toggle :selected (and (boundp 'c-subword-mode)
334 (defun c-populate-syntax-table (table)
335 "Populate the given syntax table as necessary for a C-like language.
336 This includes setting \\=' and \" as string delimiters, and setting up
337 the comment syntax to handle both line style \"//\" and block style
338 \"/*\" \"*/\" comments."
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 ?< "." table)
347 (modify-syntax-entry ?> "." table)
348 (modify-syntax-entry ?& "." table)
349 (modify-syntax-entry ?| "." table)
350 (modify-syntax-entry ?\' "\"" table)
351 (modify-syntax-entry ?\240 "." table)
353 ;; Set up block and line oriented comments. The new C
354 ;; standard mandates both comment styles even in C, so since
355 ;; all languages now require dual comments, we make this the
359 ((memq '8-bit c-emacs-features)
360 (modify-syntax-entry ?/ ". 1456" table)
361 (modify-syntax-entry ?* ". 23" table))
363 ((memq '1-bit c-emacs-features)
364 (modify-syntax-entry ?/ ". 124b" table)
365 (modify-syntax-entry ?* ". 23" table))
367 (t (error "CC Mode is incompatible with this version of Emacs")))
369 (modify-syntax-entry ?\n "> b" table)
370 ;; Give CR the same syntax as newline, for selective-display
371 (modify-syntax-entry ?\^m "> b" table))
373 (c-lang-defconst c-make-mode-syntax-table
374 "Functions that generates the mode specific syntax tables.
375 The syntax tables aren't stored directly since they're quite large."
377 (let ((table (make-syntax-table)))
378 (c-populate-syntax-table table)
379 ;; Mode specific syntaxes.
380 ,(cond ((or (c-major-mode-is 'objc-mode) (c-major-mode-is 'java-mode))
381 ;; Let '@' be part of symbols in ObjC to cope with
382 ;; its compiler directives as single keyword tokens.
383 ;; This is then necessary since it's assumed that
384 ;; every keyword is a single symbol.
385 `(modify-syntax-entry ?@ "_" table))
386 ((c-major-mode-is 'pike-mode)
387 `(modify-syntax-entry ?@ "." table)))
390 (c-lang-defconst c-mode-syntax-table
391 ;; The syntax tables in evaluated form. Only used temporarily when
392 ;; the constants in this file are evaluated.
393 t (funcall (c-lang-const c-make-mode-syntax-table)))
395 (c-lang-defconst c++-make-template-syntax-table
396 ;; A variant of `c++-mode-syntax-table' that defines `<' and `>' as
397 ;; parenthesis characters. Used temporarily when template argument
398 ;; lists are parsed. Note that this encourages incorrect parsing of
399 ;; templates since they might contain normal operators that uses the
400 ;; '<' and '>' characters. Therefore this syntax table might go
401 ;; away when CC Mode handles templates correctly everywhere. WHILE
402 ;; THIS SYNTAX TABLE IS CURRENT, `c-parse-state' MUST _NOT_ BE
405 (java c++) `(lambda ()
406 (let ((table (funcall ,(c-lang-const c-make-mode-syntax-table))))
407 (modify-syntax-entry ?< "(>" table)
408 (modify-syntax-entry ?> ")<" table)
410 (c-lang-defvar c++-template-syntax-table
411 (and (c-lang-const c++-make-template-syntax-table)
412 ;; The next eval remove a superfluous ' from '(lambda. This
413 ;; gets rid of compilation warnings.
414 (funcall (eval (c-lang-const c++-make-template-syntax-table)))))
416 (c-lang-defconst c-make-no-parens-syntax-table
417 ;; A variant of the standard syntax table which is used to find matching
418 ;; "<"s and ">"s which have been marked as parens using syntax table
419 ;; properties. The other paren characters (e.g. "{", ")" "]") are given a
420 ;; non-paren syntax here. so that the list commands will work on "< ... >"
421 ;; even when there's unbalanced other parens inside them.
423 ;; This variable is nil for languages which don't have template stuff.
424 t (if (c-lang-const c-recognize-<>-arglists)
426 ;(if (c-lang-const c-recognize-<>-arglists)
427 (let ((table (funcall ,(c-lang-const c-make-mode-syntax-table))))
428 (modify-syntax-entry ?\( "." table)
429 (modify-syntax-entry ?\) "." table)
430 (modify-syntax-entry ?\[ "." table)
431 (modify-syntax-entry ?\] "." table)
432 (modify-syntax-entry ?\{ "." table)
433 (modify-syntax-entry ?\} "." table)
435 (c-lang-defvar c-no-parens-syntax-table
436 (and (c-lang-const c-make-no-parens-syntax-table)
437 ;; See comment in `c++template-syntax-table' about the next `eval'.
438 (funcall (eval (c-lang-const c-make-no-parens-syntax-table)))))
440 (c-lang-defconst c-identifier-syntax-modifications
441 "A list that describes the modifications that should be done to the
442 mode syntax table to get a syntax table that matches all identifiers
443 and keywords as words.
445 The list is just like the one used in `font-lock-defaults': Each
446 element is a cons where the car is the character to modify and the cdr
447 the new syntax, as accepted by `modify-syntax-entry'."
448 ;; The $ character is not allowed in most languages (one exception
449 ;; is Java which allows it for legacy reasons) but we still classify
450 ;; it as an identifier character since it's often used in various
451 ;; machine generated identifiers.
452 t '((?_ . "w") (?$ . "w"))
453 (objc java) (append '((?@ . "w"))
454 (c-lang-const c-identifier-syntax-modifications))
456 (c-lang-defvar c-identifier-syntax-modifications
457 (c-lang-const c-identifier-syntax-modifications))
459 (c-lang-defvar c-identifier-syntax-table
460 (let ((table (copy-syntax-table (c-mode-var "mode-syntax-table")))
461 (mods c-identifier-syntax-modifications)
466 (modify-syntax-entry (car mod) (cdr mod) table))
468 "Syntax table built on the mode syntax table but additionally
469 classifies symbol constituents like `_' and `$' as word constituents,
470 so that all identifiers are recognized as words.")
472 (c-lang-defconst c-get-state-before-change-functions
473 ;; For documentation see the following c-lang-defvar of the same name.
474 ;; The value here may be a list of functions or a single function.
476 c++ '(c-extend-region-for-CPP
477 c-before-change-check-<>-operators
478 c-invalidate-macro-cache)
479 (c objc) '(c-extend-region-for-CPP c-invalidate-macro-cache)
480 ;; java 'c-before-change-check-<>-operators
481 awk 'c-awk-record-region-clear-NL)
482 (c-lang-defvar c-get-state-before-change-functions
483 (let ((fs (c-lang-const c-get-state-before-change-functions)))
487 "If non-nil, a list of functions called from c-before-change-hook.
488 Typically these will record enough state to allow
489 `c-before-font-lock-function' to extend the region to fontify,
490 and may do such things as removing text-properties which must be
493 These functions will be run in the order given. Each of them
494 takes 2 parameters, the BEG and END supplied to every
495 before-change function; on entry, the buffer will have been
496 widened and match-data will have been saved; point is undefined
497 on both entry and exit; the return value is ignored.
499 The functions are called even when font locking isn't enabled.
501 When the mode is initialized, the functions are called with
502 parameters \(point-min) and \(point-max).")
504 (c-lang-defconst c-before-font-lock-functions
505 ;; For documentation see the following c-lang-defvar of the same name.
506 ;; The value here may be a list of functions or a single function.
507 t 'c-change-expand-fl-region
508 (c objc) '(c-neutralize-syntax-in-and-mark-CPP
509 c-change-expand-fl-region)
510 c++ '(c-neutralize-syntax-in-and-mark-CPP
511 c-restore-<>-properties
512 c-change-expand-fl-region)
513 java '(c-restore-<>-properties
514 c-change-expand-fl-region)
515 awk 'c-awk-extend-and-syntax-tablify-region)
516 (c-lang-defvar c-before-font-lock-functions
517 (let ((fs (c-lang-const c-before-font-lock-functions)))
521 "If non-nil, a list of functions called just before font locking.
522 Typically they will extend the region about to be fontified \(see
523 below) and will set `syntax-table' text properties on the region.
525 These functions will be run in the order given. Each of them
526 takes 3 parameters, the BEG, END, and OLD-LEN supplied to every
527 after-change function; point is undefined on both entry and exit;
528 on entry, the buffer will have been widened and match-data will
529 have been saved; the return value is ignored.
531 The functions may extend the region to be fontified by setting the
532 buffer local variables c-new-BEG and c-new-END.
534 The functions are called even when font locking is disabled.
536 When the mode is initialized, these functions are called with
537 parameters \(point-min), \(point-max) and <buffer size>.")
539 (c-lang-defconst c-before-context-fontification-functions
540 t 'c-context-expand-fl-region
542 ;; For documentation see the following c-lang-defvar of the same name.
543 ;; The value here may be a list of functions or a single function.
544 (c-lang-defvar c-before-context-fontification-functions
545 (let ((fs (c-lang-const c-before-context-fontification-functions)))
549 "If non-nil, a list of functions called just before context (or
550 other non-change) fontification is done. Typically they will
553 These functions will be run in the order given. Each of them
554 takes 2 parameters, the BEG and END of the region to be
555 fontified. Point is undefined on both entry and exit. On entry,
556 the buffer will have been widened and match-data will have been
557 saved; the return value is a cons of the adjusted
558 region, (NEW-BEG . NEW-END).")
561 ;;; Syntactic analysis ("virtual semicolons") for line-oriented languages (AWK).
562 (c-lang-defconst c-at-vsemi-p-fn
563 "Contains a function \"Is there a virtual semicolon at POS or point?\".
564 Such a function takes one optional parameter, a buffer position (defaults to
565 point), and returns nil or t. This variable contains nil for languages which
566 don't have EOL terminated statements. "
568 (c c++ objc) 'c-at-macro-vsemi-p
569 awk 'c-awk-at-vsemi-p)
570 (c-lang-defvar c-at-vsemi-p-fn (c-lang-const c-at-vsemi-p-fn))
572 (c-lang-defconst c-vsemi-status-unknown-p-fn
573 "Contains a function \"are we unsure whether there is a virtual semicolon on this line?\".
574 The (admittedly kludgy) purpose of such a function is to prevent an infinite
575 recursion in c-beginning-of-statement-1 when point starts at a `while' token.
576 The function MUST NOT UNDER ANY CIRCUMSTANCES call c-beginning-of-statement-1,
577 even indirectly. This variable contains nil for languages which don't have
578 EOL terminated statements."
580 (c c++ objc) 'c-macro-vsemi-status-unknown-p
581 awk 'c-awk-vsemi-status-unknown-p)
582 (c-lang-defvar c-vsemi-status-unknown-p-fn
583 (c-lang-const c-vsemi-status-unknown-p-fn))
586 ;;; Lexer-level syntax (identifiers, tokens etc).
588 (c-lang-defconst c-has-bitfields
589 "Whether the language has bitfield declarations."
592 (c-lang-defvar c-has-bitfields (c-lang-const c-has-bitfields))
594 (c-lang-defconst c-modified-constant
595 "Regexp that matches a “modified” constant literal such as \"L\\='a\\='\",
596 a “long character”. In particular, this recognizes forms of constant
597 which `c-backward-sexp' needs to be called twice to move backwards over."
599 (c c++ objc) "L'\\([^\\'\t\f\n\r]\\|\\\\.\\)'")
600 ;; FIXME!!! Extend this to cover strings, if needed. 2008-04-11
601 (c-lang-defvar c-modified-constant (c-lang-const c-modified-constant))
603 (c-lang-defconst c-symbol-start
604 "Regexp that matches the start of a symbol, i.e. any identifier or
605 keyword. It's unspecified how far it matches. Does not contain a \\|
606 operator at the top level."
607 t (concat "[" c-alpha "_]")
608 java (concat "[" c-alpha "_@]")
609 objc (concat "[" c-alpha "_@]")
610 pike (concat "[" c-alpha "_`]"))
611 (c-lang-defvar c-symbol-start (c-lang-const c-symbol-start))
613 (c-lang-defconst c-symbol-chars
614 "Set of characters that can be part of a symbol.
615 This is of the form that fits inside [ ] in a regexp."
616 ;; Pike note: With the backquote identifiers this would include most
617 ;; operator chars too, but they are handled with other means instead.
618 t (concat c-alnum "_$")
619 objc (concat c-alnum "_$@"))
620 (c-lang-defvar c-symbol-chars (c-lang-const c-symbol-chars))
622 (c-lang-defconst c-symbol-key
623 "Regexp matching identifiers and keywords (with submatch 0). Assumed
624 to match if `c-symbol-start' matches on the same position."
625 t (concat (c-lang-const c-symbol-start)
626 "[" (c-lang-const c-symbol-chars) "]*")
628 ;; Use the value from C here since the operator backquote is
629 ;; covered by the other alternative.
630 (c-lang-const c-symbol-key c)
632 (c-make-keywords-re nil
633 (c-lang-const c-overloadable-operators))))
634 (c-lang-defvar c-symbol-key (c-lang-const c-symbol-key))
636 (c-lang-defconst c-symbol-key-depth
637 ;; Number of regexp grouping parens in `c-symbol-key'.
638 t (regexp-opt-depth (c-lang-const c-symbol-key)))
640 (c-lang-defconst c-nonsymbol-chars
641 "This is the set of chars that can't be part of a symbol, i.e. the
642 negation of `c-symbol-chars'."
643 t (concat "^" (c-lang-const c-symbol-chars)))
644 (c-lang-defvar c-nonsymbol-chars (c-lang-const c-nonsymbol-chars))
646 (c-lang-defconst c-nonsymbol-key
647 "Regexp that matches any character that can't be part of a symbol.
648 It's usually appended to other regexps to avoid matching a prefix.
649 It's assumed to not contain any submatchers."
650 ;; The same thing regarding Unicode identifiers applies here as to
652 t (concat "[" (c-lang-const c-nonsymbol-chars) "]"))
654 (c-lang-defconst c-identifier-ops
655 "The operators that make up fully qualified identifiers. nil in
656 languages that don't have such things. See `c-operators' for a
657 description of the format. Binary operators can concatenate symbols,
658 e.g. \"::\" in \"A::B::C\". Prefix operators can precede identifiers,
659 e.g. \"~\" in \"~A::B\". Other types of operators aren't supported.
661 This value is by default merged into `c-operators'."
663 c++ '((prefix "~" "??-" "compl")
666 ;; Java has "." to concatenate identifiers but it's also used for
667 ;; normal indexing. There's special code in the Java font lock
668 ;; rules to fontify qualified identifiers based on the standard
669 ;; naming conventions. We still define "." here to make
670 ;; `c-forward-name' move over as long names as possible which is
671 ;; necessary to e.g. handle throws clauses correctly.
672 java '((left-assoc "."))
673 idl '((left-assoc "::")
675 pike '((left-assoc "::")
679 (c-lang-defconst c-opt-identifier-concat-key
680 ;; Appendable adorned regexp matching the operators that join
681 ;; symbols to fully qualified identifiers, or nil in languages that
682 ;; don't have such things.
684 ;; This was a docstring constant in 5.30. It still works but is now
685 ;; considered internal - change `c-identifier-ops' instead.
686 t (let ((ops (c-filter-ops (c-lang-const c-identifier-ops)
687 '(left-assoc right-assoc)
690 (c-make-keywords-re 'appendable ops))))
691 (c-lang-defvar c-opt-identifier-concat-key
692 (c-lang-const c-opt-identifier-concat-key)
695 (c-lang-defconst c-opt-identifier-concat-key-depth
696 ;; Number of regexp grouping parens in `c-opt-identifier-concat-key'.
697 t (regexp-opt-depth (c-lang-const c-opt-identifier-concat-key)))
699 (c-lang-defconst c-opt-identifier-prefix-key
700 ;; Appendable adorned regexp matching operators that might precede
701 ;; an identifier and that are part of the identifier in that case.
702 ;; nil in languages without such things.
703 t (let ((ops (c-filter-ops (c-lang-const c-identifier-ops)
707 (c-make-keywords-re 'appendable ops))))
709 (c-lang-defconst c-after-id-concat-ops
710 "Operators that can occur after a binary operator on `c-identifier-ops'
711 in identifiers. nil in languages that don't have such things.
713 Operators here should also have appropriate entries in `c-operators' -
714 it's not taken care of by default."
716 ;; '~' for destructors in C++, '*' for member pointers.
718 ;; In Java we recognize '*' to deal with "foo.bar.*" that can occur
719 ;; in import declarations. (This will also match bogus things like
720 ;; "foo.*bar" but we don't bother.)
723 (c-lang-defconst c-opt-after-id-concat-key
724 ;; Regexp that must match the token after
725 ;; `c-opt-identifier-concat-key' for it to be considered an
726 ;; identifier concatenation operator (which e.g. causes the
727 ;; preceding identifier to be fontified as a reference). Assumed to
728 ;; be a string if `c-opt-identifier-concat-key' is.
730 ;; This was a docstring constant in 5.30. It still works but is now
731 ;; considered internal - change `c-after-id-concat-ops' instead.
732 t (concat (c-lang-const c-symbol-start)
733 (if (c-lang-const c-after-id-concat-ops)
734 (concat "\\|" (c-make-keywords-re 'appendable
735 (c-lang-const c-after-id-concat-ops)))
738 (c-lang-defconst c-identifier-start
739 "Regexp that matches the start of an (optionally qualified) identifier.
740 It should also match all keywords. It's unspecified how far it
742 t (concat (c-lang-const c-symbol-start)
743 (if (c-lang-const c-opt-identifier-prefix-key)
745 (c-lang-const c-opt-identifier-prefix-key))
747 (c-lang-defvar c-identifier-start (c-lang-const c-identifier-start))
749 (c-lang-defconst c-identifier-key
750 "Regexp matching a fully qualified identifier, like \"A::B::c\" in
751 C++. It does not recognize the full range of syntactic whitespace
752 between the tokens; `c-forward-name' has to be used for that. It
753 should also not match identifiers containing parenthesis groupings,
754 e.g. identifiers with template arguments such as \"A<X,Y>\" in C++."
755 ;; This regexp is more complex than strictly necessary to ensure
756 ;; that it can be matched with a minimum of backtracking.
757 t (concat (if (c-lang-const c-opt-identifier-prefix-key)
760 (c-lang-const c-opt-identifier-prefix-key)
761 (c-lang-const c-simple-ws) "*"
764 "\\(" (c-lang-const c-symbol-key) "\\)"
765 (if (c-lang-const c-opt-identifier-concat-key)
768 (c-lang-const c-simple-ws) "*"
769 (c-lang-const c-opt-identifier-concat-key)
770 (c-lang-const c-simple-ws) "*"
771 (if (c-lang-const c-after-id-concat-ops)
774 (c-make-keywords-re 'appendable
775 (c-lang-const c-after-id-concat-ops))
777 ;; For flexibility, consider the symbol match
778 ;; optional if we've hit a
779 ;; `c-after-id-concat-ops' operator. This is
780 ;; also necessary to handle the "*" that can
781 ;; end import declaration identifiers in Java.
783 (c-lang-const c-simple-ws) "*"
784 "\\(" (c-lang-const c-symbol-key) "\\)"
787 "\\(" (c-lang-const c-symbol-key) "\\)"
789 (concat "\\(" (c-lang-const c-symbol-key) "\\)"))
792 (c-lang-defvar c-identifier-key (c-lang-const c-identifier-key))
794 (c-lang-defconst c-identifier-last-sym-match
795 ;; This was a docstring constant in 5.30 but it's no longer used.
796 ;; It's only kept to avoid breaking third party code.
798 ;; Used to identify the submatch in `c-identifier-key' that
799 ;; surrounds the last symbol in the qualified identifier. It's a
800 ;; list of submatch numbers, of which the first that has a match is
801 ;; taken. It's assumed that at least one does when the regexp has
805 (c-lang-defconst c-string-escaped-newlines
806 "Set if the language support backslash escaped newlines inside string
810 (c-lang-defvar c-string-escaped-newlines
811 (c-lang-const c-string-escaped-newlines))
813 (c-lang-defconst c-multiline-string-start-char
814 "Set if the language supports multiline string literals without escaped
815 newlines. If t, all string literals are multiline. If a character,
816 only literals where the open quote is immediately preceded by that
817 literal are multiline."
820 (c-lang-defvar c-multiline-string-start-char
821 (c-lang-const c-multiline-string-start-char))
823 (c-lang-defconst c-opt-cpp-symbol
824 "The symbol which starts preprocessor constructs when in the margin."
827 (c-lang-defvar c-opt-cpp-symbol (c-lang-const c-opt-cpp-symbol))
829 (c-lang-defconst c-opt-cpp-prefix
830 "Regexp matching the prefix of a cpp directive in the languages that
831 normally use that macro preprocessor. Tested at bol or at boi.
832 Assumed to not contain any submatches or \\| operators."
833 ;; TODO (ACM, 2005-04-01). Amend the following to recognize escaped NLs;
834 ;; amend all uses of c-opt-cpp-prefix which count regexp-depth.
837 (c-lang-defvar c-opt-cpp-prefix (c-lang-const c-opt-cpp-prefix))
839 (c-lang-defconst c-anchored-cpp-prefix
840 "Regexp matching the prefix of a cpp directive anchored to BOL,
841 in the languages that have a macro preprocessor."
842 t "^\\s *\\(#\\)\\s *"
844 (c-lang-defvar c-anchored-cpp-prefix (c-lang-const c-anchored-cpp-prefix))
846 (c-lang-defconst c-opt-cpp-start
847 "Regexp matching the prefix of a cpp directive including the directive
848 name, or nil in languages without preprocessor support. The first
849 submatch surrounds the directive name."
850 t (if (c-lang-const c-opt-cpp-prefix)
851 (concat (c-lang-const c-opt-cpp-prefix)
852 "\\([" c-alnum "]+\\)"))
853 ;; Pike, being a scripting language, recognizes hash-bangs too.
854 pike (concat (c-lang-const c-opt-cpp-prefix)
855 "\\([" c-alnum "]+\\|!\\)"))
856 (c-lang-defvar c-opt-cpp-start (c-lang-const c-opt-cpp-start))
858 (c-lang-defconst c-cpp-message-directives
859 "List of cpp directives (without the prefix) that are followed by a
861 t (if (c-lang-const c-opt-cpp-prefix)
863 (c c++ objc pike) '("error" "warning"))
865 (c-lang-defconst c-cpp-include-directives
866 "List of cpp directives (without the prefix) that are followed by a
867 file name in angle brackets or quotes."
868 t (if (c-lang-const c-opt-cpp-prefix)
870 objc '("include" "import"))
872 (c-lang-defconst c-opt-cpp-macro-define
873 "Cpp directive (without the prefix) that is followed by a macro
874 definition, or nil if the language doesn't have any."
875 t (if (c-lang-const c-opt-cpp-prefix)
877 (c-lang-defvar c-opt-cpp-macro-define
878 (c-lang-const c-opt-cpp-macro-define))
880 (c-lang-defconst c-opt-cpp-macro-define-start
881 ;; Regexp matching everything up to the macro body of a cpp define, or the
882 ;; end of the logical line if there is none. Submatch 1 is the name of the
883 ;; macro. Set if c-opt-cpp-macro-define is.
884 t (if (c-lang-const c-opt-cpp-macro-define)
885 (concat (c-lang-const c-opt-cpp-prefix)
886 (c-lang-const c-opt-cpp-macro-define)
887 "[ \t]+\\(\\(\\sw\\|_\\)+\\)\\(([^)]*)\\)?"
889 "\\([ \t]\\|\\\\\n\\)*")))
890 (c-lang-defvar c-opt-cpp-macro-define-start
891 (c-lang-const c-opt-cpp-macro-define-start))
893 (c-lang-defconst c-opt-cpp-macro-define-id
894 ;; Regexp matching everything up to the end of the identifier defined
896 t (if (c-lang-const c-opt-cpp-macro-define)
897 (concat (c-lang-const c-opt-cpp-prefix) ; #
898 (c-lang-const c-opt-cpp-macro-define) ; define
899 "[ \t]+\\(\\sw\\|_\\)+")))
900 (c-lang-defvar c-opt-cpp-macro-define-id
901 (c-lang-const c-opt-cpp-macro-define-id))
903 (c-lang-defconst c-cpp-expr-directives
904 "List of cpp directives (without the prefix) that are followed by an
906 t (if (c-lang-const c-opt-cpp-prefix)
909 (c-lang-defconst c-cpp-expr-intro-re
910 "Regexp which matches the start of a CPP directive which contains an
911 expression, or nil if there aren't any in the language."
912 t (if (c-lang-const c-cpp-expr-directives)
914 (c-lang-const c-opt-cpp-prefix)
915 (c-make-keywords-re t (c-lang-const c-cpp-expr-directives)))))
916 (c-lang-defvar c-cpp-expr-intro-re
917 (c-lang-const c-cpp-expr-intro-re))
919 (c-lang-defconst c-cpp-expr-functions
920 "List of functions in cpp expressions."
921 t (if (c-lang-const c-opt-cpp-prefix)
923 pike '("defined" "efun" "constant"))
925 (c-lang-defconst c-assignment-operators
926 "List of all assignment operators."
927 t '("=" "*=" "/=" "%=" "+=" "-=" ">>=" "<<=" "&=" "^=" "|=")
928 java (append (c-lang-const c-assignment-operators)
930 c++ (append (c-lang-const c-assignment-operators)
931 '("and_eq" "or_eq" "xor_eq" "??!=" "??'="))
934 (c-lang-defconst c-operators
935 "List describing all operators, along with their precedence and
936 associativity. The order in the list corresponds to the precedence of
937 the operators: The operators in each element are a group with the same
938 precedence, and the group has higher precedence than the groups in all
939 following elements. The car of each element describes the type of the
940 operator group, and the cdr is a list of the operator tokens in it.
941 The operator group types are:
943 `prefix' Unary prefix operators.
944 `postfix' Unary postfix operators.
946 Unary postfix operators if and only if the chars have
948 `left-assoc' Binary left associative operators (i.e. a+b+c means (a+b)+c).
949 `right-assoc' Binary right associative operators (i.e. a=b=c means a=(b=c)).
950 `right-assoc-sequence'
951 Right associative operator that constitutes of a
952 sequence of tokens that separate expressions. All the
953 tokens in the group are in this case taken as
954 describing the sequence in one such operator, and the
955 order between them is therefore significant.
957 Operators containing a character with paren syntax are taken to match
958 with a corresponding open/close paren somewhere else. A postfix
959 operator with close paren syntax is taken to end a postfix expression
960 started somewhere earlier, rather than start a new one at point. Vice
961 versa for prefix operators with open paren syntax.
963 Note that operators like \".\" and \"->\" which in language references
964 often are described as postfix operators are considered binary here,
965 since CC Mode treats every identifier as an expression."
967 ;; There's currently no code in CC Mode that exploits all the info
968 ;; in this variable; precedence, associativity etc are present as a
969 ;; preparation for future work.
971 ;; FIXME!!! C++11's "auto" operators "=" and "->" need to go in here
972 ;; somewhere. 2012-03-24.
975 ,@(when (c-lang-const c-opt-cpp-prefix)
977 ,@(when (c-major-mode-is '(c-mode c++-mode))
980 ,@(when (c-major-mode-is '(c-mode c++-mode))
981 '("%:%:" "??=??=")))))
984 ,@(c-lang-const c-identifier-ops)
985 ,@(cond ((or (c-major-mode-is 'c++-mode) (c-major-mode-is 'java-mode))
986 `((postfix-if-paren "<" ">"))) ; Templates.
987 ((c-major-mode-is 'pike-mode)
988 `((prefix "global" "predef")))
989 ((c-major-mode-is 'java-mode)
990 `((prefix "super"))))
993 ,@(when (c-major-mode-is 'c++-mode)
994 ;; The following need special treatment.
995 `((prefix "dynamic_cast" "static_cast"
996 "reinterpret_cast" "const_cast" "typeid"
999 ,@(unless (c-major-mode-is 'java-mode)
1001 (postfix "++" "--" "[" "]" "(" ")"
1002 ,@(when (c-major-mode-is '(c-mode c++-mode))
1003 '("<:" ":>" "??(" "??)")))
1006 (prefix "++" "--" "+" "-" "!" "~"
1007 ,@(when (c-major-mode-is 'c++-mode) '("not" "compl"))
1008 ,@(when (c-major-mode-is '(c-mode c++-mode))
1009 '("*" "&" "sizeof" "??-"))
1010 ,@(when (c-major-mode-is 'objc-mode)
1011 '("@selector" "@protocol" "@encode"))
1012 ;; The following need special treatment.
1013 ,@(cond ((c-major-mode-is 'c++-mode)
1015 ((c-major-mode-is 'java-mode)
1017 ((c-major-mode-is 'pike-mode)
1018 '("class" "lambda" "catch" "throw" "gauge")))
1020 ,@(when (c-major-mode-is 'pike-mode)
1021 '("[" "]"))) ; Type cast.
1023 ;; Member selection.
1024 ,@(when (c-major-mode-is 'c++-mode)
1025 `((left-assoc ".*" "->*")))
1028 (left-assoc "*" "/" "%")
1031 (left-assoc "+" "-")
1034 (left-assoc "<<" ">>"
1035 ,@(when (c-major-mode-is 'java-mode)
1039 (left-assoc "<" ">" "<=" ">="
1040 ,@(when (c-major-mode-is 'java-mode)
1044 (left-assoc "==" "!="
1045 ,@(when (c-major-mode-is 'c++-mode) '("not_eq")))
1049 ,@(when (c-major-mode-is 'c++-mode) '("bitand")))
1051 ;; Bitwise exclusive or.
1053 ,@(when (c-major-mode-is '(c-mode c++-mode))
1055 ,@(when (c-major-mode-is 'c++-mode) '("xor")))
1059 ,@(when (c-major-mode-is '(c-mode c++-mode))
1061 ,@(when (c-major-mode-is 'c++-mode) '("bitor")))
1065 ,@(when (c-major-mode-is 'c++-mode) '("and")))
1069 ,@(when (c-major-mode-is '(c-mode c++-mode))
1071 ,@(when (c-major-mode-is 'c++-mode) '("or")))
1074 (right-assoc-sequence "?" ":")
1077 (right-assoc ,@(c-lang-const c-assignment-operators))
1080 ,@(when (c-major-mode-is 'c++-mode)
1081 '((prefix "throw")))
1086 ;; IDL got its own definition since it has a much smaller operator
1087 ;; set than the other languages.
1088 idl `(;; Preprocessor.
1092 ,@(c-lang-const c-identifier-ops)
1094 (prefix "+" "-" "~")
1096 (left-assoc "*" "/" "%")
1098 (left-assoc "+" "-")
1100 (left-assoc "<<" ">>")
1108 (c-lang-defconst c-operator-list
1109 ;; The operators as a flat list (without duplicates).
1110 t (c-filter-ops (c-lang-const c-operators) t t))
1112 (c-lang-defconst c-overloadable-operators
1113 "List of the operators that are overloadable, in their \"identifier
1114 form\". See also `c-op-identifier-prefix'."
1116 c++ '("new" "delete" ;; Can be followed by "[]" but we ignore that.
1118 "^" "??'" "xor" "&" "bitand" "|" "??!" "bitor" "~" "??-" "compl"
1119 "!" "=" "<" ">" "+=" "-=" "*=" "/=" "%=" "^="
1120 "??'=" "xor_eq" "&=" "and_eq" "|=" "??!=" "or_eq"
1121 "<<" ">>" ">>=" "<<=" "==" "!=" "not_eq" "<=" ">="
1122 "&&" "and" "||" "??!??!" "or" "++" "--" "," "->*" "->"
1123 "()" "[]" "<::>" "??(??)")
1124 ;; These work like identifiers in Pike.
1125 pike '("`+" "`-" "`&" "`|" "`^" "`<<" "`>>" "`*" "`/" "`%" "`~"
1126 "`==" "`<" "`>" "`!" "`[]" "`[]=" "`->" "`->=" "`()" "``+"
1127 "``-" "``&" "``|" "``^" "``<<" "``>>" "``*" "``/" "``%"
1130 (c-lang-defconst c-overloadable-operators-regexp
1131 ;; Regexp tested after an "operator" token in C++.
1133 c++ (c-make-keywords-re nil (c-lang-const c-overloadable-operators)))
1134 (c-lang-defvar c-overloadable-operators-regexp
1135 (c-lang-const c-overloadable-operators-regexp))
1137 (c-lang-defconst c-opt-op-identifier-prefix
1138 "Regexp matching the token before the ones in
1139 `c-overloadable-operators' when operators are specified in their
1140 \"identifier form\". This typically matches \"operator\" in C++ where
1141 operator functions are specified as e.g. \"operator +\". It's nil in
1142 languages without operator functions or where the complete operator
1143 identifier is listed in `c-overloadable-operators'.
1145 This regexp is assumed to not match any non-operator identifier."
1147 c++ (c-make-keywords-re t '("operator")))
1148 (c-lang-defvar c-opt-op-identifier-prefix
1149 (c-lang-const c-opt-op-identifier-prefix))
1151 ;; Note: the following alias is an old name which was a mis-spelling. It has
1152 ;; been corrected above and throughout cc-engine.el. It will be removed at
1153 ;; some release very shortly in the future. ACM, 2006-04-14.
1154 (defvaralias 'c-opt-op-identitier-prefix 'c-opt-op-identifier-prefix)
1155 (make-obsolete-variable 'c-opt-op-identitier-prefix 'c-opt-op-identifier-prefix
1156 "CC Mode 5.31.4, 2006-04-14")
1158 (c-lang-defconst c-other-op-syntax-tokens
1159 "List of the tokens made up of characters in the punctuation or
1160 parenthesis syntax classes that have uses other than as expression
1162 t '("{" "}" "(" ")" "[" "]" ";" ":" "," "=" "/*" "*/" "//")
1163 (c c++ pike) (append '("#" "##" ; Used by cpp.
1165 (c-lang-const c-other-op-syntax-tokens))
1166 (c c++) (append '("*") (c-lang-const c-other-op-syntax-tokens))
1167 c++ (append '("&" "<%" "%>" "<:" ":>" "%:" "%:%:")
1168 (c-lang-const c-other-op-syntax-tokens))
1169 objc (append '("#" "##" ; Used by cpp.
1171 (c-lang-const c-other-op-syntax-tokens))
1172 idl (append '("#" "##") ; Used by cpp.
1173 (c-lang-const c-other-op-syntax-tokens))
1174 pike (append '("..")
1175 (c-lang-const c-other-op-syntax-tokens)
1176 (c-lang-const c-overloadable-operators))
1177 awk '("{" "}" "(" ")" "[" "]" ";" "," "=" "/"))
1179 (c-lang-defconst c-all-op-syntax-tokens
1180 ;; List of all tokens in the punctuation and parenthesis syntax
1182 t (c--delete-duplicates (append (c-lang-const c-other-op-syntax-tokens)
1183 (c-lang-const c-operator-list))
1184 :test 'string-equal))
1186 (c-lang-defconst c-nonsymbol-token-char-list
1187 ;; List containing all chars not in the word, symbol or
1188 ;; syntactically irrelevant syntax classes, i.e. all punctuation,
1189 ;; parenthesis and string delimiter chars.
1190 t (c-with-syntax-table (c-lang-const c-mode-syntax-table)
1191 ;; Only go through the chars in the printable ASCII range. No
1192 ;; language so far has 8-bit or widestring operators.
1193 (let (list (char 32))
1195 (or (memq (char-syntax char) '(?w ?_ ?< ?> ?\ ))
1196 (setq list (cons (c-int-to-char char) list)))
1197 (setq char (1+ char)))
1200 (c-lang-defconst c-nonsymbol-token-regexp
1201 ;; Regexp matching all tokens in the punctuation and parenthesis
1202 ;; syntax classes. Note that this also matches ".", which can start
1204 t (c-make-keywords-re nil
1205 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1207 "\\`\\(\\s.\\)+\\'")))
1208 (c-lang-defvar c-nonsymbol-token-regexp
1209 (c-lang-const c-nonsymbol-token-regexp))
1211 (c-lang-defconst c-assignment-op-regexp
1212 ;; Regexp matching all assignment operators and only them. The
1213 ;; beginning of the first submatch is used to detect the end of the
1214 ;; token, along with the end of the whole match.
1215 t (if (c-lang-const c-assignment-operators)
1217 ;; Need special case for "=" since it's a prefix of "==".
1220 (c-make-keywords-re nil
1221 (c--set-difference (c-lang-const c-assignment-operators)
1223 :test 'string-equal)))
1225 (c-lang-defvar c-assignment-op-regexp
1226 (c-lang-const c-assignment-op-regexp))
1228 (c-lang-defconst c-<>-multichar-token-regexp
1229 ;; Regexp matching all tokens containing "<" or ">" which are longer
1231 t (c-make-keywords-re nil
1232 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1235 (c-lang-defvar c-<>-multichar-token-regexp
1236 (c-lang-const c-<>-multichar-token-regexp))
1238 (c-lang-defconst c-<-op-cont-regexp
1239 ;; Regexp matching the second and subsequent characters of all
1240 ;; multicharacter tokens that begin with "<".
1241 t (c-make-keywords-re nil
1242 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1245 (lambda (op) (substring op 1)))))
1246 (c-lang-defvar c-<-op-cont-regexp (c-lang-const c-<-op-cont-regexp))
1248 (c-lang-defconst c->-op-cont-tokens
1249 ;; A list of second and subsequent characters of all multicharacter tokens
1250 ;; that begin with ">".
1251 t (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1254 (lambda (op) (substring op 1)))
1255 java (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1257 "\\`>[^>]\\|\\`>>[^>]"
1258 (lambda (op) (substring op 1))))
1260 (c-lang-defconst c->-op-cont-regexp
1261 ;; Regexp matching the second and subsequent characters of all
1262 ;; multicharacter tokens that begin with ">".
1263 t (c-make-keywords-re nil (c-lang-const c->-op-cont-tokens)))
1264 (c-lang-defvar c->-op-cont-regexp (c-lang-const c->-op-cont-regexp))
1266 (c-lang-defconst c->-op-without->-cont-regexp
1267 ;; Regexp matching the second and subsequent characters of all
1268 ;; multicharacter tokens that begin with ">" except for those beginning with
1270 t (c-make-keywords-re nil
1272 (c-lang-const c->-op-cont-tokens)
1273 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1276 (lambda (op) (substring op 1)))
1277 :test 'string-equal)))
1278 (c-lang-defvar c->-op-without->-cont-regexp
1279 (c-lang-const c->-op-without->-cont-regexp))
1281 (c-lang-defconst c-multichar->-op-not->>-regexp
1282 ;; Regexp matching multichar tokens containing ">", except ">>"
1283 t (c-make-keywords-re nil
1285 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1288 (c-lang-defvar c-multichar->-op-not->>-regexp
1289 (c-lang-const c-multichar->-op-not->>-regexp))
1291 (c-lang-defconst c-stmt-delim-chars
1292 ;; The characters that should be considered to bound statements. To
1293 ;; optimize `c-crosses-statement-barrier-p' somewhat, it's assumed to
1294 ;; begin with "^" to negate the set. If ? : operators should be
1295 ;; detected then the string must end with "?:".
1297 (c-lang-defvar c-stmt-delim-chars (c-lang-const c-stmt-delim-chars))
1299 (c-lang-defconst c-stmt-delim-chars-with-comma
1300 ;; Variant of `c-stmt-delim-chars' that additionally contains ','.
1302 (c-lang-defvar c-stmt-delim-chars-with-comma
1303 (c-lang-const c-stmt-delim-chars-with-comma))
1305 (c-lang-defconst c-auto-ops
1306 ;; Ops which signal C++11's new auto uses.
1309 (c-lang-defconst c-auto-ops-re
1310 t (c-make-keywords-re nil (c-lang-const c-auto-ops)))
1311 (c-lang-defvar c-auto-ops-re (c-lang-const c-auto-ops-re))
1313 (c-lang-defconst c-haskell-op
1314 ;; Op used in the new C++11 auto function definition, indicating type.
1317 (c-lang-defconst c-haskell-op-re
1318 t (c-make-keywords-re nil (c-lang-const c-haskell-op)))
1319 (c-lang-defvar c-haskell-op-re (c-lang-const c-haskell-op-re))
1321 ;;; Syntactic whitespace.
1323 (c-lang-defconst c-simple-ws
1324 "Regexp matching an ordinary whitespace character.
1325 Does not contain a \\| operator at the top level."
1326 ;; "\\s " is not enough since it doesn't match line breaks.
1327 t "\\(\\s \\|[\n\r]\\)")
1329 (c-lang-defconst c-simple-ws-depth
1330 ;; Number of regexp grouping parens in `c-simple-ws'.
1331 t (regexp-opt-depth (c-lang-const c-simple-ws)))
1333 (c-lang-defconst c-line-comment-starter
1334 "String that starts line comments, or nil if such don't exist.
1335 Line comments are always terminated by newlines. At least one of
1336 `c-block-comment-starter' and this one is assumed to be set.
1338 Note that it's currently not enough to set this to support a new
1339 comment style. Other stuff like the syntax table must also be set up
1343 (c-lang-defvar c-line-comment-starter (c-lang-const c-line-comment-starter))
1345 (c-lang-defconst c-block-comment-starter
1346 "String that starts block comments, or nil if such don't exist.
1347 Block comments are ended by `c-block-comment-ender', which is assumed
1348 to be set if this is. At least one of `c-line-comment-starter' and
1349 this one is assumed to be set.
1351 Note that it's currently not enough to set this to support a new
1352 comment style. Other stuff like the syntax table must also be set up
1357 (c-lang-defconst c-block-comment-ender
1358 "String that ends block comments, or nil if such don't exist.
1360 Note that it's currently not enough to set this to support a new
1361 comment style. Other stuff like the syntax table must also be set up
1366 (c-lang-defconst c-comment-start-regexp
1367 ;; Regexp to match the start of any type of comment.
1368 t (let ((re (c-make-keywords-re nil
1369 (list (c-lang-const c-line-comment-starter)
1370 (c-lang-const c-block-comment-starter)))))
1371 (if (memq 'gen-comment-delim c-emacs-features)
1372 (concat re "\\|\\s!")
1374 (c-lang-defvar c-comment-start-regexp (c-lang-const c-comment-start-regexp))
1376 (c-lang-defconst c-block-comment-start-regexp
1377 ;; Regexp which matches the start of a block comment (if such exists in the
1379 t (if (c-lang-const c-block-comment-starter)
1380 (regexp-quote (c-lang-const c-block-comment-starter))
1382 (c-lang-defvar c-block-comment-start-regexp
1383 (c-lang-const c-block-comment-start-regexp))
1385 (c-lang-defconst c-line-comment-start-regexp
1386 ;; Regexp which matches the start of a line comment (if such exists in the
1387 ;; language; it does in all 7 CC Mode languages).
1388 t (if (c-lang-const c-line-comment-starter)
1389 (regexp-quote (c-lang-const c-line-comment-starter))
1391 (c-lang-defvar c-line-comment-start-regexp
1392 (c-lang-const c-line-comment-start-regexp))
1394 (c-lang-defconst c-literal-start-regexp
1395 ;; Regexp to match the start of comments and string literals.
1396 t (concat (c-lang-const c-comment-start-regexp)
1398 (if (memq 'gen-string-delim c-emacs-features)
1401 (c-lang-defvar c-literal-start-regexp (c-lang-const c-literal-start-regexp))
1403 (c-lang-defconst c-doc-comment-start-regexp
1404 "Regexp to match the start of documentation comments."
1406 ;; From font-lock.el: `doxygen' uses /*! while others use /**.
1407 (c c++ objc) "/\\*[*!]"
1410 (c-lang-defvar c-doc-comment-start-regexp
1411 (c-lang-const c-doc-comment-start-regexp))
1413 (c-lang-defconst comment-start
1414 "String that starts comments inserted with M-; etc.
1415 `comment-start' is initialized from this."
1416 ;; Default: Prefer line comments to block comments, and pad with a space.
1417 t (concat (or (c-lang-const c-line-comment-starter)
1418 (c-lang-const c-block-comment-starter))
1420 ;; In C we still default to the block comment style since line
1421 ;; comments aren't entirely portable.
1423 (c-lang-setvar comment-start (c-lang-const comment-start))
1425 (c-lang-defconst comment-end
1426 "String that ends comments inserted with M-; etc.
1427 `comment-end' is initialized from this."
1428 ;; Default: Use block comment style if comment-start uses block
1429 ;; comments, and pad with a space in that case.
1430 t (if (string-match (concat "\\`\\("
1431 (c-lang-const c-block-comment-start-regexp)
1433 (c-lang-const comment-start))
1434 (concat " " (c-lang-const c-block-comment-ender))
1436 (c-lang-setvar comment-end (c-lang-const comment-end))
1438 (c-lang-defconst comment-start-skip
1439 "Regexp to match the start of a comment plus everything up to its body.
1440 `comment-start-skip' is initialized from this."
1441 ;; Default: Allow the last char of the comment starter(s) to be
1442 ;; repeated, then allow any amount of horizontal whitespace.
1445 (mapcar (lambda (cs)
1447 (concat (regexp-quote cs) "+")))
1448 (list (c-lang-const c-line-comment-starter)
1449 (c-lang-const c-block-comment-starter)))
1452 (c-lang-setvar comment-start-skip (c-lang-const comment-start-skip))
1454 (c-lang-defconst comment-end-can-be-escaped
1455 "When non-nil, escaped EOLs inside comments are valid.
1456 This works in Emacs >= 25.1."
1459 (c-lang-setvar comment-end-can-be-escaped
1460 (c-lang-const comment-end-can-be-escaped))
1462 (c-lang-defconst c-syntactic-ws-start
1463 ;; Regexp matching any sequence that can start syntactic whitespace.
1464 ;; The only uncertain case is '#' when there are cpp directives.
1466 (c-make-keywords-re nil
1467 (append (list (c-lang-const c-line-comment-starter)
1468 (c-lang-const c-block-comment-starter)
1469 (when (c-lang-const c-opt-cpp-prefix)
1473 (when (memq 'gen-comment-delim c-emacs-features)
1475 (c-lang-defvar c-syntactic-ws-start (c-lang-const c-syntactic-ws-start))
1477 (c-lang-defconst c-syntactic-ws-end
1478 ;; Regexp matching any single character that might end syntactic whitespace.
1480 (c-make-keywords-re nil
1481 (append (when (c-lang-const c-block-comment-ender)
1484 (elt (c-lang-const c-block-comment-ender)
1486 (c-lang-const c-block-comment-ender)))))))
1488 (when (memq 'gen-comment-delim c-emacs-features)
1490 (c-lang-defvar c-syntactic-ws-end (c-lang-const c-syntactic-ws-end))
1492 (c-lang-defconst c-unterminated-block-comment-regexp
1493 ;; Regexp matching an unterminated block comment that doesn't
1494 ;; contain line breaks, or nil in languages without block comments.
1495 ;; Does not contain a \| operator at the top level.
1496 t (when (c-lang-const c-block-comment-starter)
1498 (regexp-quote (c-lang-const c-block-comment-starter))
1499 ;; It's messy to cook together a regexp that matches anything
1500 ;; but c-block-comment-ender.
1501 (let ((end (c-lang-const c-block-comment-ender)))
1502 (cond ((= (length end) 1)
1503 (concat "[^" end "\n\r]*"))
1505 (concat "[^" (substring end 0 1) "\n\r]*"
1507 (regexp-quote (substring end 0 1)) "+"
1509 ;; The quoting rules inside char classes are silly. :P
1510 (cond ((= (elt end 0) (elt end 1))
1511 (concat (substring end 0 1) "\n\r"))
1512 ((= (elt end 1) ?\])
1513 (concat (substring end 1 2) "\n\r"
1514 (substring end 0 1)))
1516 (concat (substring end 0 1) "\n\r"
1517 (substring end 1 2))))
1519 "[^" (substring end 0 1) "\n\r]*"
1522 (error "Can't handle a block comment ender of length %s"
1525 (c-lang-defconst c-block-comment-regexp
1526 ;; Regexp matching a block comment that doesn't contain line breaks,
1527 ;; or nil in languages without block comments. The reason we don't
1528 ;; allow line breaks is to avoid going very far and risk running out
1529 ;; of regexp stack; this regexp is intended to handle only short
1530 ;; comments that might be put in the middle of limited constructs
1531 ;; like declarations. Does not contain a \| operator at the top
1533 t (when (c-lang-const c-unterminated-block-comment-regexp)
1535 (c-lang-const c-unterminated-block-comment-regexp)
1536 (let ((end (c-lang-const c-block-comment-ender)))
1537 (cond ((= (length end) 1)
1540 (concat (regexp-quote (substring end 0 1)) "+"
1541 (regexp-quote (substring end 1 2))))
1543 (error "Can't handle a block comment ender of length %s"
1546 (c-lang-defconst c-nonwhite-syntactic-ws
1547 ;; Regexp matching a piece of syntactic whitespace that isn't a
1548 ;; sequence of simple whitespace characters. As opposed to
1549 ;; `c-(forward|backward)-syntactic-ws', this doesn't regard cpp
1550 ;; directives as syntactic whitespace.
1551 t (c-concat-separated
1552 (list (when (c-lang-const c-line-comment-starter)
1553 (concat (regexp-quote (c-lang-const c-line-comment-starter))
1555 (c-lang-const c-block-comment-regexp)
1557 (when (memq 'gen-comment-delim c-emacs-features)
1561 (c-lang-defconst c-syntactic-ws
1562 ;; Regexp matching syntactic whitespace, including possibly the
1563 ;; empty string. As opposed to `c-(forward|backward)-syntactic-ws',
1564 ;; this doesn't regard cpp directives as syntactic whitespace. Does
1565 ;; not contain a \| operator at the top level.
1566 t (concat (c-lang-const c-simple-ws) "*"
1568 (concat "\\(" (c-lang-const c-nonwhite-syntactic-ws) "\\)"
1569 (c-lang-const c-simple-ws) "*")
1572 (c-lang-defconst c-syntactic-ws-depth
1573 ;; Number of regexp grouping parens in `c-syntactic-ws'.
1574 t (regexp-opt-depth (c-lang-const c-syntactic-ws)))
1576 (c-lang-defconst c-nonempty-syntactic-ws
1577 ;; Regexp matching syntactic whitespace, which is at least one
1578 ;; character long. As opposed to `c-(forward|backward)-syntactic-ws',
1579 ;; this doesn't regard cpp directives as syntactic whitespace. Does
1580 ;; not contain a \| operator at the top level.
1582 (c-lang-const c-simple-ws)
1584 (c-lang-const c-nonwhite-syntactic-ws)
1587 (c-lang-defconst c-nonempty-syntactic-ws-depth
1588 ;; Number of regexp grouping parens in `c-nonempty-syntactic-ws'.
1589 t (regexp-opt-depth (c-lang-const c-nonempty-syntactic-ws)))
1591 (c-lang-defconst c-single-line-syntactic-ws
1592 ;; Regexp matching syntactic whitespace without any line breaks. As
1593 ;; opposed to `c-(forward|backward)-syntactic-ws', this doesn't
1594 ;; regard cpp directives as syntactic whitespace. Does not contain
1595 ;; a \| operator at the top level.
1596 t (if (c-lang-const c-block-comment-regexp)
1598 (c-lang-const c-block-comment-regexp)
1602 (c-lang-defconst c-single-line-syntactic-ws-depth
1603 ;; Number of regexp grouping parens in `c-single-line-syntactic-ws'.
1604 t (regexp-opt-depth (c-lang-const c-single-line-syntactic-ws)))
1606 (c-lang-defconst c-syntactic-eol
1607 ;; Regexp that matches when there is no syntactically significant
1608 ;; text before eol. Macros are regarded as syntactically
1609 ;; significant text here.
1610 t (concat (c-lang-const c-single-line-syntactic-ws)
1611 ;; Match eol (possibly inside a block comment or preceded
1612 ;; by a line continuation backslash), or the beginning of a
1613 ;; line comment. Note: This has to be modified for awk
1614 ;; where line comments start with '#'.
1617 (list (when (c-lang-const c-line-comment-starter)
1618 (regexp-quote (c-lang-const c-line-comment-starter)))
1619 (when (c-lang-const c-unterminated-block-comment-regexp)
1620 (concat (c-lang-const c-unterminated-block-comment-regexp)
1626 (c-lang-defvar c-syntactic-eol (c-lang-const c-syntactic-eol))
1631 ;; The Emacs variables beginning-of-defun-function and end-of-defun-function
1632 ;; will be set so that commands like `mark-defun' and `narrow-to-defun' work
1633 ;; right. In older Emacsen, the key sequences C-M-a and C-M-e are, however,
1634 ;; bound directly to the CC Mode functions, allowing optimization for large n.
1635 ;; From Emacs 23, this isn't necessary any more, since n is passed to the two
1637 (c-lang-defconst beginning-of-defun-function
1638 "Function to which beginning-of-defun-function will be set."
1639 t 'c-beginning-of-defun
1640 awk 'c-awk-beginning-of-defun)
1641 (c-lang-setvar beginning-of-defun-function
1642 (c-lang-const beginning-of-defun-function))
1644 (c-lang-defconst end-of-defun-function
1645 "Function to which end-of-defun-function will be set."
1647 awk 'c-awk-end-of-defun)
1648 (c-lang-setvar end-of-defun-function (c-lang-const end-of-defun-function))
1650 ;;; In-comment text handling.
1652 (c-lang-defconst c-paragraph-start
1653 "Regexp to append to `paragraph-start'."
1655 java "\\(@[a-zA-Z]+\\>\\|$\\)" ; For Javadoc.
1656 pike "\\(@[a-zA-Z_-]+\\>\\([^{]\\|$\\)\\|$\\)") ; For Pike refdoc.
1657 (c-lang-defvar c-paragraph-start (c-lang-const c-paragraph-start))
1659 (c-lang-defconst c-paragraph-separate
1660 "Regexp to append to `paragraph-separate'."
1662 pike (c-lang-const c-paragraph-start))
1663 (c-lang-defvar c-paragraph-separate (c-lang-const c-paragraph-separate))
1668 ;; Note: All and only all language constants containing keyword lists
1669 ;; should end with "-kwds"; they're automatically collected into the
1670 ;; `c-kwds-lang-consts' list below and used to build `c-keywords' etc.
1672 (c-lang-defconst c-primitive-type-kwds
1673 "Primitive type keywords. As opposed to the other keyword lists, the
1674 keywords listed here are fontified with the type face instead of the
1677 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1678 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1679 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1682 Do not try to modify this list for end user customizations; the
1683 `*-font-lock-extra-types' variable, where `*' is the mode prefix, is
1684 the appropriate place for that."
1685 t '("char" "double" "float" "int" "long" "short" "signed"
1688 '("_Bool" "_Complex" "_Imaginary") ; Conditionally defined in C99.
1689 (c-lang-const c-primitive-type-kwds))
1691 '("bool" "wchar_t" "char16_t" "char32_t")
1692 (c-lang-const c-primitive-type-kwds))
1693 ;; Objective-C extends C, but probably not the new stuff in C99.
1695 '("id" "Class" "SEL" "IMP" "BOOL")
1696 (c-lang-const c-primitive-type-kwds))
1697 java '("boolean" "byte" "char" "double" "float" "int" "long" "short" "void")
1698 idl '("Object" "ValueBase" "any" "boolean" "char" "double" "fixed" "float"
1699 "long" "octet" "sequence" "short" "string" "void" "wchar" "wstring"
1702 ;; The following can't really end a type, but we have to specify them
1703 ;; here due to the assumption in `c-primitive-type-prefix-kwds'. It
1704 ;; doesn't matter that much.
1705 "unsigned" "strong")
1706 pike '(;; this_program isn't really a keyword, but it's practically
1707 ;; used as a builtin type.
1708 "array" "float" "function" "int" "mapping" "mixed" "multiset"
1709 "object" "program" "string" "this_program" "void"))
1711 (c-lang-defconst c-primitive-type-key
1712 ;; An adorned regexp that matches `c-primitive-type-kwds'.
1713 t (c-make-keywords-re t (c-lang-const c-primitive-type-kwds)))
1714 (c-lang-defvar c-primitive-type-key (c-lang-const c-primitive-type-key))
1716 (c-lang-defconst c-primitive-type-prefix-kwds
1717 "Keywords that might act as prefixes for primitive types. Assumed to
1718 be a subset of `c-primitive-type-kwds'."
1720 (c c++) '("long" "short" "signed" "unsigned")
1721 idl '("long" "unsigned"
1725 (c-lang-defconst c-typedef-kwds
1726 "Prefix keyword(s) like \"typedef\" which make a type declaration out
1727 of a variable declaration."
1731 (c-lang-defconst c-typedef-key
1732 ;; Adorned regexp matching `c-typedef-kwds'.
1733 t (c-make-keywords-re t (c-lang-const c-typedef-kwds)))
1734 (c-lang-defvar c-typedef-key (c-lang-const c-typedef-key))
1736 (c-lang-defconst c-typeof-kwds
1737 "Keywords followed by a parenthesized expression, which stands for
1738 the type of that expression."
1740 c '("typeof") ; longstanding GNU C(++) extension.
1741 c++ '("decltype" "typeof"))
1743 (c-lang-defconst c-typeof-key
1744 ;; Adorned regexp matching `c-typeof-kwds'.
1745 t (c-make-keywords-re t (c-lang-const c-typeof-kwds)))
1746 (c-lang-defvar c-typeof-key (c-lang-const c-typeof-key))
1748 (c-lang-defconst c-type-prefix-kwds
1749 "Keywords where the following name - if any - is a type name, and
1750 where the keyword together with the symbol works as a type in
1753 Note that an alternative if the second part doesn't hold is
1754 `c-type-list-kwds'. Keywords on this list are typically also present
1755 on one of the `*-decl-kwds' lists."
1757 c '("struct" "union" "enum")
1758 c++ (append '("class" "typename")
1759 (c-lang-const c-type-prefix-kwds c)))
1761 (c-lang-defconst c-type-prefix-key
1762 ;; Adorned regexp matching `c-type-prefix-kwds'.
1763 t (c-make-keywords-re t (c-lang-const c-type-prefix-kwds)))
1764 (c-lang-defvar c-type-prefix-key (c-lang-const c-type-prefix-key))
1766 (c-lang-defconst c-type-modifier-kwds
1767 "Type modifier keywords. These can occur almost anywhere in types
1768 but they don't build a type of themselves. Unlike the keywords on
1769 `c-primitive-type-kwds', they are fontified with the keyword face and
1772 c '("const" "restrict" "volatile")
1773 c++ '("const" "constexpr" "noexcept" "volatile" "throw" "final" "override")
1774 objc '("const" "volatile"))
1776 (c-lang-defconst c-opt-type-modifier-key
1777 ;; Adorned regexp matching `c-type-modifier-kwds', or nil in
1778 ;; languages without such keywords.
1779 t (and (c-lang-const c-type-modifier-kwds)
1780 (c-make-keywords-re t (c-lang-const c-type-modifier-kwds))))
1781 (c-lang-defvar c-opt-type-modifier-key (c-lang-const c-opt-type-modifier-key))
1783 (c-lang-defconst c-opt-type-component-key
1784 ;; An adorned regexp that matches `c-primitive-type-prefix-kwds' and
1785 ;; `c-type-modifier-kwds', or nil in languages without any of them.
1786 t (and (or (c-lang-const c-primitive-type-prefix-kwds)
1787 (c-lang-const c-type-modifier-kwds))
1788 (c-make-keywords-re t
1789 (append (c-lang-const c-primitive-type-prefix-kwds)
1790 (c-lang-const c-type-modifier-kwds)))))
1791 (c-lang-defvar c-opt-type-component-key
1792 (c-lang-const c-opt-type-component-key))
1794 (c-lang-defconst c-type-start-kwds
1795 ;; All keywords that can start a type (i.e. are either a type prefix
1796 ;; or a complete type).
1797 t (c--delete-duplicates (append (c-lang-const c-primitive-type-kwds)
1798 (c-lang-const c-type-prefix-kwds)
1799 (c-lang-const c-type-modifier-kwds))
1800 :test 'string-equal))
1802 (c-lang-defconst c-class-decl-kwds
1803 "Keywords introducing declarations where the following block (if any)
1804 contains another declaration level that should be considered a class.
1806 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1807 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1808 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1811 Note that presence on this list does not automatically treat the
1812 following identifier as a type; the keyword must also be present on
1813 `c-type-prefix-kwds' or `c-type-list-kwds' to accomplish that."
1815 c '("struct" "union")
1816 c++ '("class" "struct" "union")
1817 objc '("struct" "union"
1818 "@interface" "@implementation" "@protocol")
1819 java '("class" "@interface" "interface")
1820 idl '("component" "eventtype" "exception" "home" "interface" "struct"
1823 "storagehome" "storagetype"
1825 "catalog" "executor" "manages" "segment")
1828 (c-lang-defconst c-class-key
1829 ;; Regexp matching the start of a class.
1830 t (c-make-keywords-re t (c-lang-const c-class-decl-kwds)))
1831 (c-lang-defvar c-class-key (c-lang-const c-class-key))
1833 (c-lang-defconst c-brace-list-decl-kwds
1834 "Keywords introducing declarations where the following block (if
1835 any) is a brace list.
1837 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1838 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1839 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1844 (c-lang-defconst c-brace-list-key
1845 ;; Regexp matching the start of declarations where the following
1846 ;; block is a brace list.
1847 t (c-make-keywords-re t (c-lang-const c-brace-list-decl-kwds)))
1848 (c-lang-defvar c-brace-list-key (c-lang-const c-brace-list-key))
1850 (c-lang-defconst c-after-brace-list-decl-kwds
1851 "Keywords that might follow keywords in `c-brace-list-decl-kwds'
1852 and precede the opening brace."
1854 c++ '("class" "struct"))
1856 (c-lang-defconst c-after-brace-list-key
1857 ;; Regexp matching keywords that can fall between a brace-list
1858 ;; keyword and the associated brace list.
1859 t (c-make-keywords-re t (c-lang-const c-after-brace-list-decl-kwds)))
1860 (c-lang-defvar c-after-brace-list-key (c-lang-const c-after-brace-list-key))
1862 (c-lang-defconst c-recognize-post-brace-list-type-p
1863 "Set to t when we recognize a colon and then a type after an enum,
1864 e.g., enum foo : int { A, B, C };"
1867 (c-lang-defvar c-recognize-post-brace-list-type-p
1868 (c-lang-const c-recognize-post-brace-list-type-p))
1870 (c-lang-defconst c-other-block-decl-kwds
1871 "Keywords where the following block (if any) contains another
1872 declaration level that should not be considered a class. For every
1873 keyword here, CC Mode will add a set of special syntactic symbols for
1874 those blocks. E.g. if the keyword is \"foo\" then there will be
1875 `foo-open', `foo-close', and `infoo' symbols.
1877 The intention is that this category should be used for block
1878 constructs that aren't related to object orientation concepts like
1879 classes (which thus also include e.g. interfaces, templates,
1880 contracts, structs, etc). The more pragmatic distinction is that
1881 while most want some indentation inside classes, it's fairly common
1882 that they don't want it in some of these constructs, so it should be
1883 simple to configure that differently from classes. See also
1884 `c-class-decl-kwds'.
1886 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1887 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1888 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1891 (c objc) '("extern")
1892 c++ '("namespace" "extern")
1897 (c-lang-defconst c-other-decl-block-key
1898 ;; Regexp matching the start of blocks besides classes that contain
1899 ;; another declaration level.
1900 t (c-make-keywords-re t (c-lang-const c-other-block-decl-kwds)))
1901 (c-lang-defvar c-other-decl-block-key (c-lang-const c-other-decl-block-key))
1903 (c-lang-defvar c-other-decl-block-key-in-symbols-alist
1907 (if (string= elt "extern")
1909 (intern (concat "in" elt)))))
1910 (c-lang-const c-other-block-decl-kwds))
1911 "Alist associating keywords in c-other-decl-block-decl-kwds with
1912 their matching \"in\" syntactic symbols.")
1914 (c-lang-defconst c-typedef-decl-kwds
1915 "Keywords introducing declarations where the identifier(s) being
1918 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1919 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1920 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1922 ;; Default to `c-class-decl-kwds' and `c-brace-list-decl-kwds'
1923 ;; (since e.g. "Foo" is a type that's being defined in "class Foo
1925 t (append (c-lang-const c-class-decl-kwds)
1926 (c-lang-const c-brace-list-decl-kwds))
1927 ;; Languages that have a "typedef" construct.
1928 (c c++ objc idl pike) (append (c-lang-const c-typedef-decl-kwds)
1930 ;; Unlike most other languages, exception names are not handled as
1931 ;; types in IDL since they only can occur in "raises" specs.
1932 idl (delete "exception" (append (c-lang-const c-typedef-decl-kwds) nil)))
1934 (c-lang-defconst c-typedef-decl-key
1935 t (c-make-keywords-re t (c-lang-const c-typedef-decl-kwds)))
1936 (c-lang-defvar c-typedef-decl-key (c-lang-const c-typedef-decl-key))
1938 (c-lang-defconst c-typeless-decl-kwds
1939 "Keywords introducing declarations where the \(first) identifier
1940 \(declarator) follows directly after the keyword, without any type.
1942 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1943 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1944 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1946 ;; Default to `c-class-decl-kwds' and `c-brace-list-decl-kwds'
1947 ;; (since e.g. "Foo" is the identifier being defined in "class Foo
1949 t (append (c-lang-const c-class-decl-kwds)
1950 (c-lang-const c-brace-list-decl-kwds))
1951 c++ (append (c-lang-const c-typeless-decl-kwds) '("auto")) ; C++11.
1952 ;; Note: "manages" for CORBA CIDL clashes with its presence on
1953 ;; `c-type-list-kwds' for IDL.
1954 idl (append (c-lang-const c-typeless-decl-kwds)
1955 '("factory" "finder" "native"
1960 pike (append (c-lang-const c-class-decl-kwds)
1963 (c-lang-defconst c-modifier-kwds
1964 "Keywords that can prefix normal declarations of identifiers
1965 \(and typically act as flags). Things like argument declarations
1966 inside function headers are also considered declarations in this
1969 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1970 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1971 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1974 (c c++) '("auto" "extern" "inline" "register" "static")
1975 c++ (append '("explicit" "friend" "mutable" "template" "thread_local"
1977 (c-lang-const c-modifier-kwds))
1978 objc '("auto" "bycopy" "byref" "extern" "in" "inout" "oneway" "out" "static")
1979 ;; FIXME: Some of those below ought to be on `c-other-decl-kwds' instead.
1980 idl '("abstract" "attribute" "const" "consumes" "custom" "emits" "import"
1981 "in" "inout" "local" "multiple" "oneway" "out" "private" "provides"
1982 "public" "publishes" "readonly" "typeid" "typeprefix" "uses"
1986 "bindsTo" "delegatesTo" "implements" "proxy" "storedOn")
1987 ;; Note: "const" is not used in Java, but it's still a reserved keyword.
1988 java '("abstract" "const" "default" "final" "native" "private" "protected"
1989 "public" "static" "strictfp" "synchronized" "transient" "volatile")
1990 pike '("final" "inline" "local" "nomask" "optional" "private" "protected"
1991 "public" "static" "variant"))
1993 (c-lang-defconst c-other-decl-kwds
1994 "Keywords that can start or prefix any declaration level construct,
1995 besides those on `c-class-decl-kwds', `c-brace-list-decl-kwds',
1996 `c-other-block-decl-kwds', `c-typedef-decl-kwds',
1997 `c-typeless-decl-kwds' and `c-modifier-kwds'.
1999 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
2000 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
2001 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
2004 objc '("@class" "@end" "@defs")
2005 java '("import" "package")
2006 pike '("import" "inherit"))
2008 (c-lang-defconst c-decl-start-kwds
2009 "Keywords that always start declarations, wherever they occur.
2010 This can be used for declarations that aren't recognized by the normal
2011 combination of `c-decl-prefix-re' and `c-decl-start-re'."
2013 ;; Classes can be declared anywhere in a Pike expression.
2016 (c-lang-defconst c-decl-hangon-kwds
2017 "Keywords that can occur anywhere in a declaration level construct.
2018 This is used for self-contained things that can be tacked on anywhere
2019 on a declaration and that should be ignored to be able to recognize it
2020 correctly. Typical cases are compiler extensions like
2021 \"__attribute__\" or \"__declspec\":
2023 __declspec(noreturn) void foo();
2024 class __declspec(dllexport) classname {...};
2025 void foo() __attribute__((noreturn));
2027 Note that unrecognized plain symbols are skipped anyway if they occur
2028 before the type, so such things are not necessary to mention here.
2029 Mentioning them here is necessary only if they can occur in other
2030 places, or if they are followed by a construct that must be skipped
2031 over \(like the parens in the \"__attribute__\" and \"__declspec\"
2032 examples above). In the last case, they alse need to be present on
2033 one of `c-type-list-kwds', `c-ref-list-kwds',
2034 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
2035 `c-<>-type-kwds', or `c-<>-arglist-kwds'."
2036 ;; NB: These are currently not recognized in all parts of a
2037 ;; declaration. Specifically, they aren't recognized in the middle
2038 ;; of multi-token types, inside declarators, and between the
2039 ;; identifier and the arglist paren of a function declaration.
2041 ;; FIXME: This ought to be user customizable since compiler stuff
2042 ;; like this usually is wrapped in project specific macros. (It'd
2043 ;; of course be even better if we could cope without knowing this.)
2045 (c c++) '(;; GCC extension.
2050 (c-lang-defconst c-decl-hangon-key
2051 ;; Adorned regexp matching `c-decl-hangon-kwds'.
2052 t (c-make-keywords-re t (c-lang-const c-decl-hangon-kwds)))
2053 (c-lang-defvar c-decl-hangon-key (c-lang-const c-decl-hangon-key))
2055 (c-lang-defconst c-prefix-spec-kwds
2056 ;; All keywords that can occur in the preamble of a declaration.
2057 ;; They typically occur before the type, but they are also matched
2058 ;; after presumptive types since we often can't be sure that
2059 ;; something is a type or just some sort of macro in front of the
2060 ;; declaration. They might be ambiguous with types or type
2062 t (c--delete-duplicates (append (c-lang-const c-class-decl-kwds)
2063 (c-lang-const c-brace-list-decl-kwds)
2064 (c-lang-const c-other-block-decl-kwds)
2065 (c-lang-const c-typedef-decl-kwds)
2066 (c-lang-const c-typeless-decl-kwds)
2067 (c-lang-const c-modifier-kwds)
2068 (c-lang-const c-other-decl-kwds)
2069 (c-lang-const c-decl-start-kwds)
2070 (c-lang-const c-decl-hangon-kwds))
2071 :test 'string-equal))
2073 (c-lang-defconst c-prefix-spec-kwds-re
2074 ;; Adorned regexp of `c-prefix-spec-kwds'.
2075 t (c-make-keywords-re t (c-lang-const c-prefix-spec-kwds)))
2077 (c-lang-defvar c-prefix-spec-kwds-re (c-lang-const c-prefix-spec-kwds-re))
2079 (c-lang-defconst c-specifier-key
2080 ;; Adorned regexp of the keywords in `c-prefix-spec-kwds' that aren't
2081 ;; ambiguous with types or type prefixes. These are the keywords (like
2082 ;; extern, namespace, but NOT template) that can modify a declaration.
2083 t (c-make-keywords-re t
2084 (c--set-difference (c-lang-const c-prefix-spec-kwds)
2085 (append (c-lang-const c-type-start-kwds)
2086 (c-lang-const c-<>-arglist-kwds))
2087 :test 'string-equal)))
2088 (c-lang-defvar c-specifier-key (c-lang-const c-specifier-key))
2090 (c-lang-defconst c-postfix-spec-kwds
2091 ;; Keywords that can occur after argument list of a function header
2092 ;; declaration, i.e. in the "K&R region".
2093 t (append (c-lang-const c-postfix-decl-spec-kwds)
2094 (c-lang-const c-decl-hangon-kwds)))
2096 (c-lang-defconst c-not-decl-init-keywords
2097 ;; Adorned regexp matching all keywords that can't appear at the
2098 ;; start of a declaration.
2099 t (c-make-keywords-re t
2100 (c--set-difference (c-lang-const c-keywords)
2101 (append (c-lang-const c-type-start-kwds)
2102 (c-lang-const c-prefix-spec-kwds)
2103 (c-lang-const c-typeof-kwds))
2104 :test 'string-equal)))
2105 (c-lang-defvar c-not-decl-init-keywords
2106 (c-lang-const c-not-decl-init-keywords))
2108 (c-lang-defconst c-not-primitive-type-keywords
2109 "List of all keywords apart from primitive types (like \"int\")."
2110 t (c--set-difference (c-lang-const c-keywords)
2111 (c-lang-const c-primitive-type-kwds)
2112 :test 'string-equal)
2113 ;; The "more" for C++ is the QT keyword (as in "more slots:").
2114 ;; This variable is intended for use in c-beginning-of-statement-1.
2115 c++ (append (c-lang-const c-not-primitive-type-keywords) '("more")))
2117 (c-lang-defconst c-not-primitive-type-keywords-regexp
2118 t (c-make-keywords-re t
2119 (c-lang-const c-not-primitive-type-keywords)))
2120 (c-lang-defvar c-not-primitive-type-keywords-regexp
2121 (c-lang-const c-not-primitive-type-keywords-regexp))
2123 (c-lang-defconst c-protection-kwds
2124 "Access protection label keywords in classes."
2126 c++ '("private" "protected" "public")
2127 objc '("@private" "@protected" "@public"))
2129 (c-lang-defconst c-block-decls-with-vars
2130 "Keywords introducing declarations that can contain a block which
2131 might be followed by variable declarations, e.g. like \"foo\" in
2132 \"class Foo { ... } foo;\". So if there is a block in a declaration
2133 like that, it ends with the following `;' and not right away.
2135 The keywords on list are assumed to also be present on one of the
2136 `*-decl-kwds' lists."
2138 (c objc) '("struct" "union" "enum" "typedef")
2139 c++ '("class" "struct" "union" "enum" "typedef"))
2141 (c-lang-defconst c-opt-block-decls-with-vars-key
2142 ;; Regexp matching the `c-block-decls-with-vars' keywords, or nil in
2143 ;; languages without such constructs.
2144 t (and (c-lang-const c-block-decls-with-vars)
2145 (c-make-keywords-re t (c-lang-const c-block-decls-with-vars))))
2146 (c-lang-defvar c-opt-block-decls-with-vars-key
2147 (c-lang-const c-opt-block-decls-with-vars-key))
2149 (c-lang-defconst c-postfix-decl-spec-kwds
2150 "Keywords introducing extra declaration specifiers in the region
2151 between the header and the body \(i.e. the \"K&R-region\") in
2154 java '("extends" "implements" "throws")
2155 idl '("context" "getraises" "manages" "primarykey" "raises" "setraises"
2158 "as" "const" "implements" "of" "ref"))
2160 (c-lang-defconst c-postfix-decl-spec-key
2161 ;; Regexp matching the keywords in `c-postfix-decl-spec-kwds'.
2162 t (c-make-keywords-re t (c-lang-const c-postfix-decl-spec-kwds)))
2163 (c-lang-defvar c-postfix-decl-spec-key
2164 (c-lang-const c-postfix-decl-spec-key))
2166 (c-lang-defconst c-nonsymbol-sexp-kwds
2167 "Keywords that may be followed by a nonsymbol sexp before whatever
2168 construct it's part of continues."
2170 (c c++ objc) '("extern"))
2172 (c-lang-defconst c-type-list-kwds
2173 "Keywords that may be followed by a comma separated list of type
2174 identifiers, where each optionally can be prefixed by keywords. (Can
2175 also be used for the special case when the list can contain only one
2178 Assumed to be mutually exclusive with `c-ref-list-kwds'. There's no
2179 reason to put keywords on this list if they are on `c-type-prefix-kwds'.
2180 There's also no reason to add keywords that prefixes a normal
2181 declaration consisting of a type followed by a declarator (list), so
2182 the keywords on `c-modifier-kwds' should normally not be listed here
2185 Note: Use `c-typeless-decl-kwds' for keywords followed by a function
2186 or variable identifier (that's being defined)."
2190 java '("import" "new" "extends" "super" "implements" "throws")
2191 idl '("manages" "native" "primarykey" "supports"
2193 "as" "implements" "of" "scope")
2196 (c-lang-defconst c-ref-list-kwds
2197 "Keywords that may be followed by a comma separated list of
2198 reference (i.e. namespace/scope/module) identifiers, where each
2199 optionally can be prefixed by keywords. (Can also be used for the
2200 special case when the list can contain only one element.) Assumed to
2201 be mutually exclusive with `c-type-list-kwds'.
2203 Note: Use `c-typeless-decl-kwds' for keywords followed by a function
2204 or variable identifier (that's being defined)."
2208 idl '("import" "module"
2213 (c-lang-defconst c-colon-type-list-kwds
2214 "Keywords that may be followed (not necessarily directly) by a colon
2215 and then a comma separated list of type identifiers, where each
2216 optionally can be prefixed by keywords. (Can also be used for the
2217 special case when the list can contain only one element.)"
2219 c++ '("class" "struct")
2220 idl '("component" "eventtype" "home" "interface" "valuetype"
2222 "storagehome" "storagetype"))
2224 (c-lang-defconst c-colon-type-list-re
2225 "Regexp matched after the keywords in `c-colon-type-list-kwds' to skip
2226 forward to the colon. The end of the match is assumed to be directly
2227 after the colon, so the regexp should end with \":\". Must be a
2228 regexp if `c-colon-type-list-kwds' isn't nil."
2229 t (if (c-lang-const c-colon-type-list-kwds)
2230 ;; Disallow various common punctuation chars that can't come
2231 ;; before the ":" that starts the inherit list after "class"
2232 ;; or "struct" in C++. (Also used as default for other
2234 "[^][{}();,/#=:]*:"))
2235 (c-lang-defvar c-colon-type-list-re (c-lang-const c-colon-type-list-re))
2237 (c-lang-defconst c-paren-nontype-kwds
2238 "Keywords that may be followed by a parenthesis expression that doesn't
2239 contain type identifiers."
2241 (c c++) '(;; GCC extension.
2246 (c-lang-defconst c-paren-type-kwds
2247 "Keywords that may be followed by a parenthesis expression containing
2248 type identifiers separated by arbitrary tokens."
2253 pike '("array" "function" "int" "mapping" "multiset" "object" "program"))
2255 (c-lang-defconst c-paren-any-kwds
2256 t (c--delete-duplicates (append (c-lang-const c-paren-nontype-kwds)
2257 (c-lang-const c-paren-type-kwds))
2258 :test 'string-equal))
2260 (c-lang-defconst c-<>-type-kwds
2261 "Keywords that may be followed by an angle bracket expression
2262 containing type identifiers separated by \",\". The difference from
2263 `c-<>-arglist-kwds' is that unknown names are taken to be types and
2264 not other identifiers. `c-recognize-<>-arglists' is assumed to be set
2272 (c-lang-defconst c-<>-arglist-kwds
2273 "Keywords that can be followed by a C++ style template arglist; see
2274 `c-recognize-<>-arglists' for details. That language constant is
2275 assumed to be set if this isn't nil."
2278 idl '("fixed" "string" "wstring"))
2280 (c-lang-defconst c-<>-sexp-kwds
2281 ;; All keywords that can be followed by an angle bracket sexp.
2282 t (c--delete-duplicates (append (c-lang-const c-<>-type-kwds)
2283 (c-lang-const c-<>-arglist-kwds))
2284 :test 'string-equal))
2286 (c-lang-defconst c-opt-<>-sexp-key
2287 ;; Adorned regexp matching keywords that can be followed by an angle
2288 ;; bracket sexp. Always set when `c-recognize-<>-arglists' is.
2289 t (c-make-keywords-re t (c-lang-const c-<>-sexp-kwds)))
2290 (c-lang-defvar c-opt-<>-sexp-key (c-lang-const c-opt-<>-sexp-key))
2292 (c-lang-defconst c-brace-id-list-kwds
2293 "Keywords that may be followed by a brace block containing a comma
2294 separated list of identifier definitions, i.e. like the list of
2295 identifiers that follows the type in a normal declaration."
2296 t (c-lang-const c-brace-list-decl-kwds))
2298 (c-lang-defconst c-block-stmt-1-kwds
2299 "Statement keywords followed directly by a substatement."
2301 c++ '("do" "else" "try")
2302 objc '("do" "else" "@finally" "@try")
2303 java '("do" "else" "finally" "try")
2306 (c-lang-defconst c-block-stmt-1-key
2307 ;; Regexp matching the start of any statement followed directly by a
2308 ;; substatement (doesn't match a bare block, however).
2309 t (c-make-keywords-re t (c-lang-const c-block-stmt-1-kwds)))
2310 (c-lang-defvar c-block-stmt-1-key (c-lang-const c-block-stmt-1-key))
2312 (c-lang-defconst c-block-stmt-1-2-kwds
2313 "Statement keywords optionally followed by a paren sexp.
2314 Keywords here should also be in `c-block-stmt-1-kwds'."
2318 (c-lang-defconst c-block-stmt-1-2-key
2319 ;; Regexp matching the start of a statement which may be followed by a
2320 ;; paren sexp and will then be followed by a substatement.
2321 t (c-make-keywords-re t (c-lang-const c-block-stmt-1-2-kwds)))
2322 (c-lang-defvar c-block-stmt-1-2-key (c-lang-const c-block-stmt-1-2-key))
2324 (c-lang-defconst c-block-stmt-2-kwds
2325 "Statement keywords followed by a paren sexp and then by a substatement."
2326 t '("for" "if" "switch" "while")
2327 c++ '("for" "if" "switch" "while" "catch")
2328 objc '("for" "if" "switch" "while" "@catch" "@synchronized")
2329 java '("for" "if" "switch" "while" "catch" "synchronized")
2331 pike '("for" "if" "switch" "while" "foreach")
2332 awk '("for" "if" "while"))
2334 (c-lang-defconst c-block-stmt-2-key
2335 ;; Regexp matching the start of any statement followed by a paren sexp
2336 ;; and then by a substatement.
2337 t (c-make-keywords-re t (c-lang-const c-block-stmt-2-kwds)))
2338 (c-lang-defvar c-block-stmt-2-key (c-lang-const c-block-stmt-2-key))
2340 (c-lang-defconst c-block-stmt-kwds
2341 ;; Union of `c-block-stmt-1-kwds' and `c-block-stmt-2-kwds'.
2342 t (c--delete-duplicates (append (c-lang-const c-block-stmt-1-kwds)
2343 (c-lang-const c-block-stmt-2-kwds))
2344 :test 'string-equal))
2346 (c-lang-defconst c-opt-block-stmt-key
2347 ;; Regexp matching the start of any statement that has a
2348 ;; substatement (except a bare block). Nil in languages that
2349 ;; don't have such constructs.
2350 t (if (or (c-lang-const c-block-stmt-1-kwds)
2351 (c-lang-const c-block-stmt-2-kwds))
2352 (c-make-keywords-re t
2353 (append (c-lang-const c-block-stmt-1-kwds)
2354 (c-lang-const c-block-stmt-2-kwds)))))
2355 (c-lang-defvar c-opt-block-stmt-key (c-lang-const c-opt-block-stmt-key))
2357 (c-lang-defconst c-simple-stmt-kwds
2358 "Statement keywords followed by an expression or nothing."
2359 t '("break" "continue" "goto" "return")
2360 objc '("break" "continue" "goto" "return" "@throw")
2361 ;; Note: `goto' is not valid in Java, but the keyword is still reserved.
2362 java '("break" "continue" "goto" "return" "throw")
2364 pike '("break" "continue" "return")
2365 awk '(;; Not sure about "delete", "exit", "getline", etc. ; ACM 2002/5/30
2366 "break" "continue" "return" "delete" "exit" "getline" "next"
2367 "nextfile" "print" "printf"))
2369 (c-lang-defconst c-simple-stmt-key
2370 ;; Adorned regexp matching `c-simple-stmt-kwds'.
2371 t (c-make-keywords-re t (c-lang-const c-simple-stmt-kwds)))
2372 (c-lang-defvar c-simple-stmt-key (c-lang-const c-simple-stmt-key))
2374 (c-lang-defconst c-paren-stmt-kwds
2375 "Statement keywords followed by a parenthesis expression that
2376 nevertheless contains a list separated with `;' and not `,'."
2380 (c-lang-defconst c-paren-stmt-key
2381 ;; Adorned regexp matching `c-paren-stmt-kwds'.
2382 t (c-make-keywords-re t (c-lang-const c-paren-stmt-kwds)))
2383 (c-lang-defvar c-paren-stmt-key (c-lang-const c-paren-stmt-key))
2385 (c-lang-defconst c-asm-stmt-kwds
2386 "Statement keywords followed by an assembler expression."
2388 (c c++) '("asm" "__asm__")) ;; Not standard, but common.
2390 (c-lang-defconst c-opt-asm-stmt-key
2391 ;; Regexp matching the start of an assembler statement. Nil in
2392 ;; languages that don't support that.
2393 t (if (c-lang-const c-asm-stmt-kwds)
2394 (c-make-keywords-re t (c-lang-const c-asm-stmt-kwds))))
2395 (c-lang-defvar c-opt-asm-stmt-key (c-lang-const c-opt-asm-stmt-key))
2397 (c-lang-defconst c-case-kwds
2398 "The keyword(s) which introduce a \"case\" like construct.
2399 This construct is \"<keyword> <expression> :\"."
2403 (c-lang-defconst c-case-kwds-regexp
2404 ;; Adorned regexp matching any "case"-like keyword.
2405 t (c-make-keywords-re t (c-lang-const c-case-kwds)))
2406 (c-lang-defvar c-case-kwds-regexp (c-lang-const c-case-kwds-regexp))
2408 (c-lang-defconst c-label-kwds
2409 "Keywords introducing colon terminated labels in blocks."
2410 t '("case" "default"))
2412 (c-lang-defconst c-label-kwds-regexp
2413 ;; Adorned regexp matching any keyword that introduces a label.
2414 t (c-make-keywords-re t (c-lang-const c-label-kwds)))
2415 (c-lang-defvar c-label-kwds-regexp (c-lang-const c-label-kwds-regexp))
2417 (c-lang-defconst c-before-label-kwds
2418 "Keywords that might be followed by a label identifier."
2420 (java pike) (append '("break" "continue")
2421 (c-lang-const c-before-label-kwds))
2425 (c-lang-defconst c-constant-kwds
2426 "Keywords for constants."
2428 c '("NULL" ;; Not a keyword, but practically works as one.
2429 "false" "true") ; Defined in C99.
2432 (c-lang-const c-constant-kwds c))
2433 objc '("nil" "Nil" "YES" "NO" "NS_DURING" "NS_HANDLER" "NS_ENDHANDLER")
2434 idl '("TRUE" "FALSE")
2435 java '("true" "false" "null") ; technically "literals", not keywords
2436 pike '("UNDEFINED")) ;; Not a keyword, but practically works as one.
2438 (c-lang-defconst c-primary-expr-kwds
2439 "Keywords besides constants and operators that start primary expressions."
2441 c++ '("operator" "this")
2442 objc '("super" "self")
2444 pike '("this")) ;; Not really a keyword, but practically works as one.
2446 (c-lang-defconst c-expr-kwds
2447 ;; Keywords that can occur anywhere in expressions. Built from
2448 ;; `c-primary-expr-kwds' and all keyword operators in `c-operators'.
2449 t (c--delete-duplicates
2450 (append (c-lang-const c-primary-expr-kwds)
2451 (c-filter-ops (c-lang-const c-operator-list)
2453 "\\`\\(\\w\\|\\s_\\)+\\'"))
2454 :test 'string-equal))
2456 (c-lang-defconst c-lambda-kwds
2457 "Keywords that start lambda constructs, i.e. function definitions in
2462 (c-lang-defconst c-inexpr-block-kwds
2463 "Keywords that start constructs followed by statement blocks which can
2464 be used in expressions \(the gcc extension for this in C and C++ is
2465 handled separately by `c-recognize-paren-inexpr-blocks')."
2467 pike '("catch" "gauge"))
2469 (c-lang-defconst c-inexpr-class-kwds
2470 "Keywords that can start classes inside expressions."
2475 (c-lang-defconst c-inexpr-brace-list-kwds
2476 "Keywords that can start brace list blocks inside expressions.
2477 Note that Java specific rules are currently applied to tell this from
2478 `c-inexpr-class-kwds'."
2482 (c-lang-defconst c-opt-inexpr-brace-list-key
2483 ;; Regexp matching the start of a brace list in an expression, or
2484 ;; nil in languages that don't have such things. This should not
2485 ;; match brace lists recognized through `c-special-brace-lists'.
2486 t (and (c-lang-const c-inexpr-brace-list-kwds)
2487 (c-make-keywords-re t (c-lang-const c-inexpr-brace-list-kwds))))
2488 (c-lang-defvar c-opt-inexpr-brace-list-key
2489 (c-lang-const c-opt-inexpr-brace-list-key))
2491 (c-lang-defconst c-decl-block-key
2492 ;; Regexp matching keywords in any construct that contain another
2493 ;; declaration level, i.e. that isn't followed by a function block
2494 ;; or brace list. When the first submatch matches, it's an
2495 ;; unambiguous construct, otherwise it's an ambiguous match that
2496 ;; might also be the return type of a function declaration.
2497 t (let* ((decl-kwds (append (c-lang-const c-class-decl-kwds)
2498 (c-lang-const c-other-block-decl-kwds)
2499 (c-lang-const c-inexpr-class-kwds)))
2500 (unambiguous (c--set-difference decl-kwds
2501 (c-lang-const c-type-start-kwds)
2502 :test 'string-equal))
2503 (ambiguous (c--intersection decl-kwds
2504 (c-lang-const c-type-start-kwds)
2505 :test 'string-equal)))
2507 (concat (c-make-keywords-re t unambiguous)
2509 (c-make-keywords-re t ambiguous))
2510 (c-make-keywords-re t unambiguous))))
2511 (c-lang-defvar c-decl-block-key (c-lang-const c-decl-block-key))
2513 (c-lang-defconst c-bitfield-kwds
2514 "Keywords that can introduce bitfields."
2516 (c c++ objc) '("char" "int" "long" "signed" "unsigned"))
2518 (c-lang-defconst c-opt-bitfield-key
2519 ;; Regexp matching the start of a bitfield (not uniquely), or nil in
2520 ;; languages without bitfield support.
2522 (c c++) (c-make-keywords-re t (c-lang-const c-bitfield-kwds)))
2523 (c-lang-defvar c-opt-bitfield-key (c-lang-const c-opt-bitfield-key))
2525 (c-lang-defconst c-other-kwds
2526 "Keywords not accounted for by any other `*-kwds' language constant."
2529 ;; In CORBA CIDL: (These are declaration keywords that never
2530 ;; can start a declaration.)
2531 "entity" "process" "service" "session" "storage"))
2534 ;;; Constants built from keywords.
2536 ;; Note: No `*-kwds' language constants may be defined below this point.
2539 (defconst c-kwds-lang-consts
2540 ;; List of all the language constants that contain keyword lists.
2542 (mapatoms (lambda (sym)
2543 (when (and (boundp sym)
2544 (string-match "-kwds\\'" (symbol-name sym)))
2545 ;; Make the list of globally interned symbols
2546 ;; instead of ones interned in `c-lang-constants'.
2547 (setq list (cons (intern (symbol-name sym)) list))))
2551 (c-lang-defconst c-keywords
2552 ;; All keywords as a list.
2553 t (c--delete-duplicates
2554 (c-lang-defconst-eval-immediately
2555 `(append ,@(mapcar (lambda (kwds-lang-const)
2556 `(c-lang-const ,kwds-lang-const))
2559 :test 'string-equal))
2561 (c-lang-defconst c-keywords-regexp
2562 ;; All keywords as an adorned regexp.
2563 t (c-make-keywords-re t (c-lang-const c-keywords)))
2564 (c-lang-defvar c-keywords-regexp (c-lang-const c-keywords-regexp))
2566 (c-lang-defconst c-keyword-member-alist
2567 ;; An alist with all the keywords in the cars. The cdr for each
2568 ;; keyword is a list of the symbols for the `*-kwds' lists that
2570 t (let ((kwd-list-alist
2571 (c-lang-defconst-eval-immediately
2572 `(list ,@(mapcar (lambda (kwds-lang-const)
2573 `(cons ',kwds-lang-const
2574 (c-lang-const ,kwds-lang-const)))
2575 c-kwds-lang-consts))))
2576 lang-const kwd-list kwd
2578 (while kwd-list-alist
2579 (setq lang-const (caar kwd-list-alist)
2580 kwd-list (cdar kwd-list-alist)
2581 kwd-list-alist (cdr kwd-list-alist))
2583 (setq kwd (car kwd-list)
2584 kwd-list (cdr kwd-list))
2585 (unless (setq elem (assoc kwd result-alist))
2586 (setq result-alist (cons (setq elem (list kwd)) result-alist)))
2587 (unless (memq lang-const (cdr elem))
2588 (setcdr elem (cons lang-const (cdr elem))))))
2591 (c-lang-defvar c-keywords-obarray
2592 ;; An obarray containing all keywords as symbols. The property list
2593 ;; of each symbol has a non-nil entry for the specific `*-kwds'
2594 ;; lists it's a member of.
2596 ;; E.g. to see whether the string str contains a keyword on
2597 ;; `c-class-decl-kwds', one can do like this:
2598 ;; (get (intern-soft str c-keyword-obarray) 'c-class-decl-kwds)
2599 ;; Which preferably is written using the associated functions in
2601 ;; (c-keyword-member (c-keyword-sym str) 'c-class-decl-kwds)
2603 ;; The obarray is not stored directly as a language constant since
2604 ;; the printed representation for obarrays used in .elc files isn't
2607 (let* ((alist (c-lang-const c-keyword-member-alist))
2609 (obarray (make-vector (* (length alist) 2) 0)))
2611 (setq kwd (caar alist)
2612 lang-const-list (cdar alist)
2614 (setplist (intern kwd obarray)
2615 ;; Emacs has an odd bug that causes `mapcan' to fail
2616 ;; with unintelligible errors. (XEmacs works.)
2617 ;; (2015-06-24): This bug has not yet been fixed.
2618 ;;(mapcan (lambda (lang-const)
2619 ;; (list lang-const t))
2621 (apply 'nconc (mapcar (lambda (lang-const)
2622 (list lang-const t))
2626 (c-lang-defconst c-regular-keywords-regexp
2627 ;; Adorned regexp matching all keywords that should be fontified
2628 ;; with the keywords face. I.e. that aren't types or constants.
2629 t (c-make-keywords-re t
2630 (c--set-difference (c-lang-const c-keywords)
2631 (append (c-lang-const c-primitive-type-kwds)
2632 (c-lang-const c-constant-kwds))
2633 :test 'string-equal)))
2634 (c-lang-defvar c-regular-keywords-regexp
2635 (c-lang-const c-regular-keywords-regexp))
2637 (c-lang-defconst c-primary-expr-regexp
2638 ;; Regexp matching the start of any primary expression, i.e. any
2639 ;; literal, symbol, prefix operator, and '('. It doesn't need to
2640 ;; exclude keywords; they are excluded afterwards unless the second
2641 ;; submatch matches. If the first but not the second submatch
2642 ;; matches then it is an ambiguous primary expression; it could also
2643 ;; be a match of e.g. an infix operator. (The case with ambiguous
2644 ;; keyword operators isn't handled.)
2646 t (let* ((prefix-ops
2647 (c-filter-ops (c-lang-const c-operators)
2650 ;; Filter out the special case prefix
2651 ;; operators that are close parens.
2652 (not (string-match "\\s)" op)))))
2654 (nonkeyword-prefix-ops
2655 (c-filter-ops prefix-ops
2657 "\\`\\(\\s.\\|\\s(\\|\\s)\\)+\\'"))
2660 (c-filter-ops (c-lang-const c-operators)
2665 right-assoc-sequence)
2668 (unambiguous-prefix-ops (c--set-difference nonkeyword-prefix-ops
2670 :test 'string-equal))
2671 (ambiguous-prefix-ops (c--intersection nonkeyword-prefix-ops
2673 :test 'string-equal)))
2677 ;; Take out all symbol class operators from `prefix-ops' and make the
2678 ;; first submatch from them together with `c-primary-expr-kwds'.
2679 (c-make-keywords-re t
2680 (append (c-lang-const c-primary-expr-kwds)
2681 (c--set-difference prefix-ops nonkeyword-prefix-ops
2682 :test 'string-equal)))
2685 ;; Match all ambiguous operators.
2686 (c-make-keywords-re nil
2687 (c--intersection nonkeyword-prefix-ops in-or-postfix-ops
2688 :test 'string-equal))
2692 ;; Now match all other symbols.
2693 (c-lang-const c-symbol-start)
2696 ;; The chars that can start integer and floating point
2701 ;; The unambiguous operators from `prefix-ops'.
2702 (c-make-keywords-re nil
2703 (c--set-difference nonkeyword-prefix-ops in-or-postfix-ops
2704 :test 'string-equal))
2707 ;; Match string and character literals.
2709 (if (memq 'gen-string-delim c-emacs-features)
2712 (c-lang-defvar c-primary-expr-regexp (c-lang-const c-primary-expr-regexp))
2715 ;;; Additional constants for parser-level constructs.
2717 (c-lang-defconst c-decl-start-colon-kwd-re
2718 "Regexp matching a keyword that is followed by a colon, where
2719 the whole construct can precede a declaration.
2720 E.g. \"public:\" in C++."
2722 c++ (c-make-keywords-re t (c-lang-const c-protection-kwds)))
2723 (c-lang-defvar c-decl-start-colon-kwd-re
2724 (c-lang-const c-decl-start-colon-kwd-re))
2726 (c-lang-defconst c-decl-prefix-re
2727 "Regexp matching something that might precede a declaration, cast or
2728 label, such as the last token of a preceding statement or declaration.
2729 This is used in the common situation where a declaration or cast
2730 doesn't start with any specific token that can be searched for.
2732 The regexp should not match bob; that is done implicitly. It can't
2733 require a match longer than one token. The end of the token is taken
2734 to be at the end of the first submatch, which is assumed to always
2735 match. It's undefined whether identifier syntax (see
2736 `c-identifier-syntax-table') is in effect or not. This regexp is
2737 assumed to be a superset of `c-label-prefix-re' if
2738 `c-recognize-colon-labels' is set.
2740 Besides this, `c-decl-start-kwds' is used to find declarations.
2742 Note: This variable together with `c-decl-start-re' and
2743 `c-decl-start-kwds' is only used to detect \"likely\"
2744 declaration/cast/label starts. I.e. they might produce more matches
2745 but should not miss anything (or else it's necessary to use text
2746 properties - see the next note). Wherever they match, the following
2747 construct is analyzed to see if it indeed is a declaration, cast or
2748 label. That analysis is not cheap, so it's important that not too
2749 many false matches are triggered.
2751 Note: If a declaration/cast/label start can't be detected with this
2752 variable, it's necessary to use the `c-type' text property with the
2753 value `c-decl-end' on the last char of the last token preceding the
2754 declaration. See the comment blurb at the start of cc-engine.el for
2757 ;; We match a sequence of characters to skip over things like \"};\"
2758 ;; more quickly. We match ")" in C for K&R region declarations, and
2759 ;; in all languages except Java for when a cpp macro definition
2760 ;; begins with a declaration.
2762 java "\\([{}(;,<]+\\)"
2763 ;; Match "<" in C++ to get the first argument in a template arglist.
2764 ;; In that case there's an additional check in `c-find-decl-spots'
2765 ;; that it got open paren syntax. Match ":" to aid in picking up
2766 ;; "public:", etc. This involves additional checks in
2767 ;; `c-find-decl-prefix-search' to prevent a match of identifiers
2769 c++ "\\([{}();:,<]+\\)"
2770 ;; Additionally match the protection directives in Objective-C.
2771 ;; Note that this doesn't cope with the longer directives, which we
2772 ;; would have to match from start to end since they don't end with
2773 ;; any easily recognized characters.
2774 objc (concat "\\([{}();,]+\\|"
2775 (c-make-keywords-re nil (c-lang-const c-protection-kwds))
2777 ;; Pike is like C but we also match "[" for multiple value
2778 ;; assignments and type casts.
2779 pike "\\([{}()[;,]+\\)")
2780 (c-lang-defvar c-decl-prefix-re (c-lang-const c-decl-prefix-re)
2783 (c-lang-defconst c-decl-start-re
2784 "Regexp matching the start of any declaration, cast or label.
2785 It's used on the token after the one `c-decl-prefix-re' matched. This
2786 regexp should not try to match those constructs accurately as it's
2787 only used as a sieve to avoid spending more time checking other
2789 t (c-lang-const c-identifier-start))
2790 (c-lang-defvar c-decl-start-re (c-lang-const c-decl-start-re))
2792 (c-lang-defconst c-decl-prefix-or-start-re
2793 ;; Regexp matching something that might precede or start a
2794 ;; declaration, cast or label.
2796 ;; If the first submatch matches, it's taken to match the end of a
2797 ;; token that might precede such a construct, e.g. ';', '}' or '{'.
2798 ;; It's built from `c-decl-prefix-re'.
2800 ;; If the first submatch did not match, the match of the whole
2801 ;; regexp is taken to be at the first token in the declaration.
2802 ;; `c-decl-start-re' is not checked in this case.
2804 ;; Design note: The reason the same regexp is used to match both
2805 ;; tokens that precede declarations and start them is to avoid an
2806 ;; extra regexp search from the previous declaration spot in
2807 ;; `c-find-decl-spots'. Users of `c-find-decl-spots' also count on
2808 ;; that it finds all declaration/cast/label starts in approximately
2809 ;; linear order, so we can't do the searches in two separate passes.
2810 t (if (c-lang-const c-decl-start-kwds)
2811 (concat (c-lang-const c-decl-prefix-re)
2813 (c-make-keywords-re t (c-lang-const c-decl-start-kwds)))
2814 (c-lang-const c-decl-prefix-re)))
2815 (c-lang-defvar c-decl-prefix-or-start-re
2816 (c-lang-const c-decl-prefix-or-start-re)
2819 (c-lang-defconst c-cast-parens
2820 ;; List containing the paren characters that can open a cast, or nil in
2821 ;; languages without casts.
2822 t (c-filter-ops (c-lang-const c-operators)
2825 (lambda (op) (elt op 0))))
2826 (c-lang-defvar c-cast-parens (c-lang-const c-cast-parens))
2828 (c-lang-defconst c-block-prefix-disallowed-chars
2829 "List of syntactically relevant characters that never can occur before
2830 the open brace in any construct that contains a brace block, e.g. in
2831 the \"class Foo: public Bar\" part of:
2833 class Foo: public Bar {int x();} a, *b;
2835 If parens can occur, the chars inside those aren't filtered with this
2838 `<' and `>' should be disallowed even if angle bracket arglists can
2839 occur. That since the search function needs to stop at them anyway to
2840 ensure they are given paren syntax.
2842 This is used to skip backward from the open brace to find the region
2843 in which to look for a construct like \"class\", \"enum\",
2844 \"namespace\" or whatever. That skipping should be as tight as
2845 possible for good performance."
2847 ;; Default to all chars that only occurs in nonsymbol tokens outside
2849 t (c--set-difference
2850 (c-lang-const c-nonsymbol-token-char-list)
2851 (c-filter-ops (append (c-lang-const c-identifier-ops)
2853 (c-lang-const c-after-id-concat-ops))))
2858 (while (string-match "\\(\\s.\\|\\s(\\|\\s)\\)"
2860 (setq res (cons (aref op (match-beginning 1)) res)
2864 ;; Allow cpp operations (where applicable).
2865 t (if (c-lang-const c-opt-cpp-prefix)
2866 (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2868 (c-lang-const c-block-prefix-disallowed-chars))
2870 ;; Allow ':' for inherit list starters.
2871 (c++ objc idl) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2874 ;; Allow ',' for multiple inherits.
2875 (c++ java) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2878 ;; Allow parentheses for anonymous inner classes in Java and class
2879 ;; initializer lists in Pike.
2880 (java pike) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2883 ;; Allow '"' for extern clauses (e.g. extern "C" {...}).
2884 (c c++ objc) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2887 (c-lang-defconst c-block-prefix-charset
2888 ;; `c-block-prefix-disallowed-chars' as an inverted charset suitable
2889 ;; for `c-syntactic-skip-backward'.
2890 t (c-make-bare-char-alt (c-lang-const c-block-prefix-disallowed-chars) t))
2891 (c-lang-defvar c-block-prefix-charset (c-lang-const c-block-prefix-charset))
2893 (c-lang-defconst c-type-decl-prefix-key
2894 "Regexp matching any declarator operator that might precede the
2895 identifier in a declaration, e.g. the \"*\" in \"char *argv\". This
2896 regexp should match \"(\" if parentheses are valid in declarators.
2897 The end of the first submatch is taken as the end of the operator.
2898 Identifier syntax is in effect when this is matched \(see
2899 `c-identifier-syntax-table')."
2900 t (if (c-lang-const c-type-modifier-kwds)
2901 (concat (regexp-opt (c-lang-const c-type-modifier-kwds) t) "\\>")
2902 ;; Default to a regexp that never matches.
2904 ;; Check that there's no "=" afterwards to avoid matching tokens
2906 (c objc) (concat "\\("
2909 (c-lang-const c-type-decl-prefix-key)
2915 (c-lang-const c-type-decl-prefix-key)
2918 ;; If this matches there's special treatment in
2919 ;; `c-font-lock-declarators' and
2920 ;; `c-font-lock-declarations' that check for a
2921 ;; complete name followed by ":: *".
2922 (c-lang-const c-identifier-start)
2926 pike "\\(\\*\\)\\([^=]\\|$\\)")
2927 (c-lang-defvar c-type-decl-prefix-key (c-lang-const c-type-decl-prefix-key)
2930 (c-lang-defconst c-type-decl-suffix-key
2931 "Regexp matching the declarator operators that might follow after the
2932 identifier in a declaration, e.g. the \"[\" in \"char argv[]\". This
2933 regexp should match \")\" if parentheses are valid in declarators. If
2934 it matches an open paren of some kind, the type declaration check
2935 continues at the corresponding close paren, otherwise the end of the
2936 first submatch is taken as the end of the operator. Identifier syntax
2937 is in effect when this is matched (see `c-identifier-syntax-table')."
2938 ;; Default to a regexp that matches `c-type-modifier-kwds' and a
2939 ;; function argument list parenthesis.
2940 t (if (c-lang-const c-type-modifier-kwds)
2942 (regexp-opt (c-lang-const c-type-modifier-kwds) t) "\\>"
2945 (c c++ objc) (concat
2948 (if (c-lang-const c-type-modifier-kwds)
2951 ;; "throw" in `c-type-modifier-kwds' is followed
2952 ;; by a parenthesis list, but no extra measures
2953 ;; are necessary to handle that.
2954 (regexp-opt (c-lang-const c-type-modifier-kwds) t)
2960 (c-lang-defvar c-type-decl-suffix-key (c-lang-const c-type-decl-suffix-key)
2963 (c-lang-defconst c-after-suffixed-type-decl-key
2964 "This regexp is matched after a declarator expression where
2965 `c-type-decl-suffix-key' has matched. If it matches then the
2966 construct is taken as a declaration. It's typically used to match the
2967 beginning of a function body or whatever might occur after the
2968 function header in a function declaration or definition. It's
2969 undefined whether identifier syntax (see `c-identifier-syntax-table')
2970 is in effect or not.
2972 Note that it's used in cases like after \"foo (bar)\" so it should
2973 only match when it's certain that it's a declaration, e.g., \"{\" but
2974 not \",\" or \";\"."
2976 ;; If K&R style declarations should be recognized then one could
2977 ;; consider to match the start of any symbol since we want to match
2978 ;; the start of the first declaration in the "K&R region". That
2979 ;; could however produce false matches on code like "FOO(bar) x"
2980 ;; where FOO is a cpp macro, so it's better to leave it out and rely
2981 ;; on the other heuristics in that case.
2982 t (if (c-lang-const c-postfix-spec-kwds)
2983 ;; Add on the keywords in `c-postfix-spec-kwds'.
2984 (concat (c-lang-const c-after-suffixed-type-decl-key)
2986 (c-make-keywords-re t (c-lang-const c-postfix-spec-kwds)))
2987 (c-lang-const c-after-suffixed-type-decl-key))
2988 ;; Also match the colon that starts a base class initializer list in
2989 ;; C++. That can be confused with a function call before the colon
2990 ;; in a ? : operator, but we count on that `c-decl-prefix-re' won't
2991 ;; match before such a thing (as a declaration-level construct;
2992 ;; matches inside arglist contexts are already excluded).
2994 (c-lang-defvar c-after-suffixed-type-decl-key
2995 (c-lang-const c-after-suffixed-type-decl-key)
2998 (c-lang-defconst c-after-suffixed-type-maybe-decl-key
2999 ;; Regexp that in addition to `c-after-suffixed-type-decl-key'
3000 ;; matches ";" and ",".
3001 t (concat "\\(" (c-lang-const c-after-suffixed-type-decl-key) "\\)"
3003 (c-lang-defvar c-after-suffixed-type-maybe-decl-key
3004 (c-lang-const c-after-suffixed-type-maybe-decl-key))
3006 (c-lang-defconst c-opt-type-concat-key
3007 "Regexp matching operators that concatenate types, e.g. the \"|\" in
3008 \"int|string\" in Pike. The end of the first submatch is taken as the
3009 end of the operator. nil in languages without such operators. It's
3010 undefined whether identifier syntax (see `c-identifier-syntax-table')
3011 is in effect or not."
3013 pike "\\([|.&]\\)\\($\\|[^|.&]\\)")
3014 (c-lang-defvar c-opt-type-concat-key (c-lang-const c-opt-type-concat-key)
3017 (c-lang-defconst c-opt-type-suffix-key
3018 "Regexp matching operators that might follow after a type, or nil in
3019 languages that don't have such operators. The end of the first
3020 submatch is taken as the end of the operator. This should not match
3021 things like C++ template arglists if `c-recognize-<>-arglists' is set.
3022 It's undefined whether identifier syntax (see `c-identifier-syntax-table')
3023 is in effect or not."
3025 (c c++ objc pike) "\\(\\.\\.\\.\\)"
3026 java (concat "\\(\\[" (c-lang-const c-simple-ws) "*\\]\\|\\.\\.\\.\\)"))
3027 (c-lang-defvar c-opt-type-suffix-key (c-lang-const c-opt-type-suffix-key))
3029 (c-lang-defvar c-known-type-key
3030 ;; Regexp matching the known type identifiers. This is initialized
3031 ;; from the type keywords and `*-font-lock-extra-types'. The first
3032 ;; submatch is the one that matches the type. Note that this regexp
3033 ;; assumes that symbol constituents like '_' and '$' have word
3036 (when (boundp (c-mode-symbol "font-lock-extra-types"))
3037 (c-mode-var "font-lock-extra-types")))
3039 (delq nil (mapcar (lambda (re)
3040 (when (string-match "[][.*+?^$\\]" re)
3044 (delq nil (mapcar (lambda (re)
3045 (unless (string-match "[][.*+?^$\\]" re)
3050 (append (list (c-make-keywords-re nil
3051 (append (c-lang-const c-primitive-type-kwds)
3057 (c-lang-defconst c-special-brace-lists
3058 "List of open- and close-chars that makes up a pike-style brace list,
3059 i.e. for a ([ ]) list there should be a cons (?\\[ . ?\\]) in this
3062 pike '((?{ . ?}) (?\[ . ?\]) (?< . ?>)))
3063 (c-lang-defvar c-special-brace-lists (c-lang-const c-special-brace-lists))
3065 (c-lang-defconst c-recognize-knr-p
3066 "Non-nil means K&R style argument declarations are valid."
3069 (c-lang-defvar c-recognize-knr-p (c-lang-const c-recognize-knr-p))
3071 (c-lang-defconst c-recognize-typeless-decls
3072 "Non-nil means function declarations without return type should be
3073 recognized. That can introduce an ambiguity with parenthesized macro
3074 calls before a brace block. This setting does not affect declarations
3075 that are preceded by a declaration starting keyword, so
3076 e.g. `c-typeless-decl-kwds' may still be used when it's set to nil."
3078 (c c++ objc java) t)
3079 (c-lang-defvar c-recognize-typeless-decls
3080 (c-lang-const c-recognize-typeless-decls))
3082 (c-lang-defconst c-recognize-<>-arglists
3083 "Non-nil means C++ style template arglists should be handled. More
3084 specifically, this means a comma separated list of types or
3085 expressions surrounded by \"<\" and \">\". It's always preceded by an
3086 identifier or one of the keywords on `c-<>-type-kwds' or
3087 `c-<>-arglist-kwds'. If there's an identifier before then the whole
3088 expression is considered to be a type."
3089 t (or (consp (c-lang-const c-<>-type-kwds))
3090 (consp (c-lang-const c-<>-arglist-kwds)))
3091 java t) ; 2008-10-19. This is crude. The syntax for java
3092 ; generics is not yet coded in CC Mode.
3093 (c-lang-defvar c-recognize-<>-arglists (c-lang-const c-recognize-<>-arglists))
3095 (c-lang-defconst c-<>-notable-chars-re
3096 "A regexp matching any single character notable inside a <...> construct.
3097 This must include \"<\" and \">\", and should include \",\", and
3098 any character which cannot be valid inside such a construct.
3099 This is used in `c-forward-<>-arglist-recur' to try to detect
3100 sequences of tokens which cannot be a template/generic construct.
3101 When \"(\" is present, that defun will attempt to parse a
3102 parenthesized expression inside the template. When \")\" is
3103 present it will treat an unbalanced closing paren as a sign of
3104 the invalidity of the putative template construct."
3107 (c-lang-defvar c-<>-notable-chars-re (c-lang-const c-<>-notable-chars-re))
3109 (c-lang-defconst c-enums-contain-decls
3110 "Non-nil means that an enum structure can contain declarations."
3113 (c-lang-defvar c-enums-contain-decls (c-lang-const c-enums-contain-decls))
3115 (c-lang-defconst c-recognize-paren-inits
3116 "Non-nil means that parenthesis style initializers exist,
3117 i.e. constructs like
3121 in addition to the more classic
3126 (c-lang-defvar c-recognize-paren-inits (c-lang-const c-recognize-paren-inits))
3128 (c-lang-defconst c-recognize-paren-inexpr-blocks
3129 "Non-nil to recognize gcc style in-expression blocks,
3130 i.e. compound statements surrounded by parentheses inside expressions."
3133 (c-lang-defvar c-recognize-paren-inexpr-blocks
3134 (c-lang-const c-recognize-paren-inexpr-blocks))
3136 (c-lang-defconst c-opt-<>-arglist-start
3137 ;; Regexp matching the start of angle bracket arglists in languages
3138 ;; where `c-recognize-<>-arglists' is set. Does not exclude
3139 ;; keywords outside `c-<>-arglist-kwds'. The first submatch is
3140 ;; assumed to surround the preceding symbol. The whole match is
3141 ;; assumed to end directly after the opening "<".
3142 t (if (c-lang-const c-recognize-<>-arglists)
3144 (c-lang-const c-symbol-key)
3146 (c-lang-const c-syntactic-ws)
3148 (c-lang-defvar c-opt-<>-arglist-start (c-lang-const c-opt-<>-arglist-start))
3150 (c-lang-defconst c-opt-<>-arglist-start-in-paren
3151 ;; Regexp that in addition to `c-opt-<>-arglist-start' matches close
3152 ;; parens. The first submatch is assumed to surround
3153 ;; `c-opt-<>-arglist-start'.
3154 t (if (c-lang-const c-opt-<>-arglist-start)
3156 (c-lang-const c-opt-<>-arglist-start)
3158 (c-lang-defvar c-opt-<>-arglist-start-in-paren
3159 (c-lang-const c-opt-<>-arglist-start-in-paren))
3161 (c-lang-defconst c-opt-postfix-decl-spec-key
3162 ;; Regexp matching the beginning of a declaration specifier in the
3163 ;; region between the header and the body of a declaration.
3165 ;; TODO: This is currently not used uniformly; c++-mode and
3166 ;; java-mode each have their own ways of using it.
3169 (c-lang-const c-simple-ws) "*"
3170 "\\(virtual" (c-lang-const c-simple-ws) "+\\)?\\("
3171 (c-make-keywords-re nil (c-lang-const c-protection-kwds))
3172 "\\)" (c-lang-const c-simple-ws) "+"
3173 "\\(" (c-lang-const c-symbol-key) "\\)")
3174 java (c-make-keywords-re t (c-lang-const c-postfix-spec-kwds)))
3175 (c-lang-defvar c-opt-postfix-decl-spec-key
3176 (c-lang-const c-opt-postfix-decl-spec-key))
3178 (c-lang-defconst c-recognize-colon-labels
3179 "Non-nil if generic labels ending with \":\" should be recognized.
3180 That includes labels in code and access keys in classes. This does
3181 not apply to labels recognized by `c-label-kwds' and
3182 `c-opt-extra-label-key'."
3184 (c c++ objc java pike) t)
3185 (c-lang-defvar c-recognize-colon-labels
3186 (c-lang-const c-recognize-colon-labels))
3188 (c-lang-defconst c-label-prefix-re
3189 "Regexp like `c-decl-prefix-re' that matches any token that can precede
3190 a generic colon label. Not used if `c-recognize-colon-labels' is
3193 (c-lang-defvar c-label-prefix-re
3194 (c-lang-const c-label-prefix-re))
3196 (c-lang-defconst c-nonlabel-token-key
3197 "Regexp matching things that can't occur in generic colon labels,
3198 neither in a statement nor in a declaration context. The regexp is
3199 tested at the beginning of every sexp in a suspected label,
3200 i.e. before \":\". Only used if `c-recognize-colon-labels' is set."
3202 ;; All keywords except `c-label-kwds' and `c-protection-kwds'.
3203 (c-make-keywords-re t
3204 (c--set-difference (c-lang-const c-keywords)
3205 (append (c-lang-const c-label-kwds)
3206 (c-lang-const c-protection-kwds))
3207 :test 'string-equal)))
3208 ;; Don't allow string literals, except in AWK. Character constants are OK.
3209 (c objc java pike idl) (concat "\"\\|"
3210 (c-lang-const c-nonlabel-token-key))
3211 ;; Also check for open parens in C++, to catch member init lists in
3212 ;; constructors. We normally allow it so that macros with arguments
3214 c++ (concat "\\s(\\|\"\\|" (c-lang-const c-nonlabel-token-key)))
3215 (c-lang-defvar c-nonlabel-token-key (c-lang-const c-nonlabel-token-key))
3217 (c-lang-defconst c-nonlabel-token-2-key
3218 "Regexp matching things that can't occur two symbols before a colon in
3219 a label construct. This catches C++'s inheritance construct \"class foo
3220 : bar\". Only used if `c-recognize-colon-labels' is set."
3221 t "\\<\\>" ; matches nothing
3222 c++ (c-make-keywords-re t '("class")))
3223 (c-lang-defvar c-nonlabel-token-2-key (c-lang-const c-nonlabel-token-2-key))
3225 (c-lang-defconst c-opt-extra-label-key
3226 "Optional regexp matching labels.
3227 Normally, labels are detected according to `c-nonlabel-token-key',
3228 `c-decl-prefix-re' and `c-nonlabel-decl-prefix-re'. This regexp can
3229 be used if there are additional labels that aren't recognized that
3232 objc (c-make-keywords-re t (c-lang-const c-protection-kwds)))
3233 (c-lang-defvar c-opt-extra-label-key (c-lang-const c-opt-extra-label-key))
3235 (c-lang-defconst c-opt-friend-key
3236 ;; Regexp describing friend declarations classes, or nil in
3237 ;; languages that don't have such things.
3239 ;; TODO: Ought to use `c-prefix-spec-kwds-re' or similar, and the
3240 ;; template skipping isn't done properly. This will disappear soon.
3242 c++ (concat "friend" (c-lang-const c-simple-ws) "+"
3245 (c-lang-const c-simple-ws) "*"
3247 (c-lang-const c-simple-ws) "*"
3249 (c-lang-const c-simple-ws) "+")))
3250 (c-lang-defvar c-opt-friend-key (c-lang-const c-opt-friend-key))
3252 (c-lang-defconst c-opt-method-key
3253 ;; Special regexp to match the start of Objective-C methods. The
3254 ;; first submatch is assumed to end after the + or - key.
3257 ;; TODO: Ought to use a better method than anchoring on bol.
3260 (c-lang-const c-simple-ws) "*"
3261 (concat "\\(" ; Return type.
3263 (c-lang-const c-simple-ws) "*"
3265 "\\(" (c-lang-const c-symbol-key) "\\)"))
3266 (c-lang-defvar c-opt-method-key (c-lang-const c-opt-method-key))
3268 (c-lang-defconst c-type-decl-end-used
3269 ;; Must be set in buffers where the `c-type' text property might be
3270 ;; used with the value `c-decl-end'.
3272 ;; `c-decl-end' is used to mark the ends of labels and access keys
3273 ;; to make interactive refontification work better.
3274 t (or (c-lang-const c-recognize-colon-labels)
3275 (and (c-lang-const c-label-kwds) t))
3276 ;; `c-decl-end' is used to mark the end of the @-style directives in
3279 (c-lang-defvar c-type-decl-end-used (c-lang-const c-type-decl-end-used))
3281 (c-lang-defconst c-maybe-decl-faces
3282 "List of faces that might be put at the start of a type when
3283 `c-font-lock-declarations' runs. This must be evaluated (with `eval') at
3284 runtime to get the actual list of faces. This ensures that face name
3285 aliases in Emacs are resolved."
3288 c-reference-face-name
3289 font-lock-keyword-face)
3290 java (append (c-lang-const c-maybe-decl-faces)
3291 '(font-lock-preprocessor-face)))
3292 (c-lang-defvar c-maybe-decl-faces (c-lang-const c-maybe-decl-faces))
3295 ;;; Wrap up the `c-lang-defvar' system.
3297 ;; Compile in the list of language variables that has been collected
3298 ;; with the `c-lang-defvar' and `c-lang-setvar' macros. Note that the
3299 ;; first element of each is nil.
3300 (defconst c-lang-variable-inits (cc-eval-when-compile c-lang-variable-inits))
3301 (defconst c-emacs-variable-inits (cc-eval-when-compile c-emacs-variable-inits))
3303 ;; Make the `c-lang-setvar' variables buffer local in the current buffer.
3304 ;; These are typically standard emacs variables such as `comment-start'.
3305 (defmacro c-make-emacs-variables-local ()
3307 ,@(mapcar (lambda (init)
3308 `(make-local-variable ',(car init)))
3309 (cdr c-emacs-variable-inits))))
3311 (defun c-make-init-lang-vars-fun (mode)
3312 "Create a function that initializes all the language dependent variables
3315 This function should be evaluated at compile time, so that the
3316 function it returns is byte compiled with all the evaluated results
3317 from the language constants. Use the `c-init-language-vars' macro to
3318 accomplish that conveniently."
3320 (if (cc-bytecomp-is-compiling)
3321 ;; No need to byte compile this lambda since the byte compiler is
3322 ;; smart enough to detect the `funcall' construct in the
3323 ;; `c-init-language-vars' macro below and compile it all straight
3324 ;; into the function that contains `c-init-language-vars'.
3327 ;; This let sets up the context for `c-mode-var' and similar
3328 ;; that could be in the result from `c--macroexpand-all'.
3329 (let ((c-buffer-is-cc-mode ',mode)
3330 current-var source-eval)
3331 (c-make-emacs-variables-local)
3334 (if (eq c-version-sym ',c-version-sym)
3335 (setq ,@(let ((c-buffer-is-cc-mode mode)
3336 (c-lang-const-expansion 'immediate))
3337 ;; `c-lang-const' will expand to the evaluated
3338 ;; constant immediately in `c--macroexpand-all'
3342 `(current-var ',(car init)
3343 ,(car init) ,(c--macroexpand-all
3345 ;; Note: The following `append' copies the
3346 ;; first argument. That list is small, so
3347 ;; this doesn't matter too much.
3348 (append (cdr c-emacs-variable-inits)
3349 (cdr c-lang-variable-inits)))))
3351 ;; This diagnostic message isn't useful for end
3352 ;; users, so it's disabled.
3353 ;;(unless (get ',mode 'c-has-warned-lang-consts)
3354 ;; (message ,(concat "%s compiled with CC Mode %s "
3355 ;; "but loaded with %s - evaluating "
3356 ;; "language constants from source")
3357 ;; ',mode ,c-version c-version)
3358 ;; (put ',mode 'c-has-warned-lang-consts t))
3360 (setq source-eval t)
3361 (let ((init ',(append (cdr c-emacs-variable-inits)
3362 (cdr c-lang-variable-inits))))
3363 (dolist (var-init init)
3364 (setq current-var (car var-init))
3365 (set (car var-init) (eval (cadr var-init))))))
3369 (message "Eval error in the `c-lang-defvar' or `c-lang-setvar' for `%s'%s: %S"
3373 (fallback source eval - %s compiled with CC Mode %s but loaded with %s)"
3374 ',mode ,c-version c-version)
3377 (signal (car err) (cdr err)))))))
3379 ;; Being evaluated from source. Always use the dynamic method to
3380 ;; work well when `c-lang-defvar's in this file are reevaluated
3384 (let ((c-buffer-is-cc-mode ',mode)
3385 (init (append (cdr c-emacs-variable-inits)
3386 (cdr c-lang-variable-inits)))
3388 (c-make-emacs-variables-local)
3391 (dolist (var-init init)
3392 (setq current-var (car var-init))
3393 (set (car var-init) (eval (cadr var-init))))
3398 "Eval error in the `c-lang-defvar' or `c-lang-setver' for `%s' (source eval): %S"
3400 (signal (car err) (cdr err)))))))
3403 (defmacro c-init-language-vars (mode)
3404 "Initialize all the language dependent variables for the given mode.
3405 This macro is expanded at compile time to a form tailored for the mode
3406 in question, so MODE must be a constant. Therefore MODE is not
3407 evaluated and should not be quoted."
3408 `(funcall ,(c-make-init-lang-vars-fun mode)))
3411 (cc-provide 'cc-langs)
3414 ;; indent-tabs-mode: t
3417 ;;; cc-langs.el ends here