searching.texi 75.5 KB
Newer Older
Glenn Morris's avatar
Glenn Morris committed
1 2
@c -*-texinfo-*-
@c This is part of the GNU Emacs Lisp Reference Manual.
Paul Eggert's avatar
Paul Eggert committed
3
@c Copyright (C) 1990-1995, 1998-1999, 2001-2016 Free Software
4
@c Foundation, Inc.
Glenn Morris's avatar
Glenn Morris committed
5
@c See the file elisp.texi for copying conditions.
6
@node Searching and Matching
Glenn Morris's avatar
Glenn Morris committed
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
@chapter Searching and Matching
@cindex searching

  GNU Emacs provides two ways to search through a buffer for specified
text: exact string searches and regular expression searches.  After a
regular expression search, you can examine the @dfn{match data} to
determine which text matched the whole regular expression or various
portions of it.

@menu
* String Search::         Search for an exact match.
* Searching and Case::    Case-independent or case-significant searching.
* Regular Expressions::   Describing classes of strings.
* Regexp Search::         Searching for a match for a regexp.
* POSIX Regexps::         Searching POSIX-style for the longest match.
* Match Data::            Finding out which part of the text matched,
                            after a string or regexp search.
Glenn Morris's avatar
Glenn Morris committed
24
* Search and Replace::    Commands that loop, searching and replacing.
Glenn Morris's avatar
Glenn Morris committed
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
* Standard Regexps::      Useful regexps for finding sentences, pages,...
@end menu

  The @samp{skip-chars@dots{}} functions also perform a kind of searching.
@xref{Skipping Characters}.  To search for changes in character
properties, see @ref{Property Search}.

@node String Search
@section Searching for Strings
@cindex string search

  These are the primitive functions for searching through the text in a
buffer.  They are meant for use in programs, but you may call them
interactively.  If you do so, they prompt for the search string; the
arguments @var{limit} and @var{noerror} are @code{nil}, and @var{repeat}
40 41
is 1.  For more details on interactive searching, @pxref{Search,,
Searching and Replacement, emacs, The GNU Emacs Manual}.
Glenn Morris's avatar
Glenn Morris committed
42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73

  These search functions convert the search string to multibyte if the
buffer is multibyte; they convert the search string to unibyte if the
buffer is unibyte.  @xref{Text Representations}.

@deffn Command search-forward string &optional limit noerror repeat
This function searches forward from point for an exact match for
@var{string}.  If successful, it sets point to the end of the occurrence
found, and returns the new value of point.  If no match is found, the
value and side effects depend on @var{noerror} (see below).

In the following example, point is initially at the beginning of the
line.  Then @code{(search-forward "fox")} moves point after the last
letter of @samp{fox}:

@example
@group
---------- Buffer: foo ----------
@point{}The quick brown fox jumped over the lazy dog.
---------- Buffer: foo ----------
@end group

@group
(search-forward "fox")
     @result{} 20

---------- Buffer: foo ----------
The quick brown fox@point{} jumped over the lazy dog.
---------- Buffer: foo ----------
@end group
@end example

74 75
The argument @var{limit} specifies the bound to the search, and should
be a position in the current buffer.  No match extending after
Glenn Morris's avatar
Glenn Morris committed
76 77 78 79 80 81 82 83 84
that position is accepted.  If @var{limit} is omitted or @code{nil}, it
defaults to the end of the accessible portion of the buffer.

@kindex search-failed
What happens when the search fails depends on the value of
@var{noerror}.  If @var{noerror} is @code{nil}, a @code{search-failed}
error is signaled.  If @var{noerror} is @code{t}, @code{search-forward}
returns @code{nil} and does nothing.  If @var{noerror} is neither
@code{nil} nor @code{t}, then @code{search-forward} moves point to the
85 86 87 88 89 90 91 92
upper bound and returns @code{nil}.
@c I see no prospect of this ever changing, and frankly the current
@c behavior seems better, so there seems no need to mention this.
@ignore
(It would be more consistent now to return the new position of point
in that case, but some existing programs may depend on a value of
@code{nil}.)
@end ignore
Glenn Morris's avatar
Glenn Morris committed
93 94 95 96 97

The argument @var{noerror} only affects valid searches which fail to
find a match.  Invalid arguments cause errors regardless of
@var{noerror}.

98 99 100 101 102 103 104 105
If @var{repeat} is a positive number @var{n}, it serves as a repeat
count: the search is repeated @var{n} times, each time starting at the
end of the previous time's match.  If these successive searches
succeed, the function succeeds, moving point and returning its new
value.  Otherwise the search fails, with results depending on the
value of @var{noerror}, as described above.  If @var{repeat} is a
negative number -@var{n}, it serves as a repeat count of @var{n} for a
search in the opposite (backward) direction.
Glenn Morris's avatar
Glenn Morris committed
106 107 108 109
@end deffn

@deffn Command search-backward string &optional limit noerror repeat
This function searches backward from point for @var{string}.  It is
110 111 112
like @code{search-forward}, except that it searches backwards rather
than forwards.  Backward searches leave point at the beginning of the
match.
Glenn Morris's avatar
Glenn Morris committed
113 114 115
@end deffn

@deffn Command word-search-forward string &optional limit noerror repeat
Paul Eggert's avatar
Paul Eggert committed
116
This function searches forward from point for a word match for
Glenn Morris's avatar
Glenn Morris committed
117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139
@var{string}.  If it finds a match, it sets point to the end of the
match found, and returns the new value of point.

Word matching regards @var{string} as a sequence of words, disregarding
punctuation that separates them.  It searches the buffer for the same
sequence of words.  Each word must be distinct in the buffer (searching
for the word @samp{ball} does not match the word @samp{balls}), but the
details of punctuation and spacing are ignored (searching for @samp{ball
boy} does match @samp{ball.  Boy!}).

In this example, point is initially at the beginning of the buffer; the
search leaves it between the @samp{y} and the @samp{!}.

@example
@group
---------- Buffer: foo ----------
@point{}He said "Please!  Find
the ball boy!"
---------- Buffer: foo ----------
@end group

@group
(word-search-forward "Please find the ball, boy.")
Juri Linkov's avatar
Juri Linkov committed
140
     @result{} 39
Glenn Morris's avatar
Glenn Morris committed
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160

---------- Buffer: foo ----------
He said "Please!  Find
the ball boy@point{}!"
---------- Buffer: foo ----------
@end group
@end example

If @var{limit} is non-@code{nil}, it must be a position in the current
buffer; it specifies the upper bound to the search.  The match found
must not extend after that position.

If @var{noerror} is @code{nil}, then @code{word-search-forward} signals
an error if the search fails.  If @var{noerror} is @code{t}, then it
returns @code{nil} instead of signaling an error.  If @var{noerror} is
neither @code{nil} nor @code{t}, it moves point to @var{limit} (or the
end of the accessible portion of the buffer) and returns @code{nil}.

If @var{repeat} is non-@code{nil}, then the search is repeated that many
times.  Point is positioned at the end of the last match.
161 162

@findex word-search-regexp
Juri Linkov's avatar
Juri Linkov committed
163
Internally, @code{word-search-forward} and related functions use the
164 165
function @code{word-search-regexp} to convert @var{string} to a
regular expression that ignores punctuation.
Glenn Morris's avatar
Glenn Morris committed
166 167
@end deffn

168 169
@deffn Command word-search-forward-lax string &optional limit noerror repeat
This command is identical to @code{word-search-forward}, except that
Juri Linkov's avatar
Juri Linkov committed
170 171 172 173
the beginning or the end of @var{string} need not match a word
boundary, unless @var{string} begins or ends in whitespace.
For instance, searching for @samp{ball boy} matches @samp{ball boyee},
but does not match @samp{balls boy}.
174 175
@end deffn

Glenn Morris's avatar
Glenn Morris committed
176 177 178 179 180 181 182
@deffn Command word-search-backward string &optional limit noerror repeat
This function searches backward from point for a word match to
@var{string}.  This function is just like @code{word-search-forward}
except that it searches backward and normally leaves point at the
beginning of the match.
@end deffn

183 184
@deffn Command word-search-backward-lax string &optional limit noerror repeat
This command is identical to @code{word-search-backward}, except that
Juri Linkov's avatar
Juri Linkov committed
185 186
the beginning or the end of @var{string} need not match a word
boundary, unless @var{string} begins or ends in whitespace.
187 188
@end deffn

Glenn Morris's avatar
Glenn Morris committed
189 190 191 192 193 194 195 196 197 198 199 200 201 202
@node Searching and Case
@section Searching and Case
@cindex searching and case

  By default, searches in Emacs ignore the case of the text they are
searching through; if you specify searching for @samp{FOO}, then
@samp{Foo} or @samp{foo} is also considered a match.  This applies to
regular expressions, too; thus, @samp{[aB]} would match @samp{a} or
@samp{A} or @samp{b} or @samp{B}.

  If you do not want this feature, set the variable
@code{case-fold-search} to @code{nil}.  Then all letters must match
exactly, including case.  This is a buffer-local variable; altering the
variable affects only the current buffer.  (@xref{Intro to
203 204 205
Buffer-Local}.)  Alternatively, you may change the default value.
In Lisp code, you will more typically use @code{let} to bind
@code{case-fold-search} to the desired value.
Glenn Morris's avatar
Glenn Morris committed
206 207

  Note that the user-level incremental search feature handles case
208 209 210 211
distinctions differently.  When the search string contains only lower
case letters, the search ignores case, but when the search string
contains one or more upper case letters, the search becomes
case-sensitive.  But this has nothing to do with the searching
212 213
functions used in Lisp code.  @xref{Incremental Search,,, emacs,
The GNU Emacs Manual}.
Glenn Morris's avatar
Glenn Morris committed
214 215 216 217

@defopt case-fold-search
This buffer-local variable determines whether searches should ignore
case.  If the variable is @code{nil} they do not ignore case; otherwise
218
(and by default) they do ignore case.
Glenn Morris's avatar
Glenn Morris committed
219 220
@end defopt

221
@defopt case-replace
222
This variable determines whether the higher-level replacement
223 224 225 226 227 228 229 230 231
functions should preserve case.  If the variable is @code{nil}, that
means to use the replacement text verbatim.  A non-@code{nil} value
means to convert the case of the replacement text according to the
text being replaced.

This variable is used by passing it as an argument to the function
@code{replace-match}.  @xref{Replacing Match}.
@end defopt

Glenn Morris's avatar
Glenn Morris committed
232 233 234 235 236
@node Regular Expressions
@section Regular Expressions
@cindex regular expression
@cindex regexp

237
  A @dfn{regular expression}, or @dfn{regexp} for short, is a pattern that
Glenn Morris's avatar
Glenn Morris committed
238 239 240 241 242 243
denotes a (possibly infinite) set of strings.  Searching for matches for
a regexp is a very powerful operation.  This section explains how to write
regexps; the following section says how to search for them.

@findex re-builder
@cindex regular expressions, developing
244
  For interactive development of regular expressions, you
Glenn Morris's avatar
Glenn Morris committed
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
can use the @kbd{M-x re-builder} command.  It provides a convenient
interface for creating regular expressions, by giving immediate visual
feedback in a separate buffer.  As you edit the regexp, all its
matches in the target buffer are highlighted.  Each parenthesized
sub-expression of the regexp is shown in a distinct face, which makes
it easier to verify even very complex regexps.

@menu
* Syntax of Regexps::       Rules for writing regular expressions.
* Regexp Example::          Illustrates regular expression syntax.
* Regexp Functions::        Functions for operating on regular expressions.
@end menu

@node Syntax of Regexps
@subsection Syntax of Regular Expressions
260 261
@cindex regexp syntax
@cindex syntax of regular expressions
Glenn Morris's avatar
Glenn Morris committed
262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278

  Regular expressions have a syntax in which a few characters are
special constructs and the rest are @dfn{ordinary}.  An ordinary
character is a simple regular expression that matches that character
and nothing else.  The special characters are @samp{.}, @samp{*},
@samp{+}, @samp{?}, @samp{[}, @samp{^}, @samp{$}, and @samp{\}; no new
special characters will be defined in the future.  The character
@samp{]} is special if it ends a character alternative (see later).
The character @samp{-} is special inside a character alternative.  A
@samp{[:} and balancing @samp{:]} enclose a character class inside a
character alternative.  Any other character appearing in a regular
expression is ordinary, unless a @samp{\} precedes it.

  For example, @samp{f} is not a special character, so it is ordinary, and
therefore @samp{f} is a regular expression that matches the string
@samp{f} and no other string.  (It does @emph{not} match the string
@samp{fg}, but it does match a @emph{part} of that string.)  Likewise,
279
@samp{o} is a regular expression that matches only @samp{o}.
Glenn Morris's avatar
Glenn Morris committed
280 281 282 283

  Any two regular expressions @var{a} and @var{b} can be concatenated.  The
result is a regular expression that matches a string if @var{a} matches
some amount of the beginning of that string and @var{b} matches the rest of
284
the string.
Glenn Morris's avatar
Glenn Morris committed
285 286 287 288 289 290 291 292 293 294 295 296 297 298

  As a simple example, we can concatenate the regular expressions @samp{f}
and @samp{o} to get the regular expression @samp{fo}, which matches only
the string @samp{fo}.  Still trivial.  To do something more powerful, you
need to use one of the special regular expression constructs.

@menu
* Regexp Special::      Special characters in regular expressions.
* Char Classes::        Character classes used in regular expressions.
* Regexp Backslash::    Backslash-sequences in regular expressions.
@end menu

@node Regexp Special
@subsubsection Special Characters in Regular Expressions
299
@cindex regexp, special characters in
Glenn Morris's avatar
Glenn Morris committed
300 301 302 303 304 305 306 307 308 309 310

  Here is a list of the characters that are special in a regular
expression.

@need 800
@table @asis
@item @samp{.}@: @r{(Period)}
@cindex @samp{.} in regexp
is a special character that matches any single character except a newline.
Using concatenation, we can make regular expressions like @samp{a.b}, which
matches any three-character string that begins with @samp{a} and ends with
311
@samp{b}.
Glenn Morris's avatar
Glenn Morris committed
312 313 314 315 316 317 318 319 320 321 322 323

@item @samp{*}
@cindex @samp{*} in regexp
is not a construct by itself; it is a postfix operator that means to
match the preceding regular expression repetitively as many times as
possible.  Thus, @samp{o*} matches any number of @samp{o}s (including no
@samp{o}s).

@samp{*} always applies to the @emph{smallest} possible preceding
expression.  Thus, @samp{fo*} has a repeating @samp{o}, not a repeating
@samp{fo}.  It matches @samp{f}, @samp{fo}, @samp{foo}, and so on.

324
@cindex backtracking and regular expressions
Glenn Morris's avatar
Glenn Morris committed
325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360
The matcher processes a @samp{*} construct by matching, immediately, as
many repetitions as can be found.  Then it continues with the rest of
the pattern.  If that fails, backtracking occurs, discarding some of the
matches of the @samp{*}-modified construct in the hope that that will
make it possible to match the rest of the pattern.  For example, in
matching @samp{ca*ar} against the string @samp{caaar}, the @samp{a*}
first tries to match all three @samp{a}s; but the rest of the pattern is
@samp{ar} and there is only @samp{r} left to match, so this try fails.
The next alternative is for @samp{a*} to match only two @samp{a}s.  With
this choice, the rest of the regexp matches successfully.

@strong{Warning:} Nested repetition operators can run for an
indefinitely long time, if they lead to ambiguous matching.  For
example, trying to match the regular expression @samp{\(x+y*\)*a}
against the string @samp{xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxz} could
take hours before it ultimately fails.  Emacs must try each way of
grouping the @samp{x}s before concluding that none of them can work.
Even worse, @samp{\(x*\)*} can match the null string in infinitely
many ways, so it causes an infinite loop.  To avoid these problems,
check nested repetitions carefully, to make sure that they do not
cause combinatorial explosions in backtracking.

@item @samp{+}
@cindex @samp{+} in regexp
is a postfix operator, similar to @samp{*} except that it must match
the preceding expression at least once.  So, for example, @samp{ca+r}
matches the strings @samp{car} and @samp{caaaar} but not the string
@samp{cr}, whereas @samp{ca*r} matches all three strings.

@item @samp{?}
@cindex @samp{?} in regexp
is a postfix operator, similar to @samp{*} except that it must match the
preceding expression either once or not at all.  For example,
@samp{ca?r} matches @samp{car} or @samp{cr}; nothing else.

@item @samp{*?}, @samp{+?}, @samp{??}
361
@cindex non-greedy repetition characters in regexp
362
These are @dfn{non-greedy} variants of the operators @samp{*}, @samp{+}
Glenn Morris's avatar
Glenn Morris committed
363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383
and @samp{?}.  Where those operators match the largest possible
substring (consistent with matching the entire containing expression),
the non-greedy variants match the smallest possible substring
(consistent with matching the entire containing expression).

For example, the regular expression @samp{c[ad]*a} when applied to the
string @samp{cdaaada} matches the whole string; but the regular
expression @samp{c[ad]*?a}, applied to that same string, matches just
@samp{cda}.  (The smallest possible match here for @samp{[ad]*?} that
permits the whole expression to match is @samp{d}.)

@item @samp{[ @dots{} ]}
@cindex character alternative (in regexp)
@cindex @samp{[} in regexp
@cindex @samp{]} in regexp
is a @dfn{character alternative}, which begins with @samp{[} and is
terminated by @samp{]}.  In the simplest case, the characters between
the two brackets are what this character alternative can match.

Thus, @samp{[ad]} matches either one @samp{a} or one @samp{d}, and
@samp{[ad]*} matches any string composed of just @samp{a}s and @samp{d}s
Chong Yidong's avatar
Chong Yidong committed
384
(including the empty string).  It follows that @samp{c[ad]*r}
Glenn Morris's avatar
Glenn Morris committed
385 386 387 388 389 390 391 392 393
matches @samp{cr}, @samp{car}, @samp{cdr}, @samp{caddaar}, etc.

You can also include character ranges in a character alternative, by
writing the starting and ending characters with a @samp{-} between them.
Thus, @samp{[a-z]} matches any lower-case @acronym{ASCII} letter.
Ranges may be intermixed freely with individual characters, as in
@samp{[a-z$%.]}, which matches any lower case @acronym{ASCII} letter
or @samp{$}, @samp{%} or period.

394 395 396 397
If @code{case-fold-search} is non-@code{nil}, @samp{[a-z]} also
matches upper-case letters.  Note that a range like @samp{[a-z]} is
not affected by the locale's collation sequence, it always represents
a sequence in @acronym{ASCII} order.
398
@c This wasn't obvious to me, since, e.g., the grep manual "Character
Glenn Morris's avatar
Comment  
Glenn Morris committed
399 400 401
@c Classes and Bracket Expressions" specifically notes the opposite
@c behavior.  But by experiment Emacs seems unaffected by LC_COLLATE
@c in this regard.
402 403

Note also that the usual regexp special characters are not special inside a
Glenn Morris's avatar
Glenn Morris committed
404 405 406 407 408 409 410
character alternative.  A completely different set of characters is
special inside character alternatives: @samp{]}, @samp{-} and @samp{^}.

To include a @samp{]} in a character alternative, you must make it the
first character.  For example, @samp{[]a]} matches @samp{]} or @samp{a}.
To include a @samp{-}, write @samp{-} as the first or last character of
the character alternative, or put it after a range.  Thus, @samp{[]-]}
411 412 413
matches both @samp{]} and @samp{-}.  (As explained below, you cannot
use @samp{\]} to include a @samp{]} inside a character alternative,
since @samp{\} is not special there.)
Glenn Morris's avatar
Glenn Morris committed
414 415 416 417

To include @samp{^} in a character alternative, put it anywhere but at
the beginning.

418 419
@c What if it starts with a multibyte and ends with a unibyte?
@c That doesn't seem to match anything...?
Glenn Morris's avatar
Glenn Morris committed
420 421
If a range starts with a unibyte character @var{c} and ends with a
multibyte character @var{c2}, the range is divided into two parts: one
422 423 424
spans the unibyte characters @samp{@var{c}..?\377}, the other the
multibyte characters @samp{@var{c1}..@var{c2}}, where @var{c1} is the
first character of the charset to which @var{c2} belongs.
Glenn Morris's avatar
Glenn Morris committed
425

Chong Yidong's avatar
Chong Yidong committed
426
A character alternative can also specify named character classes
427 428 429 430 431
(@pxref{Char Classes}).  This is a POSIX feature.  For example,
@samp{[[:ascii:]]} matches any @acronym{ASCII} character.
Using a character class is equivalent to mentioning each of the
characters in that class; but the latter is not feasible in practice,
since some classes include thousands of different characters.
Glenn Morris's avatar
Glenn Morris committed
432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447

@item @samp{[^ @dots{} ]}
@cindex @samp{^} in regexp
@samp{[^} begins a @dfn{complemented character alternative}.  This
matches any character except the ones specified.  Thus,
@samp{[^a-z0-9A-Z]} matches all characters @emph{except} letters and
digits.

@samp{^} is not special in a character alternative unless it is the first
character.  The character following the @samp{^} is treated as if it
were first (in other words, @samp{-} and @samp{]} are not special there).

A complemented character alternative can match a newline, unless newline is
mentioned as one of the characters not to match.  This is in contrast to
the handling of regexps in programs such as @code{grep}.

Chong Yidong's avatar
Chong Yidong committed
448 449 450 451
You can specify named character classes, just like in character
alternatives.  For instance, @samp{[^[:ascii:]]} matches any
non-@acronym{ASCII} character.  @xref{Char Classes}.

Glenn Morris's avatar
Glenn Morris committed
452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494
@item @samp{^}
@cindex beginning of line in regexp
When matching a buffer, @samp{^} matches the empty string, but only at the
beginning of a line in the text being matched (or the beginning of the
accessible portion of the buffer).  Otherwise it fails to match
anything.  Thus, @samp{^foo} matches a @samp{foo} that occurs at the
beginning of a line.

When matching a string instead of a buffer, @samp{^} matches at the
beginning of the string or after a newline character.

For historical compatibility reasons, @samp{^} can be used only at the
beginning of the regular expression, or after @samp{\(}, @samp{\(?:}
or @samp{\|}.

@item @samp{$}
@cindex @samp{$} in regexp
@cindex end of line in regexp
is similar to @samp{^} but matches only at the end of a line (or the
end of the accessible portion of the buffer).  Thus, @samp{x+$}
matches a string of one @samp{x} or more at the end of a line.

When matching a string instead of a buffer, @samp{$} matches at the end
of the string or before a newline character.

For historical compatibility reasons, @samp{$} can be used only at the
end of the regular expression, or before @samp{\)} or @samp{\|}.

@item @samp{\}
@cindex @samp{\} in regexp
has two functions: it quotes the special characters (including
@samp{\}), and it introduces additional special constructs.

Because @samp{\} quotes special characters, @samp{\$} is a regular
expression that matches only @samp{$}, and @samp{\[} is a regular
expression that matches only @samp{[}, and so on.

Note that @samp{\} also has special meaning in the read syntax of Lisp
strings (@pxref{String Type}), and must be quoted with @samp{\}.  For
example, the regular expression that matches the @samp{\} character is
@samp{\\}.  To write a Lisp string that contains the characters
@samp{\\}, Lisp syntax requires you to quote each @samp{\} with another
@samp{\}.  Therefore, the read syntax for a regular expression matching
495
@samp{\} is @code{"\\\\"}.
Glenn Morris's avatar
Glenn Morris committed
496 497 498 499 500 501 502
@end table

@strong{Please note:} For historical compatibility, special characters
are treated as ordinary ones if they are in contexts where their special
meanings make no sense.  For example, @samp{*foo} treats @samp{*} as
ordinary since there is no preceding expression on which the @samp{*}
can act.  It is poor practice to depend on this behavior; quote the
503
special character anyway, regardless of where it appears.
Glenn Morris's avatar
Glenn Morris committed
504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543

As a @samp{\} is not special inside a character alternative, it can
never remove the special meaning of @samp{-} or @samp{]}.  So you
should not quote these characters when they have no special meaning
either.  This would not clarify anything, since backslashes can
legitimately precede these characters where they @emph{have} special
meaning, as in @samp{[^\]} (@code{"[^\\]"} for Lisp string syntax),
which matches any single character except a backslash.

In practice, most @samp{]} that occur in regular expressions close a
character alternative and hence are special.  However, occasionally a
regular expression may try to match a complex pattern of literal
@samp{[} and @samp{]}.  In such situations, it sometimes may be
necessary to carefully parse the regexp from the start to determine
which square brackets enclose a character alternative.  For example,
@samp{[^][]]} consists of the complemented character alternative
@samp{[^][]} (which matches any single character that is not a square
bracket), followed by a literal @samp{]}.

The exact rules are that at the beginning of a regexp, @samp{[} is
special and @samp{]} not.  This lasts until the first unquoted
@samp{[}, after which we are in a character alternative; @samp{[} is
no longer special (except when it starts a character class) but @samp{]}
is special, unless it immediately follows the special @samp{[} or that
@samp{[} followed by a @samp{^}.  This lasts until the next special
@samp{]} that does not end a character class.  This ends the character
alternative and restores the ordinary syntax of regular expressions;
an unquoted @samp{[} is special again and a @samp{]} not.

@node Char Classes
@subsubsection Character Classes
@cindex character classes in regexp

  Here is a table of the classes you can use in a character alternative,
and what they mean:

@table @samp
@item [:ascii:]
This matches any @acronym{ASCII} character (codes 0--127).
@item [:alnum:]
544 545 546 547
This matches any letter or digit.  For multibyte characters, it
matches characters whose Unicode @samp{general-category} property
(@pxref{Character Properties}) indicates they are alphabetic or
decimal number characters.
Glenn Morris's avatar
Glenn Morris committed
548
@item [:alpha:]
549 550 551 552
This matches any letter.  For multibyte characters, it matches
characters whose Unicode @samp{general-category} property
(@pxref{Character Properties}) indicates they are alphabetic
characters.
Glenn Morris's avatar
Glenn Morris committed
553 554 555 556 557 558 559 560
@item [:blank:]
This matches space and tab only.
@item [:cntrl:]
This matches any @acronym{ASCII} control character.
@item [:digit:]
This matches @samp{0} through @samp{9}.  Thus, @samp{[-+[:digit:]]}
matches any digit, as well as @samp{+} and @samp{-}.
@item [:graph:]
561
This matches graphic characters---everything except whitespace,
562 563 564 565
@acronym{ASCII} and non-@acronym{ASCII} control characters,
surrogates, and codepoints unassigned by Unicode, as indicated by the
Unicode @samp{general-category} property (@pxref{Character
Properties}).
Glenn Morris's avatar
Glenn Morris committed
566
@item [:lower:]
567 568 569
This matches any lower-case letter, as determined by the current case
table (@pxref{Case Tables}).  If @code{case-fold-search} is
non-@code{nil}, this also matches any upper-case letter.
Glenn Morris's avatar
Glenn Morris committed
570 571 572 573 574
@item [:multibyte:]
This matches any multibyte character (@pxref{Text Representations}).
@item [:nonascii:]
This matches any non-@acronym{ASCII} character.
@item [:print:]
575
This matches any printing character---either whitespace, or a graphic
576
character matched by @samp{[:graph:]}.
Glenn Morris's avatar
Glenn Morris committed
577 578 579 580 581 582 583 584 585
@item [:punct:]
This matches any punctuation character.  (At present, for multibyte
characters, it matches anything that has non-word syntax.)
@item [:space:]
This matches any character that has whitespace syntax
(@pxref{Syntax Class Table}).
@item [:unibyte:]
This matches any unibyte character (@pxref{Text Representations}).
@item [:upper:]
586 587 588
This matches any upper-case letter, as determined by the current case
table (@pxref{Case Tables}).  If @code{case-fold-search} is
non-@code{nil}, this also matches any lower-case letter.
Glenn Morris's avatar
Glenn Morris committed
589 590 591 592 593 594 595 596 597 598
@item [:word:]
This matches any character that has word syntax (@pxref{Syntax Class
Table}).
@item [:xdigit:]
This matches the hexadecimal digits: @samp{0} through @samp{9}, @samp{a}
through @samp{f} and @samp{A} through @samp{F}.
@end table

@node Regexp Backslash
@subsubsection Backslash Constructs in Regular Expressions
599
@cindex backslash in regular expressions
Glenn Morris's avatar
Glenn Morris committed
600 601 602

  For the most part, @samp{\} followed by any character matches only
that character.  However, there are several exceptions: certain
603 604
sequences starting with @samp{\} that have special meanings.  Here is
a table of the special @samp{\} constructs.
Glenn Morris's avatar
Glenn Morris committed
605 606 607 608 609 610 611 612

@table @samp
@item \|
@cindex @samp{|} in regexp
@cindex regexp alternative
specifies an alternative.
Two regular expressions @var{a} and @var{b} with @samp{\|} in
between form an expression that matches anything that either @var{a} or
613
@var{b} matches.
Glenn Morris's avatar
Glenn Morris committed
614 615

Thus, @samp{foo\|bar} matches either @samp{foo} or @samp{bar}
616
but no other string.
Glenn Morris's avatar
Glenn Morris committed
617 618 619

@samp{\|} applies to the largest possible surrounding expressions.  Only a
surrounding @samp{\( @dots{} \)} grouping can limit the grouping power of
620
@samp{\|}.
Glenn Morris's avatar
Glenn Morris committed
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640

If you need full backtracking capability to handle multiple uses of
@samp{\|}, use the POSIX regular expression functions (@pxref{POSIX
Regexps}).

@item \@{@var{m}\@}
is a postfix operator that repeats the previous pattern exactly @var{m}
times.  Thus, @samp{x\@{5\@}} matches the string @samp{xxxxx}
and nothing else.  @samp{c[ad]\@{3\@}r} matches string such as
@samp{caaar}, @samp{cdddr}, @samp{cadar}, and so on.

@item \@{@var{m},@var{n}\@}
is a more general postfix operator that specifies repetition with a
minimum of @var{m} repeats and a maximum of @var{n} repeats.  If @var{m}
is omitted, the minimum is 0; if @var{n} is omitted, there is no
maximum.

For example, @samp{c[ad]\@{1,2\@}r} matches the strings @samp{car},
@samp{cdr}, @samp{caar}, @samp{cadr}, @samp{cdar}, and @samp{cddr}, and
nothing else.@*
Glenn Morris's avatar
Glenn Morris committed
641 642
@samp{\@{0,1\@}} or @samp{\@{,1\@}} is equivalent to @samp{?}.@*
@samp{\@{0,\@}} or @samp{\@{,\@}} is equivalent to @samp{*}.@*
Glenn Morris's avatar
Glenn Morris committed
643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675
@samp{\@{1,\@}} is equivalent to @samp{+}.

@item \( @dots{} \)
@cindex @samp{(} in regexp
@cindex @samp{)} in regexp
@cindex regexp grouping
is a grouping construct that serves three purposes:

@enumerate
@item
To enclose a set of @samp{\|} alternatives for other operations.  Thus,
the regular expression @samp{\(foo\|bar\)x} matches either @samp{foox}
or @samp{barx}.

@item
To enclose a complicated expression for the postfix operators @samp{*},
@samp{+} and @samp{?} to operate on.  Thus, @samp{ba\(na\)*} matches
@samp{ba}, @samp{bana}, @samp{banana}, @samp{bananana}, etc., with any
number (zero or more) of @samp{na} strings.

@item
To record a matched substring for future reference with
@samp{\@var{digit}} (see below).
@end enumerate

This last application is not a consequence of the idea of a
parenthetical grouping; it is a separate feature that was assigned as a
second meaning to the same @samp{\( @dots{} \)} construct because, in
practice, there was usually no conflict between the two meanings.  But
occasionally there is a conflict, and that led to the introduction of
shy groups.

@item \(?: @dots{} \)
676 677 678
@cindex shy groups
@cindex non-capturing group
@cindex unnumbered group
Stefan Monnier's avatar
Stefan Monnier committed
679
@cindex @samp{(?:} in regexp
Glenn Morris's avatar
Glenn Morris committed
680 681 682
is the @dfn{shy group} construct.  A shy group serves the first two
purposes of an ordinary group (controlling the nesting of other
operators), but it does not get a number, so you cannot refer back to
683 684 685 686
its value with @samp{\@var{digit}}.  Shy groups are particularly
useful for mechanically-constructed regular expressions, because they
can be added automatically without altering the numbering of ordinary,
non-shy groups.
Glenn Morris's avatar
Glenn Morris committed
687

688 689
Shy groups are also called @dfn{non-capturing} or @dfn{unnumbered
groups}.
Glenn Morris's avatar
Glenn Morris committed
690 691 692 693 694 695

@item \(?@var{num}: @dots{} \)
is the @dfn{explicitly numbered group} construct.  Normal groups get
their number implicitly, based on their position, which can be
inconvenient.  This construct allows you to force a particular group
number.  There is no particular restriction on the numbering,
696 697
e.g., you can have several groups with the same number in which case
the last one to match (i.e., the rightmost match) will win.
Glenn Morris's avatar
Glenn Morris committed
698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729
Implicitly numbered groups always get the smallest integer larger than
the one of any previous group.

@item \@var{digit}
matches the same text that matched the @var{digit}th occurrence of a
grouping (@samp{\( @dots{} \)}) construct.

In other words, after the end of a group, the matcher remembers the
beginning and end of the text matched by that group.  Later on in the
regular expression you can use @samp{\} followed by @var{digit} to
match that same text, whatever it may have been.

The strings matching the first nine grouping constructs appearing in
the entire regular expression passed to a search or matching function
are assigned numbers 1 through 9 in the order that the open
parentheses appear in the regular expression.  So you can use
@samp{\1} through @samp{\9} to refer to the text matched by the
corresponding grouping constructs.

For example, @samp{\(.*\)\1} matches any newline-free string that is
composed of two identical halves.  The @samp{\(.*\)} matches the first
half, which may be anything, but the @samp{\1} that follows must match
the same exact text.

If a @samp{\( @dots{} \)} construct matches more than once (which can
happen, for instance, if it is followed by @samp{*}), only the last
match is recorded.

If a particular grouping construct in the regular expression was never
matched---for instance, if it appears inside of an alternative that
wasn't used, or inside of a repetition that repeated zero times---then
the corresponding @samp{\@var{digit}} construct never matches
Glenn Morris's avatar
Glenn Morris committed
730
anything.  To use an artificial example, @samp{\(foo\(b*\)\|lose\)\2}
Glenn Morris's avatar
Glenn Morris committed
731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
cannot match @samp{lose}: the second alternative inside the larger
group matches it, but then @samp{\2} is undefined and can't match
anything.  But it can match @samp{foobb}, because the first
alternative matches @samp{foob} and @samp{\2} matches @samp{b}.

@item \w
@cindex @samp{\w} in regexp
matches any word-constituent character.  The editor syntax table
determines which characters these are.  @xref{Syntax Tables}.

@item \W
@cindex @samp{\W} in regexp
matches any character that is not a word constituent.

@item \s@var{code}
@cindex @samp{\s} in regexp
matches any character whose syntax is @var{code}.  Here @var{code} is a
character that represents a syntax code: thus, @samp{w} for word
constituent, @samp{-} for whitespace, @samp{(} for open parenthesis,
etc.  To represent whitespace syntax, use either @samp{-} or a space
character.  @xref{Syntax Class Table}, for a list of syntax codes and
the characters that stand for them.

@item \S@var{code}
@cindex @samp{\S} in regexp
matches any character whose syntax is not @var{code}.

758
@cindex category, regexp search for
Glenn Morris's avatar
Glenn Morris committed
759 760 761 762
@item \c@var{c}
matches any character whose category is @var{c}.  Here @var{c} is a
character that represents a category: thus, @samp{c} for Chinese
characters or @samp{g} for Greek characters in the standard category
763 764 765 766
table.  You can see the list of all the currently defined categories
with @kbd{M-x describe-categories @key{RET}}.  You can also define
your own categories in addition to the standard ones using the
@code{define-category} function (@pxref{Categories}).
Glenn Morris's avatar
Glenn Morris committed
767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798

@item \C@var{c}
matches any character whose category is not @var{c}.
@end table

  The following regular expression constructs match the empty string---that is,
they don't use up any characters---but whether they match depends on the
context.  For all, the beginning and end of the accessible portion of
the buffer are treated as if they were the actual beginning and end of
the buffer.

@table @samp
@item \`
@cindex @samp{\`} in regexp
matches the empty string, but only at the beginning
of the buffer or string being matched against.

@item \'
@cindex @samp{\'} in regexp
matches the empty string, but only at the end of
the buffer or string being matched against.

@item \=
@cindex @samp{\=} in regexp
matches the empty string, but only at point.
(This construct is not defined when matching against a string.)

@item \b
@cindex @samp{\b} in regexp
matches the empty string, but only at the beginning or
end of a word.  Thus, @samp{\bfoo\b} matches any occurrence of
@samp{foo} as a separate word.  @samp{\bballs?\b} matches
799
@samp{ball} or @samp{balls} as a separate word.
Glenn Morris's avatar
Glenn Morris committed
800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836

@samp{\b} matches at the beginning or end of the buffer (or string)
regardless of what text appears next to it.

@item \B
@cindex @samp{\B} in regexp
matches the empty string, but @emph{not} at the beginning or
end of a word, nor at the beginning or end of the buffer (or string).

@item \<
@cindex @samp{\<} in regexp
matches the empty string, but only at the beginning of a word.
@samp{\<} matches at the beginning of the buffer (or string) only if a
word-constituent character follows.

@item \>
@cindex @samp{\>} in regexp
matches the empty string, but only at the end of a word.  @samp{\>}
matches at the end of the buffer (or string) only if the contents end
with a word-constituent character.

@item \_<
@cindex @samp{\_<} in regexp
matches the empty string, but only at the beginning of a symbol.  A
symbol is a sequence of one or more word or symbol constituent
characters.  @samp{\_<} matches at the beginning of the buffer (or
string) only if a symbol-constituent character follows.

@item \_>
@cindex @samp{\_>} in regexp
matches the empty string, but only at the end of a symbol.  @samp{\_>}
matches at the end of the buffer (or string) only if the contents end
with a symbol-constituent character.
@end table

@kindex invalid-regexp
  Not every string is a valid regular expression.  For example, a string
837
that ends inside a character alternative without a terminating @samp{]}
Glenn Morris's avatar
Glenn Morris committed
838 839 840 841 842 843 844 845 846 847 848 849 850
is invalid, and so is a string that ends with a single @samp{\}.  If
an invalid regular expression is passed to any of the search functions,
an @code{invalid-regexp} error is signaled.

@node Regexp Example
@subsection Complex Regexp Example

  Here is a complicated regexp which was formerly used by Emacs to
recognize the end of a sentence together with any whitespace that
follows.  (Nowadays Emacs uses a similar but more complex default
regexp constructed by the function @code{sentence-end}.
@xref{Standard Regexps}.)

851 852 853
  Below, we show first the regexp as a string in Lisp syntax (to
distinguish spaces from tab characters), and then the result of
evaluating it.  The string constant begins and ends with a
Glenn Morris's avatar
Glenn Morris committed
854 855 856 857 858 859 860 861 862 863 864 865 866
double-quote.  @samp{\"} stands for a double-quote as part of the
string, @samp{\\} for a backslash as part of the string, @samp{\t} for a
tab and @samp{\n} for a newline.

@example
@group
"[.?!][]\"')@}]*\\($\\| $\\|\t\\|@ @ \\)[ \t\n]*"
     @result{} "[.?!][]\"')@}]*\\($\\| $\\|  \\|@ @ \\)[
]*"
@end group
@end example

@noindent
867
In the output, tab and newline appear as themselves.
Glenn Morris's avatar
Glenn Morris committed
868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906

  This regular expression contains four parts in succession and can be
deciphered as follows:

@table @code
@item [.?!]
The first part of the pattern is a character alternative that matches
any one of three characters: period, question mark, and exclamation
mark.  The match must begin with one of these three characters.  (This
is one point where the new default regexp used by Emacs differs from
the old.  The new value also allows some non-@acronym{ASCII}
characters that end a sentence without any following whitespace.)

@item []\"')@}]*
The second part of the pattern matches any closing braces and quotation
marks, zero or more of them, that may follow the period, question mark
or exclamation mark.  The @code{\"} is Lisp syntax for a double-quote in
a string.  The @samp{*} at the end indicates that the immediately
preceding regular expression (a character alternative, in this case) may be
repeated zero or more times.

@item \\($\\|@ $\\|\t\\|@ @ \\)
The third part of the pattern matches the whitespace that follows the
end of a sentence: the end of a line (optionally with a space), or a
tab, or two spaces.  The double backslashes mark the parentheses and
vertical bars as regular expression syntax; the parentheses delimit a
group and the vertical bars separate alternatives.  The dollar sign is
used to match the end of a line.

@item [ \t\n]*
Finally, the last part of the pattern matches any additional whitespace
beyond the minimum needed to end a sentence.
@end table

@node Regexp Functions
@subsection Regular Expression Functions

  These functions operate on regular expressions.

907
@cindex quote special characters in regexp
Glenn Morris's avatar
Glenn Morris committed
908 909 910 911 912
@defun regexp-quote string
This function returns a regular expression whose only exact match is
@var{string}.  Using this regular expression in @code{looking-at} will
succeed only if the next characters in the buffer are @var{string};
using it in a search function will succeed if the text being searched
913
contains @var{string}.  @xref{Regexp Search}.
Glenn Morris's avatar
Glenn Morris committed
914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937

This allows you to request an exact string match or search when calling
a function that wants a regular expression.

@example
@group
(regexp-quote "^The cat$")
     @result{} "\\^The cat\\$"
@end group
@end example

One use of @code{regexp-quote} is to combine an exact string match with
context described as a regular expression.  For example, this searches
for the string that is the value of @var{string}, surrounded by
whitespace:

@example
@group
(re-search-forward
 (concat "\\s-" (regexp-quote string) "\\s-"))
@end group
@end example
@end defun

938
@cindex optimize regexp
Glenn Morris's avatar
Glenn Morris committed
939 940 941 942
@defun regexp-opt strings &optional paren
This function returns an efficient regular expression that will match
any of the strings in the list @var{strings}.  This is useful when you
need to make matching or searching as fast as possible---for example,
943 944 945 946
for Font Lock mode@footnote{Note that @code{regexp-opt} does not
guarantee that its result is absolutely the most efficient form
possible.  A hand-tuned regular expression can sometimes be slightly
more efficient, but is almost never worth the effort.}.
947
@c E.g., see http://debbugs.gnu.org/2816
Glenn Morris's avatar
Glenn Morris committed
948 949 950 951

If the optional argument @var{paren} is non-@code{nil}, then the
returned regular expression is always enclosed by at least one
parentheses-grouping construct.  If @var{paren} is @code{words}, then
952 953 954 955 956
that construct is additionally surrounded by @samp{\<} and @samp{\>};
alternatively, if @var{paren} is @code{symbols}, then that construct
is additionally surrounded by @samp{\_<} and @samp{\_>}
(@code{symbols} is often appropriate when matching
programming-language keywords and the like).
Glenn Morris's avatar
Glenn Morris committed
957 958 959 960 961 962

This simplified definition of @code{regexp-opt} produces a
regular expression which is equivalent to the actual value
(but not as efficient):

@example
963
(defun regexp-opt (strings &optional paren)
Glenn Morris's avatar
Glenn Morris committed
964 965 966 967 968 969 970 971 972 973
  (let ((open-paren (if paren "\\(" ""))
        (close-paren (if paren "\\)" "")))
    (concat open-paren
            (mapconcat 'regexp-quote strings "\\|")
            close-paren)))
@end example
@end defun

@defun regexp-opt-depth regexp
This function returns the total number of grouping constructs
974 975
(parenthesized expressions) in @var{regexp}.  This does not include
shy groups (@pxref{Regexp Backslash}).
Glenn Morris's avatar
Glenn Morris committed
976 977
@end defun

978 979 980 981 982 983 984 985 986 987 988 989 990
@c Supposedly an internal regexp-opt function, but table.el uses it at least.
@defun regexp-opt-charset chars
This function returns a regular expression matching a character in the
list of characters @var{chars}.

@example
(regexp-opt-charset '(?a ?b ?c ?d ?e))
     @result{} "[a-e]"
@end example
@end defun

@c Internal functions: regexp-opt-group

Glenn Morris's avatar
Glenn Morris committed
991 992 993 994 995 996 997
@node Regexp Search
@section Regular Expression Searching
@cindex regular expression searching
@cindex regexp searching
@cindex searching for regexp

  In GNU Emacs, you can search for the next match for a regular
998 999 1000 1001 1002
expression (@pxref{Syntax of Regexps}) either incrementally or not.
For incremental search commands, see @ref{Regexp Search, , Regular
Expression Search, emacs, The GNU Emacs Manual}.  Here we describe
only the search functions useful in programs.  The principal one is
@code{re-search-forward}.
Glenn Morris's avatar
Glenn Morris committed
1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104

  These search functions convert the regular expression to multibyte if
the buffer is multibyte; they convert the regular expression to unibyte
if the buffer is unibyte.  @xref{Text Representations}.

@deffn Command re-search-forward regexp &optional limit noerror repeat
This function searches forward in the current buffer for a string of
text that is matched by the regular expression @var{regexp}.  The
function skips over any amount of text that is not matched by
@var{regexp}, and leaves point at the end of the first match found.
It returns the new value of point.

If @var{limit} is non-@code{nil}, it must be a position in the current
buffer.  It specifies the upper bound to the search.  No match
extending after that position is accepted.

If @var{repeat} is supplied, it must be a positive number; the search
is repeated that many times; each repetition starts at the end of the
previous match.  If all these successive searches succeed, the search
succeeds, moving point and returning its new value.  Otherwise the
search fails.  What @code{re-search-forward} does when the search
fails depends on the value of @var{noerror}:

@table @asis
@item @code{nil}
Signal a @code{search-failed} error.
@item @code{t}
Do nothing and return @code{nil}.
@item anything else
Move point to @var{limit} (or the end of the accessible portion of the
buffer) and return @code{nil}.
@end table

In the following example, point is initially before the @samp{T}.
Evaluating the search call moves point to the end of that line (between
the @samp{t} of @samp{hat} and the newline).

@example
@group
---------- Buffer: foo ----------
I read "@point{}The cat in the hat
comes back" twice.
---------- Buffer: foo ----------
@end group

@group
(re-search-forward "[a-z]+" nil t 5)
     @result{} 27

---------- Buffer: foo ----------
I read "The cat in the hat@point{}
comes back" twice.
---------- Buffer: foo ----------
@end group
@end example
@end deffn

@deffn Command re-search-backward regexp &optional limit noerror repeat
This function searches backward in the current buffer for a string of
text that is matched by the regular expression @var{regexp}, leaving
point at the beginning of the first text found.

This function is analogous to @code{re-search-forward}, but they are not
simple mirror images.  @code{re-search-forward} finds the match whose
beginning is as close as possible to the starting point.  If
@code{re-search-backward} were a perfect mirror image, it would find the
match whose end is as close as possible.  However, in fact it finds the
match whose beginning is as close as possible (and yet ends before the
starting point).  The reason for this is that matching a regular
expression at a given spot always works from beginning to end, and
starts at a specified beginning position.

A true mirror-image of @code{re-search-forward} would require a special
feature for matching regular expressions from end to beginning.  It's
not worth the trouble of implementing that.
@end deffn

@defun string-match regexp string &optional start
This function returns the index of the start of the first match for
the regular expression @var{regexp} in @var{string}, or @code{nil} if
there is no match.  If @var{start} is non-@code{nil}, the search starts
at that index in @var{string}.

For example,

@example
@group
(string-match
 "quick" "The quick brown fox jumped quickly.")
     @result{} 4
@end group
@group
(string-match
 "quick" "The quick brown fox jumped quickly." 8)
     @result{} 27
@end group
@end example

@noindent
The index of the first character of the
string is 0, the index of the second character is 1, and so on.

1105
If this function finds a match, the index of the first character beyond
Glenn Morris's avatar
Glenn Morris committed
1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121
the match is available as @code{(match-end 0)}.  @xref{Match Data}.

@example
@group
(string-match
 "quick" "The quick brown fox jumped quickly." 8)
     @result{} 27
@end group

@group
(match-end 0)
     @result{} 32
@end group
@end example
@end defun

1122
@defun string-match-p regexp string &optional start
1123 1124
This predicate function does what @code{string-match} does, but it
avoids modifying the match data.
1125 1126
@end defun

Glenn Morris's avatar
Glenn Morris committed
1127 1128 1129
@defun looking-at regexp
This function determines whether the text in the current buffer directly
following point matches the regular expression @var{regexp}.  ``Directly
1130
following'' means precisely that: the search is ``anchored'' and it can
Glenn Morris's avatar
Glenn Morris committed
1131 1132 1133
succeed only starting with the first character following point.  The
result is @code{t} if so, @code{nil} otherwise.

1134
This function does not move point, but it does update the match data.
1135 1136
@xref{Match Data}.  If you need to test for a match without modifying
the match data, use @code{looking-at-p}, described below.
Glenn Morris's avatar
Glenn Morris committed
1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153

In this example, point is located directly before the @samp{T}.  If it
were anywhere else, the result would be @code{nil}.

@example
@group
---------- Buffer: foo ----------
I read "@point{}The cat in the hat
comes back" twice.
---------- Buffer: foo ----------

(looking-at "The cat in the hat$")
     @result{} t
@end group
@end example
@end defun

1154
@defun looking-back regexp &optional limit greedy
1155 1156
This function returns @code{t} if @var{regexp} matches the text
immediately before point (i.e., ending at point), and @code{nil} otherwise.
Glenn Morris's avatar
Glenn Morris committed
1157 1158 1159 1160 1161 1162

Because regular expression matching works only going forward, this is
implemented by searching backwards from point for a match that ends at
point.  That can be quite slow if it has to search a long distance.
You can bound the time required by specifying @var{limit}, which says
not to search before @var{limit}.  In this case, the match that is
1163
found must begin at or after @var{limit}.  Here's an example:
1164

Glenn Morris's avatar
Glenn Morris committed
1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177
@example
@group
---------- Buffer: foo ----------
I read "@point{}The cat in the hat
comes back" twice.
---------- Buffer: foo ----------

(looking-back "read \"" 3)
     @result{} t
(looking-back "read \"" 4)
     @result{} nil
@end group
@end example
1178

1179 1180 1181 1182 1183 1184
If @var{greedy} is non-@code{nil}, this function extends the match
backwards as far as possible, stopping when a single additional
previous character cannot be part of a match for regexp.  When the
match is extended, its starting position is allowed to occur before
@var{limit}.

1185 1186 1187 1188
@c http://debbugs.gnu.org/5689
As a general recommendation, try to avoid using @code{looking-back}
wherever possible, since it is slow.  For this reason, there are no
plans to add a @code{looking-back-p} function.
Glenn Morris's avatar
Glenn Morris committed
1189 1190
@end defun

1191 1192 1193 1194 1195
@defun looking-at-p regexp
This predicate function works like @code{looking-at}, but without
updating the match data.
@end defun

Glenn Morris's avatar
Glenn Morris committed
1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211
@defvar search-spaces-regexp
If this variable is non-@code{nil}, it should be a regular expression
that says how to search for whitespace.  In that case, any group of
spaces in a regular expression being searched for stands for use of
this regular expression.  However, spaces inside of constructs such as
@samp{[@dots{}]} and @samp{*}, @samp{+}, @samp{?} are not affected by
@code{search-spaces-regexp}.

Since this variable affects all regular expression search and match
constructs, you should bind it temporarily for as small as possible
a part of the code.
@end defvar

@node POSIX Regexps
@section POSIX Regular Expression Searching

1212
@cindex backtracking and POSIX regular expressions
Glenn Morris's avatar
Glenn Morris committed
1213 1214 1215 1216 1217 1218 1219 1220 1221
  The usual regular expression functions do backtracking when necessary
to handle the @samp{\|} and repetition constructs, but they continue
this only until they find @emph{some} match.  Then they succeed and
report the first match found.

  This section describes alternative search functions which perform the
full backtracking specified by the POSIX standard for regular expression
matching.  They continue backtracking until they have tried all
possibilities and found all matches, so they can report the longest
1222
match, as required by POSIX@.  This is much slower, so use these
Glenn Morris's avatar
Glenn Morris committed
1223 1224 1225
functions only when you really need the longest match.

  The POSIX search and match functions do not properly support the
1226 1227 1228
non-greedy repetition operators (@pxref{Regexp Special, non-greedy}).
This is because POSIX backtracking conflicts with the semantics of
non-greedy repetition.
Glenn Morris's avatar
Glenn Morris committed
1229

1230
@deffn Command posix-search-forward regexp &optional limit noerror repeat
Glenn Morris's avatar
Glenn Morris committed
1231 1232 1233
This is like @code{re-search-forward} except that it performs the full
backtracking specified by the POSIX standard for regular expression
matching.
1234
@end deffn
Glenn Morris's avatar
Glenn Morris committed
1235

1236
@deffn Command posix-search-backward regexp &optional limit noerror repeat
Glenn Morris's avatar
Glenn Morris committed
1237 1238 1239
This is like @code{re-search-backward} except that it performs the full
backtracking specified by the POSIX standard for regular expression
matching.
1240
@end deffn
Glenn Morris's avatar
Glenn Morris committed
1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269

@defun posix-looking-at regexp
This is like @code{looking-at} except that it performs the full
backtracking specified by the POSIX standard for regular expression
matching.
@end defun

@defun posix-string-match regexp string &optional start
This is like @code{string-match} except that it performs the full
backtracking specified by the POSIX standard for regular expression
matching.
@end defun

@node Match Data
@section The Match Data
@cindex match data

  Emacs keeps track of the start and end positions of the segments of
text found during a search; this is called the @dfn{match data}.
Thanks to the match data, you can search for a complex pattern, such
as a date in a mail message, and then extract parts of the match under
control of the pattern.

  Because the match data normally describe the most recent search only,
you must be careful not to do another search inadvertently between the
search you wish to refer back to and the use of the match data.  If you
can't avoid another intervening search, you must save and restore the
match data around it, to prevent it from being overwritten.

1270 1271
  Notice that all functions are allowed to overwrite the match data
unless they're explicitly documented not to do so.  A consequence is
Paul Eggert's avatar
Paul Eggert committed
1272
that functions that are run implicitly in the background
1273 1274 1275
(@pxref{Timers}, and @ref{Idle Timers}) should likely save and restore
the match data explicitly.

Glenn Morris's avatar
Glenn Morris committed
1276
@menu
Glenn Morris's avatar
Glenn Morris committed
1277
* Replacing Match::       Replacing a substring that was matched.
Glenn Morris's avatar
Glenn Morris committed
1278
* Simple Match Data::     Accessing single items of match data,
Glenn Morris's avatar
Glenn Morris committed
1279
                            such as where a particular subexpression started.
Glenn Morris's avatar
Glenn Morris committed
1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292
* Entire Match Data::     Accessing the entire match data at once, as a list.
* Saving Match Data::     Saving and restoring the match data.
@end menu

@node Replacing Match
@subsection Replacing the Text that Matched
@cindex replace matched text

  This function replaces all or part of the text matched by the last
search.  It works by means of the match data.

@cindex case in replacements
@defun replace-match replacement &optional fixedcase literal string subexp
Chong Yidong's avatar
Chong Yidong committed
1293 1294 1295 1296 1297 1298 1299
This function performs a replacement operation on a buffer or string.

If you did the last search in a buffer, you should omit the
@var{string} argument or specify @code{nil} for it, and make sure that
the current buffer is the one in which you performed the last search.
Then this function edits the buffer, replacing the matched text with
@var{replacement}.  It leaves point at the end of the replacement
1300
text.
Chong Yidong's avatar
Chong Yidong committed
1301 1302 1303 1304

If you performed the last search on a string, pass the same string as
@var{string}.  Then this function returns a new string, in which the
matched text is replaced by @var{replacement}.
Glenn Morris's avatar
Glenn Morris committed
1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324

If @var{fixedcase} is non-@code{nil}, then @code{replace-match} uses
the replacement text without case conversion; otherwise, it converts
the replacement text depending upon the capitalization of the text to
be replaced.  If the original text is all upper case, this converts
the replacement text to upper case.  If all words of the original text
are capitalized, this capitalizes all the words of the replacement
text.  If all the words are one-letter and they are all upper case,
they are treated as capitalized words rather than all-upper-case
words.

If @var{literal} is non-@code{nil}, then @var{replacement} is inserted
exactly as it is, the only alterations being case changes as needed.
If it is @code{nil} (the default), then the character @samp{\} is treated
specially.  If a @samp{\} appears in @var{replacement}, then it must be
part of one of the following sequences:

@table @asis
@item @samp{\&}
@cindex @samp{&} in replacement
1325
This stands for the entire text being replaced.
Glenn Morris's avatar
Glenn Morris committed
1326

1327
@item @samp{\@var{n}}, where @var{n} is a digit
Glenn Morris's avatar
Glenn Morris committed
1328
@cindex @samp{\@var{n}} in replacement
1329 1330 1331 1332
This stands for the text that matched the @var{n}th subexpression in
the original regexp.  Subexpressions are those expressions grouped
inside @samp{\(@dots{}\)}.  If the @var{n}th subexpression never
matched, an empty string is substituted.
Glenn Morris's avatar
Glenn Morris committed
1333 1334 1335

@item @samp{\\}
@cindex @samp{\} in replacement
1336 1337 1338 1339
This stands for a single @samp{\} in the replacement text.

@item @samp{\?}
This stands for itself (for compatibility with @code{replace-regexp}
1340
and related commands; @pxref{Regexp Replace,,, emacs, The GNU
1341
Emacs Manual}).
Glenn Morris's avatar
Glenn Morris committed
1342 1343
@end table

1344 1345 1346 1347 1348 1349
@noindent
Any other character following @samp{\} signals an error.

The substitutions performed by @samp{\&} and @samp{\@var{n}} occur
after case conversion, if any.  Therefore, the strings they substitute
are never case-converted.
Glenn Morris's avatar
Glenn Morris committed
1350 1351 1352 1353 1354 1355 1356 1357

If @var{subexp} is non-@code{nil}, that says to replace just
subexpression number @var{subexp} of the regexp that was matched, not
the entire match.  For example, after matching @samp{foo \(ba*r\)},
calling @code{replace-match} with 1 as @var{subexp} means to replace
just the text that matched @samp{\(ba*r\)}.
@end defun

1358 1359 1360 1361 1362 1363 1364 1365 1366 1367
@defun match-substitute-replacement replacement &optional fixedcase literal string subexp
This function returns the text that would be inserted into the buffer
by @code{replace-match}, but without modifying the buffer.  It is
useful if you want to present the user with actual replacement result,
with constructs like @samp{\@var{n}} or @samp{\&} substituted with
matched groups.  Arguments @var{replacement} and optional
@var{fixedcase}, @var{literal}, @var{string} and @var{subexp} have the
same meaning as for @code{replace-match}.
@end defun

Glenn Morris's avatar
Glenn Morris committed
1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391
@node Simple Match Data
@subsection Simple Match Data Access

  This section explains how to use the match data to find out what was
matched by the last search or match operation, if it succeeded.

  You can ask about the entire matching text, or about a particular
parenthetical subexpression of a regular expression.  The @var{count}
argument in the functions below specifies which.  If @var{count} is
zero, you are asking about the entire match.  If @var{count} is
positive, it specifies which subexpression you want.

  Recall that the subexpressions of a regular expression are those
expressions grouped with escaped parentheses, @samp{\(@dots{}\)}.  The
@var{count}th subexpression is found by counting occurrences of
@samp{\(} from the beginning of the whole regular expression.  The first
subexpression is numbered 1, the second 2, and so on.  Only regular
expressions can have subexpressions---after a simple string search, the
only information available is about the entire match.

  Every successful search sets the match data.  Therefore, you should
query the match data immediately after searching, before calling any
other function that might perform another search.  Alternatively, you
may save and restore the match data (@pxref{Saving Match Data}) around
1392 1393
the call to functions that could perform another search.  Or use the
functions that explicitly do not modify the match data;
1394
e.g., @code{string-match-p}.
Glenn Morris's avatar
Glenn Morris committed
1395

1396 1397
@c This is an old comment and presumably there is no prospect of this
@c changing now.  But still the advice stands.
Glenn Morris's avatar
Glenn Morris committed
1398
  A search which fails may or may not alter the match data.  In the
1399 1400 1401
current implementation, it does not, but we may change it in the
future.  Don't try to rely on the value of the match data after a
failing search.
Glenn Morris's avatar
Glenn Morris committed
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414

@defun match-string count &optional in-string
This function returns, as a string, the text matched in the last search
or match operation.  It returns the entire text if @var{count} is zero,
or just the portion corresponding to the @var{count}th parenthetical
subexpression, if @var{count} is positive.

If the last such operation was done against a string with
@code{string-match}, then you should pass the same string as the
argument @var{in-string}.  After a buffer search or match,
you should omit @var{in-string} or pass @code{nil} for it; but you
should make sure that the current buffer when you call
@code{match-string} is the one in which you did the searching or
1415
matching.  Failure to follow this advice will lead to incorrect results.
Glenn Morris's avatar
Glenn Morris committed
1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427

The value is @code{nil} if @var{count} is out of range, or for a
subexpression inside a @samp{\|} alternative that wasn't used or a
repetition that repeated zero times.
@end defun

@defun match-string-no-properties count &optional in-string
This function is like @code{match-string} except that the result
has no text properties.
@end defun

@defun match-beginning count
1428 1429 1430
If the last regular expression search found a match, this function
returns the position of the start of the matching text or of a
subexpression of it.
Glenn Morris's avatar
Glenn Morris committed
1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497

If @var{count} is zero, then the value is the position of the start of
the entire match.  Otherwise, @var{count} specifies a subexpression in
the regular expression, and the value of the function is the starting
position of the match for that subexpression.

The value is @code{nil} for a subexpression inside a @samp{\|}
alternative that wasn't used or a repetition that repeated zero times.
@end defun

@defun match-end count
This function is like @code{match-beginning} except that it returns the
position of the end of the match, rather than the position of the
beginning.
@end defun

  Here is an example of using the match data, with a comment showing the
positions within the text:

@example
@group
(string-match "\\(qu\\)\\(ick\\)"
              "The quick fox jumped quickly.")
              ;0123456789
     @result{} 4
@end group

@group
(match-string 0 "The quick fox jumped quickly.")
     @result{} "quick"
(match-string 1 "The quick fox jumped quickly.")
     @result{} "qu"
(match-string 2 "The quick fox jumped quickly.")
     @result{} "ick"
@end group

@group
(match-beginning 1)       ; @r{The beginning of the match}
     @result{} 4                 ;   @r{with @samp{qu} is at index 4.}
@end group

@group
(match-beginning 2)       ; @r{The beginning of the match}
     @result{} 6                 ;   @r{with @samp{ick} is at index 6.}
@end group

@group
(match-end 1)             ; @r{The end of the match}
     @result{} 6                 ;   @r{with @samp{qu} is at index 6.}

(match-end 2)             ; @r{The end of the match}
     @result{} 9                 ;   @r{with @samp{ick} is at index 9.}
@end group
@end example

  Here is another example.  Point is initially located at the beginning
of the line.  Searching moves point to between the space and the word
@samp{in}.  The beginning of the entire match is at the 9th character of
the buffer (@samp{T}), and the beginning of the match for the first
subexpression is at the 13th character (@samp{c}).

@example
@group
(list
  (re-search-forward "The \\(cat \\)")
  (match-beginning 0)
  (match-beginning 1))
1498
    @result{} (17 9 13)
Glenn Morris's avatar
Glenn Morris committed
1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521
@end group

@group
---------- Buffer: foo ----------
I read "The cat @point{}in the hat comes back" twice.
        ^   ^
        9  13
---------- Buffer: foo ----------
@end group
@end example

@noindent
(In this case, the index returned is a buffer position; the first
character of the buffer counts as 1.)

@node Entire Match Data
@subsection Accessing the Entire Match Data

  The functions @code{match-data} and @code{set-match-data} read or
write the entire match data, all at once.

@defun match-data &optional integers reuse reseat
This function returns a list of positions (markers or integers) that
1522
record all the information on the text that the last search matched.
Glenn Morris's avatar
Glenn Morris committed
1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590
Element zero is the position of the beginning of the match for the
whole expression; element one is the position of the end of the match
for the expression.  The next two elements are the positions of the
beginning and end of the match for the first subexpression, and so on.
In general, element
@ifnottex
number 2@var{n}
@end ifnottex
@tex
number {\mathsurround=0pt $2n$}
@end tex
corresponds to @code{(match-beginning @var{n})}; and
element
@ifnottex
number 2@var{n} + 1
@end ifnottex
@tex
number {\mathsurround=0pt $2n+1$}
@end tex
corresponds to @code{(match-end @var{n})}.

Normally all the elements are markers or @code{nil}, but if
@var{integers} is non-@code{nil}, that means to use integers instead
of markers.  (In that case, the buffer itself is appended as an
additional element at the end of the list, to facilitate complete
restoration of the match data.)  If the last match was done on a
string with @code{string-match}, then integers are always used,
since markers can't point into a string.

If @var{reuse} is non-@code{nil}, it should be a list.  In that case,
@code{match-data} stores the match data in @var{reuse}.  That is,
@var{reuse} is destructively modified.  @var{reuse} does not need to
have the right length.  If it is not long enough to contain the match
data, it is extended.  If it is too long, the length of @var{reuse}
stays the same, but the elements that were not used are set to
@code{nil}.  The purpose of this feature is to reduce the need for
garbage collection.

If @var{reseat} is non-@code{nil}, all markers on the @var{reuse} list
are reseated to point to nowhere.

As always, there must be no possibility of intervening searches between
the call to a search function and the call to @code{match-data} that is
intended to access the match data for that search.

@example
@group
(match-data)
     @result{}  (#<marker at 9 in foo>
          #<marker at 17 in foo>
          #<marker at 13 in foo>
          #<marker at 17 in foo>)
@end group
@end example
@end defun

@defun set-match-data match-list &optional reseat
This function sets the match data from the elements of @var{match-list},
which should be a list that was the value of a previous call to
@code{match-data}.  (More precisely, anything that has the same format
will work.)

If @var{match-list} refers to a buffer that doesn't exist, you don't get
an error; that sets the match data in a meaningless but harmless way.

If @var{reseat} is non-@code{nil}, all markers on the @var{match-list} list
are reseated to point to nowhere.

1591
@c TODO Make it properly obsolete.
Glenn Morris's avatar
Glenn Morris committed
1592 1593 1594 1595 1596 1597 1598
@findex store-match-data
@code{store-match-data} is a semi-obsolete alias for @code{set-match-data}.
@end defun

@node Saving Match Data
@subsection Saving and Restoring the Match Data

1599
  When you call a function that may search, you may need to save
Glenn Morris's avatar
Glenn Morris committed
1600 1601 1602 1603 1604 1605 1606 1607
and restore the match data around that call, if you want to preserve the
match data from an earlier search for later use.  Here is an example
that shows the problem that arises if you fail to save the match data:

@example
@group
(re-search-forward "The \\(cat \\)")
     @result{} 48
1608
(foo)                   ; @r{@code{foo} does more searching.}
Glenn Morris's avatar
Glenn Morris committed
1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700
(match-end 0)
     @result{} 61              ; @r{Unexpected result---not 48!}
@end group
@end example

  You can save and restore the match data with @code{save-match-data}:

@defmac save-match-data body@dots{}
This macro executes @var{body}, saving and restoring the match
data around it.  The return value is the value of the last form in
@var{body}.
@end defmac

  You could use @code{set-match-data} together with @code{match-data} to
imitate the effect of the special form @code{save-match-data}.  Here is
how:

@example
@group
(let ((data (match-data)))
  (unwind-protect
      @dots{}   ; @r{Ok to change the original match data.}
    (set-match-data data)))
@end group
@end example

  Emacs automatically saves and restores the match data when it runs
process filter functions (@pxref{Filter Functions}) and process
sentinels (@pxref{Sentinels}).

@ignore
  Here is a function which restores the match data provided the buffer
associated with it still exists.

@smallexample
@group
(defun restore-match-data (data)
@c It is incorrect to split the first line of a doc string.
@c If there's a problem here, it should be solved in some other way.
  "Restore the match data DATA unless the buffer is missing."
  (catch 'foo
    (let ((d data))
@end group
      (while d
        (and (car d)
             (null (marker-buffer (car d)))
@group
             ;; @file{match-data} @r{buffer is deleted.}
             (throw 'foo nil))
        (setq d (cdr d)))
      (set-match-data data))))
@end group
@end smallexample
@end ignore

@node Search and Replace
@section Search and Replace
@cindex replacement after search
@cindex searching and replacing

  If you want to find all matches for a regexp in part of the buffer,
and replace them, the best way is to write an explicit loop using
@code{re-search-forward} and @code{replace-match}, like this:

@example
(while (re-search-forward "foo[ \t]+bar" nil t)
  (replace-match "foobar"))
@end example

@noindent
@xref{Replacing Match,, Replacing the Text that Matched}, for a
description of @code{replace-match}.

  However, replacing matches in a string is more complex, especially
if you want to do it efficiently.  So Emacs provides a function to do
this.

@defun replace-regexp-in-string regexp rep string &optional fixedcase literal subexp start
This function copies @var{string} and searches it for matches for
@var{regexp}, and replaces them with @var{rep}.  It returns the
modified copy.  If @var{start} is non-@code{nil}, the search for
matches starts at that index in @var{string}, so matches starting
before that index are not changed.

This function uses @code{replace-match} to do the replacement, and it
passes the optional arguments @var{fixedcase}, @var{literal} and
@var{subexp} along to @code{replace-match}.

Instead of a string, @var{rep} can be a function.  In that case,
@code{replace-regexp-in-string} calls @var{rep} for each match,
passing the text of the match as its sole argument.  It collects the
value @var{rep} returns and passes that to @code{replace-match} as the
1701
replacement string.  The match data at this point are the result
Glenn Morris's avatar
Glenn Morris committed
1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727
of matching @var{regexp} against a substring of @var{string}.
@end defun

  If you want to write a command along the lines of @code{query-replace},
you can use @code{perform-replace} to do the work.

@defun perform-replace from-string replacements query-flag regexp-flag delimited-flag &optional repeat-count map start end
This function is the guts of @code{query-replace} and related
commands.  It searches for occurrences of @var{from-string} in the
text between positions @var{start} and @var{end} and replaces some or
all of them.  If @var{start} is @code{nil} (or omitted), point is used
instead, and the end of the buffer's accessible portion is used for
@var{end}.

If @var{query-flag} is @code{nil}, it replaces all
occurrences; otherwise, it asks the user what to do about each one.

If @var{regexp-flag} is non-@code{nil}, then @var{from-string} is
considered a regular expression; otherwise, it must match literally.  If
@var{delimited-flag} is non-@code{nil}, then only replacements
surrounded by word boundaries are considered.

The argument @var{replacements} specifies what to replace occurrences
with.  If it is a string, that string is used.  It can also be a list of
strings, to be used in cyclic order.

1728 1729
If @var{replacements} is a cons cell, @w{@code{(@var{function}
. @var{data})}}, this means to call @var{function} after each match to
Glenn Morris's avatar
Glenn Morris committed
1730 1731 1732 1733 1734 1735 1736 1737 1738
get the replacement text.  This function is called with two arguments:
@var{data}, and the number of replacements already made.

If @var{repeat-count} is non-@code{nil}, it should be an integer.  Then
it specifies how many times to use each of the strings in the
@var{replacements} list before advancing cyclically to the next one.

If @var{from-string} contains upper-case letters, then
@code{perform-replace} binds @code{case-fold-search} to @code{nil}, and
1739
it uses the @var{replacements} without altering their case.
Glenn Morris's avatar
Glenn Morris committed
1740 1741 1742 1743 1744

Normally, the keymap @code{query-replace-map} defines the possible
user responses for queries.  The argument @var{map}, if
non-@code{nil}, specifies a keymap to use instead of
@code{query-replace-map}.
1745 1746 1747 1748 1749 1750 1751

This function uses one of two functions to search for the next
occurrence of @var{from-string}.  These functions are specified by the
values of two variables: @code{replace-re-search-function} and
@code{replace-search-function}.  The former is called when the
argument @var{regexp-flag} is non-@code{nil}, the latter when it is
@code{nil}.
Glenn Morris's avatar
Glenn Morris committed
1752 1753 1754 1755 1756 1757 1758 1759 1760 1761
@end defun

@defvar query-replace-map
This variable holds a special keymap that defines the valid user
responses for @code{perform-replace} and the commands that use it, as
well as @code{y-or-n-p} and @code{map-y-or-n-p}.  This map is unusual
in two ways:

@itemize @bullet
@item
Paul Eggert's avatar
Paul Eggert committed
1762
The key bindings are not commands, just symbols that are meaningful
Glenn Morris's avatar
Glenn Morris committed
1763 1764 1765 1766 1767 1768
to the functions that use this map.

@item
Prefix keys are not supported; each key binding must be for a
single-event key sequence.  This is because the functions don't use
@code{read-key-sequence} to get the input; instead, they read a single
1769
event and look it up ``by hand''.
Glenn Morris's avatar
Glenn Morris committed
1770 1771 1772
@end itemize
@end defvar

Paul Eggert's avatar
Paul Eggert committed
1773
Here are the meaningful bindings for @code{query-replace-map}.
Glenn Morris's avatar
Glenn Morris committed
1774 1775 1776 1777 1778
Several of them are meaningful only for @code{query-replace} and
friends.

@table @code
@item act
1779
Do take the action being considered---in other words, ``yes''.
Glenn Morris's avatar
Glenn Morris committed
1780 1781

@item skip
1782
Do not take action for this question---in other words, ``no''.
Glenn Morris's avatar
Glenn Morris committed
1783 1784

@item exit
1785 1786 1787 1788 1789
Answer this question ``no'', and give up on the entire series of
questions, assuming that the answers will be ``no''.

@item exit-prefix
Like @code{exit}, but add the key that was pressed to
1790
@code{unread-command-events} (@pxref{Event Input Misc}).
Glenn Morris's avatar
Glenn Morris committed
1791 1792

@item act-and-exit
1793 1794
Answer this question ``yes'', and give up on the entire series of
questions, assuming that subsequent answers will be ``no''.
Glenn Morris's avatar
Glenn Morris committed
1795 1796

@item act-and-show
1797
Answer this question ``yes'', but show the results---don't advance yet
Glenn Morris's avatar
Glenn Morris committed
1798 1799 1800 1801
to the next question.

@item automatic
Answer this question and all subsequent questions in the series with
1802
``yes'', without further user interaction.
Glenn Morris's avatar
Glenn Morris committed
1803 1804 1805 1806 1807 1808 1809 1810

@item backup
Move back to the previous place that a question was asked about.

@item edit
Enter a recursive edit to deal with this question---instead of any
other action that would normally be taken.

1811 1812 1813
@item edit-replacement
Edit the replacement for this question in the minibuffer.

Glenn Morris's avatar
Glenn Morris committed
1814 1815 1816 1817 1818
@item delete-and-edit
Delete the text being considered, then enter a recursive edit to replace
it.

@item recenter
Chong Yidong's avatar
Chong Yidong committed
1819 1820 1821 1822 1823 1824 1825
@itemx scroll-up
@itemx scroll-down
@itemx scroll-other-window
@itemx scroll-other-window-down
Perform the specified window scroll operation, then ask the same
question again.  Only @code{y-or-n-p} and related functions use this
answer.
Glenn Morris's avatar
Glenn Morris committed
1826 1827 1828 1829 1830 1831 1832 1833 1834

@item quit
Perform a quit right away.  Only @code{y-or-n-p} and related functions