lex.c

早期freebsd实现

/*-
 * Copyright (c) 1980, 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.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 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.
 */

#ifndef lint
static char sccsid[] = "@(#)lex.c	8.1 (Berkeley) 5/31/93";
#endif /* not lint */

#include <sys/types.h>
#include <sys/ioctl.h>
#include <termios.h>
#include <errno.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#if __STDC__
# include <stdarg.h>
#else
# include <varargs.h>
#endif

#include "csh.h"
#include "extern.h"

/*
 * These lexical routines read input and form lists of words.
 * There is some involved processing here, because of the complications
 * of input buffering, and especially because of history substitution.
 */

static Char	*word __P((void));
static int	 getC1 __P((int));
static void	 getdol __P((void));
static void	 getexcl __P((int));
static struct Hist
		*findev __P((Char *, bool));
static void	 setexclp __P((Char *));
static int	 bgetc __P((void));
static void	 bfree __P((void));
static struct wordent
		*gethent __P((int));
static int	 matchs __P((Char *, Char *));
static int	 getsel __P((int *, int *, int));
static struct wordent
		*getsub __P((struct wordent *));
static Char	*subword __P((Char *, int, bool *));
static struct wordent
		*dosub __P((int, struct wordent *, bool));

/*
 * Peekc is a peek character for getC, peekread for readc.
 * There is a subtlety here in many places... history routines
 * will read ahead and then insert stuff into the input stream.
 * If they push back a character then they must push it behind
 * the text substituted by the history substitution.  On the other
 * hand in several places we need 2 peek characters.  To make this
 * all work, the history routines read with getC, and make use both
 * of ungetC and unreadc.  The key observation is that the state
 * of getC at the call of a history reference is such that calls
 * to getC from the history routines will always yield calls of
 * readc, unless this peeking is involved.  That is to say that during
 * getexcl the variables lap, exclp, and exclnxt are all zero.
 *
 * Getdol invokes history substitution, hence the extra peek, peekd,
 * which it can ungetD to be before history substitutions.
 */
static Char peekc = 0, peekd = 0;
static Char peekread = 0;

/* (Tail of) current word from ! subst */
static Char *exclp = NULL;

/* The rest of the ! subst words */
static struct wordent *exclnxt = NULL;

/* Count of remaining words in ! subst */
static int exclc = 0;

/* "Globp" for alias resubstitution */
Char *alvecp = NULL;
int aret = F_SEEK;

/*
 * Labuf implements a general buffer for lookahead during lexical operations.
 * Text which is to be placed in the input stream can be stuck here.
 * We stick parsed ahead $ constructs during initial input,
 * process id's from `$$', and modified variable values (from qualifiers
 * during expansion in sh.dol.c) here.
 */
static Char labuf[BUFSIZ];

/*
 * Lex returns to its caller not only a wordlist (as a "var" parameter)
 * but also whether a history substitution occurred.  This is used in
 * the main (process) routine to determine whether to echo, and also
 * when called by the alias routine to determine whether to keep the
 * argument list.
 */
static bool hadhist = 0;

/*
 * Avoid alias expansion recursion via \!#
 */
int     hleft;

static Char getCtmp;

#define getC(f)		((getCtmp = peekc) ? (peekc = 0, getCtmp) : getC1(f))
#define	ungetC(c)	peekc = c
#define	ungetD(c)	peekd = c

int
lex(hp)
    register struct wordent *hp;
{
    register struct wordent *wdp;
    int     c;

    btell(&lineloc);
    hp->next = hp->prev = hp;
    hp->word = STRNULL;
    hadhist = 0;
    do
	c = readc(0);
    while (c == ' ' || c == '\t');
    if (c == HISTSUB && intty)
	/* ^lef^rit	from tty is short !:s^lef^rit */
	getexcl(c);
    else
	unreadc(c);
    wdp = hp;
    /*
     * The following loop is written so that the links needed by freelex will
     * be ready and rarin to go even if it is interrupted.
     */
    do {
	register struct wordent *new;

	new = (struct wordent *) xmalloc((size_t) sizeof(*wdp));
	new->word = 0;
	new->prev = wdp;
	new->next = hp;
	wdp->next = new;
	wdp = new;
	wdp->word = word();
    } while (wdp->word[0] != '\n');
    hp->prev = wdp;
    return (hadhist);
}

void
prlex(fp, sp0)
    FILE *fp;
    struct wordent *sp0;
{
    register struct wordent *sp = sp0->next;

    for (;;) {
	(void) fprintf(fp, "%s", vis_str(sp->word));
	sp = sp->next;
	if (sp == sp0)
	    break;
	if (sp->word[0] != '\n')
	    (void) fputc(' ', fp);
    }
}

void
copylex(hp, fp)
    register struct wordent *hp;
    register struct wordent *fp;
{
    register struct wordent *wdp;

    wdp = hp;
    fp = fp->next;
    do {
	register struct wordent *new;

	new = (struct wordent *) xmalloc((size_t) sizeof(*wdp));
	new->prev = wdp;
	new->next = hp;
	wdp->next = new;
	wdp = new;
	wdp->word = Strsave(fp->word);
	fp = fp->next;
    } while (wdp->word[0] != '\n');
    hp->prev = wdp;
}

void
freelex(vp)
    register struct wordent *vp;
{
    register struct wordent *fp;

    while (vp->next != vp) {
	fp = vp->next;
	vp->next = fp->next;
	xfree((ptr_t) fp->word);
	xfree((ptr_t) fp);
    }
    vp->prev = vp;
}

static Char *
word()
{
    register Char c, c1;
    register Char *wp;
    Char    wbuf[BUFSIZ];
    register bool dolflg;
    register int i;

    wp = wbuf;
    i = BUFSIZ - 4;
loop:
    while ((c = getC(DOALL)) == ' ' || c == '\t')
	continue;
    if (cmap(c, _META | _ESC))
	switch (c) {
	case '&':
	case '|':
	case '<':
	case '>':
	    *wp++ = c;
	    c1 = getC(DOALL);
	    if (c1 == c)
		*wp++ = c1;
	    else
		ungetC(c1);
	    goto ret;

	case '#':
	    if (intty)
		break;
	    c = 0;
	    do {
		c1 = c;
		c = getC(0);
	    } while (c != '\n');
	    if (c1 == '\\')
		goto loop;
	    /* fall into ... */

	case ';':
	case '(':
	case ')':
	case '\n':
	    *wp++ = c;
	    goto ret;

	case '\\':
	    c = getC(0);
	    if (c == '\n') {
		if (onelflg == 1)
		    onelflg = 2;
		goto loop;
	    }
	    if (c != HIST)
		*wp++ = '\\', --i;
	    c |= QUOTE;
	}
    c1 = 0;
    dolflg = DOALL;
    for (;;) {
	if (c1) {
	    if (c == c1) {
		c1 = 0;
		dolflg = DOALL;
	    }
	    else if (c == '\\') {
		c = getC(0);
		if (c == HIST)
		    c |= QUOTE;
		else {
		    if (c == '\n')
			/*
			 * if (c1 == '`') c = ' '; else
			 */
			c |= QUOTE;
		    ungetC(c);
		    c = '\\';
		}
	    }
	    else if (c == '\n') {
		seterror(ERR_UNMATCHED, c1);
		ungetC(c);
		break;
	    }
	}
	else if (cmap(c, _META | _QF | _QB | _ESC)) {
	    if (c == '\\') {
		c = getC(0);
		if (c == '\n') {
		    if (onelflg == 1)
			onelflg = 2;
		    break;
		}
		if (c != HIST)
		    *wp++ = '\\', --i;
		c |= QUOTE;
	    }
	    else if (cmap(c, _QF | _QB)) {	/* '"` */
		c1 = c;
		dolflg = c == '"' ? DOALL : DOEXCL;
	    }
	    else if (c != '#' || !intty) {
		ungetC(c);
		break;
	    }
	}
	if (--i > 0) {
	    *wp++ = c;
	    c = getC(dolflg);
	}
	else {
	    seterror(ERR_WTOOLONG);
	    wp = &wbuf[1];
	    break;
	}
    }
ret:
    *wp = 0;
    return (Strsave(wbuf));
}

static int
getC1(flag)
    register int flag;
{
    register Char c;

    while (1) {
	if ((c = peekc) != '\0') {
	    peekc = 0;
	    return (c);
	}
	if (lap) {
	    if ((c = *lap++) == 0)
		lap = 0;
	    else {
		if (cmap(c, _META | _QF | _QB))
		    c |= QUOTE;
		return (c);
	    }
	}
	if ((c = peekd) != '\0') {
	    peekd = 0;
	    return (c);
	}
	if (exclp) {
	    if ((c = *exclp++) != '\0')
		return (c);
	    if (exclnxt && --exclc >= 0) {
		exclnxt = exclnxt->next;
		setexclp(exclnxt->word);
		return (' ');
	    }
	    exclp = 0;
	    exclnxt = 0;
	}
	if (exclnxt) {
	    exclnxt = exclnxt->next;
	    if (--exclc < 0)
		exclnxt = 0;
	    else
		setexclp(exclnxt->word);
	    continue;
	}
	c = readc(0);
	if (c == '$' && (flag & DODOL)) {
	    getdol();
	    continue;
	}
	if (c == HIST && (flag & DOEXCL)) {
	    getexcl(0);
	    continue;
	}
	break;
    }
    return (c);
}

static void
getdol()
{
    register Char *np, *ep;
    Char    name[4 * MAXVARLEN + 1];
    register int c;
    int     sc;
    bool    special = 0, toolong;

    np = name, *np++ = '$';
    c = sc = getC(DOEXCL);
    if (any("\t \n", c)) {
	ungetD(c);
	ungetC('$' | QUOTE);
	return;
    }
    if (c == '{')
	*np++ = c, c = getC(DOEXCL);
    if (c == '#' || c == '?')
	special++, *np++ = c, c = getC(DOEXCL);
    *np++ = c;
    switch (c) {

    case '<':
    case '$':
    case '!':
	if (special)
	    seterror(ERR_SPDOLLT);
	*np = 0;
	addla(name);
	return;

    case '\n':
	ungetD(c);
	np--;
	seterror(ERR_NEWLINE);
	*np = 0;
	addla(name);
	return;

    case '*':
	if (special)
	    seterror(ERR_SPSTAR);
	*np = 0;
	addla(name);
	return;

    default:
	toolong = 0;
	if (Isdigit(c)) {
#ifdef notdef
	    /* let $?0 pass for now */
	    if (special) {
		seterror(ERR_DIGIT);
		*np = 0;
		addla(name);
		return;
	    }
#endif
	    /* we know that np < &name[4] */
	    ep = &np[MAXVARLEN];
	    while ((c = getC(DOEXCL)) != '\0'){
		if (!Isdigit(c))
		    break;
		if (np < ep)
		    *np++ = c;
		else
		    toolong = 1;
	    }
	}
	else if (letter(c)) {
	    /* we know that np < &name[4] */
	    ep = &np[MAXVARLEN];
	    toolong = 0;
	    while ((c = getC(DOEXCL)) != '\0') {
		/* Bugfix for ${v123x} from Chris Torek, DAS DEC-90. */
		if (!letter(c) && !Isdigit(c))
		    break;
		if (np < ep)
		    *np++ = c;
		else
		    toolong = 1;
	    }
	}
	else {
	    *np = 0;
	    seterror(ERR_VARILL);
	    addla(name);
	    return;
	}
	if (toolong) {
	    seterror(ERR_VARTOOLONG);
	    *np = 0;
	    addla(name);
	    return;
	}
	break;
    }
    if (c == '[') {
	*np++ = c;
	/*
	 * Name up to here is a max of MAXVARLEN + 8.
	 */
	ep = &np[2 * MAXVARLEN + 8];
	do {
	    /*
	     * Michael Greim: Allow $ expansion to take place in selector
	     * expressions. (limits the number of characters returned)
	     */
	    c = getC(DOEXCL | DODOL);
	    if (c == '\n') {
		ungetD(c);
		np--;
		seterror(ERR_NLINDEX);
		*np = 0;
		addla(name);
		return;
	    }
	    if (np < ep)
		*np++ = c;
	} while (c != ']');
	*np = '\0';
	if (np >= ep) {
	    seterror(ERR_SELOVFL);
	    addla(name);
	    return;
	}
	c = getC(DOEXCL);
    }
    /*
     * Name up to here is a max of 2 * MAXVARLEN + 8.
     */
    if (c == ':') {
	/*