NetBSD/usr.bin/indent/lexi.c

580 lines
14 KiB
C
Raw Normal View History

/* $NetBSD: lexi.c,v 1.7 1998/08/25 20:59:38 ross Exp $ */
1997-01-09 23:18:21 +03:00
1993-04-09 16:58:42 +04:00
/*
1997-10-18 20:04:21 +04:00
* Copyright (c) 1980, 1993
* The Regents of the University of California. All rights reserved.
1993-08-01 21:54:45 +04:00
* Copyright (c) 1976 Board of Trustees of the University of Illinois.
1997-10-18 20:04:21 +04:00
* Copyright (c) 1985 Sun Microsystems, Inc.
1993-04-09 16:58:42 +04:00
* 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.
*/
#include <sys/cdefs.h>
1993-04-09 16:58:42 +04:00
#ifndef lint
1997-10-18 20:04:21 +04:00
#if 0
static char sccsid[] = "@(#)lexi.c 8.1 (Berkeley) 6/6/93";
#else
__RCSID("$NetBSD: lexi.c,v 1.7 1998/08/25 20:59:38 ross Exp $");
1997-10-18 20:04:21 +04:00
#endif
#endif /* not lint */
1993-04-09 16:58:42 +04:00
/*
* Here we have the token scanner for indent. It scans off one token and puts
* it in the global variable "token". It returns a code, indicating the type
* of token scanned.
*/
#include <stdio.h>
#include <ctype.h>
#include <stdlib.h>
#include <string.h>
#include "indent_globs.h"
#include "indent_codes.h"
#define alphanum 1
#define opchar 3
struct templ {
char *rwd;
int rwcode;
1993-04-09 16:58:42 +04:00
};
struct templ specials[1000] =
1993-04-09 16:58:42 +04:00
{
{"switch", 1},
{"case", 2},
{"break", 0},
{"struct", 3},
{"union", 3},
{"enum", 3},
{"default", 2},
{"int", 4},
{"char", 4},
{"float", 4},
{"double", 4},
{"long", 4},
{"short", 4},
{"typdef", 4},
{"unsigned", 4},
{"register", 4},
{"static", 4},
{"global", 4},
{"extern", 4},
{"void", 4},
{"goto", 0},
{"return", 0},
{"if", 5},
{"while", 5},
{"for", 5},
{"else", 6},
{"do", 6},
{"sizeof", 7},
{0, 0}
1993-04-09 16:58:42 +04:00
};
char chartype[128] =
1993-04-09 16:58:42 +04:00
{ /* this is used to facilitate the decision of
* what type (alphanumeric, operator) each
* character is */
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0,
0, 3, 0, 0, 1, 3, 3, 0,
0, 0, 3, 3, 0, 3, 0, 3,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 0, 0, 3, 3, 3, 3,
0, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 0, 0, 0, 3, 1,
0, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 0, 3, 0, 3, 0
1993-04-09 16:58:42 +04:00
};
int
lexi()
{
int unary_delim; /* this is set to 1 if the current token
1993-04-09 16:58:42 +04:00
*
* forces a following operator to be unary */
static int last_code; /* the last token type returned */
static int l_struct; /* set to 1 if the last token was 'struct' */
int code; /* internal code to be returned */
char qchar; /* the delimiter character for a string */
e_token = s_token; /* point to start of place to save token */
unary_delim = false;
ps.col_1 = ps.last_nl; /* tell world that this token started in
1993-04-09 16:58:42 +04:00
* column 1 iff the last thing scanned was nl */
ps.last_nl = false;
1993-04-09 16:58:42 +04:00
while (*buf_ptr == ' ' || *buf_ptr == '\t') { /* get rid of blanks */
ps.col_1 = false; /* leading blanks imply token is not
* in column 1 */
if (++buf_ptr >= buf_end)
fill_buffer();
}
/* Scan an alphanumeric token */
if (chartype[(int) *buf_ptr] == alphanum ||
(buf_ptr[0] == '.' && isdigit(buf_ptr[1]))) {
/*
* we have a character or number
*/
char *j; /* used for searching thru list of
1993-04-09 16:58:42 +04:00
*
* reserved words */
struct templ *p;
if (isdigit(*buf_ptr) || (buf_ptr[0] == '.' && isdigit(buf_ptr[1]))) {
int seendot = 0, seenexp = 0;
if (*buf_ptr == '0' &&
(buf_ptr[1] == 'x' || buf_ptr[1] == 'X')) {
1993-04-09 16:58:42 +04:00
*e_token++ = *buf_ptr++;
*e_token++ = *buf_ptr++;
while (isxdigit(*buf_ptr)) {
CHECK_SIZE_TOKEN;
*e_token++ = *buf_ptr++;
}
} else {
while (1) {
if (*buf_ptr == '.') {
if (seendot)
break;
else
seendot++;
}
CHECK_SIZE_TOKEN;
*e_token++ = *buf_ptr++;
if (!isdigit(*buf_ptr)
&& *buf_ptr != '.') {
if ((*buf_ptr != 'E'
&& *buf_ptr != 'e') || seenexp)
break;
else {
seenexp++;
seendot++;
CHECK_SIZE_TOKEN;
*e_token++ = *buf_ptr++;
if (*buf_ptr == '+' || *buf_ptr == '-')
*e_token++ = *buf_ptr++;
}
}
}
}
if (*buf_ptr == 'L' || *buf_ptr == 'l')
*e_token++ = *buf_ptr++;
} else
while (chartype[(int) *buf_ptr] == alphanum) { /* copy it over */
CHECK_SIZE_TOKEN;
*e_token++ = *buf_ptr++;
if (buf_ptr >= buf_end)
fill_buffer();
1993-04-09 16:58:42 +04:00
}
*e_token++ = '\0';
while (*buf_ptr == ' ' || *buf_ptr == '\t') { /* get rid of blanks */
if (++buf_ptr >= buf_end)
fill_buffer();
1993-04-09 16:58:42 +04:00
}
ps.its_a_keyword = false;
ps.sizeof_keyword = false;
if (l_struct) { /* if last token was 'struct', then this token
1993-04-09 16:58:42 +04:00
* should be treated as a declaration */
l_struct = false;
last_code = ident;
ps.last_u_d = true;
return (decl);
}
ps.last_u_d = false; /* Operator after indentifier is
* binary */
last_code = ident; /* Remember that this is the code we
* will return */
1993-04-09 16:58:42 +04:00
/*
* This loop will check if the token is a keyword.
1993-04-09 16:58:42 +04:00
*/
for (p = specials; (j = p->rwd) != 0; p++) {
char *p = s_token; /* point at scanned token */
if (*j++ != *p++ || *j++ != *p++)
continue; /* This test depends on the
* fact that identifiers are
* always at least 1 character
* long (ie. the first two
* bytes of the identifier are
* always meaningful) */
if (p[-1] == 0)
break; /* If its a one-character identifier */
while (*p++ == *j)
if (*j++ == 0)
goto found_keyword; /* I wish that C had a
* multi-level break... */
1993-04-09 16:58:42 +04:00
}
if (p->rwd) { /* we have a keyword */
found_keyword:
ps.its_a_keyword = true;
ps.last_u_d = true;
switch (p->rwcode) {
case 1:/* it is a switch */
return (swstmt);
case 2:/* a case or default */
return (casestmt);
case 3:/* a "struct" */
if (ps.p_l_follow)
break; /* inside parens: cast */
l_struct = true;
/*
* Next time around, we will want to know that we have had a
* 'struct'
*/
case 4:/* one of the declaration keywords */
if (ps.p_l_follow) {
ps.cast_mask |= 1 << ps.p_l_follow;
break; /* inside parens: cast */
}
last_code = decl;
return (decl);
case 5:/* if, while, for */
return (sp_paren);
case 6:/* do, else */
return (sp_nparen);
case 7:
ps.sizeof_keyword = true;
default: /* all others are treated like any
* other identifier */
return (ident);
} /* end of switch */
} /* end of if (found_it) */
if (*buf_ptr == '(' && ps.tos <= 1 && ps.ind_level == 0) {
char *tp = buf_ptr;
while (tp < buf_end)
if (*tp++ == ')' && (*tp == ';' || *tp == ','))
goto not_proc;
strncpy(ps.procname, token, sizeof ps.procname - 1);
ps.in_parameter_declaration = 1;
rparen_count = 1;
not_proc: ;
}
/*
* The following hack attempts to guess whether or not the current
* token is in fact a declaration keyword -- one that has been
* typedefd
*/
if (((*buf_ptr == '*' && buf_ptr[1] != '=') || isalpha(*buf_ptr) || *buf_ptr == '_')
&& !ps.p_l_follow
&& !ps.block_init
&& (ps.last_token == rparen || ps.last_token == semicolon ||
ps.last_token == decl ||
ps.last_token == lbrace || ps.last_token == rbrace)) {
ps.its_a_keyword = true;
ps.last_u_d = true;
last_code = decl;
return decl;
}
if (last_code == decl) /* if this is a declared variable,
* then following sign is unary */
ps.last_u_d = true; /* will make "int a -1" work */
last_code = ident;
return (ident); /* the ident is not in the list */
} /* end of procesing for alpanum character */
/* Scan a non-alphanumeric token */
*e_token++ = *buf_ptr; /* if it is only a one-character token, it is
* moved here */
*e_token = '\0';
if (++buf_ptr >= buf_end)
fill_buffer();
1993-04-09 16:58:42 +04:00
switch (*token) {
case '\n':
unary_delim = ps.last_u_d;
ps.last_nl = true; /* remember that we just had a newline */
code = (had_eof ? 0 : newline);
1993-04-09 16:58:42 +04:00
/*
* if data has been exausted, the newline is a dummy, and we should
* return code to stop
*/
break;
case '\'': /* start of quoted character */
case '"': /* start of string */
qchar = *token;
if (troff) {
e_token[-1] = '`';
if (qchar == '"')
*e_token++ = '`';
e_token = chfont(&bodyf, &stringf, e_token);
1993-04-09 16:58:42 +04:00
}
do { /* copy the string */
while (1) { /* move one character or
* [/<char>]<char> */
if (*buf_ptr == '\n') {
printf("%d: Unterminated literal\n", line_no);
goto stop_lit;
}
CHECK_SIZE_TOKEN; /* Only have to do this
* once in this loop,
* since CHECK_SIZE
* guarantees that there
* are at least 5
* entries left */
*e_token = *buf_ptr++;
if (buf_ptr >= buf_end)
fill_buffer();
if (*e_token == BACKSLASH) { /* if escape, copy extra
* char */
if (*buf_ptr == '\n') /* check for escaped
* newline */
++line_no;
if (troff) {
*++e_token = BACKSLASH;
if (*buf_ptr == BACKSLASH)
*++e_token = BACKSLASH;
}
*++e_token = *buf_ptr++;
++e_token; /* we must increment
* this again because we
* copied two chars */
if (buf_ptr >= buf_end)
fill_buffer();
} else
break; /* we copied one character */
} /* end of while (1) */
} while (*e_token++ != qchar);
if (troff) {
e_token = chfont(&stringf, &bodyf, e_token - 1);
if (qchar == '"')
*e_token++ = '\'';
1993-04-09 16:58:42 +04:00
}
stop_lit:
code = ident;
break;
case ('('):
case ('['):
unary_delim = true;
code = lparen;
break;
case (')'):
case (']'):
code = rparen;
break;
case '#':
unary_delim = ps.last_u_d;
code = preesc;
break;
case '?':
unary_delim = true;
code = question;
break;
case (':'):
code = colon;
unary_delim = true;
break;
case (';'):
unary_delim = true;
code = semicolon;
break;
case ('{'):
unary_delim = true;
/*
* if (ps.in_or_st) ps.block_init = 1;
*/
/* ? code = ps.block_init ? lparen : lbrace; */
code = lbrace;
break;
case ('}'):
unary_delim = true;
/* ? code = ps.block_init ? rparen : rbrace; */
code = rbrace;
break;
case 014: /* a form feed */
unary_delim = ps.last_u_d;
ps.last_nl = true; /* remember this so we can set
* 'ps.col_1' right */
code = form_feed;
break;
case (','):
unary_delim = true;
code = comma;
break;
case '.':
1993-04-09 16:58:42 +04:00
unary_delim = false;
code = period;
break;
case '-':
case '+': /* check for -, +, --, ++ */
code = (ps.last_u_d ? unary_op : binary_op);
unary_delim = true;
if (*buf_ptr == token[0]) {
/* check for doubled character */
*e_token++ = *buf_ptr++;
/* buffer overflow will be checked at end of loop */
if (last_code == ident || last_code == rparen) {
code = (ps.last_u_d ? unary_op : postop);
/* check for following ++ or -- */
unary_delim = false;
}
} else
if (*buf_ptr == '=')
/* check for operator += */
*e_token++ = *buf_ptr++;
else
if (*buf_ptr == '>') {
/* check for operator -> */
*e_token++ = *buf_ptr++;
if (!pointer_as_binop) {
unary_delim = false;
code = unary_op;
ps.want_blank = false;
}
}
break; /* buffer overflow will be checked at end of
1993-04-09 16:58:42 +04:00
* switch */
case '=':
if (ps.in_or_st)
ps.block_init = 1;
1993-04-09 16:58:42 +04:00
#ifdef undef
if (chartype[*buf_ptr] == opchar) { /* we have two char
* assignment */
e_token[-1] = *buf_ptr++;
if ((e_token[-1] == '<' || e_token[-1] == '>') && e_token[-1] == *buf_ptr)
*e_token++ = *buf_ptr++;
*e_token++ = '='; /* Flip =+ to += */
*e_token = 0;
}
1993-04-09 16:58:42 +04:00
#else
if (*buf_ptr == '=') { /* == */
*e_token++ = '='; /* Flip =+ to += */
buf_ptr++;
*e_token = 0;
}
1993-04-09 16:58:42 +04:00
#endif
code = binary_op;
unary_delim = true;
break;
/* can drop thru!!! */
case '>':
case '<':
case '!': /* ops like <, <<, <=, !=, etc */
if (*buf_ptr == '>' || *buf_ptr == '<' || *buf_ptr == '=') {
*e_token++ = *buf_ptr;
if (++buf_ptr >= buf_end)
fill_buffer();
}
if (*buf_ptr == '=')
*e_token++ = *buf_ptr++;
code = (ps.last_u_d ? unary_op : binary_op);
unary_delim = true;
break;
default:
if (token[0] == '/' && *buf_ptr == '*') {
/* it is start of comment */
*e_token++ = '*';
if (++buf_ptr >= buf_end)
fill_buffer();
code = comment;
unary_delim = ps.last_u_d;
break;
}
while (*(e_token - 1) == *buf_ptr || *buf_ptr == '=') {
/*
* handle ||, &&, etc, and also things as in int *****i
*/
*e_token++ = *buf_ptr;
if (++buf_ptr >= buf_end)
fill_buffer();
}
code = (ps.last_u_d ? unary_op : binary_op);
unary_delim = true;
1993-04-09 16:58:42 +04:00
} /* end of switch */
if (code != newline) {
l_struct = false;
last_code = code;
1993-04-09 16:58:42 +04:00
}
if (buf_ptr >= buf_end) /* check for input buffer empty */
1993-04-09 16:58:42 +04:00
fill_buffer();
ps.last_u_d = unary_delim;
*e_token = '\0'; /* null terminate the token */
return (code);
1993-04-09 16:58:42 +04:00
}
/*
* Add the given keyword to the keyword table, using val as the keyword type
*/
void
1993-04-09 16:58:42 +04:00
addkey(key, val)
char *key;
int val;
1993-04-09 16:58:42 +04:00
{
struct templ *p = specials;
while (p->rwd)
if (p->rwd[0] == key[0] && strcmp(p->rwd, key) == 0)
return;
else
p++;
if (p >= specials + sizeof specials / sizeof specials[0])
return; /* For now, table overflows are silently
1993-04-09 16:58:42 +04:00
* ignored */
p->rwd = key;
p->rwcode = val;
p[1].rwd = 0;
p[1].rwcode = 0;
1993-04-09 16:58:42 +04:00
}