2 @node Programs, Running, Text, Top
3 @chapter Editing Programs
7 Emacs has many commands designed to understand the syntax of programming
8 languages such as Lisp and C. These commands can:
12 Move over or kill balanced expressions or @dfn{sexps} (@pxref{Lists}).
14 Move over or mark top-level balanced expressions (@dfn{defuns}, in Lisp;
17 Show how parentheses balance (@pxref{Matching}).
19 Insert, kill, or align comments (@pxref{Comments}).
21 Follow the usual indentation conventions of the language
25 The commands available for words, sentences, and paragraphs are useful in
26 editing code even though their canonical application is for editing human
27 language text. Most symbols contain words (@pxref{Words}); sentences can
28 be found in strings and comments (@pxref{Sentences}). Paragraphs per se
29 are not present in code, but the paragraph commands are useful anyway,
30 because Lisp mode and C mode define paragraphs to begin and end at blank
31 lines (@pxref{Paragraphs}). Judicious use of blank lines to make the
32 program clearer also provides interesting chunks of text for the
33 paragraph commands to work on.
35 The selective display feature is useful for looking at the overall
36 structure of a function (@pxref{Selective Display}). This feature causes
37 only the lines that are indented less than a specified amount to appear
41 * Program Modes:: Major modes for editing programs.
42 * Lists:: Expressions with balanced parentheses.
43 There are editing commands to operate on them.
44 * Defuns:: Each program is made up of separate functions.
45 There are editing commands to operate on them.
46 * Grinding:: Adjusting indentation to show the nesting.
47 * Matching:: Insertion of a close-delimiter flashes matching open.
48 * Comments:: Inserting, illing and aligning comments.
49 * Balanced Editing:: Inserting two matching parentheses at once, etc.
50 * Lisp Completion:: Completion on symbol names in Lisp code.
51 * Documentation:: Getting documentation of functions you plan to call.
52 * Change Log:: Maintaining a change history for your program.
53 * Tags:: Go direct to any function in your program in one
54 command. Tags remembers which file it is in.
55 * Fortran:: Fortran mode and its special features.
56 * Asm Mode:: Asm mode and its special features.
59 @node Program Modes, Lists, Programs, Programs
60 @section Major Modes for Programming Languages
65 Emacs has several major modes for the programming languages Lisp, Scheme (a
66 variant of Lisp), C, Fortran, and Muddle. Ideally, a major mode should be
67 implemented for each programming language you might want to edit with
68 Emacs; but often the mode for one language can serve for other
69 syntactically similar languages. The language modes that exist are those
70 that someone decided to take the trouble to write.
72 There are several variants of Lisp mode, which differ in the way they
73 interface to Lisp execution. @xref{Lisp Modes}.
75 Each of the programming language modes defines the @key{TAB} key to run
76 an indentation function that knows the indentation conventions of that
77 language and updates the current line's indentation accordingly. For
78 example, in C mode @key{TAB} is bound to @code{c-indent-line}. @key{LFD}
79 is normally defined to do @key{RET} followed by @key{TAB}; thus it, too,
80 indents in a mode-specific fashion.
83 @findex backward-delete-char-untabify
84 In most programming languages, indentation is likely to vary from line to
85 line. So the major modes for those languages rebind @key{DEL} to treat a
86 tab as if it were the equivalent number of spaces (using the command
87 @code{backward-delete-char-untabify}). This makes it possible to rub out
88 indentation one column at a time without worrying whether it is made up of
89 spaces or tabs. In these modes, use @kbd{C-b C-d} to delete a tab
90 character before point.
92 Programming language modes define paragraphs to be separated only by
93 blank lines, so that the paragraph commands remain useful. Auto Fill mode,
94 if enabled in a programming language major mode, indents the new lines
99 @vindex lisp-mode-hook
100 @vindex emacs-lisp-mode-hook
101 @vindex lisp-interaction-mode-hook
102 @vindex scheme-mode-hook
103 @vindex muddle-mode-hook
104 Turning on a major mode calls a user-supplied function called the
105 @dfn{mode hook}, which is the value of a Lisp variable. For example,
106 turning on C mode calls the value of the variable @code{c-mode-hook} if
107 that value exists and is non-@code{nil}. Mode hook variables for other
108 programming language modes include @code{lisp-mode-hook},
109 @code{emacs-lisp-mode-hook}, @code{lisp-interaction-mode-hook},
110 @code{scheme-mode-hook}, and @code{muddle-mode-hook}. The mode hook
111 function receives no arguments.@refill
113 @node Lists, Defuns, Program Modes, Programs
114 @section Lists and Sexps
117 By convention, Emacs keys for dealing with balanced expressions are
118 usually @kbd{Control-Meta-} characters. They tend to be analogous in
119 function to their @kbd{Control-} and @kbd{Meta-} equivalents. These commands
120 are usually thought of as pertaining to expressions in programming
121 languages, but can be useful with any language in which some sort of
122 parentheses exist (including English).
127 The commands fall into two classes. Some commands deal only with
128 @dfn{lists} (parenthetical groupings). They see nothing except
129 parentheses, brackets, braces (depending on what must balance in the
130 language you are working with), and escape characters that might be used
133 The other commands deal with expressions or @dfn{sexps}. The word `sexp'
134 is derived from @dfn{s-expression}, the term for a symbolic expression in
135 Lisp. In Emacs, the notion of `sexp' is not limited to Lisp. It
136 refers to an expression in the language your program is written in.
137 Each programming language has its own major mode, which customizes the
138 syntax tables so that expressions in that language count as sexps.
140 Sexps typically include symbols, numbers, and string constants, as well
141 as anything contained in parentheses, brackets, or braces.
143 In languages that use prefix and infix operators, such as C, it is not
144 possible for all expressions to be sexps. For example, C mode does not
145 recognize @samp{foo + bar} as an sexp, even though it @i{is} a C expression;
146 it recognizes @samp{foo} as one sexp and @samp{bar} as another, with the
147 @samp{+} as punctuation between them. This is a fundamental ambiguity:
148 both @samp{foo + bar} and @samp{foo} are legitimate choices for the sexp to
149 move over if point is at the @samp{f}. Note that @samp{(foo + bar)} is a
152 Some languages have obscure forms of syntax for expressions that nobody
153 has bothered to make Emacs understand properly.
155 @c doublewidecommands
158 Move forward over an sexp (@code{forward-sexp}).
160 Move backward over an sexp (@code{backward-sexp}).
162 Kill sexp forward (@code{kill-sexp}).
164 Move up and backward in list structure (@code{backward-up-list}).
166 Move down and forward in list structure (@code{down-list}).
168 Move forward over a list (@code{forward-list}).
170 Move backward over a list (@code{backward-list}).
172 Transpose expressions (@code{transpose-sexps}).
174 Put mark after following expression (@code{mark-sexp}).
180 @findex backward-sexp
181 To move forward over an sexp, use @kbd{C-M-f} (@code{forward-sexp}). If
182 the first significant character after point is an opening delimiter
183 (@samp{(} in Lisp; @samp{(}, @samp{[}, or @samp{@{} in C), @kbd{C-M-f}
184 moves past the matching closing delimiter. If the character begins a
185 symbol, string, or number, @kbd{C-M-f} moves over that. If the character
186 after point is a closing delimiter, @kbd{C-M-f} just moves past it. (This
187 last is not really moving across an sexp; it is an exception which is
188 included in the definition of @kbd{C-M-f} because it is as useful a
189 behavior as anyone can think of for that situation.)@refill
191 The command @kbd{C-M-b} (@code{backward-sexp}) moves backward over a
192 sexp. The detailed rules are like those above for @kbd{C-M-f}, but with
193 directions reversed. If there are any prefix characters (single quote,
194 back quote, and comma, in Lisp) preceding the sexp, @kbd{C-M-b} moves back
197 @kbd{C-M-f} or @kbd{C-M-b} with an argument repeats that operation the
198 specified number of times; with a negative argument, it moves in the
201 In languages such as C where the comment-terminator can be recognized,
202 the sexp commands move across comments as if they were whitespace. In
203 Lisp and other languages where comments run until the end of a line, it
204 is very difficult to ignore comments when parsing backwards; therefore,
205 in such languages the sexp commands treat the text of comments as if it
210 Killing an sexp at a time can be done with @kbd{C-M-k} (@code{kill-sexp}).
211 @kbd{C-M-k} kills the characters that @kbd{C-M-f} would move over.
216 @findex backward-list
217 The @dfn{list commands}, @kbd{C-M-n} (@code{forward-list}) and
218 @kbd{C-M-p} (@code{backward-list}), move over lists like the sexp
219 commands but skip over any number of other kinds of sexps (symbols,
220 strings, etc). In some situations, these commands are useful because
221 they usually ignore comments, since the comments usually do not contain
226 @findex backward-up-list
228 @kbd{C-M-n} and @kbd{C-M-p} stay at the same level in parentheses, when
229 that is possible. To move @i{up} one (or @var{n}) levels, use @kbd{C-M-u}
230 (@code{backward-up-list}).
231 @kbd{C-M-u} moves backward up past one unmatched opening delimiter. A
232 positive argument serves as a repeat count; a negative argument reverses
233 direction of motion and also requests repetition, so it moves forward and
234 up one or more levels.@refill
236 To move @i{down} in list structure, use @kbd{C-M-d}
237 (@code{down-list}). In Lisp mode, where @samp{(} is the only opening
238 delimiter, this is nearly the same as searching for a @samp{(}. An
239 argument specifies the number of levels of parentheses to go down.
241 @cindex transposition
243 @findex transpose-sexps
244 @kbd{C-M-t} (@code{transpose-sexps}) drags the previous sexp across
245 the next one. An argument serves as a repeat count, and a negative
246 argument drags backwards (thus canceling out the effect of @kbd{C-M-t} with
247 a positive argument). An argument of zero, rather than doing nothing,
248 transposes the sexps ending after point and the mark.
252 To make the region be the next sexp in the buffer, use @kbd{C-M-@@}
253 (@code{mark-sexp}) which sets the mark at the same place that
254 @kbd{C-M-f} would move to. @kbd{C-M-@@} takes arguments like
255 @kbd{C-M-f}. In particular, a negative argument is useful for putting
256 the mark at the beginning of the previous sexp.
258 The list and sexp commands' understanding of syntax is completely
259 controlled by the syntax table. Any character can, for example, be
260 declared to be an opening delimiter and act like an open parenthesis.
263 @node Defuns, Grinding, Lists, Programs
267 In Emacs, a parenthetical grouping at the top level in the buffer is
268 called a @dfn{defun}. The name derives from the fact that most
269 top-level lists in Lisp are instances of the special form
270 @code{defun}, but Emacs calls any top-level parenthetical
271 grouping counts a defun regardless of its contents or
272 the programming language. For example, in C, the body of a
273 function definition is a defun.
275 @c doublewidecommands
278 Move to beginning of current or preceding defun
279 (@code{beginning-of-defun}).
281 Move to end of current or following defun (@code{end-of-defun}).
283 Put region around whole current or following defun (@code{mark-defun}).
289 @findex beginning-of-defun
292 The commands to move to the beginning and end of the current defun are
293 @kbd{C-M-a} (@code{beginning-of-defun}) and @kbd{C-M-e} (@code{end-of-defun}).
295 To operate on the current defun, use @kbd{C-M-h} (@code{mark-defun})
296 which puts point at the beginning and the mark at the end of the current
297 or next defun. This is the easiest way to prepare for moving the defun
298 to a different place. In C mode, @kbd{C-M-h} runs the function
299 @code{mark-c-function}, which is almost the same as @code{mark-defun},
300 but which backs up over the argument declarations, function name, and
301 returned data type so that the entire C function is inside the region.
303 @findex compile-defun
304 To compile and evaluate the current defun, use @kbd{M-x compile-defun}.
305 This function prints the results in the minibuffer. If you include an
306 argument, it inserts the value in the current buffer after the defun.
308 Emacs assumes that any open-parenthesis found in the leftmost column is
309 the start of a defun. Therefore, @i{never put an open-parenthesis at the
310 left margin in a Lisp file unless it is the start of a top level list.
311 Never put an open-brace or other opening delimiter at the beginning of a
312 line of C code unless it starts the body of a function.} The most likely
313 problem case is when you want an opening delimiter at the start of a line
314 inside a string. To avoid trouble, put an escape character (@samp{\} in C
315 and Emacs Lisp, @samp{/} in some other Lisp dialects) before the opening
316 delimiter. It will not affect the contents of the string.
318 The original Emacs found defuns by moving upward a
319 level of parentheses until there were no more levels to go up. This
320 required scanning back to the beginning of the buffer for every
321 function. To speed this up, Emacs was changed to assume
322 that any @samp{(} (or other character assigned the syntactic class of
323 opening-delimiter) at the left margin is the start of a defun. This
324 heuristic is nearly always right; however, it mandates the convention
327 @node Grinding, Matching, Defuns, Programs
328 @section Indentation for Programs
332 The best way to keep a program properly indented (``ground'') is to
333 use Emacs to re-indent it as you change the program. Emacs has commands
334 to indent properly either a single line, a specified number of lines, or
335 all of the lines inside a single parenthetical grouping.
339 * Multi-line Indent:: Commands to reindent many lines at once.
340 * Lisp Indent:: Specifying how each Lisp function should be indented.
341 * C Indent:: Choosing an indentation style for C code.
344 @node Basic Indent, Multi-line Indent, Grinding, Grinding
345 @subsection Basic Program Indentation Commands
350 Adjust indentation of current line.
352 Equivalent to @key{RET} followed by @key{TAB} (@code{newline-and-indent}).
356 @findex c-indent-line
357 @findex lisp-indent-line
358 The basic indentation command is @key{TAB}, which gives the current
359 line the correct indentation as determined from the previous lines. The
360 function that @key{TAB} runs depends on the major mode; it is
361 @code{lisp-indent-line} in Lisp mode, @code{c-indent-line} in C mode,
362 etc. These functions understand different syntaxes for different
363 languages, but they all do about the same thing. @key{TAB} in any
364 programming language major mode inserts or deletes whitespace at the
365 beginning of the current line, independent of where point is in the
366 line. If point is inside the whitespace at the beginning of the line,
367 @key{TAB} leaves it at the end of that whitespace; otherwise, @key{TAB}
368 leaves point fixed with respect to the characters around it.
370 Use @kbd{C-q @key{TAB}} to insert a tab at point.
373 @findex newline-and-indent
374 When entering a large amount of new code, use @key{LFD}
375 (@code{newline-and-indent}), which is equivalent to a @key{RET} followed
376 by a @key{TAB}. @key{LFD} creates a blank line, then gives it the
377 appropriate indentation.
379 @key{TAB} indents the second and following lines of the body of a
380 parenthetical grouping each under the preceding one; therefore, if you
381 alter one line's indentation to be nonstandard, the lines below tend
382 to follow it. This is the right behavior in cases where the standard
383 result of @key{TAB} does not look good.
385 Remember that Emacs assumes that an open-parenthesis, open-brace, or
386 other opening delimiter at the left margin (including the indentation
387 routines) is the start of a function. You should therefore never have
388 an opening delimiter in column zero that is not the beginning of a
389 function, not even inside a string. This restriction is vital for
390 making the indentation commands fast. @xref{Defuns}, for more
391 information on this behavior.
393 @node Multi-line Indent, Lisp Indent, Basic Indent, Grinding
394 @subsection Indenting Several Lines
396 Several commands are available to re-indent several lines of code
397 which have been altered or moved to a different level in a list
403 Re-indent all the lines within one list (@code{indent-sexp}).
405 Shift an entire list rigidly sideways so that its first line
406 is properly indented.
408 Re-indent all lines in the region (@code{indent-region}).
414 To re-indent the contents of a single list, position point before the
415 beginning of it and type @kbd{C-M-q}. This key is bound to
416 @code{indent-sexp} in Lisp mode, @code{indent-c-exp} in C mode, and
417 bound to other suitable functions in other modes. The indentation of
418 the line the sexp starts on is not changed; therefore, only the relative
419 indentation within the list, and not its position, is changed. To
420 correct the position as well, type a @key{TAB} before @kbd{C-M-q}.
423 If the relative indentation within a list is correct but the
424 indentation of its beginning is not, go to the line on which the list
425 begins and type @kbd{C-u @key{TAB}}. When you give @key{TAB} a numeric
426 argument, it moves all the lines in the group, starting on the current
427 line, sideways the same amount that the current line moves. The command
428 does not move lines that start inside strings, or C
429 preprocessor lines when in C mode.
432 @findex indent-region
433 Another way to specify a range to be re-indented is with point and
434 mark. The command @kbd{C-M-\} (@code{indent-region}) applies @key{TAB}
435 to every line whose first character is between point and mark.
437 @node Lisp Indent, C Indent, Multi-line Indent, Grinding
438 @subsection Customizing Lisp Indentation
439 @cindex customization
441 The indentation pattern for a Lisp expression can depend on the function
442 called by the expression. For each Lisp function, you can choose among
443 several predefined patterns of indentation, or define an arbitrary one with
446 The standard pattern of indentation is as follows: the second line of the
447 expression is indented under the first argument, if that is on the same
448 line as the beginning of the expression; otherwise, the second line is
449 indented underneath the function name. Each following line is indented
450 under the previous line whose nesting depth is the same.
452 @vindex lisp-indent-offset
453 If the variable @code{lisp-indent-offset} is non-@code{nil}, it overrides
454 the usual indentation pattern for the second line of an expression, so that
455 such lines are always indented @code{lisp-indent-offset} more columns than
458 @vindex lisp-body-indention
459 Certain functions override the standard pattern. Functions
460 whose names start with @code{def} always indent the second line by
461 @code{lisp-body-indention} extra columns beyond the open-parenthesis
462 starting the expression.
464 Individual functions can override the standard pattern in various
465 ways, according to the @code{lisp-indent-function} property of the
466 function name. (Note: @code{lisp-indent-function} was formerly called
467 @code{lisp-indent-hook}). There are four possibilities for this
472 This is the same as no property; the standard indentation pattern is used.
474 The pattern used for function names that start with @code{def} is used for
476 @item a number, @var{number}
477 The first @var{number} arguments of the function are
478 @dfn{distinguished} arguments; the rest are considered the @dfn{body}
479 of the expression. A line in the expression is indented according to
480 whether the first argument on it is distinguished or not. If the
481 argument is part of the body, the line is indented @code{lisp-body-indent}
482 more columns than the open-parenthesis starting the containing
483 expression. If the argument is distinguished and is either the first
484 or second argument, it is indented @i{twice} that many extra columns.
485 If the argument is distinguished and not the first or second argument,
486 the standard pattern is followed for that line.
487 @item a symbol, @var{symbol}
488 @var{symbol} should be a function name; that function is called to
489 calculate the indentation of a line within this expression. The
490 function receives two arguments:
493 The value returned by @code{parse-partial-sexp} (a Lisp primitive for
494 indentation and nesting computation) when it parses up to the
495 beginning of this line.
497 The position at which the line being indented begins.
500 It should return either a number, which is the number of columns of
501 indentation for that line, or a list whose first element is such a
502 number. The difference between returning a number and returning a list
503 is that a number says that all following lines at the same nesting level
504 should be indented just like this one; a list says that following lines
505 might call for different indentations. This makes a difference when the
506 indentation is computed by @kbd{C-M-q}; if the value is a number,
507 @kbd{C-M-q} need not recalculate indentation for the following lines
508 until the end of the list.
511 @node C Indent,, Lisp Indent, Grinding
512 @subsection Customizing C Indentation
514 Two variables control which commands perform C indentation and when.
516 @vindex c-auto-newline
517 If @code{c-auto-newline} is non-@code{nil}, newlines are inserted both
518 before and after braces that you insert and after colons and semicolons.
519 Correct C indentation is done on all the lines that are made this way.
521 @vindex c-tab-always-indent
522 If @code{c-tab-always-indent} is non-@code{nil}, the @key{TAB} command
523 in C mode does indentation only if point is at the left margin or within
524 the line's indentation. If there is non-whitespace to the left of point,
525 @key{TAB} just inserts a tab character in the buffer. Normally,
526 this variable is @code{nil}, and @key{TAB} always reindents the current line.
528 C does not have anything analogous to particular function names for which
529 special forms of indentation are desirable. However, it has a different
530 need for customization facilities: many different styles of C indentation
533 There are six variables you can set to control the style that Emacs C
538 Indentation of C statements within surrounding block. The surrounding
539 block's indentation is the indentation of the line on which the
541 @item c-continued-statement-offset
542 Extra indentation given to a substatement, such as the then-clause of
543 an @code{if} or body of a @code{while}.
545 Extra indentation for lines that start with an open brace.
546 @item c-brace-imaginary-offset
547 An open brace following other text is treated as if it were this far
548 to the right of the start of its line.
549 @item c-argdecl-indent
550 Indentation level of declarations of C function arguments.
552 Extra indentation for a line that is a label, case, or default.
555 @vindex c-indent-level
556 The variable @code{c-indent-level} controls the indentation for C
557 statements with respect to the surrounding block. In the example:
565 the difference in indentation between the lines is @code{c-indent-level}.
566 Its standard value is 2.
568 If the open-brace beginning the compound statement is not at the beginning
569 of its line, the @code{c-indent-level} is added to the indentation of the
570 line, not the column of the open-brace. For example,
578 One popular indentation style is that which results from setting
579 @code{c-indent-level} to 8 and putting open-braces at the end of a line
580 in this way. Another popular style prefers to put the open-brace on a
583 @vindex c-brace-imaginary-offset
584 In fact, the value of the variable @code{c-brace-imaginary-offset} is
585 also added to the indentation of such a statement. Normally this variable
586 is zero. Think of this variable as the imaginary position of the open
587 brace, relative to the first non-blank character on the line. By setting
588 the variable to 4 and @code{c-indent-level} to 0, you can get this style:
596 When @code{c-indent-level} is zero, the statements inside most braces
597 line up exactly under the open brace. An exception are braces in column
598 zero, like those surrounding a function's body. The statements inside
599 those braces are not placed at column zero. Instead,
600 @code{c-brace-offset} and @code{c-continued-statement-offset} (see
601 below) are added to produce a typical offset between brace levels, and
602 the statements are indented that far.
604 @vindex c-continued-statement-offset
605 @code{c-continued-statement-offset} controls the extra indentation for
606 a line that starts within a statement (but not within parentheses or
607 brackets). These lines are usually statements inside other statements,
608 like the then-clauses of @code{if} statements and the bodies of
609 @code{while} statements. The @code{c-continued-statement-offset}
610 parameter determines the difference in indentation between the two lines in:
618 The default value for @code{c-continued-statement-offset} is 2. Some
619 popular indentation styles correspond to a value of zero for
620 @code{c-continued-statement-offset}.
622 @vindex c-brace-offset
623 @code{c-brace-offset} is the extra indentation given to a line that
624 starts with an open-brace. Its standard value is zero;
641 If you set @code{c-brace-offset} to 4, the first example becomes:
648 @vindex c-argdecl-indent
649 @code{c-argdecl-indent} controls the indentation of declarations of the
650 arguments of a C function. It is absolute: argument declarations receive
651 exactly @code{c-argdecl-indent} spaces. The standard value is 5 and
652 results in code like this:
661 @vindex c-label-offset
662 @code{c-label-offset} is the extra indentation given to a line that
663 contains a label, a case statement, or a @code{default:} statement. Its
664 standard value is @minus{}2 and results in code like this:
673 If @code{c-label-offset} were zero, the same code would be indented as:
682 This example assumes that the other variables above also have their
685 Using the indentation style produced by the default settings of the
686 variables just discussed and putting open braces on separate lines
687 produces clear and readable files. For an example, look at any of the C
688 source files of XEmacs.
690 @node Matching, Comments, Grinding, Programs
691 @section Automatic Display of Matching Parentheses
692 @cindex matching parentheses
695 The Emacs parenthesis-matching feature shows you automatically how
696 parentheses match in the text. Whenever a self-inserting character that
697 is a closing delimiter is typed, the cursor moves momentarily to the
698 location of the matching opening delimiter, provided that is visible on
699 the screen. If it is not on the screen, some text starting with that
700 opening delimiter is displayed in the echo area. Either way, you see
701 the grouping you are closing off.
703 In Lisp, automatic matching applies only to parentheses. In C, it
704 also applies to braces and brackets. Emacs knows which characters to regard
705 as matching delimiters based on the syntax table set by the major
708 If the opening delimiter and closing delimiter are mismatched---as
709 in @samp{[x)}---the echo area displays a warning message. The
710 correct matches are specified in the syntax table.
712 @vindex blink-matching-paren
713 @vindex blink-matching-paren-distance
714 Two variables control parenthesis matching displays.
715 @code{blink-matching-paren} turns the feature on or off. The default is
716 @code{t} (match display is on); @code{nil} turns it off.
717 @code{blink-matching-paren-distance} specifies how many characters back
718 Emacs searches to find a matching opening delimiter. If the match is
719 not found in the specified region, scanning stops, and nothing is
720 displayed. This prevents wasting lots of time scanning when there is no
721 match. The default is 4000.
723 @node Comments, Balanced Editing, Matching, Programs
724 @section Manipulating Comments
728 @findex indent-for-comment
730 The comment commands insert, kill and align comments.
735 Insert or align comment (@code{indent-for-comment}).
737 Set comment column (@code{set-comment-column}).
739 Kill comment on current line (@code{kill-comment}).
741 Like @key{RET} followed by inserting and aligning a comment
742 (@code{indent-new-comment-line}).
745 The command that creates a comment is @kbd{Meta-;}
746 (@code{indent-for-comment}). If there is no comment already on the
747 line, a new comment is created and aligned at a specific column called
748 the @dfn{comment column}. Emacs creates the comment by inserting the
749 string at the value of @code{comment-start}; see below. Point is left
750 after that string. If the text of the line extends past the comment
751 column, indentation is done to a suitable boundary (usually, at least
752 one space is inserted). If the major mode has specified a string to
753 terminate comments, that string is inserted after point, to keep the
756 You can also use @kbd{Meta-;} to align an existing comment. If a line
757 already contains the string that starts comments, @kbd{M-;} just moves
758 point after it and re-indents it to the conventional place. Exception:
759 comments starting in column 0 are not moved.
761 Some major modes have special rules for indenting certain kinds of
762 comments in certain contexts. For example, in Lisp code, comments which
763 start with two semicolons are indented as if they were lines of code,
764 instead of at the comment column. Comments which start with three
765 semicolons are supposed to start at the left margin. Emacs understands
766 these conventions by indenting a double-semicolon comment using @key{TAB}
767 and by not changing the indentation of a triple-semicolon comment at all.
770 ;; This function is just an example.
771 ;;; Here either two or three semicolons are appropriate.
773 ;;; And now, the first part of the function:
774 ;; The following line adds one.
775 (1+ x)) ; This line adds one.
778 In C code, a comment preceded on its line by nothing but whitespace
779 is indented like a line of code.
781 Even when an existing comment is properly aligned, @kbd{M-;} is still
782 useful for moving directly to the start of the comment.
786 @kbd{C-u - C-x ;} (@code{kill-comment}) kills the comment on the
787 current line, if there is one. The indentation before the start of the
788 comment is killed as well. If there does not appear to be a comment in
789 the line, nothing happens. To reinsert the comment on another line,
790 move to the end of that line, type first @kbd{C-y}, and then @kbd{M-;}
791 to realign the comment. Note that @kbd{C-u - C-x ;} is not a distinct
792 key; it is @kbd{C-x ;} (@code{set-comment-column}) with a negative
793 argument. That command is programmed to call @code{kill-comment} when
794 called with a negative argument. However, @code{kill-comment} is a
795 valid command which you could bind directly to a key if you wanted to.
797 @subsection Multiple Lines of Comments
801 @cindex Auto Fill mode
802 @findex indent-new-comment-line
803 If you are typing a comment and want to continue it on another line,
804 use the command @kbd{Meta-@key{LFD}} (@code{indent-new-comment-line}),
805 which terminates the comment you are typing, creates a new blank line
806 afterward, and begins a new comment indented under the old one. If
807 Auto Fill mode is on and you go past the fill column while typing, the
808 comment is continued in just this fashion. If point is
809 not at the end of the line when you type @kbd{M-@key{LFD}}, the text on
810 the rest of the line becomes part of the new comment line.
812 @subsection Options Controlling Comments
814 @vindex comment-column
816 @findex set-comment-column
817 The comment column is stored in the variable @code{comment-column}. You
818 can explicitly set it to a number. Alternatively, the command @kbd{C-x ;}
819 (@code{set-comment-column}) sets the comment column to the column point is
820 at. @kbd{C-u C-x ;} sets the comment column to match the last comment
821 before point in the buffer, and then calls @kbd{Meta-;} to align the
822 current line's comment under the previous one. Note that @kbd{C-u - C-x ;}
823 runs the function @code{kill-comment} as described above.
825 @code{comment-column} is a per-buffer variable; altering the variable
826 affects only the current buffer. You can also change the default value.
827 @xref{Locals}. Many major modes initialize this variable
828 for the current buffer.
830 @vindex comment-start-skip
831 The comment commands recognize comments based on the regular expression
832 that is the value of the variable @code{comment-start-skip}. This regexp
833 should not match the null string. It may match more than the comment
834 starting delimiter in the strictest sense of the word; for example, in C
835 mode the value of the variable is @code{@t{"/\\*+ *"}}, which matches extra
836 stars and spaces after the @samp{/*} itself. (Note that @samp{\\} is
837 needed in Lisp syntax to include a @samp{\} in the string, which is needed
838 to deny the first star its special meaning in regexp syntax. @xref{Regexps}.)
840 @vindex comment-start
842 When a comment command makes a new comment, it inserts the value of
843 @code{comment-start} to begin it. The value of @code{comment-end} is
844 inserted after point and will follow the text you will insert
845 into the comment. In C mode, @code{comment-start} has the value
846 @w{@code{"/* "}} and @code{comment-end} has the value @w{@code{" */"}}.
848 @vindex comment-multi-line
849 @code{comment-multi-line} controls how @kbd{M-@key{LFD}}
850 (@code{indent-new-comment-line}) behaves when used inside a comment. If
851 @code{comment-multi-line} is @code{nil}, as it normally is, then
852 @kbd{M-@key{LFD}} terminates the comment on the starting line and starts
853 a new comment on the new following line. If @code{comment-multi-line}
854 is not @code{nil}, then @kbd{M-@key{LFD}} sets up the new following line
855 as part of the same comment that was found on the starting line. This
856 is done by not inserting a terminator on the old line and not inserting
857 a starter on the new line. In languages where multi-line comments are legal,
858 the value you choose for this variable is a matter of taste.
860 @vindex comment-indent-hook
861 The variable @code{comment-indent-hook} should contain a function that
862 is called to compute the indentation for a newly inserted comment or for
863 aligning an existing comment. Major modes set this variable differently.
864 The function is called with no arguments, but with point at the
865 beginning of the comment, or at the end of a line if a new comment is to
866 be inserted. The function should return the column in which the comment
867 ought to start. For example, in Lisp mode, the indent hook function
868 bases its decision on the number of semicolons that begin an existing
869 comment and on the code in the preceding lines.
871 @node Balanced Editing, Lisp Completion, Comments, Programs
872 @section Editing Without Unbalanced Parentheses
876 Put parentheses around next sexp(s) (@code{insert-parentheses}).
878 Move past next close parenthesis and re-indent
879 (@code{move-over-close-and-reindent}).
884 @findex insert-parentheses
885 @findex move-over-close-and-reindent
886 The commands @kbd{M-(} (@code{insert-parentheses}) and @kbd{M-)}
887 (@code{move-over-close-@*and-reindent}) are designed to facilitate a style of
888 editing which keeps parentheses balanced at all times. @kbd{M-(} inserts a
889 pair of parentheses, either together as in @samp{()}, or, if given an
890 argument, around the next several sexps, and leaves point after the open
891 parenthesis. Instead of typing @kbd{( F O O )}, you can type @kbd{M-( F O
892 O}, which has the same effect except for leaving the cursor before the
893 close parenthesis. You can then type @kbd{M-)}, which moves past the
894 close parenthesis, deletes any indentation preceding it (in this example
895 there is none), and indents with @key{LFD} after it.
897 @node Lisp Completion, Documentation, Balanced Editing, Programs
898 @section Completion for Lisp Symbols
899 @cindex completion (symbol names)
901 Completion usually happens in the minibuffer. An exception is
902 completion for Lisp symbol names, which is available in all buffers.
905 @findex lisp-complete-symbol
906 The command @kbd{M-@key{TAB}} (@code{lisp-complete-symbol}) takes the
907 partial Lisp symbol before point to be an abbreviation, and compares it
908 against all non-trivial Lisp symbols currently known to Emacs. Any
909 additional characters that they all have in common are inserted at point.
910 Non-trivial symbols are those that have function definitions, values, or
913 If there is an open-parenthesis immediately before the beginning of
914 the partial symbol, only symbols with function definitions are considered
917 If the partial name in the buffer has more than one possible completion
918 and they have no additional characters in common, a list of all possible
919 completions is displayed in another window.
921 @node Documentation, Change Log, Lisp Completion, Programs
922 @section Documentation Commands
925 @findex describe-function
927 @findex describe-variable
928 As you edit Lisp code to be run in Emacs, you can use the commands
929 @kbd{C-h f} (@code{describe-function}) and @kbd{C-h v}
930 (@code{describe-variable}) to print documentation of functions and
931 variables you want to call. These commands use the minibuffer to
932 read the name of a function or variable to document, and display the
933 documentation in a window.
935 For extra convenience, these commands provide default arguments based on
936 the code in the neighborhood of point. @kbd{C-h f} sets the default to the
937 function called in the innermost list containing point. @kbd{C-h v} uses
938 the symbol name around or adjacent to point as its default.
941 The @kbd{M-x manual-entry} command gives you access to documentation
942 on Unix commands, system calls, and libraries. The command reads a
943 topic as an argument, and displays the Unix manual page for that topic.
944 @code{manual-entry} always searches all 8 sections of the
945 manual and concatenates all the entries it finds. For example,
946 the topic @samp{termcap} finds the description of the termcap library
947 from section 3, followed by the description of the termcap data base
950 @node Change Log, Tags, Documentation, Programs
954 @findex add-change-log-entry
955 The Emacs command @kbd{M-x add-change-log-entry} helps you keep a record
956 of when and why you have changed a program. It assumes that you have a
957 file in which you write a chronological sequence of entries describing
958 individual changes. The default is to store the change entries in a file
959 called @file{ChangeLog} in the same directory as the file you are editing.
960 The same @file{ChangeLog} file therefore records changes for all the files
963 A change log entry starts with a header line that contains your name
964 and the current date. Except for these header lines, every line in the
965 change log starts with a tab. One entry can describe several changes;
966 each change starts with a line starting with a tab and a star. @kbd{M-x
967 add-change-log-entry} visits the change log file and creates a new entry
968 unless the most recent entry is for today's date and your name. In
969 either case, it adds a new line to start the description of another
970 change just after the header line of the entry. When @kbd{M-x
971 add-change-log-entry} is finished, all is prepared for you to edit in
972 the description of what you changed and how. You must then save the
973 change log file yourself.
975 The change log file is always visited in Indented Text mode, which means
976 that @key{LFD} and auto-filling indent each new line like the previous
977 line. This is convenient for entering the contents of an entry, which must
978 be indented. @xref{Text Mode}.
980 Here is an example of the formatting conventions used in the change log
984 Wed Jun 26 19:29:32 1985 Richard M. Stallman (rms at mit-prep)
986 * xdisp.c (try_window_id):
987 If C-k is done at end of next-to-last line,
988 this fn updates window_end_vpos and cannot leave
989 window_end_pos nonnegative (it is zero, in fact).
990 If display is preempted before lines are output,
991 this is inconsistent. Fix by setting
992 blank_end_of_window to nonzero.
994 Tue Jun 25 05:25:33 1985 Richard M. Stallman (rms at mit-prep)
997 Call the auto fill hook if appropriate.
999 * xdisp.c (try_window_id):
1000 If point is found by compute_motion after xp, record that
1001 permanently. If display_text_line sets point position wrong
1002 (case where line is killed, point is at eob and that line is
1003 not displayed), set it again in final compute_motion.
1006 @node Tags, Fortran, Change Log, Programs
1007 @section Tags Tables
1010 A @dfn{tags table} is a description of how a multi-file program is
1011 broken up into files. It lists the names of the component files and the
1012 names and positions of the functions (or other named subunits) in each
1013 file. Grouping the related files makes it possible to search or replace
1014 through all the files with one command. Recording the function names
1015 and positions makes possible the @kbd{M-.} command which finds the
1016 definition of a function by looking up which of the files it is in.
1018 Tags tables are stored in files called @dfn{tags table files}. The
1019 conventional name for a tags table file is @file{TAGS}.
1021 Each entry in the tags table records the name of one tag, the name of the
1022 file that the tag is defined in (implicitly), and the position in that file
1023 of the tag's definition.
1025 Just what names from the described files are recorded in the tags table
1026 depends on the programming language of the described file. They
1027 normally include all functions and subroutines, and may also include
1028 global variables, data types, and anything else convenient. Each name
1029 recorded is called a @dfn{tag}.
1032 * Tag Syntax:: Tag syntax for various types of code and text files.
1033 * Create Tags Table:: Creating a tags table with @code{etags}.
1034 * Select Tags Table:: How to visit a tags table.
1035 * Find Tag:: Commands to find the definition of a specific tag.
1036 * Tags Search:: Using a tags table for searching and replacing.
1037 * List Tags:: Listing and finding tags defined in a file.
1041 @subsection Source File Tag Syntax
1043 Here is how tag syntax is defined for the most popular languages:
1047 In C code, any C function or typedef is a tag, and so are definitions of
1048 @code{struct}, @code{union} and @code{enum}. @code{#define} macro
1049 definitions and @code{enum} constants are also tags, unless you specify
1050 @samp{--no-defines} when making the tags table. Similarly, global
1051 variables are tags, unless you specify @samp{--no-globals}. Use of
1052 @samp{--no-globals} and @samp{--no-defines} can make the tags table file
1056 In C++ code, in addition to all the tag constructs of C code, member
1057 functions are also recognized, and optionally member variables if you
1058 use the @samp{--members} option. Tags for variables and functions in
1059 classes are named @samp{@var{class}::@var{variable}} and
1060 @samp{@var{class}::@var{function}}.
1063 In Java code, tags include all the constructs recognized in C++, plus
1064 the @code{extends} and @code{implements} constructs. Tags for variables
1065 and functions in classes are named @samp{@var{class}.@var{variable}} and
1066 @samp{@var{class}.@var{function}}.
1069 In La@TeX{} text, the argument of any of the commands @code{\chapter},
1070 @code{\section}, @code{\subsection}, @code{\subsubsection},
1071 @code{\eqno}, @code{\label}, @code{\ref}, @code{\cite}, @code{\bibitem},
1072 @code{\part}, @code{\appendix}, @code{\entry}, or @code{\index}, is a
1075 Other commands can make tags as well, if you specify them in the
1076 environment variable @code{TEXTAGS} before invoking @code{etags}. The
1077 value of this environment variable should be a colon-separated list of
1078 commands names. For example,
1081 TEXTAGS="def:newcommand:newenvironment"
1086 specifies (using Bourne shell syntax) that the commands @samp{\def},
1087 @samp{\newcommand} and @samp{\newenvironment} also define tags.
1090 In Lisp code, any function defined with @code{defun}, any variable
1091 defined with @code{defvar} or @code{defconst}, and in general the first
1092 argument of any expression that starts with @samp{(def} in column zero, is
1096 In Scheme code, tags include anything defined with @code{def} or with a
1097 construct whose name starts with @samp{def}. They also include variables
1098 set with @code{set!} at top level in the file.
1101 Several other languages are also supported:
1105 In assembler code, labels appearing at the beginning of a line,
1106 followed by a colon, are tags.
1109 In Bison or Yacc input files, each rule defines as a tag the nonterminal
1110 it constructs. The portions of the file that contain C code are parsed
1114 In Cobol code, paragraphs names are the tags, i.e. any word starting in
1115 column 8 and followed by a full stop.
1118 In Erlang code, the tags are the functions, records, and macros defined
1122 In Fortran code, functions and subroutines are tags.
1125 In Objective C code, tags include Objective C definitions for classes,
1126 class categories, methods and protocols.
1129 In Pascal code, the tags are the functions and procedures defined in
1133 In Perl code, the tags are the procedures defined by the @code{sub}
1137 In Postscript code, the tags are the functions.
1140 In Prolog code, a tag name appears at the left margin.
1143 You can also generate tags based on regexp matching (@pxref{Create
1144 Tags Table}) to handle other formats and languages.
1146 @node Create Tags Table
1147 @subsection Creating Tags Tables
1148 @cindex @code{etags} program
1150 The @code{etags} program is used to create a tags table file. It knows
1151 the syntax of several languages, as described in
1153 the previous section.
1158 Here is how to run @code{etags}:
1161 etags @var{inputfiles}@dots{}
1165 The @code{etags} program reads the specified files, and writes a tags table
1166 named @file{TAGS} in the current working directory. @code{etags}
1167 recognizes the language used in an input file based on its file name and
1168 contents. You can specify the language with the
1169 @samp{--language=@var{name}} option, described below.
1171 If the tags table data become outdated due to changes in the files
1172 described in the table, the way to update the tags table is the same way it
1173 was made in the first place. It is not necessary to do this often.
1175 If the tags table fails to record a tag, or records it for the wrong
1176 file, then Emacs cannot possibly find its definition. However, if the
1177 position recorded in the tags table becomes a little bit wrong (due to
1178 some editing in the file that the tag definition is in), the only
1179 consequence is a slight delay in finding the tag. Even if the stored
1180 position is very wrong, Emacs will still find the tag, but it must
1181 search the entire file for it.
1183 So you should update a tags table when you define new tags that you want
1184 to have listed, or when you move tag definitions from one file to another,
1185 or when changes become substantial. Normally there is no need to update
1186 the tags table after each edit, or even every day.
1188 One tags table can effectively include another. Specify the included
1189 tags file name with the @samp{--include=@var{file}} option when creating
1190 the file that is to include it. The latter file then acts as if it
1191 contained all the files specified in the included file, as well as the
1192 files it directly contains.
1194 If you specify the source files with relative file names when you run
1195 @code{etags}, the tags file will contain file names relative to the
1196 directory where the tags file was initially written. This way, you can
1197 move an entire directory tree containing both the tags file and the
1198 source files, and the tags file will still refer correctly to the source
1201 If you specify absolute file names as arguments to @code{etags}, then
1202 the tags file will contain absolute file names. This way, the tags file
1203 will still refer to the same files even if you move it, as long as the
1204 source files remain in the same place. Absolute file names start with
1205 @samp{/}, or with @samp{@var{device}:/} on MS-DOS and Windows.
1207 When you want to make a tags table from a great number of files, you
1208 may have problems listing them on the command line, because some systems
1209 have a limit on its length. The simplest way to circumvent this limit
1210 is to tell @code{etags} to read the file names from its standard input,
1211 by typing a dash in place of the file names, like this:
1214 find . -name "*.[chCH]" -print | etags -
1217 Use the option @samp{--language=@var{name}} to specify the language
1218 explicitly. You can intermix these options with file names; each one
1219 applies to the file names that follow it. Specify
1220 @samp{--language=auto} to tell @code{etags} to resume guessing the
1221 language from the file names and file contents. Specify
1222 @samp{--language=none} to turn off language-specific processing
1223 entirely; then @code{etags} recognizes tags by regexp matching alone.
1224 @samp{etags --help} prints the list of the languages @code{etags} knows,
1225 and the file name rules for guessing the language.
1227 The @samp{--regex} option provides a general way of recognizing tags
1228 based on regexp matching. You can freely intermix it with file names.
1229 Each @samp{--regex} option adds to the preceding ones, and applies only
1230 to the following files. The syntax is:
1233 --regex=/@var{tagregexp}[/@var{nameregexp}]/
1237 where @var{tagregexp} is used to match the lines to tag. It is always
1238 anchored, that is, it behaves as if preceded by @samp{^}. If you want
1239 to account for indentation, just match any initial number of blanks by
1240 beginning your regular expression with @samp{[ \t]*}. In the regular
1241 expressions, @samp{\} quotes the next character, and @samp{\t} stands
1242 for the tab character. Note that @code{etags} does not handle the other
1243 C escape sequences for special characters.
1245 @cindex interval operator (in regexps)
1246 The syntax of regular expressions in @code{etags} is the same as in
1247 Emacs, augmented with the @dfn{interval operator}, which works as in
1248 @code{grep} and @code{ed}. The syntax of an interval operator is
1249 @samp{\@{@var{m},@var{n}\@}}, and its meaning is to match the preceding
1250 expression at least @var{m} times and up to @var{n} times.
1252 You should not match more characters with @var{tagregexp} than that
1253 needed to recognize what you want to tag. If the match is such that
1254 more characters than needed are unavoidably matched by @var{tagregexp},
1255 you may find useful to add a @var{nameregexp}, in order to narrow the tag
1256 scope. You can find some examples below.
1258 The @samp{-R} option deletes all the regexps defined with
1259 @samp{--regex} options. It applies to the file names following it, as
1260 you can see from the following example:
1263 etags --regex=/@var{reg1}/ voo.doo --regex=/@var{reg2}/ \
1264 bar.ber -R --lang=lisp los.er
1268 Here @code{etags} chooses the parsing language for @file{voo.doo} and
1269 @file{bar.ber} according to their contents. @code{etags} also uses
1270 @var{reg1} to recognize additional tags in @file{voo.doo}, and both
1271 @var{reg1} and @var{reg2} to recognize additional tags in
1272 @file{bar.ber}. @code{etags} uses the Lisp tags rules, and no regexp
1273 matching, to recognize tags in @file{los.er}.
1275 Here are some more examples. The regexps are quoted to protect them
1276 from shell interpretation.
1279 Tag the @code{DEFVAR} macros in the emacs source files:
1282 --regex='/[ \t]*DEFVAR_[A-Z_ \t(]+"\([^"]+\)"/'
1286 Tag VHDL files (this example is a single long line, broken here for
1287 formatting reasons):
1291 --regex='/[ \t]*\(ARCHITECTURE\|CONFIGURATION\) +[^ ]* +OF/'
1292 --regex='/[ \t]*\(ATTRIBUTE\|ENTITY\|FUNCTION\|PACKAGE\
1293 \( BODY\)?\|PROCEDURE\|PROCESS\|TYPE\)[ \t]+\([^ \t(]+\)/\3/'
1297 Tag TCL files (this last example shows the usage of a @var{nameregexp}):
1300 --lang=none --regex='/proc[ \t]+\([^ \t]+\)/\1/'
1303 For a list of the other available @code{etags} options, execute
1304 @code{etags --help}.
1306 @node Select Tags Table, Find Tag, Create Tags Table, Tags
1307 @subsection Selecting a Tags Table
1309 @vindex tag-table-alist
1310 At any time Emacs has one @dfn{selected} tags table, and all the commands
1311 for working with tags tables use the selected one. To select a tags table,
1312 use the variable @code{tag-table-alist}.
1314 The value of @code{tag-table-alist} is a list that determines which
1315 @code{TAGS} files should be active for a given buffer. This is not
1316 really an association list, in that all elements are checked. The car
1317 of each element of this list is a pattern against which the buffers file
1318 name is compared; if it matches, then the cdr of the list should be the
1319 name of the tags table to use. If more than one element of this list
1320 matches the buffers file name, all of the associated tags tables are
1321 used. Earlier ones are searched first.
1323 If the car of elements of this list are strings, they are treated
1324 as regular-expressions against which the file is compared (like the
1325 @code{auto-mode-alist}). If they are not strings, they are evaluated.
1326 If they evaluate to non-@code{nil}, the current buffer is considered to
1329 If the cdr of the elements of this list are strings, they are
1330 assumed to name a tags file. If they name a directory, the string
1331 @file{tags} is appended to them to get the file name. If they are not
1332 strings, they are evaluated and must return an appropriate string.
1337 (setq tag-table-alist
1338 '(("/usr/src/public/perl/" . "/usr/src/public/perl/perl-3.0/")
1339 ("\\.el$" . "/usr/local/emacs/src/")
1340 ("/jbw/gnu/" . "/usr15/degree/stud/jbw/gnu/")
1341 ("" . "/usr/local/emacs/src/")
1345 The example defines the tags table alist in the following way:
1349 Anything in the directory @file{/usr/src/public/perl/}
1350 should use the @file{TAGS} file @file{/usr/src/public/perl/perl-3.0/TAGS}.
1352 Files ending in @file{.el} should use the @file{TAGS} file
1353 @file{/usr/local/emacs/src/TAGS}.
1355 Anything in or below the directory @file{/jbw/gnu/} should use the
1356 @file{TAGS} file @file{/usr15/degree/stud/jbw/gnu/TAGS}.
1359 If you had a file called @file{/usr/jbw/foo.el}, it would use both
1360 @file{TAGS} files, @* @file{/usr/local/emacs/src/TAGS} and
1361 @file{/usr15/degree/stud/jbw/gnu/TAGS} (in that order), because it
1362 matches both patterns.
1364 If the buffer-local variable @code{buffer-tag-table} is set, it names a
1365 tags table that is searched before all others when @code{find-tag} is
1366 executed from this buffer.
1368 If there is a file called @file{TAGS} in the same directory as the file
1369 in question, then that tags file will always be used as well (after the
1370 @code{buffer-tag-table} but before the tables specified by this list).
1372 If the variable @code{tags-file-name} is set, the @file{TAGS} file it names
1373 will apply to all buffers (for backwards compatibility.) It is searched
1376 @vindex tags-always-build-completion-table
1377 If the value of the variable @code{tags-always-build-completion-table}
1378 is @code{t}, the tags file will always be added to the completion table
1379 without asking first, regardless of the size of the tags file.
1381 @vindex tags-file-name
1382 @findex visit-tags-table
1383 The function @kbd{M-x visit-tags-table}, is largely made obsolete by
1384 the variable @code{tag-table-alist}, tells tags commands to use the tags
1385 table file @var{file} first. The @var{file} should be the name of a
1386 file created with the @code{etags} program. A directory name is also
1387 acceptable; it means the file @file{TAGS} in that directory. The
1388 function only stores the file name you provide in the variable
1389 @code{tags-file-name}. Emacs does not actually read in the tags table
1390 contents until you try to use them. You can set the variable explicitly
1391 instead of using @code{visit-tags-table}. The value of the variable
1392 @code{tags-file-name} is the name of the tags table used by all buffers.
1393 This is for backward compatibility, and is largely supplanted by the
1394 variable @code{tag-table-alist}.
1396 @node Find Tag, Tags Search, Select Tags Table, Tags
1397 @subsection Finding a Tag
1399 The most important thing that a tags table enables you to do is to find
1400 the definition of a specific tag.
1403 @item M-.@: @var{tag &optional other-window}
1404 Find first definition of @var{tag} (@code{find-tag}).
1406 Find next alternate definition of last tag specified.
1407 @item C-x 4 . @var{tag}
1408 Find first definition of @var{tag}, but display it in another window
1409 (@code{find-tag-other-window}).
1414 @kbd{M-.}@: (@code{find-tag}) is the command to find the definition of
1415 a specified tag. It searches through the tags table for that tag, as a
1416 string, then uses the tags table information to determine the file in
1417 which the definition is used and the approximate character position of
1418 the definition in the file. Then @code{find-tag} visits the file,
1419 moves point to the approximate character position, and starts searching
1420 ever-increasing distances away for the text that should appear at
1421 the beginning of the definition.
1423 If an empty argument is given (by typing @key{RET}), the sexp in the
1424 buffer before or around point is used as the name of the tag to find.
1425 @xref{Lists}, for information on sexps.
1427 The argument to @code{find-tag} need not be the whole tag name; it can
1428 be a substring of a tag name. However, there can be many tag names
1429 containing the substring you specify. Since @code{find-tag} works by
1430 searching the text of the tags table, it finds the first tag in the table
1431 that the specified substring appears in. To find other tags that match
1432 the substring, give @code{find-tag} a numeric argument, as in @kbd{C-u
1433 M-.}. This does not read a tag name, but continues searching the tag
1434 table's text for another tag containing the same substring last used.
1435 If your keyboard has a real @key{META} key, @kbd{M-0 M-.}@: is an easier
1436 alternative to @kbd{C-u M-.}.
1438 If the optional second argument @var{other-window} is non-@code{nil}, it uses
1439 another window to display the tag.
1440 Multiple active tags tables and completion are supported.
1442 Variables of note include the following:
1444 @vindex tag-table-alist
1445 @vindex tags-file-name
1446 @vindex tags-build-completion-table
1447 @vindex buffer-tag-table
1448 @vindex make-tags-files-invisible
1449 @vindex tag-mark-stack-max
1452 @item tag-table-alist
1453 Controls which tables apply to which buffers.
1454 @item tags-file-name
1455 Stores a default tags table.
1456 @item tags-build-completion-table
1457 Controls completion behavior.
1458 @item buffer-tag-table
1459 Specifies a buffer-local table.
1460 @item make-tags-files-invisible
1461 Sets whether tags tables should be very hidden.
1462 @item tag-mark-stack-max
1463 Specifies how many tags-based hops to remember.
1467 @findex find-tag-other-window
1468 Like most commands that can switch buffers, @code{find-tag} has another
1469 similar command that displays the new buffer in another window. @kbd{C-x 4
1470 .}@: invokes the function @code{find-tag-other-window}. (This key sequence
1471 ends with a period.)
1473 Emacs comes with a tags table file @file{TAGS} (in the directory
1474 containing Lisp libraries) that includes all the Lisp libraries and all
1475 the C sources of Emacs. By specifying this file with @code{visit-tags-table}
1476 and then using @kbd{M-.}@: you can quickly look at the source of any Emacs
1479 @node Tags Search, List Tags, Find Tag, Tags
1480 @subsection Searching and Replacing with Tags Tables
1482 The commands in this section visit and search all the files listed in the
1483 selected tags table, one by one. For these commands, the tags table serves
1484 only to specify a sequence of files to search. A related command is
1485 @kbd{M-x grep} (@pxref{Compilation}).
1488 @item M-x tags-search @key{RET} @var{regexp} @key{RET}
1489 Search for @var{regexp} through the files in the selected tags
1491 @item M-x tags-query-replace @key{RET} @var{regexp} @key{RET} @var{replacement} @key{RET}
1492 Perform a @code{query-replace-regexp} on each file in the selected tags table.
1494 Restart one of the commands above, from the current location of point
1495 (@code{tags-loop-continue}).
1499 @kbd{M-x tags-search} reads a regexp using the minibuffer, then
1500 searches for matches in all the files in the selected tags table, one
1501 file at a time. It displays the name of the file being searched so you
1502 can follow its progress. As soon as it finds an occurrence,
1503 @code{tags-search} returns.
1506 @findex tags-loop-continue
1507 Having found one match, you probably want to find all the rest. To find
1508 one more match, type @kbd{M-,} (@code{tags-loop-continue}) to resume the
1509 @code{tags-search}. This searches the rest of the current buffer, followed
1510 by the remaining files of the tags table.@refill
1512 @findex tags-query-replace
1513 @kbd{M-x tags-query-replace} performs a single
1514 @code{query-replace-regexp} through all the files in the tags table. It
1515 reads a regexp to search for and a string to replace with, just like
1516 ordinary @kbd{M-x query-replace-regexp}. It searches much like @kbd{M-x
1517 tags-search}, but repeatedly, processing matches according to your
1518 input. @xref{Replace}, for more information on query replace.
1520 It is possible to get through all the files in the tags table with a
1521 single invocation of @kbd{M-x tags-query-replace}. But often it is
1522 useful to exit temporarily, which you can do with any input event that
1523 has no special query replace meaning. You can resume the query replace
1524 subsequently by typing @kbd{M-,}; this command resumes the last tags
1525 search or replace command that you did.
1527 The commands in this section carry out much broader searches than the
1528 @code{find-tag} family. The @code{find-tag} commands search only for
1529 definitions of tags that match your substring or regexp. The commands
1530 @code{tags-search} and @code{tags-query-replace} find every occurrence
1531 of the regexp, as ordinary search commands and replace commands do in
1534 These commands create buffers only temporarily for the files that they
1535 have to search (those which are not already visited in Emacs buffers).
1536 Buffers in which no match is found are quickly killed; the others
1539 It may have struck you that @code{tags-search} is a lot like
1540 @code{grep}. You can also run @code{grep} itself as an inferior of
1541 Emacs and have Emacs show you the matching lines one by one. This works
1542 much like running a compilation; finding the source locations of the
1543 @code{grep} matches works like finding the compilation errors.
1546 If you wish to process all the files in a selected tags table, but
1547 @kbd{M-x tags-search} and @kbd{M-x tags-query-replace} are not giving
1548 you the desired result, you can use @kbd{M-x next-file}.
1551 @item C-u M-x next-file
1552 With a numeric argument, regardless of its value, visit the first
1553 file in the tags table and prepare to advance sequentially by files.
1555 Visit the next file in the selected tags table.
1558 @node List Tags,, Tags Search, Tags
1559 @subsection Tags Table Inquiries
1563 Display a list of the tags defined in a specific program file.
1564 @item M-x tags-apropos
1565 Display a list of all tags matching a specified regexp.
1569 @kbd{M-x list-tags} reads the name of one of the files described by the
1570 selected tags table, and displays a list of all the tags defined in that
1571 file. The ``file name'' argument is really just a string to compare
1572 against the names recorded in the tags table; it is read as a string rather
1573 than a file name. Therefore, completion and defaulting are not
1574 available, and you must enter the string the same way it appears in the tag
1575 table. Do not include a directory as part of the file name unless the file
1576 name recorded in the tags table contains that directory.
1578 @findex tags-apropos
1579 @kbd{M-x tags-apropos} is like @code{apropos} for tags. It reads a regexp,
1580 then finds all the tags in the selected tags table whose entries match that
1581 regexp, and displays the tag names found.
1583 @node Fortran, Asm Mode, Tags, Programs
1584 @section Fortran Mode
1585 @cindex Fortran mode
1587 Fortran mode provides special motion commands for Fortran statements and
1588 subprograms, and indentation commands that understand Fortran conventions
1589 of nesting, line numbers, and continuation statements.
1591 Special commands for comments are provided because Fortran comments are
1592 unlike those of other languages.
1594 Built-in abbrevs optionally save typing when you insert Fortran keywords.
1596 @findex fortran-mode
1597 Use @kbd{M-x fortran-mode} to switch to this major mode. Doing so calls
1598 the value of @code{fortran-mode-hook} as a function of no arguments if
1599 that variable has a non-@code{nil} value.
1602 * Motion: Fortran Motion. Moving point by statements or subprograms.
1603 * Indent: Fortran Indent. Indentation commands for Fortran.
1604 * Comments: Fortran Comments. Inserting and aligning comments.
1605 * Columns: Fortran Columns. Measuring columns for valid Fortran.
1606 * Abbrev: Fortran Abbrev. Built-in abbrevs for Fortran keywords.
1609 Fortran mode was contributed by Michael Prange.
1611 @node Fortran Motion, Fortran Indent, Fortran, Fortran
1612 @subsection Motion Commands
1614 Fortran mode provides special commands to move by subprograms (functions
1615 and subroutines) and by statements. There is also a command to put the
1616 region around one subprogram, which is convenient for killing it or moving it.
1618 @kindex C-M-a (Fortran mode)
1619 @kindex C-M-e (Fortran mode)
1620 @kindex C-M-h (Fortran mode)
1621 @kindex C-c C-p (Fortran mode)
1622 @kindex C-c C-n (Fortran mode)
1623 @findex beginning-of-fortran-subprogram
1624 @findex end-of-fortran-subprogram
1625 @findex mark-fortran-subprogram
1626 @findex fortran-previous-statement
1627 @findex fortran-next-statement
1631 Move to beginning of subprogram@*
1632 (@code{beginning-of-fortran-subprogram}).
1634 Move to end of subprogram (@code{end-of-fortran-subprogram}).
1636 Put point at beginning of subprogram and mark at end
1637 (@code{mark-fortran-subprogram}).
1639 Move to beginning of current or next statement
1640 (@code{fortran-next-@*statement}).
1642 Move to beginning of current or previous statement
1643 (@code{fortran-@*previous-statement}).
1646 @node Fortran Indent, Fortran Comments, Fortran Motion, Fortran
1647 @subsection Fortran Indentation
1649 Special commands and features are available for indenting Fortran
1650 code. They make sure various syntactic entities (line numbers, comment line
1651 indicators, and continuation line flags) appear in the columns that are
1652 required for standard Fortran.
1655 * Commands: ForIndent Commands. Commands for indenting Fortran.
1656 * Numbers: ForIndent Num. How line numbers auto-indent.
1657 * Conv: ForIndent Conv. Conventions you must obey to avoid trouble.
1658 * Vars: ForIndent Vars. Variables controlling Fortran indent style.
1661 @node ForIndent Commands, ForIndent Num, Fortran Indent, Fortran Indent
1662 @subsubsection Fortran Indentation Commands
1666 Indent the current line (@code{fortran-indent-line}).
1668 Break the current line and set up a continuation line.
1670 Indent all the lines of the subprogram point is in
1671 (@code{fortran-indent-subprogram}).
1674 @findex fortran-indent-line
1675 @key{TAB} is redefined by Fortran mode to reindent the current line for
1676 Fortran (@code{fortran-indent-line}). Line numbers and continuation
1677 markers are indented to their required columns, and the body of the
1678 statement is independently indented, based on its nesting in the program.
1680 @kindex C-M-q (Fortran mode)
1681 @findex fortran-indent-subprogram
1682 The key @kbd{C-M-q} is redefined as @code{fortran-indent-subprogram}, a
1683 command that reindents all the lines of the Fortran subprogram (function or
1684 subroutine) containing point.
1686 @kindex M-LFD (Fortran mode)
1687 @findex fortran-split-line
1688 The key @kbd{M-@key{LFD}} is redefined as @code{fortran-split-line}, a
1689 command to split a line in the appropriate fashion for Fortran. In a
1690 non-comment line, the second half becomes a continuation line and is
1691 indented accordingly. In a comment line, both halves become separate
1694 @node ForIndent Num, ForIndent Conv, ForIndent Commands, Fortran Indent
1695 @subsubsection Line Numbers and Continuation
1697 If a number is the first non-whitespace in the line, it is assumed to be
1698 a line number and is moved to columns 0 through 4. (Columns are always
1699 counted from 0 in XEmacs.) If the text on the line starts with the
1700 conventional Fortran continuation marker @samp{$}, it is moved to column 5.
1701 If the text begins with any non whitespace character in column 5, it is
1702 assumed to be an unconventional continuation marker and remains in column
1705 @vindex fortran-line-number-indent
1706 Line numbers of four digits or less are normally indented one space.
1707 This amount is controlled by the variable @code{fortran-line-number-indent},
1708 which is the maximum indentation a line number can have. Line numbers
1709 are indented to right-justify them to end in column 4 unless that would
1710 require more than the maximum indentation. The default value of the
1713 @vindex fortran-electric-line-number
1714 Simply inserting a line number is enough to indent it according to these
1715 rules. As each digit is inserted, the indentation is recomputed. To turn
1716 off this feature, set the variable @code{fortran-electric-line-number} to
1717 @code{nil}. Then inserting line numbers is like inserting anything else.
1719 @node ForIndent Conv, ForIndent Vars, ForIndent Num, Fortran Indent
1720 @subsubsection Syntactic Conventions
1722 Fortran mode assumes that you follow certain conventions that simplify
1723 the task of understanding a Fortran program well enough to indent it
1726 @vindex fortran-continuation-char
1729 Two nested @samp{do} loops never share a @samp{continue} statement.
1732 The same character appears in column 5 of all continuation lines. It
1733 is the value of the variable @code{fortran-continuation-char}.
1734 By default, this character is @samp{$}.
1738 If you fail to follow these conventions, the indentation commands may
1739 indent some lines unaesthetically. However, a correct Fortran program will
1740 retain its meaning when reindented even if the conventions are not
1743 @node ForIndent Vars,, ForIndent Conv, Fortran Indent
1744 @subsubsection Variables for Fortran Indentation
1746 @vindex fortran-do-indent
1747 @vindex fortran-if-indent
1748 @vindex fortran-continuation-indent
1749 @vindex fortran-check-all-num-for-matching-do
1750 @vindex fortran-minimum-statement-indent
1751 Several additional variables control how Fortran indentation works.
1754 @item fortran-do-indent
1755 Extra indentation within each level of @samp{do} statement (the default is 3).
1757 @item fortran-if-indent
1758 Extra indentation within each level of @samp{if} statement (the default is 3).
1760 @item fortran-continuation-indent
1761 Extra indentation for bodies of continuation lines (the default is 5).
1763 @item fortran-check-all-num-for-matching-do
1764 If this is @code{nil}, indentation assumes that each @samp{do}
1765 statement ends on a @samp{continue} statement. Therefore, when
1766 computing indentation for a statement other than @samp{continue}, it
1767 can save time by not checking for a @samp{do} statement ending there.
1768 If this is non-@code{nil}, indenting any numbered statement must check
1769 for a @samp{do} that ends there. The default is @code{nil}.
1771 @item fortran-minimum-statement-indent
1772 Minimum indentation for Fortran statements. For standard Fortran,
1773 this is 6. Statement bodies are always indented at least this much.
1776 @node Fortran Comments, Fortran Columns, Fortran Indent, Fortran
1777 @subsection Comments
1779 The usual Emacs comment commands assume that a comment can follow a line
1780 of code. In Fortran, the standard comment syntax requires an entire line
1781 to be just a comment. Therefore, Fortran mode replaces the standard Emacs
1782 comment commands and defines some new variables.
1784 Fortran mode can also handle a non-standard comment syntax where comments
1785 start with @samp{!} and can follow other text. Because only some Fortran
1786 compilers accept this syntax, Fortran mode will not insert such comments
1787 unless you have specified to do so in advance by setting the variable
1788 @code{comment-start} to @samp{"!"} (@pxref{Variables}).
1792 Align comment or insert new comment (@code{fortran-comment-indent}).
1795 Applies to nonstandard @samp{!} comments only.
1798 Turn all lines of the region into comments, or (with arg)
1799 turn them back into real code (@code{fortran-comment-region}).
1802 @kbd{M-;} in Fortran mode is redefined as the command
1803 @code{fortran-comment-indent}. Like the usual @kbd{M-;} command,
1804 it recognizes an existing comment and aligns its text appropriately.
1805 If there is no existing comment, a comment is inserted and aligned.
1807 Inserting and aligning comments is not the same in Fortran mode as in
1808 other modes. When a new comment must be inserted, a full-line comment is
1809 inserted if the current line is blank. On a non-blank line, a
1810 non-standard @samp{!} comment is inserted if you previously specified
1811 you wanted to use them. Otherwise a full-line comment is inserted on a
1812 new line before the current line.
1814 Non-standard @samp{!} comments are aligned like comments in other
1815 languages, but full-line comments are aligned differently. In a
1816 standard full-line comment, the comment delimiter itself must always
1817 appear in column zero. What can be aligned is the text within the
1818 comment. You can choose from three styles of alignment by setting the
1819 variable @code{fortran-comment-indent-style} to one of these values:
1821 @vindex fortran-comment-indent-style
1822 @vindex fortran-comment-line-column
1825 The text is aligned at a fixed column, which is the value of
1826 @code{fortran-comment-line-column}. This is the default.
1828 The text is aligned as if it were a line of code, but with an
1829 additional @code{fortran-comment-line-column} columns of indentation.
1831 Text in full-line columns is not moved automatically.
1834 @vindex fortran-comment-indent-char
1835 You can also specify the character to be used to indent within
1836 full-line comments by setting the variable @code{fortran-comment-indent-char}
1837 to the character you want to use.
1839 @vindex comment-line-start
1840 @vindex comment-line-start-skip
1841 Fortran mode introduces two variables @code{comment-line-start} and
1842 @code{comment-line-start-skip}, which do for full-line comments what
1843 @code{comment-start} and @code{comment-start-skip} do for
1844 ordinary text-following comments. Normally these are set properly by
1845 Fortran mode, so you do not need to change them.
1847 The normal Emacs comment command @kbd{C-x ;} has not been redefined.
1848 It can therefore be used if you use @samp{!} comments, but is useless in
1849 Fortran mode otherwise.
1851 @kindex C-c ; (Fortran mode)
1852 @findex fortran-comment-region
1853 @vindex fortran-comment-region
1854 The command @kbd{C-c ;} (@code{fortran-comment-region}) turns all the
1855 lines of the region into comments by inserting the string @samp{C$$$} at
1856 the front of each one. With a numeric arg, the region is turned back into
1857 live code by deleting @samp{C$$$} from the front of each line. You can
1858 control the string used for the comments by setting the variable
1859 @code{fortran-comment-region}. Note that here we have an example of a
1860 command and a variable with the same name; the two uses of the name never
1861 conflict because in Lisp and in Emacs it is always clear from the context
1862 which one is referred to.
1864 @node Fortran Columns, Fortran Abbrev, Fortran Comments, Fortran
1869 Displays a ``column ruler'' momentarily above the current line
1870 (@code{fortran-column-ruler}).
1872 Splits the current window horizontally so that it is 72 columns wide.
1873 This may help you avoid going over that limit (@code{fortran-window-create}).
1876 @kindex C-c C-r (Fortran mode)
1877 @findex fortran-column-ruler
1878 The command @kbd{C-c C-r} (@code{fortran-column-ruler}) shows a column
1879 ruler above the current line. The comment ruler consists of two lines
1880 of text that show you the locations of columns with special significance
1881 in Fortran programs. Square brackets show the limits of the columns for
1882 line numbers, and curly brackets show the limits of the columns for the
1883 statement body. Column numbers appear above them.
1885 Note that the column numbers count from zero, as always in XEmacs. As
1886 a result, the numbers may not be those you are familiar with; but the
1887 actual positions in the line are standard Fortran.
1889 The text used to display the column ruler is the value of the variable
1890 @code{fortran-comment-ruler}. By changing this variable, you can change
1893 @kindex C-c C-w (Fortran mode)
1894 @findex fortran-window-create
1895 For even more help, use @kbd{C-c C-w} (@code{fortran-window-create}), a
1896 command which splits the current window horizontally, resulting in a window 72
1897 columns wide. When you edit in this window, you can immediately see
1898 when a line gets too wide to be correct Fortran.
1900 @node Fortran Abbrev,, Fortran Columns, Fortran
1901 @subsection Fortran Keyword Abbrevs
1903 Fortran mode provides many built-in abbrevs for common keywords and
1904 declarations. These are the same sort of abbrevs that you can define
1905 yourself. To use them, you must turn on Abbrev mode. @pxref{Abbrevs}.
1907 The built-in abbrevs are unusual in one way: they all start with a
1908 semicolon. You cannot normally use semicolon in an abbrev, but Fortran
1909 mode makes this possible by changing the syntax of semicolon to ``word
1912 For example, one built-in Fortran abbrev is @samp{;c} for
1913 @samp{continue}. If you insert @samp{;c} and then insert a punctuation
1914 character such as a space or a newline, the @samp{;c} changes
1915 automatically to @samp{continue}, provided Abbrev mode is enabled.@refill
1917 Type @samp{;?} or @samp{;C-h} to display a list of all built-in
1918 Fortran abbrevs and what they stand for.
1920 @node Asm Mode,, Fortran, Programs
1924 Asm mode is a major mode for editing files of assembler code. It
1925 defines these commands:
1929 @code{tab-to-tab-stop}.
1931 Insert a newline and then indent using @code{tab-to-tab-stop}.
1933 Insert a colon and then remove the indentation from before the label
1934 preceding colon. Then do @code{tab-to-tab-stop}.
1936 Insert or align a comment.
1939 The variable @code{asm-comment-char} specifies which character
1940 starts comments in assembler syntax.