@node regcomp, string @subheading Syntax @example #include #include int regcomp(regex_t *preg, const char *pattern, int cflags); @end example @subheading Description This function is part of the implementation of @sc{posix 1003.2} regular expressions (@dfn{RE}s). @code{regcomp} compiles the regular expression contained in the @var{pattern} string, subject to the flags in @var{cflags}, and places the results in the @code{regex_t} structure pointed to by @var{preg}. (The regular expression syntax, as defined by @sc{posix 1003.2}, is described below.) The parameter @var{cflags} is the bitwise OR of zero or more of the following flags: @table @code @item REG_EXTENDED Compile modern (@dfn{extended}) REs, rather than the obsolete (@dfn{basic}) REs that are the default. @item REG_BASIC This is a synonym for 0, provided as a counterpart to @code{REG_EXTENDED} to improve readability. This is an extension, compatible with but not specified by @sc{posix 1003.2}, and should be used with caution in software intended to be portable to other systems. @item REG_NOSPEC Compile with recognition of all special characters turned off. All characters are thus considered ordinary, so the RE in @var{pattern} is a literal string. This is an extension, compatible with but not specified by @sc{posix 1003.2}, and should be used with caution in software intended to be portable to other systems. @code{REG_EXTENDED} and @code{REG_NOSPEC} may not be used in the same call to @code{regcomp}. @item REG_ICASE Compile for matching that ignores upper/lower case distinctions. See the description of regular expressions below for details of case-independent matching. @item REG_NOSUB Compile for matching that need only report success or failure, not what was matched. @item REG_NEWLINE Compile for newline-sensitive matching. By default, newline is a completely ordinary character with no special meaning in either REs or strings. With this flag, @samp{[^} bracket expressions and @samp{.} never match newline, a @samp{^} anchor matches the null string after any newline in the string in addition to its normal function, and the @samp{$} anchor matches the null string before any newline in the string in addition to its normal function. @item REG_PEND The regular expression ends, not at the first NUL, but just before the character pointed to by the @code{re_endp} member of the structure pointed to by @var{preg}. The @code{re_endp} member is of type @code{const char *}. This flag permits inclusion of NULs in the RE; they are considered ordinary characters. This is an extension, compatible with but not specified by @sc{posix 1003.2}, and should be used with caution in software intended to be portable to other systems. @end table When successful, @code{regcomp} returns 0 and fills in the structure pointed to by @var{preg}. One member of that structure (other than @code{re_endp}) is publicized: @code{re_nsub}, of type @code{size_t}, contains the number of parenthesized subexpressions within the RE (except that the value of this member is undefined if the @code{REG_NOSUB} flag was used). Note that the length of the RE does matter; in particular, there is a strong speed bonus for keeping RE length under about 30 characters, with most special characters counting roughly double. @subheading Return Value If @code{regcomp} succeeds, it returns zero; if it fails, it returns a non-zero error code, which is one of these: @table @code @item REG_BADPAT invalid regular expression @item REG_ECOLLATE invalid collating element @item REG_ECTYPE invalid character class @item REG_EESCAPE @samp{\} applied to unescapable character @item REG_ESUBREG invalid backreference number (e.g., larger than the number of parenthesized subexpressions in the RE) @item REG_EBRACK brackets [ ] not balanced @item REG_EPAREN parentheses ( ) not balanced @item REG_EBRACE braces @{ @} not balanced @item REG_BADBR invalid repetition count(s) in @{ @} @item REG_ERANGE invalid character range in [ ] @item REG_ESPACE ran out of memory (an RE like, say, @samp{((((a@{1,100@})@{1,100@})@{1,100@})@{1,100@})@{1,100@}'} will eventually run almost any existing machine out of swap space) @item REG_BADRPT ?, *, or + operand invalid @item REG_EMPTY empty (sub)expression @item REG_ASSERT ``can't happen'' (you found a bug in @code{regcomp}) @item REG_INVARG invalid argument (e.g. a negative-length string) @end table @subheading Regular Expressions' Syntax Regular expressions (@dfn{RE}s), as defined in @sc{posix 1003.2}, come in two forms: modern REs (roughly those of @code{egrep}; 1003.2 calls these @emph{extended} REs) and obsolete REs (roughly those of @code{ed}; 1003.2 @emph{basic} REs). Obsolete REs mostly exist for backward compatibility in some old programs; they will be discussed at the end. 1003.2 leaves some aspects of RE syntax and semantics open; `(*)' marks decisions on these aspects that may not be fully portable to other 1003.2 implementations. A (modern) RE is one(*) or more non-empty(*) @emph{branches}, separated by @samp{|}. It matches anything that matches one of the branches. A branch is one(*) or more @emph{pieces}, concatenated. It matches a match for the first, followed by a match for the second, etc. A piece is an @emph{atom} possibly followed by a single(*) @samp{*}, @samp{+}, @samp{?}, or @emph{bound}. An atom followed by @samp{*} matches a sequence of 0 or more matches of the atom. An atom followed by @samp{+} matches a sequence of 1 or more matches of the atom. An atom followed by @samp{?} matches a sequence of 0 or 1 matches of the atom. A @emph{bound} is @samp{@{} followed by an unsigned decimal integer, possibly followed by @samp{,} possibly followed by another unsigned decimal integer, always followed by @samp{@}}. The integers must lie between 0 and @code{RE_DUP_MAX} (255(*)) inclusive, and if there are two of them, the first may not exceed the second. An atom followed by a bound containing one integer @samp{i} and no comma matches a sequence of exactly @samp{i} matches of the atom. An atom followed by a bound containing one integer @samp{i} and a comma matches a sequence of @samp{i} or more matches of the atom. An atom followed by a bound containing two integers @samp{i} and @samp{j} matches a sequence of @samp{i} through @samp{j} (inclusive) matches of the atom. An atom is a regular expression enclosed in @samp{()} (matching a match for the regular expression), an empty set of @samp{()} (matching the null string(*)), a @emph{bracket expression} (see below), @samp{.} (matching any single character), @samp{^} (matching the null string at the beginning of a line), @samp{$} (matching the null string at the end of a line), a @samp{\} followed by one of the characters @samp{^.[$()|*+?@{\\} (matching that character taken as an ordinary character), a @samp{\} followed by any other character(*) (matching that character taken as an ordinary character, as if the @samp{\} had not been present(*)), or a single character with no other significance (matching that character). A @samp{@{} followed by a character other than a digit is an ordinary character, not the beginning of a bound(*). It is illegal to end an RE with @samp{\}. A @emph{bracket expression} is a list of characters enclosed in @samp{[]}. It normally matches any single character from the list (but see below). If the list begins with @samp{^}, it matches any single character (but see below) @emph{not} from the rest of the list. If two characters in the list are separated by @samp{-}, this is shorthand for the full @emph{range} of characters between those two (inclusive) in the collating sequence, e.g. @samp{[0-9]} in ASCII matches any decimal digit. It is illegal(*) for two ranges to share an endpoint, e.g. @samp{a-c-e}. Ranges are very collating-sequence-dependent, and portable programs should avoid relying on them. To include a literal @samp{]} in the list, make it the first character (following a possible @samp{^}). To include a literal @samp{-}, make it the first or last character, or the second endpoint of a range. To use a literal @samp{-} as the first endpoint of a range, enclose it in @samp{[.} and @samp{.]} to make it a collating element (see below). With the exception of these and some combinations using @samp{[} (see next paragraphs), all other special characters, including @samp{\}, lose their special significance within a bracket expression. Within a bracket expression, a collating element (a character, a multi-character sequence that collates as if it were a single character, or a collating-sequence name for either) enclosed in @samp{[.} and @samp{.]} stands for the sequence of characters of that collating element. The sequence is a single element of the bracket expression's list. A bracket expression containing a multi-character collating element can thus match more than one character, e.g. if the collating sequence includes a @samp{ch} collating element, then the RE @samp{[[.ch.]]*c} matches the first five characters of ``chchcc''. Within a bracket expression, a collating element enclosed in @samp{[=} and @samp{=]} is an equivalence class, standing for the sequences of characters of all collating elements equivalent to that one, including itself. (If there are no other equivalent collating elements, the treatment is as if the enclosing delimiters were @samp{[.} and @samp{.]}.) For example, if @samp{o} and @samp{^} are the members of an equivalence class, then @samp{[[=o=]]}, @samp{[[=^=]]}, and @samp{[o^]} are all synonymous. An equivalence class may @emph{not} be an endpoint of a range. Within a bracket expression, the name of a @emph{character class} enclosed in @samp{[:} and @samp{:]} stands for the list of all characters belonging to that class. Standard character class names are: @example alnum digit punct alpha graph space blank lower upper cntrl print xdigit @end example These stand for the character classes defined by @code{isalnum} (@pxref{isalnum}), @code{isdigit} (@pxref{isdigit}), @code{ispunct} (@pxref{ispunct}), @code{isalpha} (@pxref{isalpha}), @code{isgraph} (@pxref{isgraph}), @code{isspace} (@pxref{isspace}) (@code{blank} is the same as @code{space}), @code{islower} (@pxref{islower}), @code{isupper} (@pxref{isupper}), @code{iscntrl} (@pxref{iscntrl}), @code{isprint} (@pxref{isprint}), and @code{isxdigit} (@pxref{isxdigit}), respectively. A locale may provide others. A character class may not be used as an endpoint of a range. There are two special cases(*) of bracket expressions: the bracket expressions @samp{[[:<:]]} and @samp{[[:>:]]} match the null string at the beginning and end of a word respectively. A word is defined as a sequence of word characters which is neither preceded nor followed by word characters. A word character is an @code{alnum} character (as defined by @code{isalnum} library function) or an underscore. This is an extension, compatible with but not specified by @sc{posix 1003.2}, and should be used with caution in software intended to be portable to other systems. In the event that an RE could match more than one substring of a given string, the RE matches the one starting earliest in the string. If the RE could match more than one substring starting at that point, it matches the longest. Subexpressions also match the longest possible substrings, subject to the constraint that the whole match be as long as possible, with subexpressions starting earlier in the RE taking priority over ones starting later. Note that higher-level subexpressions thus take priority over their lower-level component subexpressions. Match lengths are measured in characters, not collating elements. A null string is considered longer than no match at all. For example, @samp{bb*} matches the three middle characters of ``abbbc'', @samp{(wee|week)(knights|nights)} matches all ten characters of ``weeknights'', when @samp{(.*).*} is matched against ``abc'' the parenthesized subexpression matches all three characters, and when @samp{(a*)*} is matched against ``bc'' both the whole RE and the parenthesized subexpression match the null string. If case-independent matching is specified, the effect is much as if all case distinctions had vanished from the alphabet. When an alphabetic that exists in multiple cases appears as an ordinary character outside a bracket expression, it is effectively transformed into a bracket expression containing both cases, e.g. @samp{x} becomes @samp{[xX]}. When it appears inside a bracket expression, all case counterparts of it are added to the bracket expression, so that (e.g.) @samp{[x]} becomes @samp{[xX]} and @samp{[^x]} becomes @samp{[^xX]}. No particular limit is imposed on the length of REs(*). Programs intended to be portable should not employ REs longer than 256 bytes, as an implementation can refuse to accept such REs and remain @sc{posix}-compliant. Obsolete (@emph{basic}) regular expressions differ in several respects. @samp{|}, @samp{+}, and @samp{?} are ordinary characters and there is no equivalent for their functionality. The delimiters for bounds are @samp{\@{} and @samp{\@}}, with @samp{@{} and @samp{@}} by themselves ordinary characters. The parentheses for nested subexpressions are @samp{\(} and @samp{\)}, with @samp{(} and @samp{)} by themselves ordinary characters. @samp{^} is an ordinary character except at the beginning of the RE or(*) the beginning of a parenthesized subexpression, @samp{$} is an ordinary character except at the end of the RE or(*) the end of a parenthesized subexpression, and @samp{*} is an ordinary character if it appears at the beginning of the RE or the beginning of a parenthesized subexpression (after a possible leading @samp{^}). Finally, there is one new type of atom, a @emph{back reference}: @samp{\} followed by a non-zero decimal digit @emph{d} matches the same sequence of characters matched by the @emph{d}th parenthesized subexpression (numbering subexpressions by the positions of their opening parentheses, left to right), so that (e.g.) @samp{\([bc]\)\1} matches ``bb'' or ``cc'' but not ``bc''. @subheading History This implementation of the @sc{posix} regexp functionality was written by @email{henry@@zoo.toronto.edu, Henry Spencer}. @subheading Bugs The locale is always assumed to be the default one of 1003.2, and only the collating elements etc. of that locale are available. @code{regcomp} implements bounded repetitions by macro expansion, which is costly in time and space if counts are large or bounded repetitions are nested. An RE like, say, @samp{((((a@{1,100@})@{1,100@})@{1,100@})@{1,100@})@{1,100@}}, will (eventually) run almost any existing machine out of swap space. There are suspected problems with response to obscure error conditions. Notably, certain kinds of internal overflow, produced only by truly enormous REs or by multiply nested bounded repetitions, are probably not handled well. Due to a mistake in 1003.2, things like @samp{a)b} are legal REs because @samp{)} is a special character only in the presence of a previous unmatched @samp{(}. This can't be fixed until the spec is fixed. The standard's definition of back references is vague. For example, does @samp{a\e(\e(b\e)*\e2\e)*d} match ``abbbd''? Until the standard is clarified, behavior in such cases should not be relied on. @subheading Portability @portability !ansi, posix @c ------------------------------------------------------------------------- @node regexec, string @subheading Syntax @example #include #include int regexec(const regex_t *preg, const char *string, size_t nmatch, regmatch_t pmatch[], int eflags); @end example @subheading Description @code{regexec} matches the compiled RE pointed to by @var{preg} against the @var{string}, subject to the flags in @var{eflags}, and reports results using @var{nmatch}, @var{pmatch}, and the returned value. The RE must have been compiled by a previous invocation of @code{regcomp} (@pxref{regcomp}). The compiled form is not altered during execution of @code{regexec}, so a single compiled RE can be used simultaneously by multiple threads. By default, the NUL-terminated string pointed to by @var{string} is considered to be the text of an entire line, with the NUL indicating the end of the line. (That is, any other end-of-line marker is considered to have been removed and replaced by the NUL.) The @var{eflags} argument is the bitwise OR of zero or more of the following flags: @table @code @item REG_NOTBOL The first character of the string is not the beginning of a line, so the @samp{^} anchor should not match before it. This does not affect the behavior of newlines under @code{REG_NEWLINE} (REG_NEWLINE, @pxref{regcomp}). @item REG_NOTEOL The NUL terminating the string does not end a line, so the @samp{$} anchor should not match before it. This does not affect the behavior of newlines under @code{REG_NEWLINE} (REG_NEWLINE, @pxref{regcomp}). @item REG_STARTEND The string is considered to start at @w{@code{@var{string} + @var{pmatch}[0].rm_so}} and to have a terminating @code{NUL} located at @w{@code{@var{string} + @var{pmatch}[0].rm_eo}} (there need not actually be a @code{NUL} at that location), regardless of the value of @var{nmatch}. See below for the definition of @var{pmatch} and @var{nmatch}. This is an extension, compatible with but not specified by @sc{posix 1003.2}, and should be used with caution in software intended to be portable to other systems. Note that a non-zero @code{rm_so} does not imply @code{REG_NOTBOL}; @code{REG_STARTEND} affects only the location of the string, not how it is matched. @item REG_TRACE trace execution (printed to stdout) @item REG_LARGE force large representation @item REG_BACKR force use of backref code @end table Regular Expressions' Syntax, @xref{regcomp}, for a discussion of what is matched in situations where an RE or a portion thereof could match any of several substrings of @var{string}. If @code{REG_NOSUB} was specified in the compilation of the RE (REG_NOSUB, @pxref{regcomp}), or if @var{nmatch} is 0, @code{regexec} ignores the @var{pmatch} argument (but see below for the case where @code{REG_STARTEND} is specified). Otherwise, @var{pmatch} should point to an array of @var{nmatch} structures of type @code{regmatch_t}. Such a structure has at least the members @code{rm_so} and @code{rm_eo}, both of type @code{regoff_t} (a signed arithmetic type at least as large as an @code{off_t} and a @code{ssize_t}, containing respectively the offset of the first character of a substring and the offset of the first character after the end of the substring. Offsets are measured from the beginning of the @var{string} argument given to @code{regexec}. An empty substring is denoted by equal offsets, both indicating the character following the empty substring. When @code{regexec} returns, the 0th member of the @var{pmatch} array is filled in to indicate what substring of @var{string} was matched by the entire RE. Remaining members report what substring was matched by parenthesized subexpressions within the RE; member @code{i} reports subexpression @code{i}, with subexpressions counted (starting at 1) by the order of their opening parentheses in the RE, left to right. Unused entries in the array---corresponding either to subexpressions that did not participate in the match at all, or to subexpressions that do not exist in the RE (that is, @code{@w{i > preg->re_nsub}}---have both @code{rm_so} and @code{rm_eo} set to @code{-1}. If a subexpression participated in the match several times, the reported substring is the last one it matched. (Note, as an example in particular, that when the RE @samp{(b*)+} matches ``bbb'', the parenthesized subexpression matches the three @samp{b}s and then an infinite number of empty strings following the last @samp{b}, so the reported substring is one of the empties.) If @code{REG_STARTEND} is specified in @var{eflags}, @var{pmatch} must point to at least one @code{regmatch_t} variable (even if @var{nmatch} is 0 or @code{REG_NOSUB} was specified in the compilation of the RE, REG_NOSUB, @pxref{regcomp}), to hold the input offsets for @code{REG_STARTEND}. Use for output is still entirely controlled by @var{nmatch}; if @var{nmatch} is 0 or @code{REG_NOSUB} was specified, the value of @code{pmatch[0]} will not be changed by a successful @code{regexec}. @subheading Return Value Normally, @code{regexec} returns 0 for success and the non-zero code @code{REG_NOMATCH} for failure. Other non-zero error codes may be returned in exceptional situations. The list of possible error return values is below: @table @code @item REG_ESPACE ran out of memory @item REG_BADPAT the passed argument @var{preg} doesn't point to an RE compiled by @code{regcomp} @item REG_INVARG invalid argument(s) (e.g., @w{@code{@var{string} + @var{pmatch}[0].rm_eo}} is less than @w{@code{@var{string} + @var{pmatch}[0].rm_so}}) @end table @subheading History This implementation of the @sc{posix} regexp functionality was written by @email{henry@@zoo.toronto.edu, Henry Spencer}. @subheading Bugs @code{regexec} performance is poor. @var{nmatch} exceeding 0 is expensive; @var{nmatch} exceeding 1 is worse. @code{regexec} is largely insensitive to RE complexity @emph{except} that back references are massively expensive. RE length does matter; in particular, there is a strong speed bonus for keeping RE length under about 30 characters, with most special characters counting roughly double. The implementation of word-boundary matching is a bit of a kludge, and bugs may lurk in combinations of word-boundary matching and anchoring. @subheading Portability @portability !ansi, posix @c ---------------------------------------------------------------------- @node regerror, string @subheading Syntax @example #include #include size_t regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size); @end example @subheading Description @code{regerror} maps a non-zero value of @var{errcode} from either @code{regcomp} (Return Value, @pxref{regcomp}) or @code{regexec} (Return Value, @pxref{regexec}) to a human-readable, printable message. If @var{preg} is non-@code{NULL}, the error code should have arisen from use of the variable of the type @code{regex_t} pointed to by @var{preg}, and if the error code came from @code{regcomp}, it should have been the result from the most recent @code{regcomp} using that @code{regex_t} variable. (@code{regerror} may be able to supply a more detailed message using information from the @code{regex_t} than from @var{errcode} alone.) @code{regerror} places the @code{NUL}-terminated message into the buffer pointed to by @var{errbuf}, limiting the length (including the @code{NUL}) to at most @var{errbuf_size} bytes. If the whole message won't fit, as much of it as will fit before the terminating @code{NUL} is supplied. In any case, the returned value is the size of buffer needed to hold the whole message (including terminating @code{NUL}). If @var{errbuf_size} is 0, @var{errbuf} is ignored but the return value is still correct. If the @var{errcode} given to @code{regerror} is first ORed with @code{REG_ITOA}, the ``message'' that results is the printable name of the error code, e.g. ``REG_NOMATCH'', rather than an explanation thereof. If @var{errcode} is @code{REG_ATOI}, then @var{preg} shall be non-NULL and the @code{re_endp} member of the structure it points to must point to the printable name of an error code (e.g. ``REG_ECOLLATE''); in this case, the result in @var{errbuf} is the decimal representation of the numeric value of the error code (0 if the name is not recognized). @code{REG_ITOA} and @code{REG_ATOI} are intended primarily as debugging facilities; they are extensions, compatible with but not specified by @sc{posix 1003.2}, and should be used with caution in software intended to be portable to other systems. Be warned also that they are considered experimental and changes are possible. @subheading Return Value The size of buffer needed to hold the message (including terminating @code{NUL}) is always returned, even if @var{errbuf_size} is zero. @subheading Portability @portability !ansi, posix @c -------------------------------------------------------------------------- @node regfree, string @subheading Syntax @example #include #include void regfree(regex_t *preg); @end example @subheading Description @code{regfree} frees any dynamically-allocated storage associated with the compiled RE pointed to by @var{preg}. The remaining @code{regex_t} is no longer a valid compiled RE and the effect of supplying it to @code{regexec} or @code{regerror} is undefined. @subheading Portability @portability !ansi, posix