This is Info file elisp, produced by Makeinfo-1.63 from the input file elisp.texi. This version is the edition 2.4.2 of the GNU Emacs Lisp Reference Manual. It corresponds to Emacs Version 19.34. Published by the Free Software Foundation 59 Temple Place, Suite 330 Boston, MA 02111-1307 USA Copyright (C) 1990, 1991, 1992, 1993, 1994, 1995, 1996 Free Software Foundation, Inc. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that this permission notice may be stated in a translation approved by the Foundation. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided also that the section entitled "GNU General Public License" is included exactly as in the original, and provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that the section entitled "GNU General Public License" may be included in a translation approved by the Free Software Foundation instead of in the original English. File: elisp, Node: Special Properties, Next: Format Properties, Prev: Property Search, Up: Text Properties Properties with Special Meanings -------------------------------- Here is a table of text property names that have special built-in meanings. The following section lists a few more special property names that are used to control filling. All other names have no standard meaning, and you can use them as you like. `category' If a character has a `category' property, we call it the "category" of the character. It should be a symbol. The properties of the symbol serve as defaults for the properties of the character. `face' You can use the property `face' to control the font and color of text. Its value is a face name or a list of face names. *Note Faces::, for more information. This feature may be temporary; in the future, we may replace it with other ways of specifying how to display text. `mouse-face' The property `mouse-face' is used instead of `face' when the mouse is on or near the character. For this purpose, "near" means that all text between the character and where the mouse is have the same `mouse-face' property value. `local-map' You can specify a different keymap for a portion of the text by means of a `local-map' property. The property's value for the character after point, if non-`nil', replaces the buffer's local map. *Note Active Keymaps::. `read-only' If a character has the property `read-only', then modifying that character is not allowed. Any command that would do so gets an error. Insertion next to a read-only character is an error if inserting ordinary text there would inherit the `read-only' property due to stickiness. Thus, you can control permission to insert next to read-only text by controlling the stickiness. *Note Sticky Properties::. Since changing properties counts as modifying the buffer, it is not possible to remove a `read-only' property unless you know the special trick: bind `inhibit-read-only' to a non-`nil' value and then remove the property. *Note Read Only Buffers::. `invisible' A non-`nil' `invisible' property can make a character invisible on the screen. *Note Invisible Text::, for details. `intangible' If a group of consecutive characters have equal and non-`nil' `intangible' properties, then you cannot place point between them. If you try to move point forward into the group, point actually moves to the end of the group. If you try to move point backward into the group, point actually moves to the start of the group. When the variable `inhibit-point-motion-hooks' is non-`nil', the `intangible' property is ignored. `modification-hooks' If a character has the property `modification-hooks', then its value should be a list of functions; modifying that character calls all of those functions. Each function receives two arguments: the beginning and end of the part of the buffer being modified. Note that if a particular modification hook function appears on several characters being modified by a single primitive, you can't predict how many times the function will be called. `insert-in-front-hooks' `insert-behind-hooks' The operation of inserting text in a buffer also calls the functions listed in the `insert-in-front-hooks' property of the following character and in the `insert-behind-hooks' property of the preceding character. These functions receive two arguments, the beginning and end of the inserted text. The functions are called *after* the actual insertion takes place. See also *Note Change Hooks::, for other hooks that are called when you change text in a buffer. `point-entered' `point-left' The special properties `point-entered' and `point-left' record hook functions that report motion of point. Each time point moves, Emacs compares these two property values: * the `point-left' property of the character after the old location, and * the `point-entered' property of the character after the new location. If these two values differ, each of them is called (if not `nil') with two arguments: the old value of point, and the new one. The same comparison is made for the characters before the old and new locations. The result may be to execute two `point-left' functions (which may be the same function) and/or two `point-entered' functions (which may be the same function). In any case, all the `point-left' functions are called first, followed by all the `point-entered' functions. A primitive function may examine characters at various positions without moving point to those positions. Only an actual change in the value of point runs these hook functions. - Variable: inhibit-point-motion-hooks When this variable is non-`nil', `point-left' and `point-entered' hooks are not run, and the `intangible' property has no effect. File: elisp, Node: Format Properties, Next: Sticky Properties, Prev: Special Properties, Up: Text Properties Formatted Text Properties ------------------------- These text properties affect the behavior of the fill commands. They are used for representing formatted text. *Note Filling::, and *Note Margins::. `hard' If a newline character has this property, it is a "hard" newline. The fill commands do not alter hard newlines and do not move words across them. However, this property takes effect only if the variable `use-hard-newlines' is non-`nil'. `right-margin' This property specifies an extra right margin for filling this part of the text. `left-margin' This property specifies an extra left margin for filling this part of the text. `justification' This property specifies the style of justification for filling this part of the text. File: elisp, Node: Sticky Properties, Next: Saving Properties, Prev: Format Properties, Up: Text Properties Stickiness of Text Properties ----------------------------- Self-inserting characters normally take on the same properties as the preceding character. This is called "inheritance" of properties. In a Lisp program, you can do insertion with inheritance or without, depending on your choice of insertion primitive. The ordinary text insertion functions such as `insert' do not inherit any properties. They insert text with precisely the properties of the string being inserted, and no others. This is correct for programs that copy text from one context to another--for example, into or out of the kill ring. To insert with inheritance, use the special primitives described in this section. Self-inserting characters inherit properties because they work using these primitives. When you do insertion with inheritance, *which* properties are inherited depends on two specific properties: `front-sticky' and `rear-nonsticky'. Insertion after a character inherits those of its properties that are "rear-sticky". Insertion before a character inherits those of its properties that are "front-sticky". By default, a text property is rear-sticky but not front-sticky. Thus, the default is to inherit all the properties of the preceding character, and nothing from the following character. You can request different behavior by specifying the stickiness of certain properties. If a character's `front-sticky' property is `t', then all its properties are front-sticky. If the `front-sticky' property is a list, then the sticky properties of the character are those whose names are in the list. For example, if a character has a `front-sticky' property whose value is `(face read-only)', then insertion before the character can inherit its `face' property and its `read-only' property, but no others. The `rear-nonsticky' works the opposite way. Every property is rear-sticky by default, so the `rear-nonsticky' property says which properties are *not* rear-sticky. If a character's `rear-nonsticky' property is `t', then none of its properties are rear-sticky. If the `rear-nonsticky' property is a list, properties are rear-sticky *unless* their names are in the list. When you insert text with inheritance, it inherits all the rear-sticky properties of the preceding character, and all the front-sticky properties of the following character. The previous character's properties take precedence when both sides offer different sticky values for the same property. Here are the functions that insert text with inheritance of properties: - Function: insert-and-inherit &rest STRINGS Insert the strings STRINGS, just like the function `insert', but inherit any sticky properties from the adjoining text. - Function: insert-before-markers-and-inherit &rest STRINGS Insert the strings STRINGS, just like the function `insert-before-markers', but inherit any sticky properties from the adjoining text. File: elisp, Node: Saving Properties, Next: Lazy Properties, Prev: Sticky Properties, Up: Text Properties Saving Text Properties in Files ------------------------------- You can save text properties in files, and restore text properties when inserting the files, using these two hooks: - Variable: write-region-annotate-functions This variable's value is a list of functions for `write-region' to run to encode text properties in some fashion as annotations to the text being written in the file. *Note Writing to Files::. Each function in the list is called with two arguments: the start and end of the region to be written. These functions should not alter the contents of the buffer. Instead, they should return lists indicating annotations to write in the file in addition to the text in the buffer. Each function should return a list of elements of the form `(POSITION . STRING)', where POSITION is an integer specifying the relative position in the text to be written, and STRING is the annotation to add there. Each list returned by one of these functions must be already sorted in increasing order by POSITION. If there is more than one function, `write-region' merges the lists destructively into one sorted list. When `write-region' actually writes the text from the buffer to the file, it intermixes the specified annotations at the corresponding positions. All this takes place without modifying the buffer. - Variable: after-insert-file-functions This variable holds a list of functions for `insert-file-contents' to call after inserting a file's contents. These functions should scan the inserted text for annotations, and convert them to the text properties they stand for. Each function receives one argument, the length of the inserted text; point indicates the start of that text. The function should scan that text for annotations, delete them, and create the text properties that the annotations specify. The function should return the updated length of the inserted text, as it stands after those changes. The value returned by one function becomes the argument to the next function. These functions should always return with point at the beginning of the inserted text. The intended use of `after-insert-file-functions' is for converting some sort of textual annotations into actual text properties. But other uses may be possible. We invite users to write Lisp programs to store and retrieve text properties in files, using these hooks, and thus to experiment with various data formats and find good ones. Eventually we hope users will produce good, general extensions we can install in Emacs. We suggest not trying to handle arbitrary Lisp objects as property names or property values--because a program that general is probably difficult to write, and slow. Instead, choose a set of possible data types that are reasonably flexible, and not too hard to encode. *Note Format Conversion::, for a related feature. File: elisp, Node: Lazy Properties, Next: Not Intervals, Prev: Saving Properties, Up: Text Properties Lazy Computation of Text Properties ----------------------------------- Instead of computing text properties for all the text in the buffer, you can arrange to compute the text properties for parts of the text when and if something depends on them. The primitive that extracts text from the buffer along with its properties is `buffer-substring'. Before examining the properties, this function runs the abnormal hook `buffer-access-fontify-functions'. - Variable: buffer-access-fontify-functions This variable holds a list of functions for computing text properties. Before `buffer-substring' copies the text and text properties for a portion of the buffer, it calls all the functions in this list. Each of the functions receives two arguments that specify the range of the buffer being accessed. (The buffer itself is always the current buffer.) The function `buffer-substring-no-properties' does not call these functions, since it ignores text properties anyway. In order to prevent the hook functions from being called more than once for the same part of the buffer, you can use the variable `buffer-access-fontified-property'. - Variable: buffer-access-fontified-property If this value's variable is non-`nil', it is a symbol which is used as a text property name. A non-`nil' value for that text property means, "the other text properties for this character have already been computed." If all the characters in the range specified for `buffer-substring' have a non-`nil' value for this property, `buffer-substring' does not call the `buffer-access-fontify-functions' functions. It assumes these characters already have the right text properties, and just copies the properties they already have. The normal way to use this feature is that the `buffer-access-fontify-functions' functions add this property, as well as others, to the characters they operate on. That way, they avoid being called over and over for the same text. File: elisp, Node: Not Intervals, Prev: Lazy Properties, Up: Text Properties Why Text Properties are not Intervals ------------------------------------- Some editors that support adding attributes to text in the buffer do so by letting the user specify "intervals" within the text, and adding the properties to the intervals. Those editors permit the user or the programmer to determine where individual intervals start and end. We deliberately provided a different sort of interface in Emacs Lisp to avoid certain paradoxical behavior associated with text modification. If the actual subdivision into intervals is meaningful, that means you can distinguish between a buffer that is just one interval with a certain property, and a buffer containing the same text subdivided into two intervals, both of which have that property. Suppose you take the buffer with just one interval and kill part of the text. The text remaining in the buffer is one interval, and the copy in the kill ring (and the undo list) becomes a separate interval. Then if you yank back the killed text, you get two intervals with the same properties. Thus, editing does not preserve the distinction between one interval and two. Suppose we "fix" this problem by coalescing the two intervals when the text is inserted. That works fine if the buffer originally was a single interval. But suppose instead that we have two adjacent intervals with the same properties, and we kill the text of one interval and yank it back. The same interval-coalescence feature that rescues the other case causes trouble in this one: after yanking, we have just one interval. One again, editing does not preserve the distinction between one interval and two. Insertion of text at the border between intervals also raises questions that have no satisfactory answer. However, it is easy to arrange for editing to behave consistently for questions of the form, "What are the properties of this character?" So we have decided these are the only questions that make sense; we have not implemented asking questions about where intervals start or end. In practice, you can usually use the property search functions in place of explicit interval boundaries. You can think of them as finding the boundaries of intervals, assuming that intervals are always coalesced whenever possible. *Note Property Search::. Emacs also provides explicit intervals as a presentation feature; see *Note Overlays::. File: elisp, Node: Substitution, Next: Transposition, Prev: Text Properties, Up: Text Substituting for a Character Code ================================= The following functions replace characters within a specified region based on their character codes. - Function: subst-char-in-region START END OLD-CHAR NEW-CHAR &optional NOUNDO This function replaces all occurrences of the character OLD-CHAR with the character NEW-CHAR in the region of the current buffer defined by START and END. If NOUNDO is non-`nil', then `subst-char-in-region' does not record the change for undo and does not mark the buffer as modified. This feature is used for controlling selective display (*note Selective Display::.). `subst-char-in-region' does not move point and returns `nil'. ---------- Buffer: foo ---------- This is the contents of the buffer before. ---------- Buffer: foo ---------- (subst-char-in-region 1 20 ?i ?X) => nil ---------- Buffer: foo ---------- ThXs Xs the contents of the buffer before. ---------- Buffer: foo ---------- - Function: translate-region START END TABLE This function applies a translation table to the characters in the buffer between positions START and END. The translation table TABLE is a string; `(aref TABLE OCHAR)' gives the translated character corresponding to OCHAR. If the length of TABLE is less than 256, any characters with codes larger than the length of TABLE are not altered by the translation. The return value of `translate-region' is the number of characters that were actually changed by the translation. This does not count characters that were mapped into themselves in the translation table. File: elisp, Node: Registers, Next: Change Hooks, Prev: Transposition, Up: Text Registers ========= A register is a sort of variable used in Emacs editing that can hold a marker, a string, a rectangle, a window configuration (of one frame), or a frame configuration (of all frames). Each register is named by a single character. All characters, including control and meta characters (but with the exception of `C-g'), can be used to name registers. Thus, there are 255 possible registers. A register is designated in Emacs Lisp by a character that is its name. The functions in this section return unpredictable values unless otherwise stated. - Variable: register-alist This variable is an alist of elements of the form `(NAME . CONTENTS)'. Normally, there is one element for each Emacs register that has been used. The object NAME is a character (an integer) identifying the register. The object CONTENTS is a string, marker, or list representing the register contents. A string represents text stored in the register. A marker represents a position. A list represents a rectangle; its elements are strings, one per line of the rectangle. - Function: get-register REG This function returns the contents of the register REG, or `nil' if it has no contents. - Function: set-register REG VALUE This function sets the contents of register REG to VALUE. A register can be set to any value, but the other register functions expect only certain data types. The return value is VALUE. - Command: view-register REG This command displays what is contained in register REG. - Command: insert-register REG &optional BEFOREP This command inserts contents of register REG into the current buffer. Normally, this command puts point before the inserted text, and the mark after it. However, if the optional second argument BEFOREP is non-`nil', it puts the mark before and point after. You can pass a non-`nil' second argument BEFOREP to this function interactively by supplying any prefix argument. If the register contains a rectangle, then the rectangle is inserted with its upper left corner at point. This means that text is inserted in the current line and underneath it on successive lines. If the register contains something other than saved text (a string) or a rectangle (a list), currently useless things happen. This may be changed in the future. File: elisp, Node: Transposition, Next: Registers, Prev: Substitution, Up: Text Transposition of Text ===================== This subroutine is used by the transposition commands. - Function: transpose-regions START1 END1 START2 END2 &optional LEAVE-MARKERS This function exchanges two nonoverlapping portions of the buffer. Arguments START1 and END1 specify the bounds of one portion and arguments START2 and END2 specify the bounds of the other portion. Normally, `transpose-regions' relocates markers with the transposed text; a marker previously positioned within one of the two transposed portions moves along with that portion, thus remaining between the same two characters in their new position. However, if LEAVE-MARKERS is non-`nil', `transpose-regions' does not do this--it leaves all markers unrelocated. File: elisp, Node: Change Hooks, Prev: Registers, Up: Text Change Hooks ============ These hook variables let you arrange to take notice of all changes in all buffers (or in a particular buffer, if you make them buffer-local). See also *Note Special Properties::, for how to detect changes to specific parts of the text. The functions you use in these hooks should save and restore the match data if they do anything that uses regular expressions; otherwise, they will interfere in bizarre ways with the editing operations that call them. - Variable: before-change-functions This variable holds a list of a functions to call before any buffer modification. Each function gets two arguments, the beginning and end of the region that is about to change, represented as integers. The buffer that is about to change is always the current buffer. - Variable: after-change-functions This variable holds a list of a functions to call after any buffer modification. Each function receives three arguments: the beginning and end of the region just changed, and the length of the text that existed before the change. (To get the current length, subtract the region beginning from the region end.) All three arguments are integers. The buffer that's about to change is always the current buffer. - Variable: before-change-function This obsolete variable holds one function to call before any buffer modification (or `nil' for no function). It is called just like the functions in `before-change-functions'. - Variable: after-change-function This obsolete variable holds one function to call after any buffer modification (or `nil' for no function). It is called just like the functions in `after-change-functions'. The four variables above are temporarily bound to `nil' during the time that any of these functions is running. This means that if one of these functions changes the buffer, that change won't run these functions. If you do want a hook function to make changes that run these functions, make it bind these variables back to their usual values. One inconvenient result of this protective feature is that you cannot have a function in `after-change-functions' or `before-change-functions' which changes the value of that variable. But that's not a real limitation. If you want those functions to change the list of functions to run, simply add one fixed function to the hook, and code that function to look in another variable for other functions to call. Here is an example: (setq my-own-after-change-functions nil) (defun indirect-after-change-function (beg end len) (let ((list my-own-after-change-functions)) (while list (funcall (car list) beg end len) (setq list (cdr list))))) (add-hooks 'after-change-functions 'indirect-after-change-function) - Variable: first-change-hook This variable is a normal hook that is run whenever a buffer is changed that was previously in the unmodified state. File: elisp, Node: Searching and Matching, Next: Syntax Tables, Prev: Text, Up: Top Searching and Matching ********************** 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 "match data" to determine which text matched the whole regular expression or various portions of it. * Menu: * String Search:: Search for an exact match. * Regular Expressions:: Describing classes of strings. * Regexp Search:: Searching for a match for a regexp. * POSIX Regexps:: Searching POSIX-style for the longest match. * Search and Replace:: Internals of `query-replace'. * Match Data:: Finding out which part of the text matched various parts of a regexp, after regexp search. * Searching and Case:: Case-independent or case-significant searching. * Standard Regexps:: Useful regexps for finding sentences, pages,... The `skip-chars...' functions also perform a kind of searching. *Note Skipping Characters::. File: elisp, Node: String Search, Next: Regular Expressions, Up: Searching and Matching Searching for Strings ===================== 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; LIMIT and NOERROR are set to `nil', and REPEAT is set to 1. - Command: search-forward STRING &optional LIMIT NOERROR REPEAT This function searches forward from point for an exact match for 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 NOERROR (see below). In the following example, point is initially at the beginning of the line. Then `(search-forward "fox")' moves point after the last letter of `fox': ---------- Buffer: foo ---------- -!-The quick brown fox jumped over the lazy dog. ---------- Buffer: foo ---------- (search-forward "fox") => 20 ---------- Buffer: foo ---------- The quick brown fox-!- jumped over the lazy dog. ---------- Buffer: foo ---------- The argument LIMIT specifies the upper bound to the search. (It must be a position in the current buffer.) No match extending after that position is accepted. If LIMIT is omitted or `nil', it defaults to the end of the accessible portion of the buffer. What happens when the search fails depends on the value of NOERROR. If NOERROR is `nil', a `search-failed' error is signaled. If NOERROR is `t', `search-forward' returns `nil' and does nothing. If NOERROR is neither `nil' nor `t', then `search-forward' moves point to the upper bound and returns `nil'. (It would be more consistent now to return the new position of point in that case, but some programs may depend on a value of `nil'.) If REPEAT is supplied (it must be a positive number), then the search is repeated that many 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. - Command: search-backward STRING &optional LIMIT NOERROR REPEAT This function searches backward from point for STRING. It is just like `search-forward' except that it searches backwards and leaves point at the beginning of the match. - Command: word-search-forward STRING &optional LIMIT NOERROR REPEAT This function searches forward from point for a "word" match for 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 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 `ball' does not match the word `balls'), but the details of punctuation and spacing are ignored (searching for `ball boy' does match `ball. Boy!'). In this example, point is initially at the beginning of the buffer; the search leaves it between the `y' and the `!'. ---------- Buffer: foo ---------- -!-He said "Please! Find the ball boy!" ---------- Buffer: foo ---------- (word-search-forward "Please find the ball, boy.") => 35 ---------- Buffer: foo ---------- He said "Please! Find the ball boy-!-!" ---------- Buffer: foo ---------- If LIMIT is non-`nil' (it must be a position in the current buffer), then it is the upper bound to the search. The match found must not extend after that position. If NOERROR is `nil', then `word-search-forward' signals an error if the search fails. If NOERROR is `t', then it returns `nil' instead of signaling an error. If NOERROR is neither `nil' nor `t', it moves point to LIMIT (or the end of the buffer) and returns `nil'. If REPEAT is non-`nil', then the search is repeated that many times. Point is positioned at the end of the last match. - Command: word-search-backward STRING &optional LIMIT NOERROR REPEAT This function searches backward from point for a word match to STRING. This function is just like `word-search-forward' except that it searches backward and normally leaves point at the beginning of the match. File: elisp, Node: Regular Expressions, Next: Regexp Search, Prev: String Search, Up: Searching and Matching Regular Expressions =================== A "regular expression" ("regexp", for short) is a pattern that 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. * Menu: * Syntax of Regexps:: Rules for writing regular expressions. * Regexp Example:: Illustrates regular expression syntax. File: elisp, Node: Syntax of Regexps, Next: Regexp Example, Up: Regular Expressions Syntax of Regular Expressions ----------------------------- Regular expressions have a syntax in which a few characters are special constructs and the rest are "ordinary". An ordinary character is a simple regular expression that matches that character and nothing else. The special characters are `.', `*', `+', `?', `[', `]', `^', `$', and `\'; no new special characters will be defined in the future. Any other character appearing in a regular expression is ordinary, unless a `\' precedes it. For example, `f' is not a special character, so it is ordinary, and therefore `f' is a regular expression that matches the string `f' and no other string. (It does *not* match the string `ff'.) Likewise, `o' is a regular expression that matches only `o'. Any two regular expressions A and B can be concatenated. The result is a regular expression that matches a string if A matches some amount of the beginning of that string and B matches the rest of the string. As a simple example, we can concatenate the regular expressions `f' and `o' to get the regular expression `fo', which matches only the string `fo'. Still trivial. To do something more powerful, you need to use one of the special characters. Here is a list of them: `. (Period)' is a special character that matches any single character except a newline. Using concatenation, we can make regular expressions like `a.b', which matches any three-character string that begins with `a' and ends with `b'. is not a construct by itself; it is a suffix operator that means to repeat the preceding regular expression as many times as possible. In `fo*', the `*' applies to the `o', so `fo*' matches one `f' followed by any number of `o's. The case of zero `o's is allowed: `fo*' does match `f'. `*' always applies to the *smallest* possible preceding expression. Thus, `fo*' has a repeating `o', not a repeating `fo'. The matcher processes a `*' 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 `*'-modified construct in case that makes it possible to match the rest of the pattern. For example, in matching `ca*ar' against the string `caaar', the `a*' first tries to match all three `a's; but the rest of the pattern is `ar' and there is only `r' left to match, so this try fails. The next alternative is for `a*' to match only two `a's. With this choice, the rest of the regexp matches successfully. Nested repetition operators can be extremely slow if they specify backtracking loops. For example, it could take hours for the regular expression `\(x+y*\)*a' to match the sequence `xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxz'. The slowness is because Emacs must try each imaginable way of grouping the 35 `x''s before concluding that none of them can work. To make sure your regular expressions run fast, check nested repetitions carefully. is a suffix operator similar to `*' except that the preceding expression must match at least once. So, for example, `ca+r' matches the strings `car' and `caaaar' but not the string `cr', whereas `ca*r' matches all three strings. is a suffix operator similar to `*' except that the preceding expression can match either once or not at all. For example, `ca?r' matches `car' or `cr', but does not match anyhing else. `[ ... ]' `[' begins a "character set", which is terminated by a `]'. In the simplest case, the characters between the two brackets form the set. Thus, `[ad]' matches either one `a' or one `d', and `[ad]*' matches any string composed of just `a's and `d's (including the empty string), from which it follows that `c[ad]*r' matches `cr', `car', `cdr', `caddaar', etc. The usual regular expression special characters are not special inside a character set. A completely different set of special characters exists inside character sets: `]', `-' and `^'. `-' is used for ranges of characters. To write a range, write two characters with a `-' between them. Thus, `[a-z]' matches any lower case letter. Ranges may be intermixed freely with individual characters, as in `[a-z$%.]', which matches any lower case letter or `$', `%', or a period. To include a `]' in a character set, make it the first character. For example, `[]a]' matches `]' or `a'. To include a `-', write `-' as the first character in the set, or put it immediately after a range. (You can replace one individual character C with the range `C-C' to make a place to put the `-'.) There is no way to write a set containing just `-' and `]'. To include `^' in a set, put it anywhere but at the beginning of the set. `[^ ... ]' `[^' begins a "complement character set", which matches any character except the ones specified. Thus, `[^a-z0-9A-Z]' matches all characters *except* letters and digits. `^' is not special in a character set unless it is the first character. The character following the `^' is treated as if it were first (thus, `-' and `]' are not special there). Note that a complement character set can match a newline, unless newline is mentioned as one of the characters not to match. is a special character that matches the empty string, but only at the beginning of a line in the text being matched. Otherwise it fails to match anything. Thus, `^foo' matches a `foo' that occurs at the beginning of a line. When matching a string instead of a buffer, `^' matches at the beginning of the string or after a newline character `\n'. is similar to `^' but matches only at the end of a line. Thus, `x+$' matches a string of one `x' or more at the end of a line. When matching a string instead of a buffer, `$' matches at the end of the string or before a newline character `\n'. has two functions: it quotes the special characters (including `\'), and it introduces additional special constructs. Because `\' quotes special characters, `\$' is a regular expression that matches only `$', and `\[' is a regular expression that matches only `[', and so on. Note that `\' also has special meaning in the read syntax of Lisp strings (*note String Type::.), and must be quoted with `\'. For example, the regular expression that matches the `\' character is `\\'. To write a Lisp string that contains the characters `\\', Lisp syntax requires you to quote each `\' with another `\'. Therefore, the read syntax for a regular expression matching `\' is `"\\\\"'. *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, `*foo' treats `*' as ordinary since there is no preceding expression on which the `*' can act. It is poor practice to depend on this behavior; quote the special character anyway, regardless of where it appears. For the most part, `\' followed by any character matches only that character. However, there are several exceptions: characters that, when preceded by `\', are special constructs. Such characters are always ordinary when encountered on their own. Here is a table of `\' constructs: specifies an alternative. Two regular expressions A and B with `\|' in between form an expression that matches anything that either A or B matches. Thus, `foo\|bar' matches either `foo' or `bar' but no other string. `\|' applies to the largest possible surrounding expressions. Only a surrounding `\( ... \)' grouping can limit the grouping power of `\|'. Full backtracking capability exists to handle multiple uses of `\|'. `\( ... \)' is a grouping construct that serves three purposes: 1. To enclose a set of `\|' alternatives for other operations. Thus, the regular expression `\(foo\|bar\)x' matches either `foox' or `barx'. 2. To enclose an expression for a suffix operator such as `*' to act on. Thus, `ba\(na\)*' matches `bananana', etc., with any (zero or more) number of `na' strings. 3. To record a matched substring for future reference. This last application is not a consequence of the idea of a parenthetical grouping; it is a separate feature that happens to be assigned as a second meaning to the same `\( ... \)' construct because there is no conflict in practice between the two meanings. Here is an explanation of this feature: `\DIGIT' matches the same text that matched the DIGITth occurrence of a `\( ... \)' construct. In other words, after the end of a `\( ... \)' construct. the matcher remembers the beginning and end of the text matched by that construct. Then, later on in the regular expression, you can use `\' followed by DIGIT to match that same text, whatever it may have been. The strings matching the first nine `\( ... \)' constructs appearing in a regular expression are assigned numbers 1 through 9 in the order that the open parentheses appear in the regular expression. So you can use `\1' through `\9' to refer to the text matched by the corresponding `\( ... \)' constructs. For example, `\(.*\)\1' matches any newline-free string that is composed of two identical halves. The `\(.*\)' matches the first half, which may be anything, but the `\1' that follows must match the same exact text. matches any word-constituent character. The editor syntax table determines which characters these are. *Note Syntax Tables::. matches any character that is not a word constituent. `\sCODE' matches any character whose syntax is CODE. Here CODE is a character that represents a syntax code: thus, `w' for word constituent, `-' for whitespace, `(' for open parenthesis, etc. *Note Syntax Tables::, for a list of syntax codes and the characters that stand for them. `\SCODE' matches any character whose syntax is not CODE. 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. matches the empty string, but only at the beginning of the buffer or string being matched against. matches the empty string, but only at the end of the buffer or string being matched against. matches the empty string, but only at point. (This construct is not defined when matching against a string.) matches the empty string, but only at the beginning or end of a word. Thus, `\bfoo\b' matches any occurrence of `foo' as a separate word. `\bballs?\b' matches `ball' or `balls' as a separate word. matches the empty string, but *not* at the beginning or end of a word. matches the empty string, but only at the beginning of a word. matches the empty string, but only at the end of a word. Not every string is a valid regular expression. For example, a string with unbalanced square brackets is invalid (with a few exceptions, such as `[]]'), and so is a string that ends with a single `\'. If an invalid regular expression is passed to any of the search functions, an `invalid-regexp' error is signaled. - Function: regexp-quote STRING This function returns a regular expression string that matches exactly STRING and nothing else. This allows you to request an exact string match when calling a function that wants a regular expression. (regexp-quote "^The cat$") => "\\^The cat\\$" One use of `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 `string', surrounded by whitespace: (re-search-forward (concat "\\s-" (regexp-quote string) "\\s-")) File: elisp, Node: Regexp Example, Prev: Syntax of Regexps, Up: Regular Expressions Complex Regexp Example ---------------------- Here is a complicated regexp, used by Emacs to recognize the end of a sentence together with any whitespace that follows. It is the value of the variable `sentence-end'. First, we show the regexp as a string in Lisp syntax to distinguish spaces from tab characters. The string constant begins and ends with a double-quote. `\"' stands for a double-quote as part of the string, `\\' for a backslash as part of the string, `\t' for a tab and `\n' for a newline. "[.?!][]\"')}]*\\($\\| $\\|\t\\| \\)[ \t\n]*" In contrast, if you evaluate the variable `sentence-end', you will see the following: sentence-end => "[.?!][]\"')}]*\\($\\| $\\| \\| \\)[ ]*" In this output, tab and newline appear as themselves. This regular expression contains four parts in succession and can be deciphered as follows: `[.?!]' The first part of the pattern is a character set that matches any one of three characters: period, question mark, and exclamation mark. The match must begin with one of these three characters. `[]\"')}]*' The second part of the pattern matches any closing braces and quotation marks, zero or more of them, that may follow the period, question mark or exclamation mark. The `\"' is Lisp syntax for a double-quote in a string. The `*' at the end indicates that the immediately preceding regular expression (a character set, in this case) may be repeated zero or more times. `\\($\\| $\\|\t\\| \\)' The third part of the pattern matches the whitespace that follows the end of a sentence: the end of a line, 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. `[ \t\n]*' Finally, the last part of the pattern matches any additional whitespace beyond the minimum needed to end a sentence.