rsalz@uunet.uu.net (Rich Salz) (02/10/89)
Submitted-by: Mike Haertel <mike@wheaties.ai.mit.edu>
Posting-number: Volume 17, Issue 99
Archive-name: gnugrep/part02
#! /bin/sh
# This is a shell archive. Remove anything before this line, then unpack
# it by saving it into a file and typing "sh file". To overwrite existing
# files, type "sh file -c". You can also feed this as standard input via
# unshar, or by typing "sh <file", e.g.. If this archive is complete, you
# will see the following message at the end:
# "End of shell archive."
# Contents: dfa.h grep.c
PATH=/bin:/usr/bin:/usr/ucb ; export PATH
if test -f 'dfa.h' -a "${1}" != "-c" ; then
echo shar: Will not clobber existing file \"'dfa.h'\"
else
echo shar: Extracting \"'dfa.h'\" \(22427 characters\)
sed "s/^X//" >'dfa.h' <<'END_OF_FILE'
X/* dfa.h - declarations for GNU deterministic regexp compiler
X Copyright (C) 1988 Free Software Foundation, Inc.
X Written June, 1988 by Mike Haertel
X
X NO WARRANTY
X
X BECAUSE THIS PROGRAM IS LICENSED FREE OF CHARGE, WE PROVIDE ABSOLUTELY
XNO WARRANTY, TO THE EXTENT PERMITTED BY APPLICABLE STATE LAW. EXCEPT
XWHEN OTHERWISE STATED IN WRITING, FREE SOFTWARE FOUNDATION, INC,
XRICHARD M. STALLMAN AND/OR OTHER PARTIES PROVIDE THIS PROGRAM "AS IS"
XWITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING,
XBUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
XFITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY
XAND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE
XDEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR
XCORRECTION.
X
X IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW WILL RICHARD M.
XSTALLMAN, THE FREE SOFTWARE FOUNDATION, INC., AND/OR ANY OTHER PARTY
XWHO MAY MODIFY AND REDISTRIBUTE THIS PROGRAM AS PERMITTED BELOW, BE
XLIABLE TO YOU FOR DAMAGES, INCLUDING ANY LOST PROFITS, LOST MONIES, OR
XOTHER SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
XUSE OR INABILITY TO USE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR
XDATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY THIRD PARTIES OR
XA FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS) THIS
XPROGRAM, EVEN IF YOU HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH
XDAMAGES, OR FOR ANY CLAIM BY ANY OTHER PARTY.
X
X GENERAL PUBLIC LICENSE TO COPY
X
X 1. You may copy and distribute verbatim copies of this source file
Xas you receive it, in any medium, provided that you conspicuously and
Xappropriately publish on each copy a valid copyright notice "Copyright
X (C) 1988 Free Software Foundation, Inc."; and include following the
Xcopyright notice a verbatim copy of the above disclaimer of warranty
Xand of this License. You may charge a distribution fee for the
Xphysical act of transferring a copy.
X
X 2. You may modify your copy or copies of this source file or
Xany portion of it, and copy and distribute such modifications under
Xthe terms of Paragraph 1 above, provided that you also do the following:
X
X a) cause the modified files to carry prominent notices stating
X that you changed the files and the date of any change; and
X
X b) cause the whole of any work that you distribute or publish,
X that in whole or in part contains or is a derivative of this
X program or any part thereof, to be licensed at no charge to all
X third parties on terms identical to those contained in this
X License Agreement (except that you may choose to grant more extensive
X warranty protection to some or all third parties, at your option).
X
X c) You may charge a distribution fee for the physical act of
X transferring a copy, and you may at your option offer warranty
X protection in exchange for a fee.
X
XMere aggregation of another unrelated program with this program (or its
Xderivative) on a volume of a storage or distribution medium does not bring
Xthe other program under the scope of these terms.
X
X 3. You may copy and distribute this program or any portion of it in
Xcompiled, executable or object code form under the terms of Paragraphs
X1 and 2 above provided that you do the following:
X
X a) accompany it with the complete corresponding machine-readable
X source code, which must be distributed under the terms of
X Paragraphs 1 and 2 above; or,
X
X b) accompany it with a written offer, valid for at least three
X years, to give any third party free (except for a nominal
X shipping charge) a complete machine-readable copy of the
X corresponding source code, to be distributed under the terms of
X Paragraphs 1 and 2 above; or,
X
X c) accompany it with the information you received as to where the
X corresponding source code may be obtained. (This alternative is
X allowed only for noncommercial distribution and only if you
X received the program in object code or executable form alone.)
X
XFor an executable file, complete source code means all the source code for
Xall modules it contains; but, as a special exception, it need not include
Xsource code for modules which are standard libraries that accompany the
Xoperating system on which the executable file runs.
X
X 4. You may not copy, sublicense, distribute or transfer this program
Xexcept as expressly provided under this License Agreement. Any attempt
Xotherwise to copy, sublicense, distribute or transfer this program is void and
Xyour rights to use the program under this License agreement shall be
Xautomatically terminated. However, parties who have received computer
Xsoftware programs from you with this License Agreement will not have
Xtheir licenses terminated so long as such parties remain in full compliance.
X
X 5. If you wish to incorporate parts of this program into other free
Xprograms whose distribution conditions are different, write to the Free
XSoftware Foundation at 675 Mass Ave, Cambridge, MA 02139. We have not yet
Xworked out a simple rule that can be stated here, but we will often permit
Xthis. We will be guided by the two goals of preserving the free status of
Xall derivatives our free software and of promoting the sharing and reuse of
Xsoftware.
X
X
XIn other words, you are welcome to use, share and improve this program.
XYou are forbidden to forbid anyone else to use, share and improve
Xwhat you give them. Help stamp out software-hoarding! */
X
X#ifdef __STDC__
X
X/* Missing include files for GNU C. */
X/* #include <stdlib.h> */
Xtypedef int size_t;
Xextern void *calloc(int, size_t);
Xextern void *malloc(size_t);
Xextern void *realloc(void *, size_t);
Xextern void free(void *);
X
X#ifndef USG
Xextern char *strchr(), *strrchr(), *memcpy();
X#else
Xextern char *index();
X#endif
X
Xextern char *bcopy(), *bzero();
X
X#ifdef SOMEDAY
X#define ISALNUM(c) isalnum(c)
X#define ISALPHA(c) isalpha(c)
X#else
X#define ISALNUM(c) (isascii(c) && isalnum(c))
X#define ISALPHA(c) (isascii(c) && isalpha(c))
X#endif
X
X#else /* ! __STDC__ */
X
X#define const
Xtypedef int size_t;
Xextern char *calloc(), *malloc(), *realloc();
Xextern void free();
X
X#ifndef USG
Xextern char *strchr(), *strrchr(), *memcpy();
X#else
Xextern char *index();
X#endif
X
Xextern char *bcopy(), *bzero();
X
X#define ISALNUM(c) (isascii(c) && isalnum(c))
X#define ISALPHA(c) (isascii(c) && isalpha(c))
X
X#endif /* ! __STDC__ */
X
X/* 1 means plain parentheses serve as grouping, and backslash
X parentheses are needed for literal searching.
X 0 means backslash-parentheses are grouping, and plain parentheses
X are for literal searching. */
X#define RE_NO_BK_PARENS 1
X
X/* 1 means plain | serves as the "or"-operator, and \| is a literal.
X 0 means \| serves as the "or"-operator, and | is a literal. */
X#define RE_NO_BK_VBAR 2
X
X/* 0 means plain + or ? serves as an operator, and \+, \? are literals.
X 1 means \+, \? are operators and plain +, ? are literals. */
X#define RE_BK_PLUS_QM 4
X
X/* 1 means | binds tighter than ^ or $.
X 0 means the contrary. */
X#define RE_TIGHT_VBAR 8
X
X/* 1 means treat \n as an _OR operator
X 0 means treat it as a normal character */
X#define RE_NEWLINE_OR 16
X
X/* 0 means that a special characters (such as *, ^, and $) always have
X their special meaning regardless of the surrounding context.
X 1 means that special characters may act as normal characters in some
X contexts. Specifically, this applies to:
X ^ - only special at the beginning, or after ( or |
X $ - only special at the end, or before ) or |
X *, +, ? - only special when not after the beginning, (, or | */
X#define RE_CONTEXT_INDEP_OPS 32
X
X/* Now define combinations of bits for the standard possibilities. */
X#define RE_SYNTAX_AWK (RE_NO_BK_PARENS | RE_NO_BK_VBAR | RE_CONTEXT_INDEP_OPS)
X#define RE_SYNTAX_EGREP (RE_SYNTAX_AWK | RE_NEWLINE_OR)
X#define RE_SYNTAX_GREP (RE_BK_PLUS_QM | RE_NEWLINE_OR)
X#define RE_SYNTAX_EMACS 0
X
X/* The NULL pointer. */
X#define NULL 0
X
X/* Number of bits in an unsigned char. */
X#define CHARBITS 8
X
X/* First integer value that is greater than any character code. */
X#define _NOTCHAR (1 << CHARBITS)
X
X/* INTBITS need not be exact, just a lower bound. */
X#define INTBITS (CHARBITS * sizeof (int))
X
X/* Number of ints required to hold a bit for every character. */
X#define _CHARSET_INTS ((_NOTCHAR + INTBITS - 1) / INTBITS)
X
X/* Sets of unsigned characters are stored as bit vectors in arrays of ints. */
Xtypedef int _charset[_CHARSET_INTS];
X
X/* The regexp is parsed into an array of tokens in postfix form. Some tokens
X are operators and others are terminal symbols. Most (but not all) of these
X codes are returned by the lexical analyzer. */
X#ifdef __STDC__
X
Xtypedef enum
X{
X _END = -1, /* _END is a terminal symbol that matches the
X end of input; any value of _END or less in
X the parse tree is such a symbol. Accepting
X states of the DFA are those that would have
X a transition on _END. */
X
X /* Ordinary character values are terminal symbols that match themselves. */
X
X _EMPTY = _NOTCHAR, /* _EMPTY is a terminal symbol that matches
X the empty string. */
X
X _BACKREF, /* _BACKREF is generated by \<digit>; it
X it not completely handled. If the scanner
X detects a transition on backref, it returns
X a kind of "semi-success" indicating that
X the match will have to be verified with
X a backtracking matcher. */
X
X _BEGLINE, /* _BEGLINE is a terminal symbol that matches
X the empty string if it is at the beginning
X of a line. */
X
X _ALLBEGLINE, /* _ALLBEGLINE is a terminal symbol that
X matches the empty string if it is at the
X beginning of a line; _ALLBEGLINE applies
X to the entire regexp and can only occur
X as the first token thereof. _ALLBEGLINE
X never appears in the parse tree; a _BEGLINE
X is prepended with _CAT to the entire
X regexp instead. */
X
X _ENDLINE, /* _ENDLINE is a terminal symbol that matches
X the empty string if it is at the end of
X a line. */
X
X _ALLENDLINE, /* _ALLENDLINE is to _ENDLINE as _ALLBEGLINE
X is to _BEGLINE. */
X
X _BEGWORD, /* _BEGWORD is a terminal symbol that matches
X the empty string if it is at the beginning
X of a word. */
X
X _ENDWORD, /* _ENDWORD is a terminal symbol that matches
X the empty string if it is at the end of
X a word. */
X
X _LIMWORD, /* _LIMWORD is a terminal symbol that matches
X the empty string if it is at the beginning
X or the end of a word. */
X
X _NOTLIMWORD, /* _NOTLIMWORD is a terminal symbol that
X matches the empty string if it is not at
X the beginning or end of a word. */
X
X _QMARK, /* _QMARK is an operator of one argument that
X matches zero or one occurences of its
X argument. */
X
X _STAR, /* _STAR is an operator of one argument that
X matches the Kleene closure (zero or more
X occurrences) of its argument. */
X
X _PLUS, /* _PLUS is an operator of one argument that
X matches the positive closure (one or more
X occurrences) of its argument. */
X
X _CAT, /* _CAT is an operator of two arguments that
X matches the concatenation of its
X arguments. _CAT is never returned by the
X lexical analyzer. */
X
X _OR, /* _OR is an operator of two arguments that
X matches either of its arguments. */
X
X _LPAREN, /* _LPAREN never appears in the parse tree,
X it is only a lexeme. */
X
X _RPAREN, /* _RPAREN never appears in the parse tree. */
X
X _SET /* _SET and (and any value greater) is a
X terminal symbol that matches any of a
X class of characters. */
X} _token;
X
X#else /* ! __STDC__ */
X
Xtypedef short _token;
X
X#define _END -1
X#define _EMPTY _NOTCHAR
X#define _BACKREF (_EMPTY + 1)
X#define _BEGLINE (_EMPTY + 2)
X#define _ALLBEGLINE (_EMPTY + 3)
X#define _ENDLINE (_EMPTY + 4)
X#define _ALLENDLINE (_EMPTY + 5)
X#define _BEGWORD (_EMPTY + 6)
X#define _ENDWORD (_EMPTY + 7)
X#define _LIMWORD (_EMPTY + 8)
X#define _NOTLIMWORD (_EMPTY + 9)
X#define _QMARK (_EMPTY + 10)
X#define _STAR (_EMPTY + 11)
X#define _PLUS (_EMPTY + 12)
X#define _CAT (_EMPTY + 13)
X#define _OR (_EMPTY + 14)
X#define _LPAREN (_EMPTY + 15)
X#define _RPAREN (_EMPTY + 16)
X#define _SET (_EMPTY + 17)
X
X#endif /* ! __STDC__ */
X
X/* Sets are stored in an array in the compiled regexp; the index of the
X array corresponding to a given set token is given by _SET_INDEX(t). */
X#define _SET_INDEX(t) ((t) - _SET)
X
X/* Sometimes characters can only be matched depending on the surrounding
X context. Such context decisions depend on what the previous character
X was, and the value of the current (lookahead) character. Context
X dependent constraints are encoded as 8 bit integers. Each bit that
X is set indicates that the constraint succeeds in the corresponding
X context.
X
X bit 7 - previous and current are newlines
X bit 6 - previous was newline, current isn't
X bit 5 - previous wasn't newline, current is
X bit 4 - neither previous nor current is a newline
X bit 3 - previous and current are word-constituents
X bit 2 - previous was word-constituent, current isn't
X bit 1 - previous wasn't word-constituent, current is
X bit 0 - neither previous nor current is word-constituent
X
X Word-constituent characters are those that satisfy isalnum().
X
X The macro _SUCCEEDS_IN_CONTEXT determines whether a a given constraint
X succeeds in a particular context. Prevn is true if the previous character
X was a newline, currn is true if the lookahead character is a newline.
X Prevl and currl similarly depend upon whether the previous and current
X characters are word-constituent letters. */
X#define _MATCHES_NEWLINE_CONTEXT(constraint, prevn, currn) \
X ((constraint) & 1 << ((prevn) ? 2 : 0) + ((currn) ? 1 : 0) + 4)
X#define _MATCHES_LETTER_CONTEXT(constraint, prevl, currl) \
X ((constraint) & 1 << ((prevl) ? 2 : 0) + ((currl) ? 1 : 0))
X#define _SUCCEEDS_IN_CONTEXT(constraint, prevn, currn, prevl, currl) \
X (_MATCHES_NEWLINE_CONTEXT(constraint, prevn, currn) \
X && _MATCHES_LETTER_CONTEXT(constraint, prevl, currl))
X
X/* The following macros give information about what a constraint depends on. */
X#define _PREV_NEWLINE_DEPENDENT(constraint) \
X (((constraint) & 0xc0) >> 2 != ((constraint) & 0x30))
X#define _PREV_LETTER_DEPENDENT(constraint) \
X (((constraint) & 0x0c) >> 2 != ((constraint) & 0x03))
X
X/* Tokens that match the empty string subject to some constraint actually
X work by applying that constraint to determine what may follow them,
X taking into account what has gone before. The following values are
X the constraints corresponding to the special tokens previously defined. */
X#define _NO_CONSTRAINT 0xff
X#define _BEGLINE_CONSTRAINT 0xcf
X#define _ENDLINE_CONSTRAINT 0xaf
X#define _BEGWORD_CONSTRAINT 0xf2
X#define _ENDWORD_CONSTRAINT 0xf4
X#define _LIMWORD_CONSTRAINT 0xf6
X#define _NOTLIMWORD_CONSTRAINT 0xf9
X
X/* States of the recognizer correspond to sets of positions in the parse
X tree, together with the constraints under which they may be matched.
X So a position is encoded as an index into the parse tree together with
X a constraint. */
Xtypedef struct
X{
X unsigned index:24, /* Index into the parse array. */
X constraint:8; /* Constraint for matching this position. */
X} _position;
X
X/* Sets of positions are stored as arrays. */
Xtypedef struct
X{
X _position *elems; /* Elements of this position set. */
X int nelem; /* Number of elements in this set. */
X} _position_set;
X
X/* A state of the regexp consists of a set of positions, some flags,
X and the token value of the lowest-numbered position of the state that
X contains an _END token. */
Xtypedef struct
X{
X int hash; /* Hash of the positions of this state. */
X _position_set elems; /* Positions this state could match. */
X unsigned newline:1, /* True if previous state matched newline. */
X letter:1, /* True if previous state matched a letter. */
X backref:1, /* True if this state matches a \<digit>. */
X constraint:8; /* Constraint for this state to accept. */
X int first_end; /* Token value of the first _END in elems. */
X} _dfa_state;
X
X/* If an r.e. is at most MUST_MAX characters long, we look for a string which
X must appear in it; whatever's found is dropped into the struct reg. */
X
X#define MUST_MAX 50
X
X/* A compiled regular expression. */
Xstruct regexp
X{
X /* Stuff built by the scanner. */
X _charset *charsets; /* Array of character sets for _SET tokens. */
X int cindex; /* Index for adding new charsets. */
X int calloc; /* Number of charsets currently allocated. */
X
X /* Stuff built by the parser. */
X _token *tokens; /* Postfix parse array. */
X int tindex; /* Index for adding new tokens. */
X int talloc; /* Number of tokens currently allocated. */
X int depth; /* Depth required of an evaluation stack
X used for depth-first traversal of the
X parse tree. */
X int nleaves; /* Number of leaves on the parse tree. */
X int nregexps; /* Count of parallel regexps being built
X with regparse(). */
X
X /* Stuff owned by the state builder. */
X _dfa_state *states; /* States of the regexp. */
X int sindex; /* Index for adding new states. */
X int salloc; /* Number of states currently allocated. */
X
X /* Stuff built by the structure analyzer. */
X _position_set *follows; /* Array of follow sets, indexed by position
X index. The follow of a position is the set
X of positions containing characters that
X could conceivably follow a character
X matching the given position in a string
X matching the regexp. Allocated to the
X maximum possible position index. */
X int searchflag; /* True if we are supposed to build a searching
X as opposed to an exact matcher. A searching
X matcher finds the first and shortest string
X matching a regexp anywhere in the buffer,
X whereas an exact matcher finds the longest
X string matching, but anchored to the
X beginning of the buffer. */
X
X /* Stuff owned by the executor. */
X int tralloc; /* Number of transition tables that have
X slots so far. */
X int trcount; /* Number of transition tables that have
X actually been built. */
X int **trans; /* Transition tables for states that can
X never accept. If the transitions for a
X state have not yet been computed, or the
X state could possibly accept, its entry in
X this table is NULL. */
X int **realtrans; /* Trans always points to realtrans + 1; this
X is so trans[-1] can contain NULL. */
X int **fails; /* Transition tables after failing to accept
X on a state that potentially could do so. */
X int *success; /* Table of acceptance conditions used in
X regexecute and computed in build_state. */
X int *newlines; /* Transitions on newlines. The entry for a
X newline in any transition table is always
X -1 so we can count lines without wasting
X too many cycles. The transition for a
X newline is stored separately and handled
X as a special case. Newline is also used
X as a sentinel at the end of the buffer. */
X char must[MUST_MAX];
X int mustn;
X};
X
X/* Some macros for user access to regexp internals. */
X
X/* ACCEPTING returns true if s could possibly be an accepting state of r. */
X#define ACCEPTING(s, r) ((r).states[s].constraint)
X
X/* ACCEPTS_IN_CONTEXT returns true if the given state accepts in the
X specified context. */
X#define ACCEPTS_IN_CONTEXT(prevn, currn, prevl, currl, state, reg) \
X _SUCCEEDS_IN_CONTEXT((reg).states[state].constraint, \
X prevn, currn, prevl, currl)
X
X/* FIRST_MATCHING_REGEXP returns the index number of the first of parallel
X regexps that a given state could accept. Parallel regexps are numbered
X starting at 1. */
X#define FIRST_MATCHING_REGEXP(state, reg) (-(reg).states[state].first_end)
X
X/* Entry points. */
X
X#ifdef __STDC__
X
X/* Regsyntax() takes two arguments; the first sets the syntax bits described
X earlier in this file, and the second sets the case-folding flag. */
Xextern void regsyntax(int, int);
X
X/* Compile the given string of the given length into the given struct regexp.
X Final argument is a flag specifying whether to build a searching or an
X exact matcher. */
Xextern void regcompile(const char *, size_t, struct regexp *, int);
X
X/* Execute the given struct regexp on the buffer of characters. The
X first char * points to the beginning, and the second points to the
X first character after the end of the buffer, which must be a writable
X place so a sentinel end-of-buffer marker can be stored there. The
X second-to-last argument is a flag telling whether to allow newlines to
X be part of a string matching the regexp. The next-to-last argument,
X if non-NULL, points to a place to increment every time we see a
X newline. The final argument, if non-NULL, points to a flag that will
X be set if further examination by a backtracking matcher is needed in
X order to verify backreferencing; otherwise the flag will be cleared.
X Returns NULL if no match is found, or a pointer to the first
X character after the first & shortest matching string in the buffer. */
Xextern char *regexecute(struct regexp *, char *, char *, int, int *, int *);
X
X/* Free the storage held by the components of a struct regexp. */
Xextern void regfree(struct regexp *);
X
X/* Entry points for people who know what they're doing. */
X
X/* Initialize the components of a struct regexp. */
Xextern void reginit(struct regexp *);
X
X/* Incrementally parse a string of given length into a struct regexp. */
Xextern void regparse(const char *, size_t, struct regexp *);
X
X/* Analyze a parsed regexp; second argument tells whether to build a searching
X or an exact matcher. */
Xextern void reganalyze(struct regexp *, int);
X
X/* Compute, for each possible character, the transitions out of a given
X state, storing them in an array of integers. */
Xextern void regstate(int, struct regexp *, int []);
X
X/* Error handling. */
X
X/* Regerror() is called by the regexp routines whenever an error occurs. It
X takes a single argument, a NUL-terminated string describing the error.
X The default regerror() prints the error message to stderr and exits.
X The user can provide a different regfree() if so desired. */
Xextern void regerror(const char *);
X
X#else /* ! __STDC__ */
Xextern void regsyntax(), regcompile(), regfree(), reginit(), regparse();
Xextern void reganalyze(), regstate(), regerror();
Xextern char *regexecute();
X#endif
END_OF_FILE
if test 22427 -ne `wc -c <'dfa.h'`; then
echo shar: \"'dfa.h'\" unpacked with wrong size!
fi
# end of 'dfa.h'
fi
if test -f 'grep.c' -a "${1}" != "-c" ; then
echo shar: Will not clobber existing file \"'grep.c'\"
else
echo shar: Extracting \"'grep.c'\" \(28346 characters\)
sed "s/^X//" >'grep.c' <<'END_OF_FILE'
X/* grep - print lines matching an extended regular expression
X Copyright (C) 1988 Free Software Foundation, Inc.
X Written June, 1988 by Mike Haertel
X BMG speedups added July, 1988
X by James A. Woods and Arthur David Olson
X
X NO WARRANTY
X
X BECAUSE THIS PROGRAM IS LICENSED FREE OF CHARGE, WE PROVIDE ABSOLUTELY
XNO WARRANTY, TO THE EXTENT PERMITTED BY APPLICABLE STATE LAW. EXCEPT
XWHEN OTHERWISE STATED IN WRITING, FREE SOFTWARE FOUNDATION, INC,
XRICHARD M. STALLMAN AND/OR OTHER PARTIES PROVIDE THIS PROGRAM "AS IS"
XWITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING,
XBUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
XFITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY
XAND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE
XDEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, REPAIR OR
XCORRECTION.
X
X IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW WILL RICHARD M.
XSTALLMAN, THE FREE SOFTWARE FOUNDATION, INC., AND/OR ANY OTHER PARTY
XWHO MAY MODIFY AND REDISTRIBUTE THIS PROGRAM AS PERMITTED BELOW, BE
XLIABLE TO YOU FOR DAMAGES, INCLUDING ANY LOST PROFITS, LOST MONIES, OR
XOTHER SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
XUSE OR INABILITY TO USE (INCLUDING BUT NOT LIMITED TO LOSS OF DATA OR
XDATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY THIRD PARTIES OR
XA FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS) THIS
XPROGRAM, EVEN IF YOU HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH
XDAMAGES, OR FOR ANY CLAIM BY ANY OTHER PARTY.
X
X GENERAL PUBLIC LICENSE TO COPY
X
X 1. You may copy and distribute verbatim copies of this source file
Xas you receive it, in any medium, provided that you conspicuously and
Xappropriately publish on each copy a valid copyright notice "Copyright
X (C) 1988 Free Software Foundation, Inc."; and include following the
Xcopyright notice a verbatim copy of the above disclaimer of warranty
Xand of this License. You may charge a distribution fee for the
Xphysical act of transferring a copy.
X
X 2. You may modify your copy or copies of this source file or
Xany portion of it, and copy and distribute such modifications under
Xthe terms of Paragraph 1 above, provided that you also do the following:
X
X a) cause the modified files to carry prominent notices stating
X that you changed the files and the date of any change; and
X
X b) cause the whole of any work that you distribute or publish,
X that in whole or in part contains or is a derivative of this
X program or any part thereof, to be licensed at no charge to all
X third parties on terms identical to those contained in this
X License Agreement (except that you may choose to grant more extensive
X warranty protection to some or all third parties, at your option).
X
X c) You may charge a distribution fee for the physical act of
X transferring a copy, and you may at your option offer warranty
X protection in exchange for a fee.
X
XMere aggregation of another unrelated program with this program (or its
Xderivative) on a volume of a storage or distribution medium does not bring
Xthe other program under the scope of these terms.
X
X 3. You may copy and distribute this program or any portion of it in
Xcompiled, executable or object code form under the terms of Paragraphs
X1 and 2 above provided that you do the following:
X
X a) accompany it with the complete corresponding machine-readable
X source code, which must be distributed under the terms of
X Paragraphs 1 and 2 above; or,
X
X b) accompany it with a written offer, valid for at least three
X years, to give any third party free (except for a nominal
X shipping charge) a complete machine-readable copy of the
X corresponding source code, to be distributed under the terms of
X Paragraphs 1 and 2 above; or,
X
X c) accompany it with the information you received as to where the
X corresponding source code may be obtained. (This alternative is
X allowed only for noncommercial distribution and only if you
X received the program in object code or executable form alone.)
X
XFor an executable file, complete source code means all the source code for
Xall modules it contains; but, as a special exception, it need not include
Xsource code for modules which are standard libraries that accompany the
Xoperating system on which the executable file runs.
X
X 4. You may not copy, sublicense, distribute or transfer this program
Xexcept as expressly provided under this License Agreement. Any attempt
Xotherwise to copy, sublicense, distribute or transfer this program is void and
Xyour rights to use the program under this License agreement shall be
Xautomatically terminated. However, parties who have received computer
Xsoftware programs from you with this License Agreement will not have
Xtheir licenses terminated so long as such parties remain in full compliance.
X
X 5. If you wish to incorporate parts of this program into other free
Xprograms whose distribution conditions are different, write to the Free
XSoftware Foundation at 675 Mass Ave, Cambridge, MA 02139. We have not yet
Xworked out a simple rule that can be stated here, but we will often permit
Xthis. We will be guided by the two goals of preserving the free status of
Xall derivatives our free software and of promoting the sharing and reuse of
Xsoftware.
X
X
XIn other words, you are welcome to use, share and improve this program.
XYou are forbidden to forbid anyone else to use, share and improve
Xwhat you give them. Help stamp out software-hoarding! */
X
X#include <ctype.h>
X#include <stdio.h>
X#ifdef USG
X#include <memory.h>
X#include <string.h>
X#else
X#include <strings.h>
X#endif
X#include "dfa.h"
X#include "regex.h"
X
X#ifdef __STDC__
Xextern getopt(int, char **, const char *);
Xextern read(int, void *, int);
Xextern open(const char *, int, ...);
Xextern void close();
X#else
Xextern char *strrchr();
X#endif
X
Xextern char *optarg;
Xextern optind, opterr;
Xextern errno;
Xextern char *sys_errlist[];
X
X#define MAX(a, b) ((a) > (b) ? (a) : (b))
X
X/* Exit status codes. */
X#define MATCHES_FOUND 0 /* Exit 0 if no errors and matches found. */
X#define NO_MATCHES_FOUND 1 /* Exit 1 if no matches were found. */
X#define ERROR 2 /* Exit 2 if some error occurred. */
X
X/* Error is set true if something awful happened. */
Xstatic int error;
X
X/* The program name for error messages. */
Xstatic char *prog;
X
X/* We do all our own buffering by hand for efficiency. */
Xstatic char *buffer; /* The buffer itself, grown as needed. */
Xstatic bufbytes; /* Number of bytes in the buffer. */
Xstatic size_t bufalloc; /* Number of bytes allocated to the buffer. */
Xstatic bufprev; /* Number of bytes that have been forgotten.
X This is used to get byte offsets from the
X beginning of the file. */
Xstatic bufread; /* Number of bytes to get with each read(). */
X
Xstatic void
Xinitialize_buffer()
X{
X bufread = 8192;
X bufalloc = bufread + bufread / 2;
X buffer = malloc(bufalloc);
X if (! buffer)
X {
X fprintf(stderr, "%s: Memory exhausted (%s)\n", prog,
X sys_errlist[errno]);
X exit(ERROR);
X }
X}
X
X/* The current input file. */
Xstatic fd;
Xstatic char *filename;
Xstatic eof;
X
X/* Fill the buffer retaining the last n bytes at the beginning of the
X newly filled buffer (for backward context). Returns the number of new
X bytes read from disk. */
Xstatic
Xfill_buffer_retaining(n)
X int n;
X{
X char *p, *q;
X int i;
X
X /* See if we need to grow the buffer. */
X if (bufalloc - n <= bufread)
X {
X while (bufalloc - n <= bufread)
X {
X bufalloc *= 2;
X bufread *= 2;
X }
X buffer = realloc(buffer, bufalloc);
X if (! buffer)
X {
X fprintf(stderr, "%s: Memory exhausted (%s)\n", prog,
X sys_errlist[errno]);
X exit(ERROR);
X }
X }
X
X bufprev += bufbytes - n;
X
X /* Shift stuff down. */
X for (i = n, p = buffer, q = p + bufbytes - n; i--; )
X *p++ = *q++;
X bufbytes = n;
X
X if (eof)
X return 0;
X
X /* Read in new stuff. */
X i = read(fd, buffer + bufbytes, bufread);
X if (i < 0)
X {
X fprintf(stderr, "%s: read on %s failed (%s)\n", prog,
X filename ? filename : "<stdin>", sys_errlist[errno]);
X error = 1;
X }
X
X /* Kludge to pretend every nonempty file ends with a newline. */
X if (i == 0 && bufbytes > 0 && buffer[bufbytes - 1] != '\n')
X {
X eof = i = 1;
X buffer[bufbytes] = '\n';
X }
X
X bufbytes += i;
X return i;
X}
X
X/* Various flags set according to the argument switches. */
Xstatic trailing_context; /* Lines of context to show after matches. */
Xstatic leading_context; /* Lines of context to show before matches. */
Xstatic byte_count; /* Precede output lines the byte count of the
X first character on the line. */
Xstatic no_filenames; /* Do not display filenames. */
Xstatic line_numbers; /* Precede output lines with line numbers. */
Xstatic silent; /* Produce no output at all. This switch
X is bogus, ever hear of /dev/null? */
Xstatic nonmatching_lines; /* Print lines that don't match the regexp. */
X
Xstatic bmgexec; /* Invoke Boyer-Moore-Gosper routines */
X
X/* The compiled regular expression lives here. */
Xstatic struct regexp reg;
X
X/* The compiled regular expression for the backtracking matcher lives here. */
Xstatic struct re_pattern_buffer regex;
X
X/* Pointer in the buffer after the last character printed. */
Xstatic char *printed_limit;
X
X/* True when printed_limit has been artifically advanced without printing
X anything. */
Xstatic int printed_limit_fake;
X
X/* Print a line at the given line number, returning the number of
X characters actually printed. Matching is true if the line is to
X be considered a "matching line". This is only meaningful if
X surrounding context is turned on. */
Xstatic
Xprint_line(p, number, matching)
X char *p;
X int number;
X int matching;
X{
X int count = 0;
X
X if (silent)
X {
X do
X ++count;
X while (*p++ != '\n');
X printed_limit_fake = 0;
X printed_limit = p;
X return count;
X }
X
X if (filename && !no_filenames)
X printf("%s%c", filename, matching ? ':' : '-');
X if (byte_count)
X printf("%d%c", p - buffer + bufprev, matching ? ':' : '-');
X if (line_numbers)
X printf("%d%c", number, matching ? ':' : '-');
X do
X {
X ++count;
X putchar(*p);
X }
X while (*p++ != '\n');
X printed_limit_fake = 0;
X printed_limit = p;
X return count;
X}
X
X/* Print matching or nonmatching lines from the current file. Returns a
X count of matching or nonmatching lines. */
Xstatic
Xgrep()
X{
X int retain = 0; /* Number of bytes to retain on next call
X to fill_buffer_retaining(). */
X char *search_limit; /* Pointer to the character after the last
X newline in the buffer. */
X char saved_char; /* Character after the last newline. */
X char *resume; /* Pointer to where to resume search. */
X int resume_index = 0; /* Count of characters to ignore after
X refilling the buffer. */
X int line_count = 1; /* Line number. */
X int try_backref; /* Set to true if we need to verify the
X match with a backtracking matcher. */
X int initial_line_count; /* Line count at beginning of last search. */
X char *match; /* Pointer to the first character after the
X string matching the regexp. */
X int match_count = 0; /* Count of matching lines. */
X char *matching_line; /* Pointer to first character of the matching
X line, or of the first line of context to
X print if context is turned on. */
X char *real_matching_line; /* Pointer to the first character of the
X real matching line. */
X char *next_line; /* Pointer to first character of the line
X following the matching line. */
X int pending_lines = 0; /* Lines of context left over from last match
X that we have to print. */
X static first_match = 1; /* True when nothing has been printed. */
X int i;
X char *tmp;
X char *execute();
X
X printed_limit_fake = 0;
X
X while (fill_buffer_retaining(retain) > 0)
X {
X /* Find the last newline in the buffer. */
X search_limit = buffer + bufbytes;
X while (search_limit > buffer && search_limit[-1] != '\n')
X --search_limit;
X if (search_limit == buffer)
X {
X retain = bufbytes;
X continue;
X }
X
X /* Save the character after the last newline so regexecute can write
X its own sentinel newline. */
X saved_char = *search_limit;
X
X /* Search the buffer for a match. */
X printed_limit = buffer;
X resume = buffer + resume_index;
X initial_line_count = line_count;
X
X while (match = execute(®, resume, search_limit, 0, &line_count, &try_backref))
X {
X ++match_count;
X
X /* Find the beginning of the matching line. */
X matching_line = match;
X while (matching_line > resume && matching_line[-1] != '\n')
X --matching_line;
X real_matching_line = matching_line;
X
X /* Find the beginning of the next line. */
X next_line = match;
X while (next_line < search_limit && *next_line++ != '\n')
X ;
X
X /* If a potential backreference is indicated, try it out with
X a backtracking matcher to make sure the line is a match. */
X if (try_backref && re_search(®ex, matching_line,
X next_line - matching_line,
X 0,
X next_line - matching_line,
X NULL) < 0)
X {
X resume = next_line;
X if (resume == search_limit)
X break;
X else
X continue;
X }
X
X /* Print leftover lines from last time. If nonmatching_lines is
X turned on, print these as if they were matching lines. */
X while (resume < matching_line && pending_lines)
X {
X resume += print_line(resume, initial_line_count++,
X nonmatching_lines);
X --pending_lines;
X }
X
X /* Print out the matching or nonmatching lines as necessary. */
X if (! nonmatching_lines)
X {
X /* Back up over leading context if necessary. */
X for (i = leading_context; matching_line > printed_limit
X && i; --i)
X {
X while (matching_line > printed_limit
X && (--matching_line)[-1] != '\n')
X ;
X --line_count;
X }
X
X /* If context is enabled, we may have to print a separator. */
X if ((leading_context || trailing_context) && !silent
X && !first_match && (printed_limit_fake || matching_line
X > printed_limit))
X printf("----------\n");
X first_match = 0;
X
X /* Print the matching line and its leading context. */
X while (matching_line < real_matching_line)
X matching_line += print_line(matching_line, line_count++, 0);
X matching_line += print_line(matching_line, line_count++, 1);
X
X /* If there's trailing context, leave some lines pending until
X next time. */
X pending_lines = trailing_context;
X }
X else if (matching_line > resume)
X {
X char *real_resume = resume;
X
X /* Back up over leading context if necessary. */
X for (i = leading_context; resume > printed_limit && i; --i)
X {
X while (resume > printed_limit && (--resume)[-1] != '\n')
X ;
X --initial_line_count;
X }
X
X /* If context is enabled, we may have to print a separator. */
X if ((leading_context || trailing_context) && !silent
X && !first_match && (printed_limit_fake || resume
X > printed_limit))
X printf("----------\n");
X first_match = 0;
X
X /* Print out the presumably matching leading context. */
X while (resume < real_resume)
X resume += print_line(resume, initial_line_count++, 0);
X
X /* Print out the nonmatching lines prior to the matching line. */
X while (resume < matching_line)
X resume += print_line(resume, initial_line_count++, 1);
X
X /* Deal with trailing context. */
X if (trailing_context)
X {
X print_line(matching_line, line_count, 0);
X pending_lines = trailing_context - 1;
X }
X
X /* Count the current line. */
X ++line_count;
X }
X else
X {
X /* The line immediately after a matching line has to be printed
X because it was pending. */
X if (pending_lines > 0)
X {
X --pending_lines;
X print_line(matching_line, line_count, 0);
X }
X ++line_count;
X }
X
X /* Resume searching at the beginning of the next line. */
X initial_line_count = line_count;
X resume = next_line;
X
X if (resume == search_limit)
X break;
X }
X
X /* Restore the saved character. */
X *search_limit = saved_char;
X
X if (! nonmatching_lines)
X {
X while (resume < search_limit && pending_lines)
X {
X resume += print_line(resume, initial_line_count++, 0);
X --pending_lines;
X }
X }
X else if (search_limit > resume)
X {
X char *initial_resume = resume;
X
X /* Back up over leading context if necessary. */
X for (i = leading_context; resume > printed_limit && i; --i)
X {
X while (resume > printed_limit && (--resume)[-1] != '\n')
X ;
X --initial_line_count;
X }
X
X /* If context is enabled, we may have to print a separator. */
X if ((leading_context || trailing_context) && !silent
X && !first_match && (printed_limit_fake || resume
X > printed_limit))
X printf("----------\n");
X first_match = 0;
X
X /* Print out all the nonmatching lines up to the search limit. */
X while (resume < initial_resume)
X resume += print_line(resume, initial_line_count++, 0);
X while (resume < search_limit)
X resume += print_line(resume, initial_line_count++, 1);
X
X pending_lines = trailing_context;
X resume_index = 0;
X retain = bufbytes - (search_limit - buffer);
X continue;
X }
X
X /* Save the trailing end of the buffer for possible use as leading
X context in the future. */
X i = leading_context;
X tmp = search_limit;
X while (tmp > printed_limit && i--)
X while (tmp > printed_limit && (--tmp)[-1] != '\n')
X ;
X resume_index = search_limit - tmp;
X retain = bufbytes - (tmp - buffer);
X if (tmp > printed_limit)
X printed_limit_fake = 1;
X }
X
X return nonmatching_lines ? line_count - match_count : match_count;
X}
X
Xvoid
Xusage_and_die()
X{
X fprintf(stderr,
X"usage: %s [-CVbchilnsvwx] [-<num>] [-AB <num>] [-f file] [-e] expr [files]\n",
X prog);
X exit(ERROR);
X}
X
Xstatic char version[] = "GNU e?grep, version 1.2";
X
Xmain(argc, argv)
X int argc;
X char **argv;
X{
X int c;
X int ignore_case = 0; /* Compile the regexp to ignore case. */
X char *the_regexp = 0; /* The regular expression. */
X int regexp_len; /* Length of the regular expression. */
X char *regexp_file = 0; /* File containing parallel regexps. */
X int count_lines = 0; /* Display only a count of matching lines. */
X int list_files = 0; /* Display only the names of matching files. */
X int whole_word = 0; /* Insist that the regexp match a word only. */
X int whole_line = 0; /* Insist on matching only whole lines. */
X int line_count = 0; /* Count of matching lines for a file. */
X int matches_found = 0; /* True if matches were found. */
X char *regex_errmesg; /* Error message from regex routines. */
X char translate[_NOTCHAR]; /* Translate table for case conversion
X (needed by the backtracking matcher). */
X
X if (prog = strrchr(argv[0], '/'))
X ++prog;
X else
X prog = argv[0];
X
X opterr = 0;
X while ((c = getopt(argc, argv, "0123456789A:B:CVbce:f:hilnsvwx")) != EOF)
X switch (c)
X {
X case '?':
X usage_and_die();
X break;
X
X case '0':
X case '1':
X case '2':
X case '3':
X case '4':
X case '5':
X case '6':
X case '7':
X case '8':
X case '9':
X trailing_context = 10 * trailing_context + c - '0';
X leading_context = 10 * leading_context + c - '0';
X break;
X
X case 'A':
X if (! sscanf(optarg, "%d", &trailing_context)
X || trailing_context < 0)
X usage_and_die();
X break;
X
X case 'B':
X if (! sscanf(optarg, "%d", &leading_context)
X || leading_context < 0)
X usage_and_die();
X break;
X
X case 'C':
X trailing_context = leading_context = 2;
X break;
X
X case 'V':
X fprintf(stderr, "%s\n", version);
X break;
X
X case 'b':
X byte_count = 1;
X break;
X
X case 'c':
X count_lines = 1;
X silent = 1;
X break;
X
X case 'e':
X /* It doesn't make sense to mix -f and -e. */
X if (regexp_file)
X usage_and_die();
X the_regexp = optarg;
X break;
X
X case 'f':
X /* It doesn't make sense to mix -f and -e. */
X if (the_regexp)
X usage_and_die();
X regexp_file = optarg;
X break;
X
X case 'h':
X no_filenames = 1;
X break;
X
X case 'i':
X ignore_case = 1;
X for (c = 0; c < _NOTCHAR; ++c)
X if (isupper(c))
X translate[c] = tolower(c);
X else
X translate[c] = c;
X regex.translate = translate;
X break;
X
X case 'l':
X list_files = 1;
X silent = 1;
X break;
X
X case 'n':
X line_numbers = 1;
X break;
X
X case 's':
X silent = 1;
X break;
X
X case 'v':
X nonmatching_lines = 1;
X break;
X
X case 'w':
X whole_word = 1;
X break;
X
X case 'x':
X whole_line = 1;
X break;
X
X default:
X /* This can't happen. */
X fprintf(stderr, "%s: getopt(3) let one by!\n", prog);
X usage_and_die();
X break;
X }
X
X /* Set the syntax depending on arg 0 and whether to ignore case. */
X if (*prog == 'e')
X {
X regsyntax(RE_SYNTAX_EGREP, ignore_case);
X re_set_syntax(RE_SYNTAX_EGREP);
X }
X else
X {
X regsyntax(RE_SYNTAX_GREP, ignore_case);
X re_set_syntax(RE_SYNTAX_GREP);
X }
X
X /* Compile the regexp according to all the options. */
X if (regexp_file)
X {
X FILE *fp = fopen(regexp_file, "r");
X int len = 256;
X int i = 0;
X
X if (! fp)
X {
X fprintf(stderr, "%s: %s: %s\n", prog, regexp_file,
X sys_errlist[errno]);
X exit(ERROR);
X }
X
X the_regexp = malloc(len);
X while ((c = getc(fp)) != EOF)
X {
X the_regexp[i++] = c;
X if (i == len)
X the_regexp = realloc(the_regexp, len *= 2);
X }
X /* Nuke the concluding newline so we won't match the empty string. */
X if (i > 0 && the_regexp[i - 1] == '\n')
X --i;
X regexp_len = i;
X }
X else if (! the_regexp)
X {
X if (optind >= argc)
X usage_and_die();
X the_regexp = argv[optind++];
X regexp_len = strlen(the_regexp);
X }
X else
X regexp_len = strlen(the_regexp);
X
X if (whole_word || whole_line)
X {
X char *n = malloc(regexp_len + 8);
X int i = 0;
X
X if (whole_line)
X n[i++] = '^';
X else
X n[i++] = '\\', n[i++] = '<';
X if (*prog != 'e')
X n[i++] = '\\';
X n[i++] = '(';
X memcpy(n + i, the_regexp, regexp_len);
X i += regexp_len;
X if (*prog != 'e')
X n[i++] = '\\';
X n[i++] = ')';
X if (whole_line)
X n[i++] = '$';
X else
X n[i++] = '\\', n[i++] = '>';
X the_regexp = n;
X regexp_len = i;
X }
X
X regcompile(the_regexp, regexp_len, ®, 1);
X
X if (regex_errmesg = re_compile_pattern(the_regexp, regexp_len, ®ex))
X regerror(regex_errmesg);
X
X /*
X Find the longest metacharacter-free string which must occur in the
X regexpr, before short-circuiting regexecute() with Boyer-Moore-Gosper.
X (Conjecture: The problem in general is NP-complete.) If there is no
X such string (like for many alternations), then default to full automaton
X search. regmust() code and heuristics [see dfa.c] courtesy
X Arthur David Olson.
X */
X if (line_numbers == 0 && nonmatching_lines == 0)
X {
X if (reg.mustn == 0 || reg.mustn == MUST_MAX ||
X strchr(reg.must, '\0') != reg.must + reg.mustn)
X bmgexec = 0;
X else
X {
X reg.must[reg.mustn] = '\0';
X if (getenv("MUSTDEBUG") != NULL)
X (void) printf("must have: \"%s\"\n", reg.must);
X bmg_setup(reg.must, ignore_case);
X bmgexec = 1;
X }
X }
X
X if (argc - optind < 2)
X no_filenames = 1;
X
X initialize_buffer();
X
X if (argc > optind)
X while (optind < argc)
X {
X bufprev = eof = 0;
X filename = argv[optind++];
X fd = open(filename, 0, 0);
X if (fd < 0)
X {
X fprintf(stderr, "%s: %s: %s\n", prog, filename,
X sys_errlist[errno]);
X error = 1;
X continue;
X }
X if (line_count = grep())
X matches_found = 1;
X close(fd);
X if (count_lines)
X if (!no_filenames)
X printf("%s:%d\n", filename, line_count);
X else
X printf("%d\n", line_count);
X else if (list_files && line_count)
X printf("%s\n", filename);
X }
X else
X {
X if (line_count = grep())
X matches_found = 1;
X if (count_lines)
X printf("%d\n", line_count);
X else if (list_files && line_count)
X printf("<stdin>\n");
X }
X
X if (error)
X exit(ERROR);
X if (matches_found)
X exit(MATCHES_FOUND);
X exit(NO_MATCHES_FOUND);
X}
X
X/* Needed by the regexp routines. This could be fancier, especially when
X dealing with parallel regexps in files. */
Xvoid
Xregerror(s)
X const char *s;
X{
X fprintf(stderr, "%s: %s\n", prog, s);
X exit(ERROR);
X}
X
X/*
X bmg_setup() and bmg_search() adapted from:
X Boyer/Moore/Gosper-assisted 'egrep' search, with delta0 table as in
X original paper (CACM, October, 1977). No delta1 or delta2. According to
X experiment (Horspool, Soft. Prac. Exp., 1982), delta2 is of minimal
X practical value. However, to improve for worst case input, integrating
X the improved Galil strategies (Apostolico/Giancarlo, Siam. J. Comput.,
X February 1986) deserves consideration.
X
X James A. Woods Copyleft (C) 1986, 1988
X NASA Ames Research Center
X*/
X
Xchar *
Xexecute(r, begin, end, newline, count, try_backref)
X struct regexp *r;
X char *begin;
X char *end;
X int newline;
X int *count;
X int *try_backref;
X{
X register char *p, *s;
X char *match;
X char *start = begin;
X char save; /* regexecute() sentinel */
X int len;
X char *bmg_search();
X
X if (!bmgexec) /* full automaton search */
X return(regexecute(r, begin, end, newline, count, try_backref));
X else
X {
X len = end - begin;
X while ((match = bmg_search((unsigned char *) start, len)) != NULL)
X {
X p = match; /* narrow search range to submatch line */
X while (p > begin && *p != '\n')
X p--;
X s = match;
X while (s < end && *s != '\n')
X s++;
X s++;
X
X save = *s;
X *s = '\0';
X match = regexecute(r, p, s, newline, count, try_backref);
X *s = save;
X
X if (match != NULL)
X return((char *) match);
X else
X {
X start = s;
X len = end - start;
X }
X }
X return(NULL);
X }
X}
X
X#include <ctype.h>
Xint delta0[256];
Xunsigned char cmap[256]; /* (un)folded characters */
Xunsigned char pattern[5000];
Xint patlen;
X
Xchar *
Xbmg_search(buffer, buflen)
X unsigned char *buffer;
X int buflen;
X{
X register unsigned char *k, *strend, *s, *buflim;
X register int t;
X int j;
X
X if (patlen > buflen)
X return NULL;
X
X buflim = buffer + buflen;
X if (buflen > patlen * 4)
X strend = buflim - patlen * 4;
X else
X strend = buffer;
X
X s = buffer;
X k = buffer + patlen - 1;
X
X for (;;)
X {
X /* The dreaded inner loop, revisited. */
X while (k < strend && (t = delta0[*k]))
X {
X k += t;
X k += delta0[*k];
X k += delta0[*k];
X }
X while (k < buflim && delta0[*k])
X ++k;
X if (k == buflim)
X break;
X
X j = patlen - 1;
X s = k;
X while (cmap[*--s] == pattern[--j])
X ;
X /*
X delta-less shortcut for literati, but
X short shrift for genetic engineers.
X */
X if (j >= 0)
X k++;
X else /* submatch */
X return ((char *)k);
X }
X return(NULL);
X}
X
Xbmg_setup(pat, folded) /* compute "boyer-moore" delta table */
X char *pat;
X int folded;
X{ /* ... HAKMEM lives ... */
X int j;
X
X patlen = strlen(pat);
X
X if (folded) /* fold case while saving pattern */
X for (j = 0; j < patlen; j++)
X pattern[j] = (isupper((int) pat[j]) ?
X (char) tolower((int) pat[j]) : pat[j]);
X else
X memcpy(pattern, pat, patlen);
X
X for (j = 0; j < 256; j++)
X {
X delta0[j] = patlen;
X cmap[j] = (char) j; /* could be done at compile time */
X }
X for (j = 0; j < patlen - 1; j++)
X delta0[pattern[j]] = patlen - j - 1;
X delta0[pattern[patlen - 1]] = 0;
X
X if (folded)
X {
X for (j = 0; j < patlen - 1; j++)
X if (islower((int) pattern[j]))
X delta0[toupper((int) pattern[j])] = patlen - j - 1;
X if (islower((int) pattern[patlen - 1]))
X delta0[toupper((int) pattern[patlen - 1])] = 0;
X for (j = 'A'; j <= 'Z'; j++)
X cmap[j] = (char) tolower((int) j);
X }
X}
X
X#ifndef USG
X
X/* (groan) compatibility */
X
Xchar *
Xstrchr(s, c)
X char *s;
X{
X return index(s, c);
X}
X
Xchar *
Xstrrchr(s, c)
X char *s;
X{
X return rindex(s, c);
X}
X
Xchar *
Xmemcpy(d, s, n)
X char *d, *s;
X{
X return bcopy(s, d, n);
X}
X
X#else
X
Xchar *
Xindex(s, c)
X char *s;
X{
X return strchr(s, c);
X}
X
Xchar *
Xbcopy(s, d, n)
X char *s, *d;
X{
X return memcpy(d, s, n);
X}
X
Xchar *
Xbzero(s, n)
X char *s;
X{
X return memset(s, 0, n);
X}
X
Xbcmp(s, t, n)
X char *s, *t;
X{
X return memcmp(s, t, n);
X}
X
X#endif
END_OF_FILE
if test 28346 -ne `wc -c <'grep.c'`; then
echo shar: \"'grep.c'\" unpacked with wrong size!
fi
# end of 'grep.c'
fi
echo shar: End of shell archive.
exit 0
--
Please send comp.sources.unix-related mail to rsalz@uunet.uu.net.