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-raw-strings
478 c-before-change-check-<>-operators
480 c-invalidate-macro-cache)
481 (c objc) '(c-extend-region-for-CPP
483 c-invalidate-macro-cache)
484 ;; java 'c-before-change-check-<>-operators
485 awk 'c-awk-record-region-clear-NL)
486 (c-lang-defvar c-get-state-before-change-functions
487 (let ((fs (c-lang-const c-get-state-before-change-functions)))
491 "If non-nil, a list of functions called from c-before-change-hook.
492 Typically these will record enough state to allow
493 `c-before-font-lock-function' to extend the region to fontify,
494 and may do such things as removing text-properties which must be
497 These functions will be run in the order given. Each of them
498 takes 2 parameters, the BEG and END supplied to every
499 before-change function; on entry, the buffer will have been
500 widened and match-data will have been saved; point is undefined
501 on both entry and exit; the return value is ignored.
503 The functions are called even when font locking isn't enabled.
505 When the mode is initialized, the functions are called with
506 parameters \(point-min) and \(point-max).")
508 (c-lang-defconst c-before-font-lock-functions
509 ;; For documentation see the following c-lang-defvar of the same name.
510 ;; The value here may be a list of functions or a single function.
511 t 'c-change-expand-fl-region
512 (c objc) '(c-extend-font-lock-region-for-macros
513 c-neutralize-syntax-in-and-mark-CPP
514 c-change-expand-fl-region)
515 c++ '(c-extend-font-lock-region-for-macros
516 c-after-change-re-mark-raw-strings
517 c-neutralize-syntax-in-and-mark-CPP
518 c-restore-<>-properties
519 c-change-expand-fl-region)
520 java '(c-restore-<>-properties
521 c-change-expand-fl-region)
522 awk 'c-awk-extend-and-syntax-tablify-region)
523 (c-lang-defvar c-before-font-lock-functions
524 (let ((fs (c-lang-const c-before-font-lock-functions)))
528 "If non-nil, a list of functions called just before font locking.
529 Typically they will extend the region about to be fontified \(see
530 below) and will set `syntax-table' text properties on the region.
532 These functions will be run in the order given. Each of them
533 takes 3 parameters, the BEG, END, and OLD-LEN supplied to every
534 after-change function; point is undefined on both entry and exit;
535 on entry, the buffer will have been widened and match-data will
536 have been saved; the return value is ignored.
538 The functions may extend the region to be fontified by setting the
539 buffer local variables c-new-BEG and c-new-END.
541 The functions are called even when font locking is disabled.
543 When the mode is initialized, these functions are called with
544 parameters \(point-min), \(point-max) and <buffer size>.")
546 (c-lang-defconst c-before-context-fontification-functions
547 t 'c-context-expand-fl-region
549 ;; For documentation see the following c-lang-defvar of the same name.
550 ;; The value here may be a list of functions or a single function.
551 (c-lang-defvar c-before-context-fontification-functions
552 (let ((fs (c-lang-const c-before-context-fontification-functions)))
556 "If non-nil, a list of functions called just before context (or
557 other non-change) fontification is done. Typically they will
560 These functions will be run in the order given. Each of them
561 takes 2 parameters, the BEG and END of the region to be
562 fontified. Point is undefined on both entry and exit. On entry,
563 the buffer will have been widened and match-data will have been
564 saved; the return value is a cons of the adjusted
565 region, (NEW-BEG . NEW-END).")
568 ;;; Syntactic analysis ("virtual semicolons") for line-oriented languages (AWK).
569 (c-lang-defconst c-at-vsemi-p-fn
570 "Contains a function \"Is there a virtual semicolon at POS or point?\".
571 Such a function takes one optional parameter, a buffer position (defaults to
572 point), and returns nil or t. This variable contains nil for languages which
573 don't have EOL terminated statements. "
575 (c c++ objc) 'c-at-macro-vsemi-p
576 awk 'c-awk-at-vsemi-p)
577 (c-lang-defvar c-at-vsemi-p-fn (c-lang-const c-at-vsemi-p-fn))
579 (c-lang-defconst c-vsemi-status-unknown-p-fn
580 "Contains a function \"are we unsure whether there is a virtual semicolon on this line?\".
581 The (admittedly kludgy) purpose of such a function is to prevent an infinite
582 recursion in c-beginning-of-statement-1 when point starts at a `while' token.
583 The function MUST NOT UNDER ANY CIRCUMSTANCES call c-beginning-of-statement-1,
584 even indirectly. This variable contains nil for languages which don't have
585 EOL terminated statements."
587 (c c++ objc) 'c-macro-vsemi-status-unknown-p
588 awk 'c-awk-vsemi-status-unknown-p)
589 (c-lang-defvar c-vsemi-status-unknown-p-fn
590 (c-lang-const c-vsemi-status-unknown-p-fn))
593 ;;; Lexer-level syntax (identifiers, tokens etc).
595 (c-lang-defconst c-has-bitfields
596 "Whether the language has bitfield declarations."
599 (c-lang-defvar c-has-bitfields (c-lang-const c-has-bitfields))
601 (c-lang-defconst c-modified-constant
602 "Regexp that matches a “modified” constant literal such as \"L\\='a\\='\",
603 a “long character”. In particular, this recognizes forms of constant
604 which `c-backward-sexp' needs to be called twice to move backwards over."
606 (c c++ objc) "L'\\([^\\'\t\f\n\r]\\|\\\\.\\)'")
607 ;; FIXME!!! Extend this to cover strings, if needed. 2008-04-11
608 (c-lang-defvar c-modified-constant (c-lang-const c-modified-constant))
610 (c-lang-defconst c-symbol-start
611 "Regexp that matches the start of a symbol, i.e. any identifier or
612 keyword. It's unspecified how far it matches. Does not contain a \\|
613 operator at the top level."
614 t (concat "[" c-alpha "_]")
615 java (concat "[" c-alpha "_@]")
616 objc (concat "[" c-alpha "_@]")
617 pike (concat "[" c-alpha "_`]"))
618 (c-lang-defvar c-symbol-start (c-lang-const c-symbol-start))
620 (c-lang-defconst c-symbol-chars
621 "Set of characters that can be part of a symbol.
622 This is of the form that fits inside [ ] in a regexp."
623 ;; Pike note: With the backquote identifiers this would include most
624 ;; operator chars too, but they are handled with other means instead.
625 t (concat c-alnum "_$")
626 objc (concat c-alnum "_$@"))
627 (c-lang-defvar c-symbol-chars (c-lang-const c-symbol-chars))
629 (c-lang-defconst c-symbol-char-key
630 "Regexp matching a sequence of at least one identifier character."
631 t (concat "[" (c-lang-const c-symbol-chars) "]+"))
632 (c-lang-defvar c-symbol-char-key (c-lang-const c-symbol-char-key))
634 (c-lang-defconst c-symbol-key
635 "Regexp matching identifiers and keywords (with submatch 0). Assumed
636 to match if `c-symbol-start' matches on the same position."
637 t (concat (c-lang-const c-symbol-start)
638 "[" (c-lang-const c-symbol-chars) "]*")
640 ;; Use the value from C here since the operator backquote is
641 ;; covered by the other alternative.
642 (c-lang-const c-symbol-key c)
644 (c-make-keywords-re nil
645 (c-lang-const c-overloadable-operators))))
646 (c-lang-defvar c-symbol-key (c-lang-const c-symbol-key))
648 (c-lang-defconst c-symbol-key-depth
649 ;; Number of regexp grouping parens in `c-symbol-key'.
650 t (regexp-opt-depth (c-lang-const c-symbol-key)))
652 (c-lang-defconst c-nonsymbol-chars
653 "This is the set of chars that can't be part of a symbol, i.e. the
654 negation of `c-symbol-chars'."
655 t (concat "^" (c-lang-const c-symbol-chars)))
656 (c-lang-defvar c-nonsymbol-chars (c-lang-const c-nonsymbol-chars))
658 (c-lang-defconst c-nonsymbol-key
659 "Regexp that matches any character that can't be part of a symbol.
660 It's usually appended to other regexps to avoid matching a prefix.
661 It's assumed to not contain any submatchers."
662 ;; The same thing regarding Unicode identifiers applies here as to
664 t (concat "[" (c-lang-const c-nonsymbol-chars) "]"))
666 (c-lang-defconst c-identifier-ops
667 "The operators that make up fully qualified identifiers. nil in
668 languages that don't have such things. See `c-operators' for a
669 description of the format. Binary operators can concatenate symbols,
670 e.g. \"::\" in \"A::B::C\". Prefix operators can precede identifiers,
671 e.g. \"~\" in \"~A::B\". Other types of operators aren't supported.
673 This value is by default merged into `c-operators'."
675 c++ '((prefix "~" "??-" "compl")
678 ;; Java has "." to concatenate identifiers but it's also used for
679 ;; normal indexing. There's special code in the Java font lock
680 ;; rules to fontify qualified identifiers based on the standard
681 ;; naming conventions. We still define "." here to make
682 ;; `c-forward-name' move over as long names as possible which is
683 ;; necessary to e.g. handle throws clauses correctly.
684 java '((left-assoc "."))
685 idl '((left-assoc "::")
687 pike '((left-assoc "::")
691 (c-lang-defconst c-opt-identifier-concat-key
692 ;; Appendable adorned regexp matching the operators that join
693 ;; symbols to fully qualified identifiers, or nil in languages that
694 ;; don't have such things.
696 ;; This was a docstring constant in 5.30. It still works but is now
697 ;; considered internal - change `c-identifier-ops' instead.
698 t (let ((ops (c-filter-ops (c-lang-const c-identifier-ops)
699 '(left-assoc right-assoc)
702 (c-make-keywords-re 'appendable ops))))
703 (c-lang-defvar c-opt-identifier-concat-key
704 (c-lang-const c-opt-identifier-concat-key)
707 (c-lang-defconst c-opt-identifier-concat-key-depth
708 ;; Number of regexp grouping parens in `c-opt-identifier-concat-key'.
709 t (regexp-opt-depth (c-lang-const c-opt-identifier-concat-key)))
711 (c-lang-defconst c-opt-identifier-prefix-key
712 ;; Appendable adorned regexp matching operators that might precede
713 ;; an identifier and that are part of the identifier in that case.
714 ;; nil in languages without such things.
715 t (let ((ops (c-filter-ops (c-lang-const c-identifier-ops)
719 (c-make-keywords-re 'appendable ops))))
721 (c-lang-defconst c-after-id-concat-ops
722 "Operators that can occur after a binary operator on `c-identifier-ops'
723 in identifiers. nil in languages that don't have such things.
725 Operators here should also have appropriate entries in `c-operators' -
726 it's not taken care of by default."
728 ;; '~' for destructors in C++, '*' for member pointers.
730 ;; In Java we recognize '*' to deal with "foo.bar.*" that can occur
731 ;; in import declarations. (This will also match bogus things like
732 ;; "foo.*bar" but we don't bother.)
735 (c-lang-defconst c-opt-after-id-concat-key
736 ;; Regexp that must match the token after
737 ;; `c-opt-identifier-concat-key' for it to be considered an
738 ;; identifier concatenation operator (which e.g. causes the
739 ;; preceding identifier to be fontified as a reference). Assumed to
740 ;; be a string if `c-opt-identifier-concat-key' is.
742 ;; This was a docstring constant in 5.30. It still works but is now
743 ;; considered internal - change `c-after-id-concat-ops' instead.
744 t (concat (c-lang-const c-symbol-start)
745 (if (c-lang-const c-after-id-concat-ops)
746 (concat "\\|" (c-make-keywords-re 'appendable
747 (c-lang-const c-after-id-concat-ops)))
750 (c-lang-defconst c-identifier-start
751 "Regexp that matches the start of an (optionally qualified) identifier.
752 It should also match all keywords. It's unspecified how far it
754 t (concat (c-lang-const c-symbol-start)
755 (if (c-lang-const c-opt-identifier-prefix-key)
757 (c-lang-const c-opt-identifier-prefix-key))
759 (c-lang-defvar c-identifier-start (c-lang-const c-identifier-start))
761 (c-lang-defconst c-identifier-key
762 "Regexp matching a fully qualified identifier, like \"A::B::c\" in
763 C++. It does not recognize the full range of syntactic whitespace
764 between the tokens; `c-forward-name' has to be used for that. It
765 should also not match identifiers containing parenthesis groupings,
766 e.g. identifiers with template arguments such as \"A<X,Y>\" in C++."
767 ;; This regexp is more complex than strictly necessary to ensure
768 ;; that it can be matched with a minimum of backtracking.
769 t (concat (if (c-lang-const c-opt-identifier-prefix-key)
772 (c-lang-const c-opt-identifier-prefix-key)
773 (c-lang-const c-simple-ws) "*"
776 "\\(" (c-lang-const c-symbol-key) "\\)"
777 (if (c-lang-const c-opt-identifier-concat-key)
780 (c-lang-const c-simple-ws) "*"
781 (c-lang-const c-opt-identifier-concat-key)
782 (c-lang-const c-simple-ws) "*"
783 (if (c-lang-const c-after-id-concat-ops)
786 (c-make-keywords-re 'appendable
787 (c-lang-const c-after-id-concat-ops))
789 ;; For flexibility, consider the symbol match
790 ;; optional if we've hit a
791 ;; `c-after-id-concat-ops' operator. This is
792 ;; also necessary to handle the "*" that can
793 ;; end import declaration identifiers in Java.
795 (c-lang-const c-simple-ws) "*"
796 "\\(" (c-lang-const c-symbol-key) "\\)"
799 "\\(" (c-lang-const c-symbol-key) "\\)"
801 (concat "\\(" (c-lang-const c-symbol-key) "\\)"))
804 (c-lang-defvar c-identifier-key (c-lang-const c-identifier-key))
806 (c-lang-defconst c-identifier-last-sym-match
807 ;; This was a docstring constant in 5.30 but it's no longer used.
808 ;; It's only kept to avoid breaking third party code.
810 ;; Used to identify the submatch in `c-identifier-key' that
811 ;; surrounds the last symbol in the qualified identifier. It's a
812 ;; list of submatch numbers, of which the first that has a match is
813 ;; taken. It's assumed that at least one does when the regexp has
817 (c-lang-defconst c-string-escaped-newlines
818 "Set if the language support backslash escaped newlines inside string
822 (c-lang-defvar c-string-escaped-newlines
823 (c-lang-const c-string-escaped-newlines))
825 (c-lang-defconst c-multiline-string-start-char
826 "Set if the language supports multiline string literals without escaped
827 newlines. If t, all string literals are multiline. If a character,
828 only literals where the open quote is immediately preceded by that
829 literal are multiline."
832 (c-lang-defvar c-multiline-string-start-char
833 (c-lang-const c-multiline-string-start-char))
835 (c-lang-defconst c-opt-cpp-symbol
836 "The symbol which starts preprocessor constructs when in the margin."
839 (c-lang-defvar c-opt-cpp-symbol (c-lang-const c-opt-cpp-symbol))
841 (c-lang-defconst c-opt-cpp-prefix
842 "Regexp matching the prefix of a cpp directive in the languages that
843 normally use that macro preprocessor. Tested at bol or at boi.
844 Assumed to not contain any submatches or \\| operators."
845 ;; TODO (ACM, 2005-04-01). Amend the following to recognize escaped NLs;
846 ;; amend all uses of c-opt-cpp-prefix which count regexp-depth.
849 (c-lang-defvar c-opt-cpp-prefix (c-lang-const c-opt-cpp-prefix))
851 (c-lang-defconst c-anchored-cpp-prefix
852 "Regexp matching the prefix of a cpp directive anchored to BOL,
853 in the languages that have a macro preprocessor."
854 t "^\\s *\\(#\\)\\s *"
856 (c-lang-defvar c-anchored-cpp-prefix (c-lang-const c-anchored-cpp-prefix))
858 (c-lang-defconst c-opt-cpp-start
859 "Regexp matching the prefix of a cpp directive including the directive
860 name, or nil in languages without preprocessor support. The first
861 submatch surrounds the directive name."
862 t (if (c-lang-const c-opt-cpp-prefix)
863 (concat (c-lang-const c-opt-cpp-prefix)
864 "\\([" c-alnum "]+\\)"))
865 ;; Pike, being a scripting language, recognizes hash-bangs too.
866 pike (concat (c-lang-const c-opt-cpp-prefix)
867 "\\([" c-alnum "]+\\|!\\)"))
868 (c-lang-defvar c-opt-cpp-start (c-lang-const c-opt-cpp-start))
870 (c-lang-defconst c-cpp-message-directives
871 "List of cpp directives (without the prefix) that are followed by a
873 t (if (c-lang-const c-opt-cpp-prefix)
875 (c c++ objc pike) '("error" "warning"))
877 (c-lang-defconst c-cpp-include-directives
878 "List of cpp directives (without the prefix) that are followed by a
879 file name in angle brackets or quotes."
880 t (if (c-lang-const c-opt-cpp-prefix)
882 objc '("include" "import"))
884 (c-lang-defconst c-opt-cpp-macro-define
885 "Cpp directive (without the prefix) that is followed by a macro
886 definition, or nil if the language doesn't have any."
887 t (if (c-lang-const c-opt-cpp-prefix)
889 (c-lang-defvar c-opt-cpp-macro-define
890 (c-lang-const c-opt-cpp-macro-define))
892 (c-lang-defconst c-opt-cpp-macro-define-start
893 ;; Regexp matching everything up to the macro body of a cpp define, or the
894 ;; end of the logical line if there is none. Submatch 1 is the name of the
895 ;; macro. Set if c-opt-cpp-macro-define is.
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)
899 "[ \t]+\\(\\(\\sw\\|_\\)+\\)\\(([^)]*)\\)?"
901 "\\([ \t]\\|\\\\\n\\)*")))
902 (c-lang-defvar c-opt-cpp-macro-define-start
903 (c-lang-const c-opt-cpp-macro-define-start))
905 (c-lang-defconst c-opt-cpp-macro-define-id
906 ;; Regexp matching everything up to the end of the identifier defined
908 t (if (c-lang-const c-opt-cpp-macro-define)
909 (concat (c-lang-const c-opt-cpp-prefix) ; #
910 (c-lang-const c-opt-cpp-macro-define) ; define
911 "[ \t]+\\(\\sw\\|_\\)+")))
912 (c-lang-defvar c-opt-cpp-macro-define-id
913 (c-lang-const c-opt-cpp-macro-define-id))
915 (c-lang-defconst c-cpp-expr-directives
916 "List of cpp directives (without the prefix) that are followed by an
918 t (if (c-lang-const c-opt-cpp-prefix)
921 (c-lang-defconst c-cpp-expr-intro-re
922 "Regexp which matches the start of a CPP directive which contains an
923 expression, or nil if there aren't any in the language."
924 t (if (c-lang-const c-cpp-expr-directives)
926 (c-lang-const c-opt-cpp-prefix)
927 (c-make-keywords-re t (c-lang-const c-cpp-expr-directives)))))
928 (c-lang-defvar c-cpp-expr-intro-re
929 (c-lang-const c-cpp-expr-intro-re))
931 (c-lang-defconst c-cpp-expr-functions
932 "List of functions in cpp expressions."
933 t (if (c-lang-const c-opt-cpp-prefix)
935 pike '("defined" "efun" "constant"))
937 (c-lang-defconst c-assignment-operators
938 "List of all assignment operators."
939 t '("=" "*=" "/=" "%=" "+=" "-=" ">>=" "<<=" "&=" "^=" "|=")
940 java (append (c-lang-const c-assignment-operators)
942 c++ (append (c-lang-const c-assignment-operators)
943 '("and_eq" "or_eq" "xor_eq" "??!=" "??'="))
946 (c-lang-defconst c-operators
947 "List describing all operators, along with their precedence and
948 associativity. The order in the list corresponds to the precedence of
949 the operators: The operators in each element are a group with the same
950 precedence, and the group has higher precedence than the groups in all
951 following elements. The car of each element describes the type of the
952 operator group, and the cdr is a list of the operator tokens in it.
953 The operator group types are:
955 `prefix' Unary prefix operators.
956 `postfix' Unary postfix operators.
958 Unary postfix operators if and only if the chars have
960 `left-assoc' Binary left associative operators (i.e. a+b+c means (a+b)+c).
961 `right-assoc' Binary right associative operators (i.e. a=b=c means a=(b=c)).
962 `right-assoc-sequence'
963 Right associative operator that constitutes of a
964 sequence of tokens that separate expressions. All the
965 tokens in the group are in this case taken as
966 describing the sequence in one such operator, and the
967 order between them is therefore significant.
969 Operators containing a character with paren syntax are taken to match
970 with a corresponding open/close paren somewhere else. A postfix
971 operator with close paren syntax is taken to end a postfix expression
972 started somewhere earlier, rather than start a new one at point. Vice
973 versa for prefix operators with open paren syntax.
975 Note that operators like \".\" and \"->\" which in language references
976 often are described as postfix operators are considered binary here,
977 since CC Mode treats every identifier as an expression."
979 ;; There's currently no code in CC Mode that exploits all the info
980 ;; in this variable; precedence, associativity etc are present as a
981 ;; preparation for future work.
983 ;; FIXME!!! C++11's "auto" operators "=" and "->" need to go in here
984 ;; somewhere. 2012-03-24.
987 ,@(when (c-lang-const c-opt-cpp-prefix)
989 ,@(when (c-major-mode-is '(c-mode c++-mode))
992 ,@(when (c-major-mode-is '(c-mode c++-mode))
993 '("%:%:" "??=??=")))))
996 ,@(c-lang-const c-identifier-ops)
997 ,@(cond ((or (c-major-mode-is 'c++-mode) (c-major-mode-is 'java-mode))
998 `((postfix-if-paren "<" ">"))) ; Templates.
999 ((c-major-mode-is 'pike-mode)
1000 `((prefix "global" "predef")))
1001 ((c-major-mode-is 'java-mode)
1002 `((prefix "super"))))
1005 ,@(when (c-major-mode-is 'c++-mode)
1006 ;; The following need special treatment.
1007 `((prefix "dynamic_cast" "static_cast"
1008 "reinterpret_cast" "const_cast" "typeid"
1011 ,@(unless (c-major-mode-is 'java-mode)
1013 (postfix "++" "--" "[" "]" "(" ")"
1014 ,@(when (c-major-mode-is '(c-mode c++-mode))
1015 '("<:" ":>" "??(" "??)")))
1018 (prefix "++" "--" "+" "-" "!" "~"
1019 ,@(when (c-major-mode-is 'c++-mode) '("not" "compl"))
1020 ,@(when (c-major-mode-is '(c-mode c++-mode))
1021 '("*" "&" "sizeof" "??-"))
1022 ,@(when (c-major-mode-is 'objc-mode)
1023 '("@selector" "@protocol" "@encode"))
1024 ;; The following need special treatment.
1025 ,@(cond ((c-major-mode-is 'c++-mode)
1027 ((c-major-mode-is 'java-mode)
1029 ((c-major-mode-is 'pike-mode)
1030 '("class" "lambda" "catch" "throw" "gauge")))
1032 ,@(when (c-major-mode-is 'pike-mode)
1033 '("[" "]"))) ; Type cast.
1035 ;; Member selection.
1036 ,@(when (c-major-mode-is 'c++-mode)
1037 `((left-assoc ".*" "->*")))
1040 (left-assoc "*" "/" "%")
1043 (left-assoc "+" "-")
1046 (left-assoc "<<" ">>"
1047 ,@(when (c-major-mode-is 'java-mode)
1051 (left-assoc "<" ">" "<=" ">="
1052 ,@(when (c-major-mode-is 'java-mode)
1056 (left-assoc "==" "!="
1057 ,@(when (c-major-mode-is 'c++-mode) '("not_eq")))
1061 ,@(when (c-major-mode-is 'c++-mode) '("bitand")))
1063 ;; Bitwise exclusive or.
1065 ,@(when (c-major-mode-is '(c-mode c++-mode))
1067 ,@(when (c-major-mode-is 'c++-mode) '("xor")))
1071 ,@(when (c-major-mode-is '(c-mode c++-mode))
1073 ,@(when (c-major-mode-is 'c++-mode) '("bitor")))
1077 ,@(when (c-major-mode-is 'c++-mode) '("and")))
1081 ,@(when (c-major-mode-is '(c-mode c++-mode))
1083 ,@(when (c-major-mode-is 'c++-mode) '("or")))
1086 (right-assoc-sequence "?" ":")
1089 (right-assoc ,@(c-lang-const c-assignment-operators))
1092 ,@(when (c-major-mode-is 'c++-mode)
1093 '((prefix "throw")))
1098 ;; IDL got its own definition since it has a much smaller operator
1099 ;; set than the other languages.
1100 idl `(;; Preprocessor.
1104 ,@(c-lang-const c-identifier-ops)
1106 (prefix "+" "-" "~")
1108 (left-assoc "*" "/" "%")
1110 (left-assoc "+" "-")
1112 (left-assoc "<<" ">>")
1120 (c-lang-defconst c-operator-list
1121 ;; The operators as a flat list (without duplicates).
1122 t (c-filter-ops (c-lang-const c-operators) t t))
1124 (c-lang-defconst c-overloadable-operators
1125 "List of the operators that are overloadable, in their \"identifier
1126 form\". See also `c-op-identifier-prefix'."
1128 c++ '("new" "delete" ;; Can be followed by "[]" but we ignore that.
1130 "^" "??'" "xor" "&" "bitand" "|" "??!" "bitor" "~" "??-" "compl"
1131 "!" "=" "<" ">" "+=" "-=" "*=" "/=" "%=" "^="
1132 "??'=" "xor_eq" "&=" "and_eq" "|=" "??!=" "or_eq"
1133 "<<" ">>" ">>=" "<<=" "==" "!=" "not_eq" "<=" ">="
1134 "&&" "and" "||" "??!??!" "or" "++" "--" "," "->*" "->"
1135 "()" "[]" "<::>" "??(??)")
1136 ;; These work like identifiers in Pike.
1137 pike '("`+" "`-" "`&" "`|" "`^" "`<<" "`>>" "`*" "`/" "`%" "`~"
1138 "`==" "`<" "`>" "`!" "`[]" "`[]=" "`->" "`->=" "`()" "``+"
1139 "``-" "``&" "``|" "``^" "``<<" "``>>" "``*" "``/" "``%"
1142 (c-lang-defconst c-overloadable-operators-regexp
1143 ;; Regexp tested after an "operator" token in C++.
1145 c++ (c-make-keywords-re nil (c-lang-const c-overloadable-operators)))
1146 (c-lang-defvar c-overloadable-operators-regexp
1147 (c-lang-const c-overloadable-operators-regexp))
1149 (c-lang-defconst c-opt-op-identifier-prefix
1150 "Regexp matching the token before the ones in
1151 `c-overloadable-operators' when operators are specified in their
1152 \"identifier form\". This typically matches \"operator\" in C++ where
1153 operator functions are specified as e.g. \"operator +\". It's nil in
1154 languages without operator functions or where the complete operator
1155 identifier is listed in `c-overloadable-operators'.
1157 This regexp is assumed to not match any non-operator identifier."
1159 c++ (c-make-keywords-re t '("operator")))
1160 (c-lang-defvar c-opt-op-identifier-prefix
1161 (c-lang-const c-opt-op-identifier-prefix))
1163 ;; Note: the following alias is an old name which was a mis-spelling. It has
1164 ;; been corrected above and throughout cc-engine.el. It will be removed at
1165 ;; some release very shortly in the future. ACM, 2006-04-14.
1166 (defvaralias 'c-opt-op-identitier-prefix 'c-opt-op-identifier-prefix)
1167 (make-obsolete-variable 'c-opt-op-identitier-prefix 'c-opt-op-identifier-prefix
1168 "CC Mode 5.31.4, 2006-04-14")
1170 (c-lang-defconst c-other-op-syntax-tokens
1171 "List of the tokens made up of characters in the punctuation or
1172 parenthesis syntax classes that have uses other than as expression
1174 t '("{" "}" "(" ")" "[" "]" ";" ":" "," "=" "/*" "*/" "//")
1175 (c c++ pike) (append '("#" "##" ; Used by cpp.
1177 (c-lang-const c-other-op-syntax-tokens))
1178 (c c++) (append '("*") (c-lang-const c-other-op-syntax-tokens))
1179 c++ (append '("&" "<%" "%>" "<:" ":>" "%:" "%:%:")
1180 (c-lang-const c-other-op-syntax-tokens))
1181 objc (append '("#" "##" ; Used by cpp.
1183 (c-lang-const c-other-op-syntax-tokens))
1184 idl (append '("#" "##") ; Used by cpp.
1185 (c-lang-const c-other-op-syntax-tokens))
1186 pike (append '("..")
1187 (c-lang-const c-other-op-syntax-tokens)
1188 (c-lang-const c-overloadable-operators))
1189 awk '("{" "}" "(" ")" "[" "]" ";" "," "=" "/"))
1191 (c-lang-defconst c-all-op-syntax-tokens
1192 ;; List of all tokens in the punctuation and parenthesis syntax
1194 t (c--delete-duplicates (append (c-lang-const c-other-op-syntax-tokens)
1195 (c-lang-const c-operator-list))
1196 :test 'string-equal))
1198 (c-lang-defconst c-nonsymbol-token-char-list
1199 ;; List containing all chars not in the word, symbol or
1200 ;; syntactically irrelevant syntax classes, i.e. all punctuation,
1201 ;; parenthesis and string delimiter chars.
1202 t (c-with-syntax-table (c-lang-const c-mode-syntax-table)
1203 ;; Only go through the chars in the printable ASCII range. No
1204 ;; language so far has 8-bit or widestring operators.
1205 (let (list (char 32))
1207 (or (memq (char-syntax char) '(?w ?_ ?< ?> ?\ ))
1208 (setq list (cons (c-int-to-char char) list)))
1209 (setq char (1+ char)))
1212 (c-lang-defconst c-nonsymbol-token-regexp
1213 ;; Regexp matching all tokens in the punctuation and parenthesis
1214 ;; syntax classes. Note that this also matches ".", which can start
1216 t (c-make-keywords-re nil
1217 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1219 "\\`\\(\\s.\\)+\\'")))
1220 (c-lang-defvar c-nonsymbol-token-regexp
1221 (c-lang-const c-nonsymbol-token-regexp))
1223 (c-lang-defconst c-assignment-op-regexp
1224 ;; Regexp matching all assignment operators and only them. The
1225 ;; beginning of the first submatch is used to detect the end of the
1226 ;; token, along with the end of the whole match.
1227 t (if (c-lang-const c-assignment-operators)
1229 ;; Need special case for "=" since it's a prefix of "==".
1232 (c-make-keywords-re nil
1233 (c--set-difference (c-lang-const c-assignment-operators)
1235 :test 'string-equal)))
1237 (c-lang-defvar c-assignment-op-regexp
1238 (c-lang-const c-assignment-op-regexp))
1240 (c-lang-defconst c-:$-multichar-token-regexp
1241 ;; Regexp matching all tokens ending in ":" which are longer than one char.
1242 ;; Currently (2016-01-07) only used in C++ Mode.
1243 t (c-make-keywords-re nil
1244 (c-filter-ops (c-lang-const c-operators) t ".+:$")))
1245 (c-lang-defvar c-:$-multichar-token-regexp
1246 (c-lang-const c-:$-multichar-token-regexp))
1248 (c-lang-defconst c-<>-multichar-token-regexp
1249 ;; Regexp matching all tokens containing "<" or ">" which are longer
1251 t (c-make-keywords-re nil
1252 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1255 (c-lang-defvar c-<>-multichar-token-regexp
1256 (c-lang-const c-<>-multichar-token-regexp))
1258 (c-lang-defconst c-<-op-cont-regexp
1259 ;; Regexp matching the second and subsequent characters of all
1260 ;; multicharacter tokens that begin with "<".
1261 t (c-make-keywords-re nil
1262 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1265 (lambda (op) (substring op 1)))))
1266 (c-lang-defvar c-<-op-cont-regexp (c-lang-const c-<-op-cont-regexp))
1268 (c-lang-defconst c->-op-cont-tokens
1269 ;; A list of second and subsequent characters of all multicharacter tokens
1270 ;; that begin with ">".
1271 t (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1274 (lambda (op) (substring op 1)))
1275 java (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1277 "\\`>[^>]\\|\\`>>[^>]"
1278 (lambda (op) (substring op 1))))
1280 (c-lang-defconst c->-op-cont-regexp
1281 ;; Regexp matching the second and subsequent characters of all
1282 ;; multicharacter tokens that begin with ">".
1283 t (c-make-keywords-re nil (c-lang-const c->-op-cont-tokens)))
1284 (c-lang-defvar c->-op-cont-regexp (c-lang-const c->-op-cont-regexp))
1286 (c-lang-defconst c->-op-without->-cont-regexp
1287 ;; Regexp matching the second and subsequent characters of all
1288 ;; multicharacter tokens that begin with ">" except for those beginning with
1290 t (c-make-keywords-re nil
1292 (c-lang-const c->-op-cont-tokens)
1293 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1296 (lambda (op) (substring op 1)))
1297 :test 'string-equal)))
1298 (c-lang-defvar c->-op-without->-cont-regexp
1299 (c-lang-const c->-op-without->-cont-regexp))
1301 (c-lang-defconst c-multichar->-op-not->>-regexp
1302 ;; Regexp matching multichar tokens containing ">", except ">>"
1303 t (c-make-keywords-re nil
1305 (c-filter-ops (c-lang-const c-all-op-syntax-tokens)
1308 (c-lang-defvar c-multichar->-op-not->>-regexp
1309 (c-lang-const c-multichar->-op-not->>-regexp))
1311 (c-lang-defconst c-stmt-delim-chars
1312 ;; The characters that should be considered to bound statements. To
1313 ;; optimize `c-crosses-statement-barrier-p' somewhat, it's assumed to
1314 ;; begin with "^" to negate the set. If ? : operators should be
1315 ;; detected then the string must end with "?:".
1317 (c-lang-defvar c-stmt-delim-chars (c-lang-const c-stmt-delim-chars))
1319 (c-lang-defconst c-stmt-delim-chars-with-comma
1320 ;; Variant of `c-stmt-delim-chars' that additionally contains ','.
1322 (c-lang-defvar c-stmt-delim-chars-with-comma
1323 (c-lang-const c-stmt-delim-chars-with-comma))
1325 (c-lang-defconst c-auto-ops
1326 ;; Ops which signal C++11's new auto uses.
1329 (c-lang-defconst c-auto-ops-re
1330 t (c-make-keywords-re nil (c-lang-const c-auto-ops)))
1331 (c-lang-defvar c-auto-ops-re (c-lang-const c-auto-ops-re))
1333 (c-lang-defconst c-haskell-op
1334 ;; Op used in the new C++11 auto function definition, indicating type.
1337 (c-lang-defconst c-haskell-op-re
1338 t (c-make-keywords-re nil (c-lang-const c-haskell-op)))
1339 (c-lang-defvar c-haskell-op-re (c-lang-const c-haskell-op-re))
1341 (c-lang-defconst c-pre-start-tokens
1342 "List of operators following which an apparent declaration \(e.g.
1343 \"t1 *fn (t2 *b);\") is most likely to be an actual declaration
1344 \(as opposed to an arithmetic expression)."
1346 (c-lang-defvar c-pre-start-tokens (c-lang-const c-pre-start-tokens))
1349 ;;; Syntactic whitespace.
1351 (c-lang-defconst c-simple-ws
1352 "Regexp matching an ordinary whitespace character.
1353 Does not contain a \\| operator at the top level."
1354 ;; "\\s " is not enough since it doesn't match line breaks.
1355 t "\\(\\s \\|[\n\r]\\)")
1357 (c-lang-defconst c-simple-ws-depth
1358 ;; Number of regexp grouping parens in `c-simple-ws'.
1359 t (regexp-opt-depth (c-lang-const c-simple-ws)))
1361 (c-lang-defconst c-line-comment-starter
1362 "String that starts line comments, or nil if such don't exist.
1363 Line comments are always terminated by newlines. At least one of
1364 `c-block-comment-starter' and this one is assumed to be set.
1366 Note that it's currently not enough to set this to support a new
1367 comment style. Other stuff like the syntax table must also be set up
1371 (c-lang-defvar c-line-comment-starter (c-lang-const c-line-comment-starter))
1373 (c-lang-defconst c-block-comment-starter
1374 "String that starts block comments, or nil if such don't exist.
1375 Block comments are ended by `c-block-comment-ender', which is assumed
1376 to be set if this is. At least one of `c-line-comment-starter' and
1377 this one is assumed to be set.
1379 Note that it's currently not enough to set this to support a new
1380 comment style. Other stuff like the syntax table must also be set up
1385 (c-lang-defconst c-block-comment-ender
1386 "String that ends block comments, or nil if such don't exist.
1388 Note that it's currently not enough to set this to support a new
1389 comment style. Other stuff like the syntax table must also be set up
1394 (c-lang-defconst c-comment-start-regexp
1395 ;; Regexp to match the start of any type of comment.
1396 t (let ((re (c-make-keywords-re nil
1397 (list (c-lang-const c-line-comment-starter)
1398 (c-lang-const c-block-comment-starter)))))
1399 (if (memq 'gen-comment-delim c-emacs-features)
1400 (concat re "\\|\\s!")
1402 (c-lang-defvar c-comment-start-regexp (c-lang-const c-comment-start-regexp))
1404 (c-lang-defconst c-block-comment-start-regexp
1405 ;; Regexp which matches the start of a block comment (if such exists in the
1407 t (if (c-lang-const c-block-comment-starter)
1408 (regexp-quote (c-lang-const c-block-comment-starter))
1410 (c-lang-defvar c-block-comment-start-regexp
1411 (c-lang-const c-block-comment-start-regexp))
1413 (c-lang-defconst c-line-comment-start-regexp
1414 ;; Regexp which matches the start of a line comment (if such exists in the
1415 ;; language; it does in all 7 CC Mode languages).
1416 t (if (c-lang-const c-line-comment-starter)
1417 (regexp-quote (c-lang-const c-line-comment-starter))
1419 (c-lang-defvar c-line-comment-start-regexp
1420 (c-lang-const c-line-comment-start-regexp))
1422 (c-lang-defconst c-literal-start-regexp
1423 ;; Regexp to match the start of comments and string literals.
1424 t (concat (c-lang-const c-comment-start-regexp)
1426 (if (memq 'gen-string-delim c-emacs-features)
1429 (c-lang-defvar c-literal-start-regexp (c-lang-const c-literal-start-regexp))
1431 (c-lang-defconst c-doc-comment-start-regexp
1432 "Regexp to match the start of documentation comments."
1434 ;; From font-lock.el: `doxygen' uses /*! while others use /**.
1435 (c c++ objc) "/\\*[*!]"
1438 (c-lang-defvar c-doc-comment-start-regexp
1439 (c-lang-const c-doc-comment-start-regexp))
1441 (c-lang-defconst comment-start
1442 "String that starts comments inserted with M-; etc.
1443 `comment-start' is initialized from this."
1444 ;; Default: Prefer line comments to block comments, and pad with a space.
1445 t (concat (or (c-lang-const c-line-comment-starter)
1446 (c-lang-const c-block-comment-starter))
1448 ;; In C we still default to the block comment style since line
1449 ;; comments aren't entirely portable.
1451 (c-lang-setvar comment-start (c-lang-const comment-start))
1453 (c-lang-defconst comment-end
1454 "String that ends comments inserted with M-; etc.
1455 `comment-end' is initialized from this."
1456 ;; Default: Use block comment style if comment-start uses block
1457 ;; comments, and pad with a space in that case.
1458 t (if (string-match (concat "\\`\\("
1459 (c-lang-const c-block-comment-start-regexp)
1461 (c-lang-const comment-start))
1462 (concat " " (c-lang-const c-block-comment-ender))
1464 (c-lang-setvar comment-end (c-lang-const comment-end))
1466 (c-lang-defconst comment-start-skip
1467 "Regexp to match the start of a comment plus everything up to its body.
1468 `comment-start-skip' is initialized from this."
1469 ;; Default: Allow the last char of the comment starter(s) to be
1470 ;; repeated, then allow any amount of horizontal whitespace.
1473 (mapcar (lambda (cs)
1475 (concat (regexp-quote cs) "+")))
1476 (list (c-lang-const c-line-comment-starter)
1477 (c-lang-const c-block-comment-starter)))
1480 (c-lang-setvar comment-start-skip (c-lang-const comment-start-skip))
1482 (c-lang-defconst comment-end-can-be-escaped
1483 "When non-nil, escaped EOLs inside comments are valid.
1484 This works in Emacs >= 25.1."
1487 (c-lang-setvar comment-end-can-be-escaped
1488 (c-lang-const comment-end-can-be-escaped))
1490 (c-lang-defconst c-syntactic-ws-start
1491 ;; Regexp matching any sequence that can start syntactic whitespace.
1492 ;; The only uncertain case is '#' when there are cpp directives.
1494 (c-make-keywords-re nil
1495 (append (list (c-lang-const c-line-comment-starter)
1496 (c-lang-const c-block-comment-starter)
1497 (when (c-lang-const c-opt-cpp-prefix)
1501 (when (memq 'gen-comment-delim c-emacs-features)
1503 (c-lang-defvar c-syntactic-ws-start (c-lang-const c-syntactic-ws-start))
1505 (c-lang-defconst c-syntactic-ws-end
1506 ;; Regexp matching any single character that might end syntactic whitespace.
1508 (c-make-keywords-re nil
1509 (append (when (c-lang-const c-block-comment-ender)
1512 (elt (c-lang-const c-block-comment-ender)
1514 (c-lang-const c-block-comment-ender)))))))
1516 (when (memq 'gen-comment-delim c-emacs-features)
1518 (c-lang-defvar c-syntactic-ws-end (c-lang-const c-syntactic-ws-end))
1520 (c-lang-defconst c-unterminated-block-comment-regexp
1521 ;; Regexp matching an unterminated block comment that doesn't
1522 ;; contain line breaks, or nil in languages without block comments.
1523 ;; Does not contain a \| operator at the top level.
1524 t (when (c-lang-const c-block-comment-starter)
1526 (regexp-quote (c-lang-const c-block-comment-starter))
1527 ;; It's messy to cook together a regexp that matches anything
1528 ;; but c-block-comment-ender.
1529 (let ((end (c-lang-const c-block-comment-ender)))
1530 (cond ((= (length end) 1)
1531 (concat "[^" end "\n\r]*"))
1533 (concat "[^" (substring end 0 1) "\n\r]*"
1535 (regexp-quote (substring end 0 1)) "+"
1537 ;; The quoting rules inside char classes are silly. :P
1538 (cond ((= (elt end 0) (elt end 1))
1539 (concat (substring end 0 1) "\n\r"))
1540 ((= (elt end 1) ?\])
1541 (concat (substring end 1 2) "\n\r"
1542 (substring end 0 1)))
1544 (concat (substring end 0 1) "\n\r"
1545 (substring end 1 2))))
1547 "[^" (substring end 0 1) "\n\r]*"
1550 (error "Can't handle a block comment ender of length %s"
1553 (c-lang-defconst c-block-comment-regexp
1554 ;; Regexp matching a block comment that doesn't contain line breaks,
1555 ;; or nil in languages without block comments. The reason we don't
1556 ;; allow line breaks is to avoid going very far and risk running out
1557 ;; of regexp stack; this regexp is intended to handle only short
1558 ;; comments that might be put in the middle of limited constructs
1559 ;; like declarations. Does not contain a \| operator at the top
1561 t (when (c-lang-const c-unterminated-block-comment-regexp)
1563 (c-lang-const c-unterminated-block-comment-regexp)
1564 (let ((end (c-lang-const c-block-comment-ender)))
1565 (cond ((= (length end) 1)
1568 (concat (regexp-quote (substring end 0 1)) "+"
1569 (regexp-quote (substring end 1 2))))
1571 (error "Can't handle a block comment ender of length %s"
1574 (c-lang-defconst c-nonwhite-syntactic-ws
1575 ;; Regexp matching a piece of syntactic whitespace that isn't a
1576 ;; sequence of simple whitespace characters. As opposed to
1577 ;; `c-(forward|backward)-syntactic-ws', this doesn't regard cpp
1578 ;; directives as syntactic whitespace.
1579 t (c-concat-separated
1580 (list (when (c-lang-const c-line-comment-starter)
1581 (concat (regexp-quote (c-lang-const c-line-comment-starter))
1583 (c-lang-const c-block-comment-regexp)
1585 (when (memq 'gen-comment-delim c-emacs-features)
1589 (c-lang-defconst c-syntactic-ws
1590 ;; Regexp matching syntactic whitespace, including possibly the
1591 ;; empty string. As opposed to `c-(forward|backward)-syntactic-ws',
1592 ;; this doesn't regard cpp directives as syntactic whitespace. Does
1593 ;; not contain a \| operator at the top level.
1594 t (concat (c-lang-const c-simple-ws) "*"
1596 (concat "\\(" (c-lang-const c-nonwhite-syntactic-ws) "\\)"
1597 (c-lang-const c-simple-ws) "*")
1600 (c-lang-defconst c-syntactic-ws-depth
1601 ;; Number of regexp grouping parens in `c-syntactic-ws'.
1602 t (regexp-opt-depth (c-lang-const c-syntactic-ws)))
1604 (c-lang-defconst c-nonempty-syntactic-ws
1605 ;; Regexp matching syntactic whitespace, which is at least one
1606 ;; character long. As opposed to `c-(forward|backward)-syntactic-ws',
1607 ;; this doesn't regard cpp directives as syntactic whitespace. Does
1608 ;; not contain a \| operator at the top level.
1610 (c-lang-const c-simple-ws)
1612 (c-lang-const c-nonwhite-syntactic-ws)
1615 (c-lang-defconst c-nonempty-syntactic-ws-depth
1616 ;; Number of regexp grouping parens in `c-nonempty-syntactic-ws'.
1617 t (regexp-opt-depth (c-lang-const c-nonempty-syntactic-ws)))
1619 (c-lang-defconst c-single-line-syntactic-ws
1620 ;; Regexp matching syntactic whitespace without any line breaks. As
1621 ;; opposed to `c-(forward|backward)-syntactic-ws', this doesn't
1622 ;; regard cpp directives as syntactic whitespace. Does not contain
1623 ;; a \| operator at the top level.
1624 t (if (c-lang-const c-block-comment-regexp)
1626 (c-lang-const c-block-comment-regexp)
1630 (c-lang-defconst c-single-line-syntactic-ws-depth
1631 ;; Number of regexp grouping parens in `c-single-line-syntactic-ws'.
1632 t (regexp-opt-depth (c-lang-const c-single-line-syntactic-ws)))
1634 (c-lang-defconst c-syntactic-eol
1635 ;; Regexp that matches when there is no syntactically significant
1636 ;; text before eol. Macros are regarded as syntactically
1637 ;; significant text here.
1638 t (concat (c-lang-const c-single-line-syntactic-ws)
1639 ;; Match eol (possibly inside a block comment or preceded
1640 ;; by a line continuation backslash), or the beginning of a
1641 ;; line comment. Note: This has to be modified for awk
1642 ;; where line comments start with '#'.
1645 (list (when (c-lang-const c-line-comment-starter)
1646 (regexp-quote (c-lang-const c-line-comment-starter)))
1647 (when (c-lang-const c-unterminated-block-comment-regexp)
1648 (concat (c-lang-const c-unterminated-block-comment-regexp)
1654 (c-lang-defvar c-syntactic-eol (c-lang-const c-syntactic-eol))
1659 ;; The Emacs variables beginning-of-defun-function and end-of-defun-function
1660 ;; will be set so that commands like `mark-defun' and `narrow-to-defun' work
1661 ;; right. In older Emacsen, the key sequences C-M-a and C-M-e are, however,
1662 ;; bound directly to the CC Mode functions, allowing optimization for large n.
1663 ;; From Emacs 23, this isn't necessary any more, since n is passed to the two
1665 (c-lang-defconst beginning-of-defun-function
1666 "Function to which beginning-of-defun-function will be set."
1667 t 'c-beginning-of-defun
1668 awk 'c-awk-beginning-of-defun)
1669 (c-lang-setvar beginning-of-defun-function
1670 (c-lang-const beginning-of-defun-function))
1672 (c-lang-defconst end-of-defun-function
1673 "Function to which end-of-defun-function will be set."
1675 awk 'c-awk-end-of-defun)
1676 (c-lang-setvar end-of-defun-function (c-lang-const end-of-defun-function))
1678 ;;; In-comment text handling.
1680 (c-lang-defconst c-paragraph-start
1681 "Regexp to append to `paragraph-start'."
1683 java "\\(@[a-zA-Z]+\\>\\|$\\)" ; For Javadoc.
1684 pike "\\(@[a-zA-Z_-]+\\>\\([^{]\\|$\\)\\|$\\)") ; For Pike refdoc.
1685 (c-lang-defvar c-paragraph-start (c-lang-const c-paragraph-start))
1687 (c-lang-defconst c-paragraph-separate
1688 "Regexp to append to `paragraph-separate'."
1690 pike (c-lang-const c-paragraph-start))
1691 (c-lang-defvar c-paragraph-separate (c-lang-const c-paragraph-separate))
1696 ;; Note: All and only all language constants containing keyword lists
1697 ;; should end with "-kwds"; they're automatically collected into the
1698 ;; `c-kwds-lang-consts' list below and used to build `c-keywords' etc.
1700 (c-lang-defconst c-primitive-type-kwds
1701 "Primitive type keywords. As opposed to the other keyword lists, the
1702 keywords listed here are fontified with the type face instead of the
1705 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1706 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1707 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1710 Do not try to modify this list for end user customizations; the
1711 `*-font-lock-extra-types' variable, where `*' is the mode prefix, is
1712 the appropriate place for that."
1713 t '("char" "double" "float" "int" "long" "short" "signed"
1716 '("_Bool" "_Complex" "_Imaginary") ; Conditionally defined in C99.
1717 (c-lang-const c-primitive-type-kwds))
1719 '("bool" "wchar_t" "char16_t" "char32_t")
1720 (c-lang-const c-primitive-type-kwds))
1721 ;; Objective-C extends C, but probably not the new stuff in C99.
1723 '("id" "Class" "SEL" "IMP" "BOOL")
1724 (c-lang-const c-primitive-type-kwds))
1725 java '("boolean" "byte" "char" "double" "float" "int" "long" "short" "void")
1726 idl '("Object" "ValueBase" "any" "boolean" "char" "double" "fixed" "float"
1727 "long" "octet" "sequence" "short" "string" "void" "wchar" "wstring"
1730 ;; The following can't really end a type, but we have to specify them
1731 ;; here due to the assumption in `c-primitive-type-prefix-kwds'. It
1732 ;; doesn't matter that much.
1733 "unsigned" "strong")
1734 pike '(;; this_program isn't really a keyword, but it's practically
1735 ;; used as a builtin type.
1736 "array" "float" "function" "int" "mapping" "mixed" "multiset"
1737 "object" "program" "string" "this_program" "void"))
1739 (c-lang-defconst c-primitive-type-key
1740 ;; An adorned regexp that matches `c-primitive-type-kwds'.
1741 t (c-make-keywords-re t (c-lang-const c-primitive-type-kwds)))
1742 (c-lang-defvar c-primitive-type-key (c-lang-const c-primitive-type-key))
1744 (c-lang-defconst c-primitive-type-prefix-kwds
1745 "Keywords that might act as prefixes for primitive types. Assumed to
1746 be a subset of `c-primitive-type-kwds'."
1748 (c c++) '("long" "short" "signed" "unsigned")
1749 idl '("long" "unsigned"
1753 (c-lang-defconst c-typedef-kwds
1754 "Prefix keyword(s) like \"typedef\" which make a type declaration out
1755 of a variable declaration."
1759 (c-lang-defconst c-typedef-key
1760 ;; Adorned regexp matching `c-typedef-kwds'.
1761 t (c-make-keywords-re t (c-lang-const c-typedef-kwds)))
1762 (c-lang-defvar c-typedef-key (c-lang-const c-typedef-key))
1764 (c-lang-defconst c-typeof-kwds
1765 "Keywords followed by a parenthesized expression, which stands for
1766 the type of that expression."
1768 c '("typeof") ; longstanding GNU C(++) extension.
1769 c++ '("decltype" "typeof"))
1771 (c-lang-defconst c-typeof-key
1772 ;; Adorned regexp matching `c-typeof-kwds'.
1773 t (c-make-keywords-re t (c-lang-const c-typeof-kwds)))
1774 (c-lang-defvar c-typeof-key (c-lang-const c-typeof-key))
1776 (c-lang-defconst c-type-prefix-kwds
1777 "Keywords where the following name - if any - is a type name, and
1778 where the keyword together with the symbol works as a type in
1781 Note that an alternative if the second part doesn't hold is
1782 `c-type-list-kwds'. Keywords on this list are typically also present
1783 on one of the `*-decl-kwds' lists."
1785 c '("struct" "union" "enum")
1786 c++ (append '("class" "typename")
1787 (c-lang-const c-type-prefix-kwds c)))
1789 (c-lang-defconst c-type-prefix-key
1790 ;; Adorned regexp matching `c-type-prefix-kwds'.
1791 t (c-make-keywords-re t (c-lang-const c-type-prefix-kwds)))
1792 (c-lang-defvar c-type-prefix-key (c-lang-const c-type-prefix-key))
1794 (c-lang-defconst c-type-modifier-kwds
1795 "Type modifier keywords. These can occur almost anywhere in types
1796 but they don't build a type of themselves. Unlike the keywords on
1797 `c-primitive-type-kwds', they are fontified with the keyword face and
1800 c '("const" "restrict" "volatile")
1801 c++ '("const" "noexcept" "volatile" "throw" "final" "override")
1802 objc '("const" "volatile"))
1804 (c-lang-defconst c-opt-type-modifier-key
1805 ;; Adorned regexp matching `c-type-modifier-kwds', or nil in
1806 ;; languages without such keywords.
1807 t (and (c-lang-const c-type-modifier-kwds)
1808 (c-make-keywords-re t (c-lang-const c-type-modifier-kwds))))
1809 (c-lang-defvar c-opt-type-modifier-key (c-lang-const c-opt-type-modifier-key))
1811 (c-lang-defconst c-opt-type-component-key
1812 ;; An adorned regexp that matches `c-primitive-type-prefix-kwds' and
1813 ;; `c-type-modifier-kwds', or nil in languages without any of them.
1814 t (and (or (c-lang-const c-primitive-type-prefix-kwds)
1815 (c-lang-const c-type-modifier-kwds))
1816 (c-make-keywords-re t
1817 (append (c-lang-const c-primitive-type-prefix-kwds)
1818 (c-lang-const c-type-modifier-kwds)))))
1819 (c-lang-defvar c-opt-type-component-key
1820 (c-lang-const c-opt-type-component-key))
1822 (c-lang-defconst c-type-start-kwds
1823 ;; All keywords that can start a type (i.e. are either a type prefix
1824 ;; or a complete type).
1825 t (c--delete-duplicates (append (c-lang-const c-primitive-type-kwds)
1826 (c-lang-const c-type-prefix-kwds)
1827 (c-lang-const c-type-modifier-kwds))
1828 :test 'string-equal))
1830 (c-lang-defconst c-class-decl-kwds
1831 "Keywords introducing declarations where the following block (if any)
1832 contains another declaration level that should be considered a class.
1834 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1835 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1836 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1839 Note that presence on this list does not automatically treat the
1840 following identifier as a type; the keyword must also be present on
1841 `c-type-prefix-kwds' or `c-type-list-kwds' to accomplish that."
1843 c '("struct" "union")
1844 c++ '("class" "struct" "union")
1845 objc '("struct" "union"
1846 "@interface" "@implementation" "@protocol")
1847 java '("class" "@interface" "interface")
1848 idl '("component" "eventtype" "exception" "home" "interface" "struct"
1851 "storagehome" "storagetype"
1853 "catalog" "executor" "manages" "segment")
1856 (c-lang-defconst c-class-key
1857 ;; Regexp matching the start of a class.
1858 t (c-make-keywords-re t (c-lang-const c-class-decl-kwds)))
1859 (c-lang-defvar c-class-key (c-lang-const c-class-key))
1861 (c-lang-defconst c-brace-list-decl-kwds
1862 "Keywords introducing declarations where the following block (if
1863 any) is a brace list.
1865 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1866 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1867 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1872 (c-lang-defconst c-brace-list-key
1873 ;; Regexp matching the start of declarations where the following
1874 ;; block is a brace list.
1875 t (c-make-keywords-re t (c-lang-const c-brace-list-decl-kwds)))
1876 (c-lang-defvar c-brace-list-key (c-lang-const c-brace-list-key))
1878 (c-lang-defconst c-after-brace-list-decl-kwds
1879 "Keywords that might follow keywords in `c-brace-list-decl-kwds'
1880 and precede the opening brace."
1882 c++ '("class" "struct"))
1884 (c-lang-defconst c-after-brace-list-key
1885 ;; Regexp matching keywords that can fall between a brace-list
1886 ;; keyword and the associated brace list.
1887 t (c-make-keywords-re t (c-lang-const c-after-brace-list-decl-kwds)))
1888 (c-lang-defvar c-after-brace-list-key (c-lang-const c-after-brace-list-key))
1890 (c-lang-defconst c-recognize-post-brace-list-type-p
1891 "Set to t when we recognize a colon and then a type after an enum,
1892 e.g., enum foo : int { A, B, C };"
1895 (c-lang-defvar c-recognize-post-brace-list-type-p
1896 (c-lang-const c-recognize-post-brace-list-type-p))
1898 (c-lang-defconst c-other-block-decl-kwds
1899 "Keywords where the following block (if any) contains another
1900 declaration level that should not be considered a class. For every
1901 keyword here, CC Mode will add a set of special syntactic symbols for
1902 those blocks. E.g. if the keyword is \"foo\" then there will be
1903 `foo-open', `foo-close', and `infoo' symbols.
1905 The intention is that this category should be used for block
1906 constructs that aren't related to object orientation concepts like
1907 classes (which thus also include e.g. interfaces, templates,
1908 contracts, structs, etc). The more pragmatic distinction is that
1909 while most want some indentation inside classes, it's fairly common
1910 that they don't want it in some of these constructs, so it should be
1911 simple to configure that differently from classes. See also
1912 `c-class-decl-kwds'.
1914 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1915 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1916 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1919 (c objc) '("extern")
1920 c++ '("namespace" "extern")
1925 (c-lang-defconst c-other-decl-block-key
1926 ;; Regexp matching the start of blocks besides classes that contain
1927 ;; another declaration level.
1928 t (c-make-keywords-re t (c-lang-const c-other-block-decl-kwds)))
1929 (c-lang-defvar c-other-decl-block-key (c-lang-const c-other-decl-block-key))
1931 (c-lang-defvar c-other-decl-block-key-in-symbols-alist
1935 (if (string= elt "extern")
1937 (intern (concat "in" elt)))))
1938 (c-lang-const c-other-block-decl-kwds))
1939 "Alist associating keywords in c-other-decl-block-decl-kwds with
1940 their matching \"in\" syntactic symbols.")
1942 (c-lang-defconst c-typedef-decl-kwds
1943 "Keywords introducing declarations where the identifier(s) being
1946 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1947 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1948 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1950 ;; Default to `c-class-decl-kwds' and `c-brace-list-decl-kwds'
1951 ;; (since e.g. "Foo" is a type that's being defined in "class Foo
1953 t (append (c-lang-const c-class-decl-kwds)
1954 (c-lang-const c-brace-list-decl-kwds))
1955 ;; Languages that have a "typedef" construct.
1956 (c c++ objc idl pike) (append (c-lang-const c-typedef-decl-kwds)
1958 ;; Unlike most other languages, exception names are not handled as
1959 ;; types in IDL since they only can occur in "raises" specs.
1960 idl (delete "exception" (append (c-lang-const c-typedef-decl-kwds) nil)))
1962 (c-lang-defconst c-typedef-decl-key
1963 t (c-make-keywords-re t (c-lang-const c-typedef-decl-kwds)))
1964 (c-lang-defvar c-typedef-decl-key (c-lang-const c-typedef-decl-key))
1966 (c-lang-defconst c-typeless-decl-kwds
1967 "Keywords introducing declarations where the \(first) identifier
1968 \(declarator) follows directly after the keyword, without any type.
1970 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1971 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1972 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
1974 ;; Default to `c-class-decl-kwds' and `c-brace-list-decl-kwds'
1975 ;; (since e.g. "Foo" is the identifier being defined in "class Foo
1977 t (append (c-lang-const c-class-decl-kwds)
1978 (c-lang-const c-brace-list-decl-kwds))
1979 c++ (append (c-lang-const c-typeless-decl-kwds) '("auto")) ; C++11.
1980 ;; Note: "manages" for CORBA CIDL clashes with its presence on
1981 ;; `c-type-list-kwds' for IDL.
1982 idl (append (c-lang-const c-typeless-decl-kwds)
1983 '("factory" "finder" "native"
1988 pike (append (c-lang-const c-class-decl-kwds)
1991 (c-lang-defconst c-modifier-kwds
1992 "Keywords that can prefix normal declarations of identifiers
1993 \(and typically act as flags). Things like argument declarations
1994 inside function headers are also considered declarations in this
1997 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
1998 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
1999 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
2002 (c c++) '("auto" "extern" "inline" "register" "static")
2003 c++ (append '("constexpr" "explicit" "friend" "mutable" "template"
2004 "thread_local" "using" "virtual")
2005 (c-lang-const c-modifier-kwds))
2006 objc '("auto" "bycopy" "byref" "extern" "in" "inout" "oneway" "out" "static")
2007 ;; FIXME: Some of those below ought to be on `c-other-decl-kwds' instead.
2008 idl '("abstract" "attribute" "const" "consumes" "custom" "emits" "import"
2009 "in" "inout" "local" "multiple" "oneway" "out" "private" "provides"
2010 "public" "publishes" "readonly" "typeid" "typeprefix" "uses"
2014 "bindsTo" "delegatesTo" "implements" "proxy" "storedOn")
2015 ;; Note: "const" is not used in Java, but it's still a reserved keyword.
2016 java '("abstract" "const" "default" "final" "native" "private" "protected"
2017 "public" "static" "strictfp" "synchronized" "transient" "volatile")
2018 pike '("final" "inline" "local" "nomask" "optional" "private" "protected"
2019 "public" "static" "variant"))
2021 (c-lang-defconst c-other-decl-kwds
2022 "Keywords that can start or prefix any declaration level construct,
2023 besides those on `c-class-decl-kwds', `c-brace-list-decl-kwds',
2024 `c-other-block-decl-kwds', `c-typedef-decl-kwds',
2025 `c-typeless-decl-kwds' and `c-modifier-kwds'.
2027 If any of these also are on `c-type-list-kwds', `c-ref-list-kwds',
2028 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
2029 `c-<>-type-kwds', or `c-<>-arglist-kwds' then the associated clauses
2032 objc '("@class" "@end" "@defs")
2033 java '("import" "package")
2034 pike '("import" "inherit"))
2036 (c-lang-defconst c-decl-start-kwds
2037 "Keywords that always start declarations, wherever they occur.
2038 This can be used for declarations that aren't recognized by the normal
2039 combination of `c-decl-prefix-re' and `c-decl-start-re'."
2041 ;; Classes can be declared anywhere in a Pike expression.
2044 (c-lang-defconst c-decl-hangon-kwds
2045 "Keywords that can occur anywhere in a declaration level construct.
2046 This is used for self-contained things that can be tacked on anywhere
2047 on a declaration and that should be ignored to be able to recognize it
2048 correctly. Typical cases are compiler extensions like
2049 \"__attribute__\" or \"__declspec\":
2051 __declspec(noreturn) void foo();
2052 class __declspec(dllexport) classname {...};
2053 void foo() __attribute__((noreturn));
2055 Note that unrecognized plain symbols are skipped anyway if they occur
2056 before the type, so such things are not necessary to mention here.
2057 Mentioning them here is necessary only if they can occur in other
2058 places, or if they are followed by a construct that must be skipped
2059 over \(like the parens in the \"__attribute__\" and \"__declspec\"
2060 examples above). In the last case, they alse need to be present on
2061 one of `c-type-list-kwds', `c-ref-list-kwds',
2062 `c-colon-type-list-kwds', `c-paren-nontype-kwds', `c-paren-type-kwds',
2063 `c-<>-type-kwds', or `c-<>-arglist-kwds'."
2064 ;; NB: These are currently not recognized in all parts of a
2065 ;; declaration. Specifically, they aren't recognized in the middle
2066 ;; of multi-token types, inside declarators, and between the
2067 ;; identifier and the arglist paren of a function declaration.
2069 ;; FIXME: This ought to be user customizable since compiler stuff
2070 ;; like this usually is wrapped in project specific macros. (It'd
2071 ;; of course be even better if we could cope without knowing this.)
2073 (c c++) '(;; GCC extension.
2078 (c-lang-defconst c-decl-hangon-key
2079 ;; Adorned regexp matching `c-decl-hangon-kwds'.
2080 t (c-make-keywords-re t (c-lang-const c-decl-hangon-kwds)))
2081 (c-lang-defvar c-decl-hangon-key (c-lang-const c-decl-hangon-key))
2083 (c-lang-defconst c-prefix-spec-kwds
2084 ;; All keywords that can occur in the preamble of a declaration.
2085 ;; They typically occur before the type, but they are also matched
2086 ;; after presumptive types since we often can't be sure that
2087 ;; something is a type or just some sort of macro in front of the
2088 ;; declaration. They might be ambiguous with types or type
2090 t (c--delete-duplicates (append (c-lang-const c-class-decl-kwds)
2091 (c-lang-const c-brace-list-decl-kwds)
2092 (c-lang-const c-other-block-decl-kwds)
2093 (c-lang-const c-typedef-decl-kwds)
2094 (c-lang-const c-typeless-decl-kwds)
2095 (c-lang-const c-modifier-kwds)
2096 (c-lang-const c-other-decl-kwds)
2097 (c-lang-const c-decl-start-kwds)
2098 (c-lang-const c-decl-hangon-kwds))
2099 :test 'string-equal))
2101 (c-lang-defconst c-prefix-spec-kwds-re
2102 ;; Adorned regexp of `c-prefix-spec-kwds'.
2103 t (c-make-keywords-re t (c-lang-const c-prefix-spec-kwds)))
2105 (c-lang-defvar c-prefix-spec-kwds-re (c-lang-const c-prefix-spec-kwds-re))
2107 (c-lang-defconst c-specifier-key
2108 ;; Adorned regexp of the keywords in `c-prefix-spec-kwds' that aren't
2109 ;; ambiguous with types or type prefixes. These are the keywords (like
2110 ;; extern, namespace, but NOT template) that can modify a declaration.
2111 t (c-make-keywords-re t
2112 (c--set-difference (c-lang-const c-prefix-spec-kwds)
2113 (append (c-lang-const c-type-start-kwds)
2114 (c-lang-const c-<>-arglist-kwds))
2115 :test 'string-equal)))
2116 (c-lang-defvar c-specifier-key (c-lang-const c-specifier-key))
2118 (c-lang-defconst c-postfix-spec-kwds
2119 ;; Keywords that can occur after argument list of a function header
2120 ;; declaration, i.e. in the "K&R region".
2121 t (append (c-lang-const c-postfix-decl-spec-kwds)
2122 (c-lang-const c-decl-hangon-kwds)))
2124 (c-lang-defconst c-not-decl-init-keywords
2125 ;; Adorned regexp matching all keywords that can't appear at the
2126 ;; start of a declaration.
2127 t (c-make-keywords-re t
2128 (c--set-difference (c-lang-const c-keywords)
2129 (append (c-lang-const c-type-start-kwds)
2130 (c-lang-const c-prefix-spec-kwds)
2131 (c-lang-const c-typeof-kwds))
2132 :test 'string-equal)))
2133 (c-lang-defvar c-not-decl-init-keywords
2134 (c-lang-const c-not-decl-init-keywords))
2136 (c-lang-defconst c-not-primitive-type-keywords
2137 "List of all keywords apart from primitive types (like \"int\")."
2138 t (c--set-difference (c-lang-const c-keywords)
2139 (c-lang-const c-primitive-type-kwds)
2140 :test 'string-equal)
2141 ;; The "more" for C++ is the QT keyword (as in "more slots:").
2142 ;; This variable is intended for use in c-beginning-of-statement-1.
2143 c++ (append (c-lang-const c-not-primitive-type-keywords) '("more")))
2145 (c-lang-defconst c-not-primitive-type-keywords-regexp
2146 t (c-make-keywords-re t
2147 (c-lang-const c-not-primitive-type-keywords)))
2148 (c-lang-defvar c-not-primitive-type-keywords-regexp
2149 (c-lang-const c-not-primitive-type-keywords-regexp))
2151 (c-lang-defconst c-protection-kwds
2152 "Access protection label keywords in classes."
2154 c++ '("private" "protected" "public")
2155 objc '("@private" "@protected" "@public"))
2157 (c-lang-defconst c-block-decls-with-vars
2158 "Keywords introducing declarations that can contain a block which
2159 might be followed by variable declarations, e.g. like \"foo\" in
2160 \"class Foo { ... } foo;\". So if there is a block in a declaration
2161 like that, it ends with the following `;' and not right away.
2163 The keywords on list are assumed to also be present on one of the
2164 `*-decl-kwds' lists."
2166 (c objc) '("struct" "union" "enum" "typedef")
2167 c++ '("class" "struct" "union" "enum" "typedef"))
2169 (c-lang-defconst c-opt-block-decls-with-vars-key
2170 ;; Regexp matching the `c-block-decls-with-vars' keywords, or nil in
2171 ;; languages without such constructs.
2172 t (and (c-lang-const c-block-decls-with-vars)
2173 (c-make-keywords-re t (c-lang-const c-block-decls-with-vars))))
2174 (c-lang-defvar c-opt-block-decls-with-vars-key
2175 (c-lang-const c-opt-block-decls-with-vars-key))
2177 (c-lang-defconst c-postfix-decl-spec-kwds
2178 "Keywords introducing extra declaration specifiers in the region
2179 between the header and the body \(i.e. the \"K&R-region\") in
2182 java '("extends" "implements" "throws")
2183 idl '("context" "getraises" "manages" "primarykey" "raises" "setraises"
2186 "as" "const" "implements" "of" "ref"))
2188 (c-lang-defconst c-postfix-decl-spec-key
2189 ;; Regexp matching the keywords in `c-postfix-decl-spec-kwds'.
2190 t (c-make-keywords-re t (c-lang-const c-postfix-decl-spec-kwds)))
2191 (c-lang-defvar c-postfix-decl-spec-key
2192 (c-lang-const c-postfix-decl-spec-key))
2194 (c-lang-defconst c-nonsymbol-sexp-kwds
2195 "Keywords that may be followed by a nonsymbol sexp before whatever
2196 construct it's part of continues."
2198 (c c++ objc) '("extern"))
2200 (c-lang-defconst c-type-list-kwds
2201 "Keywords that may be followed by a comma separated list of type
2202 identifiers, where each optionally can be prefixed by keywords. (Can
2203 also be used for the special case when the list can contain only one
2206 Assumed to be mutually exclusive with `c-ref-list-kwds'. There's no
2207 reason to put keywords on this list if they are on `c-type-prefix-kwds'.
2208 There's also no reason to add keywords that prefixes a normal
2209 declaration consisting of a type followed by a declarator (list), so
2210 the keywords on `c-modifier-kwds' should normally not be listed here
2213 Note: Use `c-typeless-decl-kwds' for keywords followed by a function
2214 or variable identifier (that's being defined)."
2218 java '("import" "new" "extends" "super" "implements" "throws")
2219 idl '("manages" "native" "primarykey" "supports"
2221 "as" "implements" "of" "scope")
2224 (c-lang-defconst c-ref-list-kwds
2225 "Keywords that may be followed by a comma separated list of
2226 reference (i.e. namespace/scope/module) identifiers, where each
2227 optionally can be prefixed by keywords. (Can also be used for the
2228 special case when the list can contain only one element.) Assumed to
2229 be mutually exclusive with `c-type-list-kwds'.
2231 Note: Use `c-typeless-decl-kwds' for keywords followed by a function
2232 or variable identifier (that's being defined)."
2236 idl '("import" "module"
2241 (c-lang-defconst c-colon-type-list-kwds
2242 "Keywords that may be followed (not necessarily directly) by a colon
2243 and then a comma separated list of type identifiers, where each
2244 optionally can be prefixed by keywords. (Can also be used for the
2245 special case when the list can contain only one element.)"
2247 c++ '("class" "struct")
2248 idl '("component" "eventtype" "home" "interface" "valuetype"
2250 "storagehome" "storagetype"))
2252 (c-lang-defconst c-colon-type-list-re
2253 "Regexp matched after the keywords in `c-colon-type-list-kwds' to skip
2254 forward to the colon. The end of the match is assumed to be directly
2255 after the colon, so the regexp should end with \":\". Must be a
2256 regexp if `c-colon-type-list-kwds' isn't nil."
2257 t (if (c-lang-const c-colon-type-list-kwds)
2258 ;; Disallow various common punctuation chars that can't come
2259 ;; before the ":" that starts the inherit list after "class"
2260 ;; or "struct" in C++. (Also used as default for other
2262 "[^][{}();,/#=:]*:"))
2263 (c-lang-defvar c-colon-type-list-re (c-lang-const c-colon-type-list-re))
2265 (c-lang-defconst c-paren-nontype-kwds
2266 "Keywords that may be followed by a parenthesis expression that doesn't
2267 contain type identifiers."
2269 (c c++) '(;; GCC extension.
2274 (c-lang-defconst c-paren-nontype-key
2275 t (c-make-keywords-re t (c-lang-const c-paren-nontype-kwds)))
2276 (c-lang-defvar c-paren-nontype-key (c-lang-const c-paren-nontype-key))
2278 (c-lang-defconst c-paren-type-kwds
2279 "Keywords that may be followed by a parenthesis expression containing
2280 type identifiers separated by arbitrary tokens."
2285 pike '("array" "function" "int" "mapping" "multiset" "object" "program"))
2287 (c-lang-defconst c-paren-any-kwds
2288 t (c--delete-duplicates (append (c-lang-const c-paren-nontype-kwds)
2289 (c-lang-const c-paren-type-kwds))
2290 :test 'string-equal))
2292 (c-lang-defconst c-<>-type-kwds
2293 "Keywords that may be followed by an angle bracket expression
2294 containing type identifiers separated by \",\". The difference from
2295 `c-<>-arglist-kwds' is that unknown names are taken to be types and
2296 not other identifiers. `c-recognize-<>-arglists' is assumed to be set
2304 (c-lang-defconst c-<>-arglist-kwds
2305 "Keywords that can be followed by a C++ style template arglist; see
2306 `c-recognize-<>-arglists' for details. That language constant is
2307 assumed to be set if this isn't nil."
2310 idl '("fixed" "string" "wstring"))
2312 (c-lang-defconst c-<>-sexp-kwds
2313 ;; All keywords that can be followed by an angle bracket sexp.
2314 t (c--delete-duplicates (append (c-lang-const c-<>-type-kwds)
2315 (c-lang-const c-<>-arglist-kwds))
2316 :test 'string-equal))
2318 (c-lang-defconst c-opt-<>-sexp-key
2319 ;; Adorned regexp matching keywords that can be followed by an angle
2320 ;; bracket sexp. Always set when `c-recognize-<>-arglists' is.
2321 t (c-make-keywords-re t (c-lang-const c-<>-sexp-kwds)))
2322 (c-lang-defvar c-opt-<>-sexp-key (c-lang-const c-opt-<>-sexp-key))
2324 (c-lang-defconst c-inside-<>-type-kwds
2325 "Keywords which, used inside a C++ style template arglist, introduce a type."
2327 java '("extends" "super"))
2329 (c-lang-defconst c-inside-<>-type-key
2330 t (c-make-keywords-re t (c-lang-const c-inside-<>-type-kwds)))
2331 (c-lang-defvar c-inside-<>-type-key (c-lang-const c-inside-<>-type-key))
2333 (c-lang-defconst c-brace-id-list-kwds
2334 "Keywords that may be followed by a brace block containing a comma
2335 separated list of identifier definitions, i.e. like the list of
2336 identifiers that follows the type in a normal declaration."
2337 t (c-lang-const c-brace-list-decl-kwds))
2339 (c-lang-defconst c-block-stmt-1-kwds
2340 "Statement keywords followed directly by a substatement."
2342 c++ '("do" "else" "try")
2343 objc '("do" "else" "@finally" "@try")
2344 java '("do" "else" "finally" "try")
2347 (c-lang-defconst c-block-stmt-1-key
2348 ;; Regexp matching the start of any statement followed directly by a
2349 ;; substatement (doesn't match a bare block, however).
2350 t (c-make-keywords-re t (c-lang-const c-block-stmt-1-kwds)))
2351 (c-lang-defvar c-block-stmt-1-key (c-lang-const c-block-stmt-1-key))
2353 (c-lang-defconst c-block-stmt-1-2-kwds
2354 "Statement keywords optionally followed by a paren sexp.
2355 Keywords here should also be in `c-block-stmt-1-kwds'."
2359 (c-lang-defconst c-block-stmt-1-2-key
2360 ;; Regexp matching the start of a statement which may be followed by a
2361 ;; paren sexp and will then be followed by a substatement.
2362 t (c-make-keywords-re t (c-lang-const c-block-stmt-1-2-kwds)))
2363 (c-lang-defvar c-block-stmt-1-2-key (c-lang-const c-block-stmt-1-2-key))
2365 (c-lang-defconst c-block-stmt-2-kwds
2366 "Statement keywords followed by a paren sexp and then by a substatement."
2367 t '("for" "if" "switch" "while")
2368 c++ '("for" "if" "switch" "while" "catch")
2369 objc '("for" "if" "switch" "while" "@catch" "@synchronized")
2370 java '("for" "if" "switch" "while" "catch" "synchronized")
2372 pike '("for" "if" "switch" "while" "foreach")
2373 awk '("for" "if" "while"))
2375 (c-lang-defconst c-block-stmt-2-key
2376 ;; Regexp matching the start of any statement followed by a paren sexp
2377 ;; and then by a substatement.
2378 t (c-make-keywords-re t (c-lang-const c-block-stmt-2-kwds)))
2379 (c-lang-defvar c-block-stmt-2-key (c-lang-const c-block-stmt-2-key))
2381 (c-lang-defconst c-block-stmt-kwds
2382 ;; Union of `c-block-stmt-1-kwds' and `c-block-stmt-2-kwds'.
2383 t (c--delete-duplicates (append (c-lang-const c-block-stmt-1-kwds)
2384 (c-lang-const c-block-stmt-2-kwds))
2385 :test 'string-equal))
2387 (c-lang-defconst c-opt-block-stmt-key
2388 ;; Regexp matching the start of any statement that has a
2389 ;; substatement (except a bare block). Nil in languages that
2390 ;; don't have such constructs.
2391 t (if (or (c-lang-const c-block-stmt-1-kwds)
2392 (c-lang-const c-block-stmt-2-kwds))
2393 (c-make-keywords-re t
2394 (append (c-lang-const c-block-stmt-1-kwds)
2395 (c-lang-const c-block-stmt-2-kwds)))))
2396 (c-lang-defvar c-opt-block-stmt-key (c-lang-const c-opt-block-stmt-key))
2398 (c-lang-defconst c-simple-stmt-kwds
2399 "Statement keywords followed by an expression or nothing."
2400 t '("break" "continue" "goto" "return")
2401 objc '("break" "continue" "goto" "return" "@throw")
2402 ;; Note: `goto' is not valid in Java, but the keyword is still reserved.
2403 java '("break" "continue" "goto" "return" "throw")
2405 pike '("break" "continue" "return")
2406 awk '(;; Not sure about "delete", "exit", "getline", etc. ; ACM 2002/5/30
2407 "break" "continue" "return" "delete" "exit" "getline" "next"
2408 "nextfile" "print" "printf"))
2410 (c-lang-defconst c-simple-stmt-key
2411 ;; Adorned regexp matching `c-simple-stmt-kwds'.
2412 t (c-make-keywords-re t (c-lang-const c-simple-stmt-kwds)))
2413 (c-lang-defvar c-simple-stmt-key (c-lang-const c-simple-stmt-key))
2415 (c-lang-defconst c-paren-stmt-kwds
2416 "Statement keywords followed by a parenthesis expression that
2417 nevertheless contains a list separated with `;' and not `,'."
2421 (c-lang-defconst c-paren-stmt-key
2422 ;; Adorned regexp matching `c-paren-stmt-kwds'.
2423 t (c-make-keywords-re t (c-lang-const c-paren-stmt-kwds)))
2424 (c-lang-defvar c-paren-stmt-key (c-lang-const c-paren-stmt-key))
2426 (c-lang-defconst c-asm-stmt-kwds
2427 "Statement keywords followed by an assembler expression."
2429 (c c++) '("asm" "__asm__")) ;; Not standard, but common.
2431 (c-lang-defconst c-opt-asm-stmt-key
2432 ;; Regexp matching the start of an assembler statement. Nil in
2433 ;; languages that don't support that.
2434 t (if (c-lang-const c-asm-stmt-kwds)
2435 (c-make-keywords-re t (c-lang-const c-asm-stmt-kwds))))
2436 (c-lang-defvar c-opt-asm-stmt-key (c-lang-const c-opt-asm-stmt-key))
2438 (c-lang-defconst c-case-kwds
2439 "The keyword(s) which introduce a \"case\" like construct.
2440 This construct is \"<keyword> <expression> :\"."
2444 (c-lang-defconst c-case-kwds-regexp
2445 ;; Adorned regexp matching any "case"-like keyword.
2446 t (c-make-keywords-re t (c-lang-const c-case-kwds)))
2447 (c-lang-defvar c-case-kwds-regexp (c-lang-const c-case-kwds-regexp))
2449 (c-lang-defconst c-label-kwds
2450 "Keywords introducing colon terminated labels in blocks."
2451 t '("case" "default"))
2453 (c-lang-defconst c-label-kwds-regexp
2454 ;; Adorned regexp matching any keyword that introduces a label.
2455 t (c-make-keywords-re t (c-lang-const c-label-kwds)))
2456 (c-lang-defvar c-label-kwds-regexp (c-lang-const c-label-kwds-regexp))
2458 (c-lang-defconst c-before-label-kwds
2459 "Keywords that might be followed by a label identifier."
2461 (java pike) (append '("break" "continue")
2462 (c-lang-const c-before-label-kwds))
2466 (c-lang-defconst c-constant-kwds
2467 "Keywords for constants."
2469 c '("NULL" ;; Not a keyword, but practically works as one.
2470 "false" "true") ; Defined in C99.
2473 (c-lang-const c-constant-kwds c))
2474 objc '("nil" "Nil" "YES" "NO" "NS_DURING" "NS_HANDLER" "NS_ENDHANDLER")
2475 idl '("TRUE" "FALSE")
2476 java '("true" "false" "null") ; technically "literals", not keywords
2477 pike '("UNDEFINED")) ;; Not a keyword, but practically works as one.
2479 (c-lang-defconst c-primary-expr-kwds
2480 "Keywords besides constants and operators that start primary expressions."
2482 c++ '("operator" "this")
2483 objc '("super" "self")
2485 pike '("this")) ;; Not really a keyword, but practically works as one.
2487 (c-lang-defconst c-expr-kwds
2488 ;; Keywords that can occur anywhere in expressions. Built from
2489 ;; `c-primary-expr-kwds' and all keyword operators in `c-operators'.
2490 t (c--delete-duplicates
2491 (append (c-lang-const c-primary-expr-kwds)
2492 (c-filter-ops (c-lang-const c-operator-list)
2494 "\\`\\(\\w\\|\\s_\\)+\\'"))
2495 :test 'string-equal))
2497 (c-lang-defconst c-lambda-kwds
2498 "Keywords that start lambda constructs, i.e. function definitions in
2503 (c-lang-defconst c-inexpr-block-kwds
2504 "Keywords that start constructs followed by statement blocks which can
2505 be used in expressions \(the gcc extension for this in C and C++ is
2506 handled separately by `c-recognize-paren-inexpr-blocks')."
2508 pike '("catch" "gauge"))
2510 (c-lang-defconst c-inexpr-class-kwds
2511 "Keywords that can start classes inside expressions."
2516 (c-lang-defconst c-inexpr-brace-list-kwds
2517 "Keywords that can start brace list blocks inside expressions.
2518 Note that Java specific rules are currently applied to tell this from
2519 `c-inexpr-class-kwds'."
2523 (c-lang-defconst c-opt-inexpr-brace-list-key
2524 ;; Regexp matching the start of a brace list in an expression, or
2525 ;; nil in languages that don't have such things. This should not
2526 ;; match brace lists recognized through `c-special-brace-lists'.
2527 t (and (c-lang-const c-inexpr-brace-list-kwds)
2528 (c-make-keywords-re t (c-lang-const c-inexpr-brace-list-kwds))))
2529 (c-lang-defvar c-opt-inexpr-brace-list-key
2530 (c-lang-const c-opt-inexpr-brace-list-key))
2532 (c-lang-defconst c-decl-block-key
2533 ;; Regexp matching keywords in any construct that contain another
2534 ;; declaration level, i.e. that isn't followed by a function block
2535 ;; or brace list. When the first submatch matches, it's an
2536 ;; unambiguous construct, otherwise it's an ambiguous match that
2537 ;; might also be the return type of a function declaration.
2538 t (let* ((decl-kwds (append (c-lang-const c-class-decl-kwds)
2539 (c-lang-const c-other-block-decl-kwds)
2540 (c-lang-const c-inexpr-class-kwds)))
2541 (unambiguous (c--set-difference decl-kwds
2542 (c-lang-const c-type-start-kwds)
2543 :test 'string-equal))
2544 (ambiguous (c--intersection decl-kwds
2545 (c-lang-const c-type-start-kwds)
2546 :test 'string-equal)))
2548 (concat (c-make-keywords-re t unambiguous)
2550 (c-make-keywords-re t ambiguous))
2551 (c-make-keywords-re t unambiguous))))
2552 (c-lang-defvar c-decl-block-key (c-lang-const c-decl-block-key))
2554 (c-lang-defconst c-bitfield-kwds
2555 "Keywords that can introduce bitfields."
2557 (c c++ objc) '("char" "int" "long" "signed" "unsigned"))
2559 (c-lang-defconst c-opt-bitfield-key
2560 ;; Regexp matching the start of a bitfield (not uniquely), or nil in
2561 ;; languages without bitfield support.
2563 (c c++) (c-make-keywords-re t (c-lang-const c-bitfield-kwds)))
2564 (c-lang-defvar c-opt-bitfield-key (c-lang-const c-opt-bitfield-key))
2566 (c-lang-defconst c-other-kwds
2567 "Keywords not accounted for by any other `*-kwds' language constant."
2570 ;; In CORBA CIDL: (These are declaration keywords that never
2571 ;; can start a declaration.)
2572 "entity" "process" "service" "session" "storage"))
2575 ;;; Constants built from keywords.
2577 ;; Note: No `*-kwds' language constants may be defined below this point.
2580 (defconst c-kwds-lang-consts
2581 ;; List of all the language constants that contain keyword lists.
2583 (mapatoms (lambda (sym)
2584 (when (and (boundp sym)
2585 (string-match "-kwds\\'" (symbol-name sym)))
2586 ;; Make the list of globally interned symbols
2587 ;; instead of ones interned in `c-lang-constants'.
2588 (setq list (cons (intern (symbol-name sym)) list))))
2592 (c-lang-defconst c-keywords
2593 ;; All keywords as a list.
2594 t (c--delete-duplicates
2595 (c-lang-defconst-eval-immediately
2596 `(append ,@(mapcar (lambda (kwds-lang-const)
2597 `(c-lang-const ,kwds-lang-const))
2600 :test 'string-equal))
2602 (c-lang-defconst c-keywords-regexp
2603 ;; All keywords as an adorned regexp.
2604 t (c-make-keywords-re t (c-lang-const c-keywords)))
2605 (c-lang-defvar c-keywords-regexp (c-lang-const c-keywords-regexp))
2607 (c-lang-defconst c-keyword-member-alist
2608 ;; An alist with all the keywords in the cars. The cdr for each
2609 ;; keyword is a list of the symbols for the `*-kwds' lists that
2611 t (let ((kwd-list-alist
2612 (c-lang-defconst-eval-immediately
2613 `(list ,@(mapcar (lambda (kwds-lang-const)
2614 `(cons ',kwds-lang-const
2615 (c-lang-const ,kwds-lang-const)))
2616 c-kwds-lang-consts))))
2617 lang-const kwd-list kwd
2619 (while kwd-list-alist
2620 (setq lang-const (caar kwd-list-alist)
2621 kwd-list (cdar kwd-list-alist)
2622 kwd-list-alist (cdr kwd-list-alist))
2624 (setq kwd (car kwd-list)
2625 kwd-list (cdr kwd-list))
2626 (unless (setq elem (assoc kwd result-alist))
2627 (setq result-alist (cons (setq elem (list kwd)) result-alist)))
2628 (unless (memq lang-const (cdr elem))
2629 (setcdr elem (cons lang-const (cdr elem))))))
2632 (c-lang-defvar c-keywords-obarray
2633 ;; An obarray containing all keywords as symbols. The property list
2634 ;; of each symbol has a non-nil entry for the specific `*-kwds'
2635 ;; lists it's a member of.
2637 ;; E.g. to see whether the string str contains a keyword on
2638 ;; `c-class-decl-kwds', one can do like this:
2639 ;; (get (intern-soft str c-keyword-obarray) 'c-class-decl-kwds)
2640 ;; Which preferably is written using the associated functions in
2642 ;; (c-keyword-member (c-keyword-sym str) 'c-class-decl-kwds)
2644 ;; The obarray is not stored directly as a language constant since
2645 ;; the printed representation for obarrays used in .elc files isn't
2648 (let* ((alist (c-lang-const c-keyword-member-alist))
2650 (obarray (make-vector (* (length alist) 2) 0)))
2652 (setq kwd (caar alist)
2653 lang-const-list (cdar alist)
2655 (setplist (intern kwd obarray)
2656 ;; Emacs has an odd bug that causes `mapcan' to fail
2657 ;; with unintelligible errors. (XEmacs works.)
2658 ;; (2015-06-24): This bug has not yet been fixed.
2659 ;;(mapcan (lambda (lang-const)
2660 ;; (list lang-const t))
2662 (apply 'nconc (mapcar (lambda (lang-const)
2663 (list lang-const t))
2667 (c-lang-defconst c-regular-keywords-regexp
2668 ;; Adorned regexp matching all keywords that should be fontified
2669 ;; with the keywords face. I.e. that aren't types or constants.
2670 t (c-make-keywords-re t
2671 (c--set-difference (c-lang-const c-keywords)
2672 (append (c-lang-const c-primitive-type-kwds)
2673 (c-lang-const c-constant-kwds))
2674 :test 'string-equal)))
2675 (c-lang-defvar c-regular-keywords-regexp
2676 (c-lang-const c-regular-keywords-regexp))
2678 (c-lang-defconst c-primary-expr-regexp
2679 ;; Regexp matching the start of any primary expression, i.e. any
2680 ;; literal, symbol, prefix operator, and '('. It doesn't need to
2681 ;; exclude keywords; they are excluded afterwards unless the second
2682 ;; submatch matches. If the first but not the second submatch
2683 ;; matches then it is an ambiguous primary expression; it could also
2684 ;; be a match of e.g. an infix operator. (The case with ambiguous
2685 ;; keyword operators isn't handled.)
2687 t (let* ((prefix-ops
2688 (c-filter-ops (c-lang-const c-operators)
2691 ;; Filter out the special case prefix
2692 ;; operators that are close parens.
2693 (not (string-match "\\s)" op)))))
2695 (nonkeyword-prefix-ops
2696 (c-filter-ops prefix-ops
2698 "\\`\\(\\s.\\|\\s(\\|\\s)\\)+\\'"))
2701 (c-filter-ops (c-lang-const c-operators)
2706 right-assoc-sequence)
2709 (unambiguous-prefix-ops (c--set-difference nonkeyword-prefix-ops
2711 :test 'string-equal))
2712 (ambiguous-prefix-ops (c--intersection nonkeyword-prefix-ops
2714 :test 'string-equal)))
2718 ;; Take out all symbol class operators from `prefix-ops' and make the
2719 ;; first submatch from them together with `c-primary-expr-kwds'.
2720 (c-make-keywords-re t
2721 (append (c-lang-const c-primary-expr-kwds)
2722 (c--set-difference prefix-ops nonkeyword-prefix-ops
2723 :test 'string-equal)))
2726 ;; Match all ambiguous operators.
2727 (c-make-keywords-re nil
2728 (c--intersection nonkeyword-prefix-ops in-or-postfix-ops
2729 :test 'string-equal))
2733 ;; Now match all other symbols.
2734 (c-lang-const c-symbol-start)
2737 ;; The chars that can start integer and floating point
2742 ;; The unambiguous operators from `prefix-ops'.
2743 (c-make-keywords-re nil
2744 (c--set-difference nonkeyword-prefix-ops in-or-postfix-ops
2745 :test 'string-equal))
2748 ;; Match string and character literals.
2750 (if (memq 'gen-string-delim c-emacs-features)
2753 (c-lang-defvar c-primary-expr-regexp (c-lang-const c-primary-expr-regexp))
2756 ;;; Additional constants for parser-level constructs.
2758 (c-lang-defconst c-decl-start-colon-kwd-re
2759 "Regexp matching a keyword that is followed by a colon, where
2760 the whole construct can precede a declaration.
2761 E.g. \"public:\" in C++."
2763 c++ (c-make-keywords-re t (c-lang-const c-protection-kwds)))
2764 (c-lang-defvar c-decl-start-colon-kwd-re
2765 (c-lang-const c-decl-start-colon-kwd-re))
2767 (c-lang-defconst c-decl-prefix-re
2768 "Regexp matching something that might precede a declaration, cast or
2769 label, such as the last token of a preceding statement or declaration.
2770 This is used in the common situation where a declaration or cast
2771 doesn't start with any specific token that can be searched for.
2773 The regexp should not match bob; that is done implicitly. It can't
2774 require a match longer than one token. The end of the token is taken
2775 to be at the end of the first submatch, which is assumed to always
2776 match. It's undefined whether identifier syntax (see
2777 `c-identifier-syntax-table') is in effect or not. This regexp is
2778 assumed to be a superset of `c-label-prefix-re' if
2779 `c-recognize-colon-labels' is set.
2781 Besides this, `c-decl-start-kwds' is used to find declarations.
2783 Note: This variable together with `c-decl-start-re' and
2784 `c-decl-start-kwds' is only used to detect \"likely\"
2785 declaration/cast/label starts. I.e. they might produce more matches
2786 but should not miss anything (or else it's necessary to use text
2787 properties - see the next note). Wherever they match, the following
2788 construct is analyzed to see if it indeed is a declaration, cast or
2789 label. That analysis is not cheap, so it's important that not too
2790 many false matches are triggered.
2792 Note: If a declaration/cast/label start can't be detected with this
2793 variable, it's necessary to use the `c-type' text property with the
2794 value `c-decl-end' on the last char of the last token preceding the
2795 declaration. See the comment blurb at the start of cc-engine.el for
2798 ;; We match a sequence of characters to skip over things like \"};\"
2799 ;; more quickly. We match ")" in C for K&R region declarations, and
2800 ;; in all languages except Java for when a cpp macro definition
2801 ;; begins with a declaration.
2803 java "\\([{}(;,<]+\\)"
2804 ;; Match "<" in C++ to get the first argument in a template arglist.
2805 ;; In that case there's an additional check in `c-find-decl-spots'
2806 ;; that it got open paren syntax. Match ":" to aid in picking up
2807 ;; "public:", etc. This involves additional checks in
2808 ;; `c-find-decl-prefix-search' to prevent a match of identifiers
2810 c++ "\\([{}();:,<]+\\)"
2811 ;; Additionally match the protection directives in Objective-C.
2812 ;; Note that this doesn't cope with the longer directives, which we
2813 ;; would have to match from start to end since they don't end with
2814 ;; any easily recognized characters.
2815 objc (concat "\\([{}();,]+\\|"
2816 (c-make-keywords-re nil (c-lang-const c-protection-kwds))
2818 ;; Pike is like C but we also match "[" for multiple value
2819 ;; assignments and type casts.
2820 pike "\\([{}()[;,]+\\)")
2821 (c-lang-defvar c-decl-prefix-re (c-lang-const c-decl-prefix-re)
2824 (c-lang-defconst c-decl-start-re
2825 "Regexp matching the start of any declaration, cast or label.
2826 It's used on the token after the one `c-decl-prefix-re' matched. This
2827 regexp should not try to match those constructs accurately as it's
2828 only used as a sieve to avoid spending more time checking other
2830 t (c-lang-const c-identifier-start))
2831 (c-lang-defvar c-decl-start-re (c-lang-const c-decl-start-re))
2833 (c-lang-defconst c-decl-prefix-or-start-re
2834 ;; Regexp matching something that might precede or start a
2835 ;; declaration, cast or label.
2837 ;; If the first submatch matches, it's taken to match the end of a
2838 ;; token that might precede such a construct, e.g. ';', '}' or '{'.
2839 ;; It's built from `c-decl-prefix-re'.
2841 ;; If the first submatch did not match, the match of the whole
2842 ;; regexp is taken to be at the first token in the declaration.
2843 ;; `c-decl-start-re' is not checked in this case.
2845 ;; Design note: The reason the same regexp is used to match both
2846 ;; tokens that precede declarations and start them is to avoid an
2847 ;; extra regexp search from the previous declaration spot in
2848 ;; `c-find-decl-spots'. Users of `c-find-decl-spots' also count on
2849 ;; that it finds all declaration/cast/label starts in approximately
2850 ;; linear order, so we can't do the searches in two separate passes.
2851 t (if (c-lang-const c-decl-start-kwds)
2852 (concat (c-lang-const c-decl-prefix-re)
2854 (c-make-keywords-re t (c-lang-const c-decl-start-kwds)))
2855 (c-lang-const c-decl-prefix-re)))
2856 (c-lang-defvar c-decl-prefix-or-start-re
2857 (c-lang-const c-decl-prefix-or-start-re)
2860 (c-lang-defconst c-cast-parens
2861 ;; List containing the paren characters that can open a cast, or nil in
2862 ;; languages without casts.
2863 t (c-filter-ops (c-lang-const c-operators)
2866 (lambda (op) (elt op 0))))
2867 (c-lang-defvar c-cast-parens (c-lang-const c-cast-parens))
2869 (c-lang-defconst c-block-prefix-disallowed-chars
2870 "List of syntactically relevant characters that never can occur before
2871 the open brace in any construct that contains a brace block, e.g. in
2872 the \"class Foo: public Bar\" part of:
2874 class Foo: public Bar {int x();} a, *b;
2876 If parens can occur, the chars inside those aren't filtered with this
2879 `<' and `>' should be disallowed even if angle bracket arglists can
2880 occur. That since the search function needs to stop at them anyway to
2881 ensure they are given paren syntax.
2883 This is used to skip backward from the open brace to find the region
2884 in which to look for a construct like \"class\", \"enum\",
2885 \"namespace\" or whatever. That skipping should be as tight as
2886 possible for good performance."
2888 ;; Default to all chars that only occurs in nonsymbol tokens outside
2890 t (c--set-difference
2891 (c-lang-const c-nonsymbol-token-char-list)
2892 (c-filter-ops (append (c-lang-const c-identifier-ops)
2894 (c-lang-const c-after-id-concat-ops))))
2899 (while (string-match "\\(\\s.\\|\\s(\\|\\s)\\)"
2901 (setq res (cons (aref op (match-beginning 1)) res)
2905 ;; Allow cpp operations (where applicable).
2906 t (if (c-lang-const c-opt-cpp-prefix)
2907 (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2909 (c-lang-const c-block-prefix-disallowed-chars))
2911 ;; Allow ':' for inherit list starters.
2912 (c++ objc idl) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2915 ;; Allow ',' for multiple inherits.
2916 (c++ java) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2919 ;; Allow parentheses for anonymous inner classes in Java and class
2920 ;; initializer lists in Pike.
2921 (java pike) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2924 ;; Allow '"' for extern clauses (e.g. extern "C" {...}).
2925 (c c++ objc) (c--set-difference (c-lang-const c-block-prefix-disallowed-chars)
2928 (c-lang-defconst c-block-prefix-charset
2929 ;; `c-block-prefix-disallowed-chars' as an inverted charset suitable
2930 ;; for `c-syntactic-skip-backward'.
2931 t (c-make-bare-char-alt (c-lang-const c-block-prefix-disallowed-chars) t))
2932 (c-lang-defvar c-block-prefix-charset (c-lang-const c-block-prefix-charset))
2934 (c-lang-defconst c-type-decl-prefix-key
2935 "Regexp matching any declarator operator that might precede the
2936 identifier in a declaration, e.g. the \"*\" in \"char *argv\". This
2937 regexp should match \"(\" if parentheses are valid in declarators.
2938 The end of the first submatch is taken as the end of the operator.
2939 Identifier syntax is in effect when this is matched \(see
2940 `c-identifier-syntax-table')."
2941 t (if (c-lang-const c-type-modifier-kwds)
2942 (concat (regexp-opt (c-lang-const c-type-modifier-kwds) t) "\\>")
2943 ;; Default to a regexp that never matches.
2945 ;; Check that there's no "=" afterwards to avoid matching tokens
2947 (c objc) (concat "\\("
2950 (c-lang-const c-type-decl-prefix-key)
2956 (c-lang-const c-type-decl-prefix-key)
2959 ;; If this matches there's special treatment in
2960 ;; `c-font-lock-declarators' and
2961 ;; `c-font-lock-declarations' that check for a
2962 ;; complete name followed by ":: *".
2963 (c-lang-const c-identifier-start)
2967 pike "\\(\\*\\)\\([^=]\\|$\\)")
2968 (c-lang-defvar c-type-decl-prefix-key (c-lang-const c-type-decl-prefix-key)
2971 (c-lang-defconst c-type-decl-suffix-key
2972 "Regexp matching the declarator operators that might follow after the
2973 identifier in a declaration, e.g. the \"[\" in \"char argv[]\". This
2974 regexp should match \")\" if parentheses are valid in declarators. If
2975 it matches an open paren of some kind, the type declaration check
2976 continues at the corresponding close paren, otherwise the end of the
2977 first submatch is taken as the end of the operator. Identifier syntax
2978 is in effect when this is matched (see `c-identifier-syntax-table')."
2979 ;; Default to a regexp that matches `c-type-modifier-kwds' and a
2980 ;; function argument list parenthesis.
2981 t (if (c-lang-const c-type-modifier-kwds)
2983 (regexp-opt (c-lang-const c-type-modifier-kwds) t) "\\>"
2986 (c c++ objc) (concat
2989 (if (c-lang-const c-type-modifier-kwds)
2992 ;; "throw" in `c-type-modifier-kwds' is followed
2993 ;; by a parenthesis list, but no extra measures
2994 ;; are necessary to handle that.
2995 (regexp-opt (c-lang-const c-type-modifier-kwds) t)
3001 (c-lang-defvar c-type-decl-suffix-key (c-lang-const c-type-decl-suffix-key)
3004 (c-lang-defconst c-after-suffixed-type-decl-key
3005 "This regexp is matched after a declarator expression where
3006 `c-type-decl-suffix-key' has matched. If it matches then the
3007 construct is taken as a declaration. It's typically used to match the
3008 beginning of a function body or whatever might occur after the
3009 function header in a function declaration or definition. It's
3010 undefined whether identifier syntax (see `c-identifier-syntax-table')
3011 is in effect or not.
3013 Note that it's used in cases like after \"foo (bar)\" so it should
3014 only match when it's certain that it's a declaration, e.g., \"{\" but
3015 not \",\" or \";\"."
3017 ;; If K&R style declarations should be recognized then one could
3018 ;; consider to match the start of any symbol since we want to match
3019 ;; the start of the first declaration in the "K&R region". That
3020 ;; could however produce false matches on code like "FOO(bar) x"
3021 ;; where FOO is a cpp macro, so it's better to leave it out and rely
3022 ;; on the other heuristics in that case.
3023 t (if (c-lang-const c-postfix-spec-kwds)
3024 ;; Add on the keywords in `c-postfix-spec-kwds'.
3025 (concat (c-lang-const c-after-suffixed-type-decl-key)
3027 (c-make-keywords-re t (c-lang-const c-postfix-spec-kwds)))
3028 (c-lang-const c-after-suffixed-type-decl-key))
3029 ;; Also match the colon that starts a base class initializer list in
3030 ;; C++. That can be confused with a function call before the colon
3031 ;; in a ? : operator, but we count on that `c-decl-prefix-re' won't
3032 ;; match before such a thing (as a declaration-level construct;
3033 ;; matches inside arglist contexts are already excluded).
3035 (c-lang-defvar c-after-suffixed-type-decl-key
3036 (c-lang-const c-after-suffixed-type-decl-key)
3039 (c-lang-defconst c-after-suffixed-type-maybe-decl-key
3040 ;; Regexp that in addition to `c-after-suffixed-type-decl-key'
3041 ;; matches ";" and ",".
3042 t (concat "\\(" (c-lang-const c-after-suffixed-type-decl-key) "\\)"
3044 (c-lang-defvar c-after-suffixed-type-maybe-decl-key
3045 (c-lang-const c-after-suffixed-type-maybe-decl-key))
3047 (c-lang-defconst c-opt-type-concat-key
3048 "Regexp matching operators that concatenate types, e.g. the \"|\" in
3049 \"int|string\" in Pike. The end of the first submatch is taken as the
3050 end of the operator. nil in languages without such operators. It's
3051 undefined whether identifier syntax (see `c-identifier-syntax-table')
3052 is in effect or not."
3054 pike "\\([|.&]\\)\\($\\|[^|.&]\\)")
3055 (c-lang-defvar c-opt-type-concat-key (c-lang-const c-opt-type-concat-key)
3058 (c-lang-defconst c-opt-type-suffix-key
3059 "Regexp matching operators that might follow after a type, or nil in
3060 languages that don't have such operators. The end of the first
3061 submatch is taken as the end of the operator. This should not match
3062 things like C++ template arglists if `c-recognize-<>-arglists' is set.
3063 It's undefined whether identifier syntax (see `c-identifier-syntax-table')
3064 is in effect or not."
3066 (c c++ objc pike) "\\(\\.\\.\\.\\)"
3067 java (concat "\\(\\[" (c-lang-const c-simple-ws) "*\\]\\|\\.\\.\\.\\)"))
3068 (c-lang-defvar c-opt-type-suffix-key (c-lang-const c-opt-type-suffix-key))
3070 (c-lang-defvar c-known-type-key
3071 ;; Regexp matching the known type identifiers. This is initialized
3072 ;; from the type keywords and `*-font-lock-extra-types'. The first
3073 ;; submatch is the one that matches the type. Note that this regexp
3074 ;; assumes that symbol constituents like '_' and '$' have word
3077 (when (boundp (c-mode-symbol "font-lock-extra-types"))
3078 (c-mode-var "font-lock-extra-types")))
3080 (delq nil (mapcar (lambda (re)
3081 (when (string-match "[][.*+?^$\\]" re)
3085 (delq nil (mapcar (lambda (re)
3086 (unless (string-match "[][.*+?^$\\]" re)
3091 (append (list (c-make-keywords-re nil
3092 (append (c-lang-const c-primitive-type-kwds)
3098 (c-lang-defconst c-special-brace-lists
3099 "List of open- and close-chars that makes up a pike-style brace list,
3100 i.e., for a ([ ]) list there should be a cons (?\\[ . ?\\]) in this
3103 pike '((?{ . ?}) (?\[ . ?\]) (?< . ?>)))
3104 (c-lang-defvar c-special-brace-lists (c-lang-const c-special-brace-lists))
3106 (c-lang-defconst c-recognize-knr-p
3107 "Non-nil means K&R style argument declarations are valid."
3110 (c-lang-defvar c-recognize-knr-p (c-lang-const c-recognize-knr-p))
3112 (c-lang-defconst c-recognize-typeless-decls
3113 "Non-nil means function declarations without return type should be
3114 recognized. That can introduce an ambiguity with parenthesized macro
3115 calls before a brace block. This setting does not affect declarations
3116 that are preceded by a declaration starting keyword, so
3117 e.g. `c-typeless-decl-kwds' may still be used when it's set to nil."
3119 (c c++ objc java) t)
3120 (c-lang-defvar c-recognize-typeless-decls
3121 (c-lang-const c-recognize-typeless-decls))
3123 (c-lang-defconst c-recognize-<>-arglists
3124 "Non-nil means C++ style template arglists should be handled. More
3125 specifically, this means a comma separated list of types or
3126 expressions surrounded by \"<\" and \">\". It's always preceded by an
3127 identifier or one of the keywords on `c-<>-type-kwds' or
3128 `c-<>-arglist-kwds'. If there's an identifier before then the whole
3129 expression is considered to be a type."
3130 t (or (consp (c-lang-const c-<>-type-kwds))
3131 (consp (c-lang-const c-<>-arglist-kwds)))
3132 java t) ; 2008-10-19. This is crude. The syntax for java
3133 ; generics is not yet coded in CC Mode.
3134 (c-lang-defvar c-recognize-<>-arglists (c-lang-const c-recognize-<>-arglists))
3136 (c-lang-defconst c-<>-notable-chars-re
3137 "A regexp matching any single character notable inside a <...> construct.
3138 This must include \"<\" and \">\", and should include \",\", and
3139 any character which cannot be valid inside such a construct.
3140 This is used in `c-forward-<>-arglist-recur' to try to detect
3141 sequences of tokens which cannot be a template/generic construct.
3142 When \"(\" is present, that defun will attempt to parse a
3143 parenthesized expression inside the template. When \")\" is
3144 present it will treat an unbalanced closing paren as a sign of
3145 the invalidity of the putative template construct."
3148 (c-lang-defvar c-<>-notable-chars-re (c-lang-const c-<>-notable-chars-re))
3150 (c-lang-defconst c-enums-contain-decls
3151 "Non-nil means that an enum structure can contain declarations."
3154 (c-lang-defvar c-enums-contain-decls (c-lang-const c-enums-contain-decls))
3156 (c-lang-defconst c-recognize-paren-inits
3157 "Non-nil means that parenthesis style initializers exist,
3158 i.e. constructs like
3162 in addition to the more classic
3167 (c-lang-defvar c-recognize-paren-inits (c-lang-const c-recognize-paren-inits))
3169 (c-lang-defconst c-recognize-paren-inexpr-blocks
3170 "Non-nil to recognize gcc style in-expression blocks,
3171 i.e. compound statements surrounded by parentheses inside expressions."
3174 (c-lang-defvar c-recognize-paren-inexpr-blocks
3175 (c-lang-const c-recognize-paren-inexpr-blocks))
3177 (c-lang-defconst c-opt-<>-arglist-start
3178 ;; Regexp matching the start of angle bracket arglists in languages
3179 ;; where `c-recognize-<>-arglists' is set. Does not exclude
3180 ;; keywords outside `c-<>-arglist-kwds'. The first submatch is
3181 ;; assumed to surround the preceding symbol. The whole match is
3182 ;; assumed to end directly after the opening "<".
3183 t (if (c-lang-const c-recognize-<>-arglists)
3185 (c-lang-const c-symbol-key)
3187 (c-lang-const c-syntactic-ws)
3189 (c-lang-defvar c-opt-<>-arglist-start (c-lang-const c-opt-<>-arglist-start))
3191 (c-lang-defconst c-opt-<>-arglist-start-in-paren
3192 ;; Regexp that in addition to `c-opt-<>-arglist-start' matches close
3193 ;; parens. The first submatch is assumed to surround
3194 ;; `c-opt-<>-arglist-start'.
3195 t (if (c-lang-const c-opt-<>-arglist-start)
3197 (c-lang-const c-opt-<>-arglist-start)
3199 (c-lang-defvar c-opt-<>-arglist-start-in-paren
3200 (c-lang-const c-opt-<>-arglist-start-in-paren))
3202 (c-lang-defconst c-opt-postfix-decl-spec-key
3203 ;; Regexp matching the beginning of a declaration specifier in the
3204 ;; region between the header and the body of a declaration.
3206 ;; TODO: This is currently not used uniformly; c++-mode and
3207 ;; java-mode each have their own ways of using it.
3210 (c-lang-const c-simple-ws) "*"
3211 "\\(virtual" (c-lang-const c-simple-ws) "+\\)?\\("
3212 (c-make-keywords-re nil (c-lang-const c-protection-kwds))
3213 "\\)" (c-lang-const c-simple-ws) "+"
3214 "\\(" (c-lang-const c-symbol-key) "\\)")
3215 java (c-make-keywords-re t (c-lang-const c-postfix-spec-kwds)))
3216 (c-lang-defvar c-opt-postfix-decl-spec-key
3217 (c-lang-const c-opt-postfix-decl-spec-key))
3219 (c-lang-defconst c-recognize-colon-labels
3220 "Non-nil if generic labels ending with \":\" should be recognized.
3221 That includes labels in code and access keys in classes. This does
3222 not apply to labels recognized by `c-label-kwds' and
3223 `c-opt-extra-label-key'."
3225 (c c++ objc java pike) t)
3226 (c-lang-defvar c-recognize-colon-labels
3227 (c-lang-const c-recognize-colon-labels))
3229 (c-lang-defconst c-label-prefix-re
3230 "Regexp like `c-decl-prefix-re' that matches any token that can precede
3231 a generic colon label. Not used if `c-recognize-colon-labels' is
3234 (c-lang-defvar c-label-prefix-re
3235 (c-lang-const c-label-prefix-re))
3237 (c-lang-defconst c-nonlabel-token-key
3238 "Regexp matching things that can't occur in generic colon labels,
3239 neither in a statement nor in a declaration context. The regexp is
3240 tested at the beginning of every sexp in a suspected label,
3241 i.e. before \":\". Only used if `c-recognize-colon-labels' is set."
3243 ;; All keywords except `c-label-kwds' and `c-protection-kwds'.
3244 (c-make-keywords-re t
3245 (c--set-difference (c-lang-const c-keywords)
3246 (append (c-lang-const c-label-kwds)
3247 (c-lang-const c-protection-kwds))
3248 :test 'string-equal)))
3249 ;; Don't allow string literals, except in AWK. Character constants are OK.
3250 (c objc java pike idl) (concat "\"\\|"
3251 (c-lang-const c-nonlabel-token-key))
3252 ;; Also check for open parens in C++, to catch member init lists in
3253 ;; constructors. We normally allow it so that macros with arguments
3255 c++ (concat "\\s(\\|\"\\|" (c-lang-const c-nonlabel-token-key)))
3256 (c-lang-defvar c-nonlabel-token-key (c-lang-const c-nonlabel-token-key))
3258 (c-lang-defconst c-nonlabel-token-2-key
3259 "Regexp matching things that can't occur two symbols before a colon in
3260 a label construct. This catches C++'s inheritance construct \"class foo
3261 : bar\". Only used if `c-recognize-colon-labels' is set."
3262 t "\\<\\>" ; matches nothing
3263 c++ (c-make-keywords-re t '("class")))
3264 (c-lang-defvar c-nonlabel-token-2-key (c-lang-const c-nonlabel-token-2-key))
3266 (c-lang-defconst c-opt-extra-label-key
3267 "Optional regexp matching labels.
3268 Normally, labels are detected according to `c-nonlabel-token-key',
3269 `c-decl-prefix-re' and `c-nonlabel-decl-prefix-re'. This regexp can
3270 be used if there are additional labels that aren't recognized that
3273 objc (c-make-keywords-re t (c-lang-const c-protection-kwds)))
3274 (c-lang-defvar c-opt-extra-label-key (c-lang-const c-opt-extra-label-key))
3276 (c-lang-defconst c-opt-friend-key
3277 ;; Regexp describing friend declarations classes, or nil in
3278 ;; languages that don't have such things.
3280 ;; TODO: Ought to use `c-prefix-spec-kwds-re' or similar, and the
3281 ;; template skipping isn't done properly. This will disappear soon.
3283 c++ (concat "friend" (c-lang-const c-simple-ws) "+"
3286 (c-lang-const c-simple-ws) "*"
3288 (c-lang-const c-simple-ws) "*"
3290 (c-lang-const c-simple-ws) "+")))
3291 (c-lang-defvar c-opt-friend-key (c-lang-const c-opt-friend-key))
3293 (c-lang-defconst c-opt-method-key
3294 ;; Special regexp to match the start of Objective-C methods. The
3295 ;; first submatch is assumed to end after the + or - key.
3298 ;; TODO: Ought to use a better method than anchoring on bol.
3301 (c-lang-const c-simple-ws) "*"
3302 (concat "\\(" ; Return type.
3304 (c-lang-const c-simple-ws) "*"
3306 "\\(" (c-lang-const c-symbol-key) "\\)"))
3307 (c-lang-defvar c-opt-method-key (c-lang-const c-opt-method-key))
3309 (c-lang-defconst c-type-decl-end-used
3310 ;; Must be set in buffers where the `c-type' text property might be
3311 ;; used with the value `c-decl-end'.
3313 ;; `c-decl-end' is used to mark the ends of labels and access keys
3314 ;; to make interactive refontification work better.
3315 t (or (c-lang-const c-recognize-colon-labels)
3316 (and (c-lang-const c-label-kwds) t))
3317 ;; `c-decl-end' is used to mark the end of the @-style directives in
3320 (c-lang-defvar c-type-decl-end-used (c-lang-const c-type-decl-end-used))
3322 (c-lang-defconst c-maybe-decl-faces
3323 "List of faces that might be put at the start of a type when
3324 `c-font-lock-declarations' runs. This must be evaluated (with `eval') at
3325 runtime to get the actual list of faces. This ensures that face name
3326 aliases in Emacs are resolved."
3329 c-reference-face-name
3330 font-lock-keyword-face)
3331 java (append (c-lang-const c-maybe-decl-faces)
3332 '(font-lock-preprocessor-face)))
3333 (c-lang-defvar c-maybe-decl-faces (c-lang-const c-maybe-decl-faces))
3336 ;;; Wrap up the `c-lang-defvar' system.
3338 ;; Compile in the list of language variables that has been collected
3339 ;; with the `c-lang-defvar' and `c-lang-setvar' macros. Note that the
3340 ;; first element of each is nil.
3341 (defconst c-lang-variable-inits (cc-eval-when-compile c-lang-variable-inits))
3342 (defconst c-emacs-variable-inits (cc-eval-when-compile c-emacs-variable-inits))
3344 ;; Make the `c-lang-setvar' variables buffer local in the current buffer.
3345 ;; These are typically standard emacs variables such as `comment-start'.
3346 (defmacro c-make-emacs-variables-local ()
3348 ,@(mapcar (lambda (init)
3349 `(make-local-variable ',(car init)))
3350 (cdr c-emacs-variable-inits))))
3352 (defun c-make-init-lang-vars-fun (mode)
3353 "Create a function that initializes all the language dependent variables
3356 This function should be evaluated at compile time, so that the
3357 function it returns is byte compiled with all the evaluated results
3358 from the language constants. Use the `c-init-language-vars' macro to
3359 accomplish that conveniently."
3361 (if (cc-bytecomp-is-compiling)
3362 ;; No need to byte compile this lambda since the byte compiler is
3363 ;; smart enough to detect the `funcall' construct in the
3364 ;; `c-init-language-vars' macro below and compile it all straight
3365 ;; into the function that contains `c-init-language-vars'.
3368 ;; This let sets up the context for `c-mode-var' and similar
3369 ;; that could be in the result from `c--macroexpand-all'.
3370 (let ((c-buffer-is-cc-mode ',mode)
3371 current-var source-eval)
3372 (c-make-emacs-variables-local)
3375 (if (eq c-version-sym ',c-version-sym)
3376 (setq ,@(let ((c-buffer-is-cc-mode mode)
3377 (c-lang-const-expansion 'immediate))
3378 ;; `c-lang-const' will expand to the evaluated
3379 ;; constant immediately in `c--macroexpand-all'
3383 `(current-var ',(car init)
3384 ,(car init) ,(c--macroexpand-all
3386 ;; Note: The following `append' copies the
3387 ;; first argument. That list is small, so
3388 ;; this doesn't matter too much.
3389 (append (cdr c-emacs-variable-inits)
3390 (cdr c-lang-variable-inits)))))
3392 ;; This diagnostic message isn't useful for end
3393 ;; users, so it's disabled.
3394 ;;(unless (get ',mode 'c-has-warned-lang-consts)
3395 ;; (message ,(concat "%s compiled with CC Mode %s "
3396 ;; "but loaded with %s - evaluating "
3397 ;; "language constants from source")
3398 ;; ',mode ,c-version c-version)
3399 ;; (put ',mode 'c-has-warned-lang-consts t))
3401 (setq source-eval t)
3402 (let ((init ',(append (cdr c-emacs-variable-inits)
3403 (cdr c-lang-variable-inits))))
3404 (dolist (var-init init)
3405 (setq current-var (car var-init))
3406 (set (car var-init) (eval (cadr var-init))))))
3410 (message "Eval error in the `c-lang-defvar' or `c-lang-setvar' for `%s'%s: %S"
3414 (fallback source eval - %s compiled with CC Mode %s but loaded with %s)"
3415 ',mode ,c-version c-version)
3418 (signal (car err) (cdr err)))))))
3420 ;; Being evaluated from source. Always use the dynamic method to
3421 ;; work well when `c-lang-defvar's in this file are reevaluated
3425 (let ((c-buffer-is-cc-mode ',mode)
3426 (init (append (cdr c-emacs-variable-inits)
3427 (cdr c-lang-variable-inits)))
3429 (c-make-emacs-variables-local)
3432 (dolist (var-init init)
3433 (setq current-var (car var-init))
3434 (set (car var-init) (eval (cadr var-init))))
3439 "Eval error in the `c-lang-defvar' or `c-lang-setver' for `%s' (source eval): %S"
3441 (signal (car err) (cdr err)))))))
3444 (defmacro c-init-language-vars (mode)
3445 "Initialize all the language dependent variables for the given mode.
3446 This macro is expanded at compile time to a form tailored for the mode
3447 in question, so MODE must be a constant. Therefore MODE is not
3448 evaluated and should not be quoted."
3449 `(funcall ,(c-make-init-lang-vars-fun mode)))
3452 (cc-provide 'cc-langs)
3455 ;; indent-tabs-mode: t
3458 ;;; cc-langs.el ends here