From 763ed733711a5bb66d39de4d0ed4bcaae77e2461 Mon Sep 17 00:00:00 2001 From: Ed Schouten Date: Sun, 14 Mar 2010 10:18:58 +0000 Subject: [PATCH] Trim down libcompat by removing . Erwin ran an exp-run with libcompat and removed. It turns out the regexp library is almost entirely unused. In fact, it looks like it is sometimes used by accident. Because these function names clash with libc's , some application use both and libcompat, which means they link against the wrong regex library. This commit removes the regexp library and reimplements re_comp() and re_exec() using . It seems the grammar of the regular expressions accepted by these functions is similar to POSIX EREs. After this commit, 1 low-profile port will be broken, but the maintainer already has a patch for it sitting in his mailbox. --- ObsoleteFiles.inc | 4 + include/Makefile | 2 +- include/regexp.h | 70 -- lib/libcompat/4.3/re_comp.3 | 13 +- lib/libcompat/4.3/{regex.c => re_comp.c} | 70 +- lib/libcompat/Makefile | 24 +- lib/libcompat/regexp/COPYRIGHT | 22 - lib/libcompat/regexp/README | 84 -- lib/libcompat/regexp/regerror.c | 18 - lib/libcompat/regexp/regexp.3 | 319 ------ lib/libcompat/regexp/regexp.c | 1337 ---------------------- lib/libcompat/regexp/regmagic.h | 5 - lib/libcompat/regexp/regsub.c | 85 -- 13 files changed, 48 insertions(+), 2005 deletions(-) delete mode 100644 include/regexp.h rename lib/libcompat/4.3/{regex.c => re_comp.c} (78%) delete mode 100644 lib/libcompat/regexp/COPYRIGHT delete mode 100644 lib/libcompat/regexp/README delete mode 100644 lib/libcompat/regexp/regerror.c delete mode 100644 lib/libcompat/regexp/regexp.3 delete mode 100644 lib/libcompat/regexp/regexp.c delete mode 100644 lib/libcompat/regexp/regmagic.h delete mode 100644 lib/libcompat/regexp/regsub.c diff --git a/ObsoleteFiles.inc b/ObsoleteFiles.inc index e576e9e22bde..fe0f4f21a232 100644 --- a/ObsoleteFiles.inc +++ b/ObsoleteFiles.inc @@ -14,6 +14,10 @@ # The file is partitioned: OLD_FILES first, then OLD_LIBS and OLD_DIRS last. # +# 20100314: removal of regexp.h +OLD_FILES+=usr/include/regexp.h +OLD_FILES+=usr/share/man/man3/regexp.3.gz +OLD_FILES+=usr/share/man/man3/regsub.3.gz # 20100303: actual removal of utmp.h OLD_FILES+=usr/include/utmp.h # 20100227: [ia64] removed and diff --git a/include/Makefile b/include/Makefile index 509bd47865f2..96690adb5155 100644 --- a/include/Makefile +++ b/include/Makefile @@ -17,7 +17,7 @@ INCS= a.out.h ar.h assert.h bitstring.h complex.h cpio.h _ctype.h ctype.h \ ndbm.h netconfig.h \ netdb.h nl_types.h nlist.h nss.h nsswitch.h paths.h \ printf.h proc_service.h pthread.h \ - pthread_np.h pwd.h ranlib.h readpassphrase.h regex.h regexp.h \ + pthread_np.h pwd.h ranlib.h readpassphrase.h regex.h \ res_update.h resolv.h runetype.h search.h semaphore.h setjmp.h \ signal.h spawn.h stab.h \ stdbool.h stddef.h stdio.h stdlib.h string.h stringlist.h \ diff --git a/include/regexp.h b/include/regexp.h deleted file mode 100644 index 39cdf3afc096..000000000000 --- a/include/regexp.h +++ /dev/null @@ -1,70 +0,0 @@ -/* - * Copyright (c) 1986 by University of Toronto. - * Copyright (c) 1989, 1993 - * The Regents of the University of California. All rights reserved. - * - * This code is derived from software contributed to Berkeley - * by Henry Spencer. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of the University nor the names of its contributors - * may be used to endorse or promote products derived from this software - * without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - * - * @(#)regexp.h 8.1 (Berkeley) 6/2/93 - * $FreeBSD$ - */ - -#ifndef _REGEXP_H_ -#define _REGEXP_H_ - -#ifdef __GNUC__ -#warning "this file includes which is deprecated, use instead" -#endif - -/* - * Definitions etc. for regexp(3) routines. - * - * Caveat: this is V8 regexp(3) [actually, a reimplementation thereof], - * not the System V one. - */ -#define NSUBEXP 10 -typedef struct regexp { - char *startp[NSUBEXP]; - char *endp[NSUBEXP]; - char regstart; /* Internal use only. */ - char reganch; /* Internal use only. */ - char *regmust; /* Internal use only. */ - int regmlen; /* Internal use only. */ - char program[1]; /* Unwarranted chumminess with compiler. */ -} regexp; - -#include - -__BEGIN_DECLS -regexp *regcomp(const char *); -int regexec(const regexp *, const char *); -void regsub(const regexp *, const char *, char *); -void regerror(const char *); -__END_DECLS - -#endif /* !_REGEXP_H_ */ diff --git a/lib/libcompat/4.3/re_comp.3 b/lib/libcompat/4.3/re_comp.3 index e3b595a4457a..4d970facb091 100644 --- a/lib/libcompat/4.3/re_comp.3 +++ b/lib/libcompat/4.3/re_comp.3 @@ -100,15 +100,10 @@ returns \-1 for an internal error. The .Fn re_comp function -returns one of the following strings if an error occurs: -.Bd -unfilled -offset indent -No previous regular expression, -Regular expression too long, -unmatched \e(, -missing ], -too many \e(\e) pairs, -unmatched \e). -.Ed +returns +.Dq no previous regular expression +or one of the strings generated by +.Xr regerror 3 . .Sh SEE ALSO .Xr ed 1 , .Xr egrep 1 , diff --git a/lib/libcompat/4.3/regex.c b/lib/libcompat/4.3/re_comp.c similarity index 78% rename from lib/libcompat/4.3/regex.c rename to lib/libcompat/4.3/re_comp.c index 470cab52e779..dbe57b129973 100644 --- a/lib/libcompat/4.3/regex.c +++ b/lib/libcompat/4.3/re_comp.c @@ -44,49 +44,49 @@ __FBSDID("$FreeBSD$"); static char sccsid[] = "@(#)regex.c 5.1 (Berkeley) 3/29/92"; #endif /* LIBC_SCCS and not lint */ -#include +#include #include -#include -#include -#include -#include +#include -static regexp *re_regexp; -static int re_goterr; -static char *re_errstr; +static regex_t re_regexp; +static int re_gotexp; +static char re_errstr[100]; char * -re_comp(char *s) -{ - if (s == NULL || *s == '\0') { - if (re_regexp == NULL) - return "no previous regular expression"; - return (NULL); - } - if (re_regexp) - free(re_regexp); - if (re_errstr) - free(re_errstr); - re_goterr = 0; - re_regexp = regcomp(s); - return (re_goterr ? re_errstr : NULL); -} - -int -re_exec(char *s) +re_comp(const char *s) { int rc; - re_goterr = 0; - rc = regexec(re_regexp, s); - return (re_goterr ? -1 : rc); + if (s == NULL || *s == '\0') { + if (!re_gotexp) + return __DECONST(char *, + "no previous regular expression"); + return (NULL); + } + + if (re_gotexp) { + regfree(&re_regexp); + re_gotexp = 0; + } + + rc = regcomp(&re_regexp, s, REG_EXTENDED); + if (rc == 0) { + re_gotexp = 1; + return (NULL); + } + + regerror(rc, &re_regexp, re_errstr, sizeof(re_errstr)); + re_errstr[sizeof(re_errstr) - 1] = '\0'; + return (re_errstr); } -void -regerror(const char *s) +int +re_exec(const char *s) { - re_goterr = 1; - if (re_errstr) - free(re_errstr); - re_errstr = strdup(s); + int rc; + + if (!re_gotexp) + return (-1); + rc = regexec(&re_regexp, s, 0, NULL, 0); + return (rc == 0 ? 1 : 0); } diff --git a/lib/libcompat/Makefile b/lib/libcompat/Makefile index 3a9d666c5b1d..35ced5f13aeb 100644 --- a/lib/libcompat/Makefile +++ b/lib/libcompat/Makefile @@ -1,19 +1,15 @@ # @(#)Makefile 8.1 (Berkeley) 6/4/93 # $FreeBSD$ -LIB=compat +LIB= compat CFLAGS+=-DLIBC_SCCS -DSYSLIBC_SCCS -I${.CURDIR}/../libc/locale NO_PIC= WARNS?= 0 -.PATH: ${.CURDIR}/4.1/${MACHINE_ARCH} ${.CURDIR}/4.1 \ - ${.CURDIR}/4.3/${MACHINE_ARCH} ${.CURDIR}/4.3 \ - ${.CURDIR}/4.4/${MACHINE_ARCH} ${.CURDIR}/4.4 \ - ${.CURDIR}/regexp +.PATH: ${.CURDIR}/4.1 ${.CURDIR}/4.3 ${.CURDIR}/4.4 # compat 4.1 sources -# XXX MISSING: tell.c SRCS+= ascftime.c cftime.c ftime.c getpw.c MAN+= 4.1/ftime.3 4.1/getpw.3 @@ -22,27 +18,15 @@ MAN+= 4.1/cftime.3 MLINKS+=cftime.3 ascftime.3 # compat 4.3 sources -# XXX MISSING: ecvt.c gcvt.c sibuf.c sobuf.c strout.c -SRCS+= cfree.c regex.c rexec.c +SRCS+= cfree.c re_comp.c rexec.c -# XXX MISSING: ecvt.0 MAN+= 4.3/cfree.3 4.3/re_comp.3 4.3/rexec.3 -# XXX MISSING: ecvt.3, so can't MLINK -#MLINKS+=ecvt.3 fcvt.3 ecvt.3 gcvt.3 MLINKS+=re_comp.3 re_exec.3 # compat 4.4 sources SRCS+= cuserid.c + MAN+= 4.4/cuserid.3 -# regexp sources -SRCS+= regerror.c regexp.c regsub.c - -MAN+= regexp/regexp.3 - -# XXX name clash with libc -# MLINKS+=regexp.3 regcomp.3 regexp.3 regexec.3 regexp.3 regerror.3 -MLINKS+=regexp.3 regsub.3 - .include diff --git a/lib/libcompat/regexp/COPYRIGHT b/lib/libcompat/regexp/COPYRIGHT deleted file mode 100644 index 48b3f4339171..000000000000 --- a/lib/libcompat/regexp/COPYRIGHT +++ /dev/null @@ -1,22 +0,0 @@ -This entire subtree is copyright the University of Toronto. -The following copyright notice applies to all files found here. None of -these files contain AT&T proprietary source code. -_____________________________________________________________________________ - - Copyright (c) 1986 by University of Toronto. - Written by Henry Spencer. Not derived from licensed software. - - Permission is granted to anyone to use this software for any - purpose on any computer system, and to redistribute it freely, - subject to the following restrictions: - - 1. The author is not responsible for the consequences of use of - this software, no matter how awful, even if they arise - from defects in it. - - 2. The origin of this software must not be misrepresented, either - by explicit claim or by omission. - - 3. Altered versions must be plainly marked as such, and must not - be misrepresented as being the original software. - diff --git a/lib/libcompat/regexp/README b/lib/libcompat/regexp/README deleted file mode 100644 index 37d6f51c7119..000000000000 --- a/lib/libcompat/regexp/README +++ /dev/null @@ -1,84 +0,0 @@ -This is a nearly-public-domain reimplementation of the V8 regexp(3) package. -It gives C programs the ability to use egrep-style regular expressions, and -does it in a much cleaner fashion than the analogous routines in SysV. - - Copyright (c) 1986 by University of Toronto. - Written by Henry Spencer. Not derived from licensed software. - - Permission is granted to anyone to use this software for any - purpose on any computer system, and to redistribute it freely, - subject to the following restrictions: - - 1. The author is not responsible for the consequences of use of - this software, no matter how awful, even if they arise - from defects in it. - - 2. The origin of this software must not be misrepresented, either - by explicit claim or by omission. - - 3. Altered versions must be plainly marked as such, and must not - be misrepresented as being the original software. - -Barring a couple of small items in the BUGS list, this implementation is -believed 100% compatible with V8. It should even be binary-compatible, -sort of, since the only fields in a "struct regexp" that other people have -any business touching are declared in exactly the same way at the same -location in the struct (the beginning). - -This implementation is *NOT* AT&T/Bell code, and is not derived from licensed -software. Even though U of T is a V8 licensee. This software is based on -a V8 manual page sent to me by Dennis Ritchie (the manual page enclosed -here is a complete rewrite and hence is not covered by AT&T copyright). -The software was nearly complete at the time of arrival of our V8 tape. -I haven't even looked at V8 yet, although a friend elsewhere at U of T has -been kind enough to run a few test programs using the V8 regexp(3) to resolve -a few fine points. I admit to some familiarity with regular-expression -implementations of the past, but the only one that this code traces any -ancestry to is the one published in Kernighan & Plauger (from which this -one draws ideas but not code). - -Simplistically: put this stuff into a source directory, copy regexp.h into -/usr/include, inspect Makefile for compilation options that need changing -to suit your local environment, and then do "make r". This compiles the -regexp(3) functions, compiles a test program, and runs a large set of -regression tests. If there are no complaints, then put regexp.o, regsub.o, -and regerror.o into your C library, and regexp.3 into your manual-pages -directory. - -Note that if you don't put regexp.h into /usr/include *before* compiling, -you'll have to add "-I." to CFLAGS before compiling. - -The files are: - -Makefile instructions to make everything -regexp.3 manual page -regexp.h header file, for /usr/include -regexp.c source for regcomp() and regexec() -regsub.c source for regsub() -regerror.c source for default regerror() -regmagic.h internal header file -try.c source for test program -timer.c source for timing program -tests test list for try and timer - -This implementation uses nondeterministic automata rather than the -deterministic ones found in some other implementations, which makes it -simpler, smaller, and faster at compiling regular expressions, but slower -at executing them. In theory, anyway. This implementation does employ -some special-case optimizations to make the simpler cases (which do make -up the bulk of regular expressions actually used) run quickly. In general, -if you want blazing speed you're in the wrong place. Replacing the insides -of egrep with this stuff is probably a mistake; if you want your own egrep -you're going to have to do a lot more work. But if you want to use regular -expressions a little bit in something else, you're in luck. Note that many -existing text editors use nondeterministic regular-expression implementations, -so you're in good company. - -This stuff should be pretty portable, given appropriate option settings. -If your chars have less than 8 bits, you're going to have to change the -internal representation of the automaton, although knowledge of the details -of this is fairly localized. There are no "reserved" char values except for -NUL, and no special significance is attached to the top bit of chars. -The string(3) functions are used a fair bit, on the grounds that they are -probably faster than coding the operations in line. Some attempts at code -tuning have been made, but this is invariably a bit machine-specific. diff --git a/lib/libcompat/regexp/regerror.c b/lib/libcompat/regexp/regerror.c deleted file mode 100644 index 6d0077d63429..000000000000 --- a/lib/libcompat/regexp/regerror.c +++ /dev/null @@ -1,18 +0,0 @@ -#include -#include - -void -regerror(s) -const char *s; -{ -#ifdef ERRAVAIL - error("regexp: %s", s); -#else -/* - fprintf(stderr, "regexp(3): %s\n", s); - exit(1); -*/ - return; /* let std. egrep handle errors */ -#endif - /* NOTREACHED */ -} diff --git a/lib/libcompat/regexp/regexp.3 b/lib/libcompat/regexp/regexp.3 deleted file mode 100644 index 14b3125a572d..000000000000 --- a/lib/libcompat/regexp/regexp.3 +++ /dev/null @@ -1,319 +0,0 @@ -.\" Copyright (c) 1991, 1993 -.\" The Regents of the University of California. All rights reserved. -.\" -.\" Redistribution and use in source and binary forms, with or without -.\" modification, are permitted provided that the following conditions -.\" are met: -.\" 1. Redistributions of source code must retain the above copyright -.\" notice, this list of conditions and the following disclaimer. -.\" 2. Redistributions in binary form must reproduce the above copyright -.\" notice, this list of conditions and the following disclaimer in the -.\" documentation and/or other materials provided with the distribution. -.\" 4. Neither the name of the University nor the names of its contributors -.\" may be used to endorse or promote products derived from this software -.\" without specific prior written permission. -.\" -.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND -.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE -.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE -.\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE -.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL -.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS -.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) -.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT -.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY -.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF -.\" SUCH DAMAGE. -.\" -.\" @(#)regexp.3 8.1 (Berkeley) 6/4/93 -.\" $FreeBSD$ -.\" -.Dd June 4, 1993 -.Dt REGEXP 3 -.Os -.Sh NAME -.Nm regcomp , -.Nm regexec , -.Nm regsub , -.Nm regerror -.Nd regular expression handlers -.Sh LIBRARY -.Lb libcompat -.Sh SYNOPSIS -.In regexp.h -.Ft regexp * -.Fn regcomp "const char *exp" -.Ft int -.Fn regexec "const regexp *prog" "const char *string" -.Ft void -.Fn regsub "const regexp *prog" "const char *source" "char *dest" -.Sh DESCRIPTION -.Bf Sy -This interface is made obsolete by -.Xr regex 3 . -.Ef -.Pp -The -.Fn regcomp , -.Fn regexec , -.Fn regsub , -and -.Fn regerror -functions -implement -.Xr egrep 1 Ns -style -regular expressions and supporting facilities. -.Pp -The -.Fn regcomp -function -compiles a regular expression into a structure of type -.Vt regexp , -and returns a pointer to it. -The space has been allocated using -.Xr malloc 3 -and may be released by -.Xr free 3 . -.Pp -The -.Fn regexec -function -matches a -.Dv NUL Ns -terminated -.Fa string -against the compiled regular expression -in -.Fa prog . -It returns 1 for success and 0 for failure, and adjusts the contents of -.Fa prog Ns 's -.Em startp -and -.Em endp -(see below) accordingly. -.Pp -The members of a -.Vt regexp -structure include at least the following (not necessarily in order): -.Bd -literal -offset indent -char *startp[NSUBEXP]; -char *endp[NSUBEXP]; -.Ed -.Pp -where -.Dv NSUBEXP -is defined (as 10) in the header file. -Once a successful -.Fn regexec -has been done using the -.Fn regexp , -each -.Em startp Ns - Em endp -pair describes one substring -within the -.Fa string , -with the -.Em startp -pointing to the first character of the substring and -the -.Em endp -pointing to the first character following the substring. -The 0th substring is the substring of -.Fa string -that matched the whole -regular expression. -The others are those substrings that matched parenthesized expressions -within the regular expression, with parenthesized expressions numbered -in left-to-right order of their opening parentheses. -.Pp -The -.Fn regsub -function -copies -.Fa source -to -.Fa dest , -making substitutions according to the -most recent -.Fn regexec -performed using -.Fa prog . -Each instance of `&' in -.Fa source -is replaced by the substring -indicated by -.Em startp Ns Bq -and -.Em endp Ns Bq . -Each instance of -.Sq \e Ns Em n , -where -.Em n -is a digit, is replaced by -the substring indicated by -.Em startp Ns Bq Em n -and -.Em endp Ns Bq Em n . -To get a literal `&' or -.Sq \e Ns Em n -into -.Fa dest , -prefix it with `\e'; -to get a literal `\e' preceding `&' or -.Sq \e Ns Em n , -prefix it with -another `\e'. -.Pp -The -.Fn regerror -function -is called whenever an error is detected in -.Fn regcomp , -.Fn regexec , -or -.Fn regsub . -The default -.Fn regerror -writes the string -.Fa msg , -with a suitable indicator of origin, -on the standard -error output -and invokes -.Xr exit 3 . -The -.Fn regerror -function -can be replaced by the user if other actions are desirable. -.Sh REGULAR EXPRESSION SYNTAX -A regular expression is zero or more -.Em branches , -separated by `|'. -It matches anything that matches one of the branches. -.Pp -A branch is zero or more -.Em pieces , -concatenated. -It matches a match for the first, followed by a match for the second, etc. -.Pp -A piece is an -.Em atom -possibly followed by `*', `+', or `?'. -An atom followed by `*' matches a sequence of 0 or more matches of the atom. -An atom followed by `+' matches a sequence of 1 or more matches of the atom. -An atom followed by `?' matches a match of the atom, or the null string. -.Pp -An atom is a regular expression in parentheses (matching a match for the -regular expression), a -.Em range -(see below), `.' -(matching any single character), `^' (matching the null string at the -beginning of the input string), `$' (matching the null string at the -end of the input string), a `\e' followed by a single character (matching -that character), or a single character with no other significance -(matching that character). -.Pp -A -.Em range -is a sequence of characters enclosed in `[]'. -It normally matches any single character from the sequence. -If the sequence begins with `^', -it matches any single character -.Em not -from the rest of the sequence. -If two characters in the sequence are separated by `\-', this is shorthand -for the full list of -.Tn ASCII -characters between them -(e.g.\& `[0-9]' matches any decimal digit). -To include a literal `]' in the sequence, make it the first character -(following a possible `^'). -To include a literal `\-', make it the first or last character. -.Sh AMBIGUITY -If a regular expression could match two different parts of the input string, -it will match the one which begins earliest. -If both begin in the same place but match different lengths, or match -the same length in different ways, life gets messier, as follows. -.Pp -In general, the possibilities in a list of branches are considered in -left-to-right order, the possibilities for `*', `+', and `?' are -considered longest-first, nested constructs are considered from the -outermost in, and concatenated constructs are considered leftmost-first. -The match that will be chosen is the one that uses the earliest -possibility in the first choice that has to be made. -If there is more than one choice, the next will be made in the same manner -(earliest possibility) subject to the decision on the first choice. -And so forth. -.Pp -For example, -.Sq Li (ab|a)b*c -could match -`abc' in one of two ways. -The first choice is between `ab' and `a'; since `ab' is earlier, and does -lead to a successful overall match, it is chosen. -Since the `b' is already spoken for, -the `b*' must match its last possibility\(emthe empty string\(emsince -it must respect the earlier choice. -.Pp -In the particular case where no `|'s are present and there is only one -`*', `+', or `?', the net effect is that the longest possible -match will be chosen. -So -.Sq Li ab* , -presented with `xabbbby', will match `abbbb'. -Note that if -.Sq Li ab* , -is tried against `xabyabbbz', it -will match `ab' just after `x', due to the begins-earliest rule. -(In effect, the decision on where to start the match is the first choice -to be made, hence subsequent choices must respect it even if this leads them -to less-preferred alternatives.) -.Sh RETURN VALUES -The -.Fn regcomp -function -returns -.Dv NULL -for a failure -.Pf ( Fn regerror -permitting), -where failures are syntax errors, exceeding implementation limits, -or applying `+' or `*' to a possibly-null operand. -.Sh SEE ALSO -.Xr ed 1 , -.Xr egrep 1 , -.Xr ex 1 , -.Xr expr 1 , -.Xr fgrep 1 , -.Xr grep 1 , -.Xr regex 3 -.Sh HISTORY -Both code and manual page for -.Fn regcomp , -.Fn regexec , -.Fn regsub , -and -.Fn regerror -were written at the University of Toronto -and appeared in -.Bx 4.3 tahoe . -They are intended to be compatible with the Bell V8 -.Xr regexp 3 , -but are not derived from Bell code. -.Sh BUGS -Empty branches and empty regular expressions are not portable to V8. -.Pp -The restriction against -applying `*' or `+' to a possibly-null operand is an artifact of the -simplistic implementation. -.Pp -Does not support -.Xr egrep 1 Ns 's -newline-separated branches; -neither does the V8 -.Xr regexp 3 , -though. -.Pp -Due to emphasis on -compactness and simplicity, -it is not strikingly fast. -It does give special attention to handling simple cases quickly. diff --git a/lib/libcompat/regexp/regexp.c b/lib/libcompat/regexp/regexp.c deleted file mode 100644 index 8276ffcf6866..000000000000 --- a/lib/libcompat/regexp/regexp.c +++ /dev/null @@ -1,1337 +0,0 @@ -/* - * regcomp and regexec -- regsub and regerror are elsewhere - * - * Copyright (c) 1986 by University of Toronto. - * Written by Henry Spencer. Not derived from licensed software. - * - * Permission is granted to anyone to use this software for any - * purpose on any computer system, and to redistribute it freely, - * subject to the following restrictions: - * - * 1. The author is not responsible for the consequences of use of - * this software, no matter how awful, even if they arise - * from defects in it. - * - * 2. The origin of this software must not be misrepresented, either - * by explicit claim or by omission. - * - * 3. Altered versions must be plainly marked as such, and must not - * be misrepresented as being the original software. - *** THIS IS AN ALTERED VERSION. It was altered by John Gilmore, - *** hoptoad!gnu, on 27 Dec 1986, to add \n as an alternative to | - *** to assist in implementing egrep. - *** THIS IS AN ALTERED VERSION. It was altered by John Gilmore, - *** hoptoad!gnu, on 27 Dec 1986, to add \< and \> for word-matching - *** as in BSD grep and ex. - *** THIS IS AN ALTERED VERSION. It was altered by John Gilmore, - *** hoptoad!gnu, on 28 Dec 1986, to optimize characters quoted with \. - *** THIS IS AN ALTERED VERSION. It was altered by James A. Woods, - *** ames!jaw, on 19 June 1987, to quash a regcomp() redundancy. - * - * Beware that some of this code is subtly aware of the way operator - * precedence is structured in regular expressions. Serious changes in - * regular-expression syntax might require a total rethink. - */ - -#include -__FBSDID("$FreeBSD$"); - -#include -#include -#include -#include -#include -#include -#include "collate.h" -#include "regmagic.h" - -/* - * The "internal use only" fields in regexp.h are present to pass info from - * compile to execute that permits the execute phase to run lots faster on - * simple cases. They are: - * - * regstart char that must begin a match; '\0' if none obvious - * reganch is the match anchored (at beginning-of-line only)? - * regmust string (pointer into program) that match must include, or NULL - * regmlen length of regmust string - * - * Regstart and reganch permit very fast decisions on suitable starting points - * for a match, cutting down the work a lot. Regmust permits fast rejection - * of lines that cannot possibly match. The regmust tests are costly enough - * that regcomp() supplies a regmust only if the r.e. contains something - * potentially expensive (at present, the only such thing detected is * or + - * at the start of the r.e., which can involve a lot of backup). Regmlen is - * supplied because the test in regexec() needs it and regcomp() is computing - * it anyway. - */ - -/* - * Structure for regexp "program". This is essentially a linear encoding - * of a nondeterministic finite-state machine (aka syntax charts or - * "railroad normal form" in parsing technology). Each node is an opcode - * plus a "next" pointer, possibly plus an operand. "Next" pointers of - * all nodes except BRANCH implement concatenation; a "next" pointer with - * a BRANCH on both ends of it is connecting two alternatives. (Here we - * have one of the subtle syntax dependencies: an individual BRANCH (as - * opposed to a collection of them) is never concatenated with anything - * because of operator precedence.) The operand of some types of node is - * a literal string; for others, it is a node leading into a sub-FSM. In - * particular, the operand of a BRANCH node is the first node of the branch. - * (NB this is *not* a tree structure: the tail of the branch connects - * to the thing following the set of BRANCHes.) The opcodes are: - */ - -/* definition number opnd? meaning */ -#define END 0 /* no End of program. */ -#define BOL 1 /* no Match "" at beginning of line. */ -#define EOL 2 /* no Match "" at end of line. */ -#define ANY 3 /* no Match any one character. */ -#define ANYOF 4 /* str Match any character in this string. */ -#define ANYBUT 5 /* str Match any character not in this string. */ -#define BRANCH 6 /* node Match this alternative, or the next... */ -#define BACK 7 /* no Match "", "next" ptr points backward. */ -#define EXACTLY 8 /* str Match this string. */ -#define NOTHING 9 /* no Match empty string. */ -#define STAR 10 /* node Match this (simple) thing 0 or more times. */ -#define PLUS 11 /* node Match this (simple) thing 1 or more times. */ -#define WORDA 12 /* no Match "" at wordchar, where prev is nonword */ -#define WORDZ 13 /* no Match "" at nonwordchar, where prev is word */ -#define OPEN 20 /* no Mark this point in input as start of #n. */ - /* OPEN+1 is number 1, etc. */ -#define CLOSE 30 /* no Analogous to OPEN. */ - -/* - * Opcode notes: - * - * BRANCH The set of branches constituting a single choice are hooked - * together with their "next" pointers, since precedence prevents - * anything being concatenated to any individual branch. The - * "next" pointer of the last BRANCH in a choice points to the - * thing following the whole choice. This is also where the - * final "next" pointer of each individual branch points; each - * branch starts with the operand node of a BRANCH node. - * - * BACK Normal "next" pointers all implicitly point forward; BACK - * exists to make loop structures possible. - * - * STAR,PLUS '?', and complex '*' and '+', are implemented as circular - * BRANCH structures using BACK. Simple cases (one character - * per match) are implemented with STAR and PLUS for speed - * and to minimize recursive plunges. - * - * OPEN,CLOSE ...are numbered at compile time. - */ - -/* - * A node is one char of opcode followed by two chars of "next" pointer. - * "Next" pointers are stored as two 8-bit pieces, high order first. The - * value is a positive offset from the opcode of the node containing it. - * An operand, if any, simply follows the node. (Note that much of the - * code generation knows about this implicit relationship.) - * - * Using two bytes for the "next" pointer is vast overkill for most things, - * but allows patterns to get big without disasters. - */ -#define OP(p) (*(p)) -#define NEXT(p) (((*((p)+1)&0377)<<8) + (*((p)+2)&0377)) -#define OPERAND(p) ((p) + 3) - -/* - * See regmagic.h for one further detail of program structure. - */ - - -/* - * Utility definitions. - */ -#ifndef CHARBITS -#define UCHARAT(p) ((int)*(unsigned char *)(p)) -#else -#define UCHARAT(p) ((int)*(p)&CHARBITS) -#endif - -#define FAIL(m) { regerror(m); return(NULL); } -#define ISMULT(c) ((c) == '*' || (c) == '+' || (c) == '?') - -/* - * Flags to be passed up and down. - */ -#define HASWIDTH 01 /* Known never to match null string. */ -#define SIMPLE 02 /* Simple enough to be STAR/PLUS operand. */ -#define SPSTART 04 /* Starts with * or +. */ -#define WORST 0 /* Worst case. */ - -/* - * Global work variables for regcomp(). - */ -static char *regparse; /* Input-scan pointer. */ -static int regnpar; /* () count. */ -static char regdummy; -static char *regcode; /* Code-emit pointer; ®dummy = don't. */ -static long regsize; /* Code size. */ - -/* - * Forward declarations for regcomp()'s friends. - */ -#ifndef STATIC -#define STATIC static -#endif -STATIC char *reg(); -STATIC char *regbranch(); -STATIC char *regpiece(); -STATIC char *regatom(); -STATIC char *regnode(); -STATIC char *regnext(); -STATIC void regc(); -STATIC void reginsert(); -STATIC void regtail(); -STATIC void regoptail(); -#ifdef STRCSPN -STATIC int strcspn(); -#endif - -/* - - regcomp - compile a regular expression into internal code - * - * We can't allocate space until we know how big the compiled form will be, - * but we can't compile it (and thus know how big it is) until we've got a - * place to put the code. So we cheat: we compile it twice, once with code - * generation turned off and size counting turned on, and once "for real". - * This also means that we don't allocate space until we are sure that the - * thing really will compile successfully, and we never have to move the - * code and thus invalidate pointers into it. (Note that it has to be in - * one piece because free() must be able to free it all.) - * - * Beware that the optimization-preparation code in here knows about some - * of the structure of the compiled regexp. - */ -regexp * -regcomp(exp) -const char *exp; -{ - regexp *r; - char *scan; - char *longest; - int len; - int flags; - - if (exp == NULL) - FAIL("NULL argument"); - - /* First pass: determine size, legality. */ -#ifdef notdef - if (exp[0] == '.' && exp[1] == '*') exp += 2; /* aid grep */ -#endif - regparse = (char *)exp; - regnpar = 1; - regsize = 0L; - regcode = ®dummy; - regc(MAGIC); - if (reg(0, &flags) == NULL) - return(NULL); - - /* Small enough for pointer-storage convention? */ - if (regsize >= 32767L) /* Probably could be 65535L. */ - FAIL("regexp too big"); - - /* Allocate space. */ - r = (regexp *)malloc(sizeof(regexp) + (unsigned)regsize); - if (r == NULL) - FAIL("out of space"); - - /* Second pass: emit code. */ - regparse = (char *)exp; - regnpar = 1; - regcode = r->program; - regc(MAGIC); - if (reg(0, &flags) == NULL) - return(NULL); - - /* Dig out information for optimizations. */ - r->regstart = '\0'; /* Worst-case defaults. */ - r->reganch = 0; - r->regmust = NULL; - r->regmlen = 0; - scan = r->program+1; /* First BRANCH. */ - if (OP(regnext(scan)) == END) { /* Only one top-level choice. */ - scan = OPERAND(scan); - - /* Starting-point info. */ - if (OP(scan) == EXACTLY) - r->regstart = *OPERAND(scan); - else if (OP(scan) == BOL) - r->reganch++; - - /* - * If there's something expensive in the r.e., find the - * longest literal string that must appear and make it the - * regmust. Resolve ties in favor of later strings, since - * the regstart check works with the beginning of the r.e. - * and avoiding duplication strengthens checking. Not a - * strong reason, but sufficient in the absence of others. - */ - if (flags&SPSTART) { - longest = NULL; - len = 0; - for (; scan != NULL; scan = regnext(scan)) - if (OP(scan) == EXACTLY && strlen(OPERAND(scan)) >= len) { - longest = OPERAND(scan); - len = strlen(OPERAND(scan)); - } - r->regmust = longest; - r->regmlen = len; - } - } - - return(r); -} - -/* - - reg - regular expression, i.e. main body or parenthesized thing - * - * Caller must absorb opening parenthesis. - * - * Combining parenthesis handling with the base level of regular expression - * is a trifle forced, but the need to tie the tails of the branches to what - * follows makes it hard to avoid. - */ -static char * -reg(paren, flagp) -int paren; /* Parenthesized? */ -int *flagp; -{ - char *ret; - char *br; - char *ender; - int parno; - int flags; - - *flagp = HASWIDTH; /* Tentatively. */ - - /* Make an OPEN node, if parenthesized. */ - if (paren) { - if (regnpar >= NSUBEXP) - FAIL("too many ()"); - parno = regnpar; - regnpar++; - ret = regnode(OPEN+parno); - } else - ret = NULL; - - /* Pick up the branches, linking them together. */ - br = regbranch(&flags); - if (br == NULL) - return(NULL); - if (ret != NULL) - regtail(ret, br); /* OPEN -> first. */ - else - ret = br; - if (!(flags&HASWIDTH)) - *flagp &= ~HASWIDTH; - *flagp |= flags&SPSTART; - while (*regparse == '|' || *regparse == '\n') { - regparse++; - br = regbranch(&flags); - if (br == NULL) - return(NULL); - regtail(ret, br); /* BRANCH -> BRANCH. */ - if (!(flags&HASWIDTH)) - *flagp &= ~HASWIDTH; - *flagp |= flags&SPSTART; - } - - /* Make a closing node, and hook it on the end. */ - ender = regnode((paren) ? CLOSE+parno : END); - regtail(ret, ender); - - /* Hook the tails of the branches to the closing node. */ - for (br = ret; br != NULL; br = regnext(br)) - regoptail(br, ender); - - /* Check for proper termination. */ - if (paren && *regparse++ != ')') { - FAIL("unmatched ()"); - } else if (!paren && *regparse != '\0') { - if (*regparse == ')') { - FAIL("unmatched ()"); - } else - FAIL("junk on end"); /* "Can't happen". */ - /* NOTREACHED */ - } - - return(ret); -} - -/* - - regbranch - one alternative of an | operator - * - * Implements the concatenation operator. - */ -static char * -regbranch(flagp) -int *flagp; -{ - char *ret; - char *chain; - char *latest; - int flags; - - *flagp = WORST; /* Tentatively. */ - - ret = regnode(BRANCH); - chain = NULL; - while (*regparse != '\0' && *regparse != ')' && - *regparse != '\n' && *regparse != '|') { - latest = regpiece(&flags); - if (latest == NULL) - return(NULL); - *flagp |= flags&HASWIDTH; - if (chain == NULL) /* First piece. */ - *flagp |= flags&SPSTART; - else - regtail(chain, latest); - chain = latest; - } - if (chain == NULL) /* Loop ran zero times. */ - (void) regnode(NOTHING); - - return(ret); -} - -/* - - regpiece - something followed by possible [*+?] - * - * Note that the branching code sequences used for ? and the general cases - * of * and + are somewhat optimized: they use the same NOTHING node as - * both the endmarker for their branch list and the body of the last branch. - * It might seem that this node could be dispensed with entirely, but the - * endmarker role is not redundant. - */ -static char * -regpiece(flagp) -int *flagp; -{ - char *ret; - char op; - char *next; - int flags; - - ret = regatom(&flags); - if (ret == NULL) - return(NULL); - - op = *regparse; - if (!ISMULT(op)) { - *flagp = flags; - return(ret); - } - - if (!(flags&HASWIDTH) && op != '?') - FAIL("*+ operand could be empty"); - *flagp = (op != '+') ? (WORST|SPSTART) : (WORST|HASWIDTH); - - if (op == '*' && (flags&SIMPLE)) - reginsert(STAR, ret); - else if (op == '*') { - /* Emit x* as (x&|), where & means "self". */ - reginsert(BRANCH, ret); /* Either x */ - regoptail(ret, regnode(BACK)); /* and loop */ - regoptail(ret, ret); /* back */ - regtail(ret, regnode(BRANCH)); /* or */ - regtail(ret, regnode(NOTHING)); /* null. */ - } else if (op == '+' && (flags&SIMPLE)) - reginsert(PLUS, ret); - else if (op == '+') { - /* Emit x+ as x(&|), where & means "self". */ - next = regnode(BRANCH); /* Either */ - regtail(ret, next); - regtail(regnode(BACK), ret); /* loop back */ - regtail(next, regnode(BRANCH)); /* or */ - regtail(ret, regnode(NOTHING)); /* null. */ - } else if (op == '?') { - /* Emit x? as (x|) */ - reginsert(BRANCH, ret); /* Either x */ - regtail(ret, regnode(BRANCH)); /* or */ - next = regnode(NOTHING); /* null. */ - regtail(ret, next); - regoptail(ret, next); - } - regparse++; - if (ISMULT(*regparse)) - FAIL("nested *?+"); - - return(ret); -} - -/* - - regatom - the lowest level - * - * Optimization: gobbles an entire sequence of ordinary characters so that - * it can turn them into a single node, which is smaller to store and - * faster to run. Backslashed characters are exceptions, each becoming a - * separate node; the code is simpler that way and it's not worth fixing. - */ -static char * -regatom(flagp) -int *flagp; -{ - char *ret; - int flags; - - *flagp = WORST; /* Tentatively. */ - - switch (*regparse++) { - /* FIXME: these chars only have meaning at beg/end of pat? */ - case '^': - ret = regnode(BOL); - break; - case '$': - ret = regnode(EOL); - break; - case '.': - ret = regnode(ANY); - *flagp |= HASWIDTH|SIMPLE; - break; - case '[': { - int class; - int classend; - int i; - - if (*regparse == '^') { /* Complement of range. */ - ret = regnode(ANYBUT); - regparse++; - } else - ret = regnode(ANYOF); - if (*regparse == ']' || *regparse == '-') - regc(*regparse++); - while (*regparse != '\0' && *regparse != ']') { - if (*regparse == '-') { - regparse++; - if (*regparse == ']' || *regparse == '\0') - regc('-'); - else { - class = UCHARAT(regparse-2); - classend = UCHARAT(regparse); - if (__collate_load_error) { - if (class > classend) - FAIL("invalid [] range"); - for (class++; class <= classend; class++) - regc(class); - } else { - if (__collate_range_cmp(class, classend) > 0) - FAIL("invalid [] range"); - for (i = 0; i <= UCHAR_MAX; i++) - if ( i != class - && __collate_range_cmp(class, i) <= 0 - && __collate_range_cmp(i, classend) <= 0 - ) - regc(i); - } - regparse++; - } - } else - regc(*regparse++); - } - regc('\0'); - if (*regparse != ']') - FAIL("unmatched []"); - regparse++; - *flagp |= HASWIDTH|SIMPLE; - } - break; - case '(': - ret = reg(1, &flags); - if (ret == NULL) - return(NULL); - *flagp |= flags&(HASWIDTH|SPSTART); - break; - case '\0': - case '|': - case '\n': - case ')': - FAIL("internal urp"); /* Supposed to be caught earlier. */ - break; - case '?': - case '+': - case '*': - FAIL("?+* follows nothing"); - break; - case '\\': - switch (*regparse++) { - case '\0': - FAIL("trailing \\"); - break; - case '<': - ret = regnode(WORDA); - break; - case '>': - ret = regnode(WORDZ); - break; - /* FIXME: Someday handle \1, \2, ... */ - default: - /* Handle general quoted chars in exact-match routine */ - goto de_fault; - } - break; - de_fault: - default: - /* - * Encode a string of characters to be matched exactly. - * - * This is a bit tricky due to quoted chars and due to - * '*', '+', and '?' taking the SINGLE char previous - * as their operand. - * - * On entry, the char at regparse[-1] is going to go - * into the string, no matter what it is. (It could be - * following a \ if we are entered from the '\' case.) - * - * Basic idea is to pick up a good char in ch and - * examine the next char. If it's *+? then we twiddle. - * If it's \ then we frozzle. If it's other magic char - * we push ch and terminate the string. If none of the - * above, we push ch on the string and go around again. - * - * regprev is used to remember where "the current char" - * starts in the string, if due to a *+? we need to back - * up and put the current char in a separate, 1-char, string. - * When regprev is NULL, ch is the only char in the - * string; this is used in *+? handling, and in setting - * flags |= SIMPLE at the end. - */ - { - char *regprev; - char ch; - - regparse--; /* Look at cur char */ - ret = regnode(EXACTLY); - for ( regprev = 0 ; ; ) { - ch = *regparse++; /* Get current char */ - switch (*regparse) { /* look at next one */ - - default: - regc(ch); /* Add cur to string */ - break; - - case '.': case '[': case '(': - case ')': case '|': case '\n': - case '$': case '^': - case '\0': - /* FIXME, $ and ^ should not always be magic */ - magic: - regc(ch); /* dump cur char */ - goto done; /* and we are done */ - - case '?': case '+': case '*': - if (!regprev) /* If just ch in str, */ - goto magic; /* use it */ - /* End mult-char string one early */ - regparse = regprev; /* Back up parse */ - goto done; - - case '\\': - regc(ch); /* Cur char OK */ - switch (regparse[1]){ /* Look after \ */ - case '\0': - case '<': - case '>': - /* FIXME: Someday handle \1, \2, ... */ - goto done; /* Not quoted */ - default: - /* Backup point is \, scan * point is after it. */ - regprev = regparse; - regparse++; - continue; /* NOT break; */ - } - } - regprev = regparse; /* Set backup point */ - } - done: - regc('\0'); - *flagp |= HASWIDTH; - if (!regprev) /* One char? */ - *flagp |= SIMPLE; - } - break; - } - - return(ret); -} - -/* - - regnode - emit a node - */ -static char * /* Location. */ -regnode(op) -char op; -{ - char *ret; - char *ptr; - - ret = regcode; - if (ret == ®dummy) { - regsize += 3; - return(ret); - } - - ptr = ret; - *ptr++ = op; - *ptr++ = '\0'; /* Null "next" pointer. */ - *ptr++ = '\0'; - regcode = ptr; - - return(ret); -} - -/* - - regc - emit (if appropriate) a byte of code - */ -static void -regc(b) -char b; -{ - if (regcode != ®dummy) - *regcode++ = b; - else - regsize++; -} - -/* - - reginsert - insert an operator in front of already-emitted operand - * - * Means relocating the operand. - */ -static void -reginsert(op, opnd) -char op; -char *opnd; -{ - char *src; - char *dst; - char *place; - - if (regcode == ®dummy) { - regsize += 3; - return; - } - - src = regcode; - regcode += 3; - dst = regcode; - while (src > opnd) - *--dst = *--src; - - place = opnd; /* Op node, where operand used to be. */ - *place++ = op; - *place++ = '\0'; - *place++ = '\0'; -} - -/* - - regtail - set the next-pointer at the end of a node chain - */ -static void -regtail(p, val) -char *p; -char *val; -{ - char *scan; - char *temp; - int offset; - - if (p == ®dummy) - return; - - /* Find last node. */ - scan = p; - for (;;) { - temp = regnext(scan); - if (temp == NULL) - break; - scan = temp; - } - - if (OP(scan) == BACK) - offset = scan - val; - else - offset = val - scan; - *(scan+1) = (offset>>8)&0377; - *(scan+2) = offset&0377; -} - -/* - - regoptail - regtail on operand of first argument; nop if operandless - */ -static void -regoptail(p, val) -char *p; -char *val; -{ - /* "Operandless" and "op != BRANCH" are synonymous in practice. */ - if (p == NULL || p == ®dummy || OP(p) != BRANCH) - return; - regtail(OPERAND(p), val); -} - -/* - * regexec and friends - */ - -/* - * Global work variables for regexec(). - */ -static char *reginput; /* String-input pointer. */ -static char *regbol; /* Beginning of input, for ^ check. */ -static char **regstartp; /* Pointer to startp array. */ -static char **regendp; /* Ditto for endp. */ - -/* - * Forwards. - */ -STATIC int regtry(); -STATIC int regmatch(); -STATIC int regrepeat(); - -#ifdef DEBUG -int regnarrate = 0; -void regdump(); -STATIC char *regprop(); -#endif - -/* - - regexec - match a regexp against a string - */ -int -regexec(prog, string) -const regexp *prog; -const char *string; -{ - char *s; - extern char *strchr(); - - /* Be paranoid... */ - if (prog == NULL || string == NULL) { - regerror("NULL parameter"); - return(0); - } - - /* Check validity of program. */ - if (UCHARAT(prog->program) != MAGIC) { - regerror("corrupted program"); - return(0); - } - - /* If there is a "must appear" string, look for it. */ - if (prog->regmust != NULL) { - s = (char *)string; - while ((s = strchr(s, prog->regmust[0])) != NULL) { - if (strncmp(s, prog->regmust, prog->regmlen) == 0) - break; /* Found it. */ - s++; - } - if (s == NULL) /* Not present. */ - return(0); - } - - /* Mark beginning of line for ^ . */ - regbol = (char *)string; - - /* Simplest case: anchored match need be tried only once. */ - if (prog->reganch) - return(regtry(prog, string)); - - /* Messy cases: unanchored match. */ - s = (char *)string; - if (prog->regstart != '\0') - /* We know what char it must start with. */ - while ((s = strchr(s, prog->regstart)) != NULL) { - if (regtry(prog, s)) - return(1); - s++; - } - else - /* We don't -- general case. */ - do { - if (regtry(prog, s)) - return(1); - } while (*s++ != '\0'); - - /* Failure. */ - return(0); -} - -/* - - regtry - try match at specific point - */ -static int /* 0 failure, 1 success */ -regtry(prog, string) -regexp *prog; -char *string; -{ - int i; - char **sp; - char **ep; - - reginput = string; - regstartp = prog->startp; - regendp = prog->endp; - - sp = prog->startp; - ep = prog->endp; - for (i = NSUBEXP; i > 0; i--) { - *sp++ = NULL; - *ep++ = NULL; - } - if (regmatch(prog->program + 1)) { - prog->startp[0] = string; - prog->endp[0] = reginput; - return(1); - } else - return(0); -} - -/* - - regmatch - main matching routine - * - * Conceptually the strategy is simple: check to see whether the current - * node matches, call self recursively to see whether the rest matches, - * and then act accordingly. In practice we make some effort to avoid - * recursion, in particular by going through "ordinary" nodes (that don't - * need to know whether the rest of the match failed) by a loop instead of - * by recursion. - */ -static int /* 0 failure, 1 success */ -regmatch(prog) -char *prog; -{ - char *scan; /* Current node. */ - char *next; /* Next node. */ - - scan = prog; -#ifdef DEBUG - if (scan != NULL && regnarrate) - fprintf(stderr, "%s(\n", regprop(scan)); -#endif - while (scan != NULL) { -#ifdef DEBUG - if (regnarrate) - fprintf(stderr, "%s...\n", regprop(scan)); -#endif - next = regnext(scan); - - switch (OP(scan)) { - case BOL: - if (reginput != regbol) - return(0); - break; - case EOL: - if (*reginput != '\0') - return(0); - break; - case WORDA: - /* Must be looking at a letter, digit, or _ */ - if ((!isalnum((unsigned char)*reginput)) && *reginput != '_') - return(0); - /* Prev must be BOL or nonword */ - if (reginput > regbol && - (isalnum((unsigned char)reginput[-1]) || reginput[-1] == '_')) - return(0); - break; - case WORDZ: - /* Must be looking at non letter, digit, or _ */ - if (isalnum((unsigned char)*reginput) || *reginput == '_') - return(0); - /* We don't care what the previous char was */ - break; - case ANY: - if (*reginput == '\0') - return(0); - reginput++; - break; - case EXACTLY: { - int len; - char *opnd; - - opnd = OPERAND(scan); - /* Inline the first character, for speed. */ - if (*opnd != *reginput) - return(0); - len = strlen(opnd); - if (len > 1 && strncmp(opnd, reginput, len) != 0) - return(0); - reginput += len; - } - break; - case ANYOF: - if (*reginput == '\0' || strchr(OPERAND(scan), *reginput) == NULL) - return(0); - reginput++; - break; - case ANYBUT: - if (*reginput == '\0' || strchr(OPERAND(scan), *reginput) != NULL) - return(0); - reginput++; - break; - case NOTHING: - break; - case BACK: - break; - case OPEN+1: - case OPEN+2: - case OPEN+3: - case OPEN+4: - case OPEN+5: - case OPEN+6: - case OPEN+7: - case OPEN+8: - case OPEN+9: { - int no; - char *save; - - no = OP(scan) - OPEN; - save = reginput; - - if (regmatch(next)) { - /* - * Don't set startp if some later - * invocation of the same parentheses - * already has. - */ - if (regstartp[no] == NULL) - regstartp[no] = save; - return(1); - } else - return(0); - } - break; - case CLOSE+1: - case CLOSE+2: - case CLOSE+3: - case CLOSE+4: - case CLOSE+5: - case CLOSE+6: - case CLOSE+7: - case CLOSE+8: - case CLOSE+9: { - int no; - char *save; - - no = OP(scan) - CLOSE; - save = reginput; - - if (regmatch(next)) { - /* - * Don't set endp if some later - * invocation of the same parentheses - * already has. - */ - if (regendp[no] == NULL) - regendp[no] = save; - return(1); - } else - return(0); - } - break; - case BRANCH: { - char *save; - - if (OP(next) != BRANCH) /* No choice. */ - next = OPERAND(scan); /* Avoid recursion. */ - else { - do { - save = reginput; - if (regmatch(OPERAND(scan))) - return(1); - reginput = save; - scan = regnext(scan); - } while (scan != NULL && OP(scan) == BRANCH); - return(0); - /* NOTREACHED */ - } - } - break; - case STAR: - case PLUS: { - char nextch; - int no; - char *save; - int min; - - /* - * Lookahead to avoid useless match attempts - * when we know what character comes next. - */ - nextch = '\0'; - if (OP(next) == EXACTLY) - nextch = *OPERAND(next); - min = (OP(scan) == STAR) ? 0 : 1; - save = reginput; - no = regrepeat(OPERAND(scan)); - while (no >= min) { - /* If it could work, try it. */ - if (nextch == '\0' || *reginput == nextch) - if (regmatch(next)) - return(1); - /* Couldn't or didn't -- back up. */ - no--; - reginput = save + no; - } - return(0); - } - break; - case END: - return(1); /* Success! */ - break; - default: - regerror("memory corruption"); - return(0); - break; - } - - scan = next; - } - - /* - * We get here only if there's trouble -- normally "case END" is - * the terminating point. - */ - regerror("corrupted pointers"); - return(0); -} - -/* - - regrepeat - repeatedly match something simple, report how many - */ -static int -regrepeat(p) -char *p; -{ - int count = 0; - char *scan; - char *opnd; - - scan = reginput; - opnd = OPERAND(p); - switch (OP(p)) { - case ANY: - count = strlen(scan); - scan += count; - break; - case EXACTLY: - while (*opnd == *scan) { - count++; - scan++; - } - break; - case ANYOF: - while (*scan != '\0' && strchr(opnd, *scan) != NULL) { - count++; - scan++; - } - break; - case ANYBUT: - while (*scan != '\0' && strchr(opnd, *scan) == NULL) { - count++; - scan++; - } - break; - default: /* Oh dear. Called inappropriately. */ - regerror("internal foulup"); - count = 0; /* Best compromise. */ - break; - } - reginput = scan; - - return(count); -} - -/* - - regnext - dig the "next" pointer out of a node - */ -static char * -regnext(p) -char *p; -{ - int offset; - - if (p == ®dummy) - return(NULL); - - offset = NEXT(p); - if (offset == 0) - return(NULL); - - if (OP(p) == BACK) - return(p-offset); - else - return(p+offset); -} - -#ifdef DEBUG - -STATIC char *regprop(); - -/* - - regdump - dump a regexp onto stdout in vaguely comprehensible form - */ -void -regdump(r) -regexp *r; -{ - char *s; - char op = EXACTLY; /* Arbitrary non-END op. */ - char *next; - extern char *strchr(); - - - s = r->program + 1; - while (op != END) { /* While that wasn't END last time... */ - op = OP(s); - printf("%2d%s", s-r->program, regprop(s)); /* Where, what. */ - next = regnext(s); - if (next == NULL) /* Next ptr. */ - printf("(0)"); - else - printf("(%d)", (s-r->program)+(next-s)); - s += 3; - if (op == ANYOF || op == ANYBUT || op == EXACTLY) { - /* Literal string, where present. */ - while (*s != '\0') { - putchar(*s); - s++; - } - s++; - } - putchar('\n'); - } - - /* Header fields of interest. */ - if (r->regstart != '\0') - printf("start `%c' ", r->regstart); - if (r->reganch) - printf("anchored "); - if (r->regmust != NULL) - printf("must have \"%s\"", r->regmust); - printf("\n"); -} - -/* - - regprop - printable representation of opcode - */ -static char * -regprop(op) -char *op; -{ - char *p; - static char buf[50]; - - (void) strcpy(buf, ":"); - - switch (OP(op)) { - case BOL: - p = "BOL"; - break; - case EOL: - p = "EOL"; - break; - case ANY: - p = "ANY"; - break; - case ANYOF: - p = "ANYOF"; - break; - case ANYBUT: - p = "ANYBUT"; - break; - case BRANCH: - p = "BRANCH"; - break; - case EXACTLY: - p = "EXACTLY"; - break; - case NOTHING: - p = "NOTHING"; - break; - case BACK: - p = "BACK"; - break; - case END: - p = "END"; - break; - case OPEN+1: - case OPEN+2: - case OPEN+3: - case OPEN+4: - case OPEN+5: - case OPEN+6: - case OPEN+7: - case OPEN+8: - case OPEN+9: - sprintf(buf+strlen(buf), "OPEN%d", OP(op)-OPEN); - p = NULL; - break; - case CLOSE+1: - case CLOSE+2: - case CLOSE+3: - case CLOSE+4: - case CLOSE+5: - case CLOSE+6: - case CLOSE+7: - case CLOSE+8: - case CLOSE+9: - sprintf(buf+strlen(buf), "CLOSE%d", OP(op)-CLOSE); - p = NULL; - break; - case STAR: - p = "STAR"; - break; - case PLUS: - p = "PLUS"; - break; - case WORDA: - p = "WORDA"; - break; - case WORDZ: - p = "WORDZ"; - break; - default: - regerror("corrupted opcode"); - break; - } - if (p != NULL) - (void) strcat(buf, p); - return(buf); -} -#endif - -/* - * The following is provided for those people who do not have strcspn() in - * their C libraries. They should get off their butts and do something - * about it; at least one public-domain implementation of those (highly - * useful) string routines has been published on Usenet. - */ -#ifdef STRCSPN -/* - * strcspn - find length of initial segment of s1 consisting entirely - * of characters not from s2 - */ - -static int -strcspn(s1, s2) -char *s1; -char *s2; -{ - char *scan1; - char *scan2; - int count; - - count = 0; - for (scan1 = s1; *scan1 != '\0'; scan1++) { - for (scan2 = s2; *scan2 != '\0';) /* ++ moved down. */ - if (*scan1 == *scan2++) - return(count); - count++; - } - return(count); -} -#endif diff --git a/lib/libcompat/regexp/regmagic.h b/lib/libcompat/regexp/regmagic.h deleted file mode 100644 index 5acf4478ff71..000000000000 --- a/lib/libcompat/regexp/regmagic.h +++ /dev/null @@ -1,5 +0,0 @@ -/* - * The first byte of the regexp internal "program" is actually this magic - * number; the start node begins in the second byte. - */ -#define MAGIC 0234 diff --git a/lib/libcompat/regexp/regsub.c b/lib/libcompat/regexp/regsub.c deleted file mode 100644 index 4ab921cf6f20..000000000000 --- a/lib/libcompat/regexp/regsub.c +++ /dev/null @@ -1,85 +0,0 @@ -/* - * regsub - * - * Copyright (c) 1986 by University of Toronto. - * Written by Henry Spencer. Not derived from licensed software. - * - * Permission is granted to anyone to use this software for any - * purpose on any computer system, and to redistribute it freely, - * subject to the following restrictions: - * - * 1. The author is not responsible for the consequences of use of - * this software, no matter how awful, even if they arise - * from defects in it. - * - * 2. The origin of this software must not be misrepresented, either - * by explicit claim or by omission. - * - * 3. Altered versions must be plainly marked as such, and must not - * be misrepresented as being the original software. - */ - -#include -__FBSDID("$FreeBSD$"); - -#include -#include -#include -#include "regmagic.h" - -#ifndef CHARBITS -#define UCHARAT(p) ((int)*(unsigned char *)(p)) -#else -#define UCHARAT(p) ((int)*(p)&CHARBITS) -#endif - -/* - - regsub - perform substitutions after a regexp match - */ -void -regsub(prog, source, dest) -const regexp *prog; -const char *source; -char *dest; -{ - char *src; - char *dst; - char c; - int no; - int len; - extern char *strncpy(); - - if (prog == NULL || source == NULL || dest == NULL) { - regerror("NULL parm to regsub"); - return; - } - if (UCHARAT(prog->program) != MAGIC) { - regerror("damaged regexp fed to regsub"); - return; - } - - src = (char *)source; - dst = dest; - while ((c = *src++) != '\0') { - if (c == '&') - no = 0; - else if (c == '\\' && '0' <= *src && *src <= '9') - no = *src++ - '0'; - else - no = -1; - if (no < 0) { /* Ordinary character. */ - if (c == '\\' && (*src == '\\' || *src == '&')) - c = *src++; - *dst++ = c; - } else if (prog->startp[no] != NULL && prog->endp[no] != NULL) { - len = prog->endp[no] - prog->startp[no]; - (void) strncpy(dst, prog->startp[no], len); - dst += len; - if (len != 0 && *(dst-1) == '\0') { /* strncpy hit NUL. */ - regerror("damaged match string"); - return; - } - } - } - *dst++ = '\0'; -}