/* Token-reader for Bison's input parser,
- Copyright (C) 1984, 1986, 1989 Free Software Foundation, Inc.
+ Copyright (C) 1984, 1986, 1989, 1992, 2000 Free Software Foundation, Inc.
-This file is part of Bison, the GNU Compiler Compiler.
+ This file is part of Bison, the GNU Compiler Compiler.
-Bison is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
+ Bison is free software; you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation; either version 2, or (at your option)
+ any later version.
-Bison is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-GNU General Public License for more details.
+ Bison is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
-You should have received a copy of the GNU General Public License
-along with Bison; see the file COPYING. If not, write to
-the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA. */
+ You should have received a copy of the GNU General Public License
+ along with Bison; see the file COPYING. If not, write to
+ the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
+ Boston, MA 02111-1307, USA. */
-
-/*
- lex() is the entry point. It is called from reader.c.
- It returns one of the token-type codes defined in lex.h.
- When an identifier is seen, the code IDENTIFIER is returned
- and the name is looked up in the symbol table using symtab.c;
- symval is set to a pointer to the entry found. */
-
-#include <stdio.h>
-#include <ctype.h>
#include "system.h"
+#include "getargs.h"
#include "files.h"
+#include "getopt.h" /* for optarg */
#include "symtab.h"
#include "lex.h"
-#include "new.h"
+#include "alloc.h"
+#include "complain.h"
+/*spec_outfile is declared in files.h, for -o */
-extern int lineno;
extern int translations;
-int parse_percent_token();
-
-extern void fatals();
-extern void fatal();
+/* functions from main.c */
+extern char *printable_version PARAMS ((int));
/* Buffer for storing the current token. */
char *token_buffer;
/* Allocated size of token_buffer, not including space for terminator. */
-static int maxtoken;
+int maxtoken;
bucket *symval;
int numval;
void
-init_lex()
+init_lex (void)
{
maxtoken = 100;
token_buffer = NEW2 (maxtoken + 1, char);
}
-static char *
-grow_token_buffer (p)
- char *p;
+char *
+grow_token_buffer (char *p)
{
int offset = p - token_buffer;
maxtoken *= 2;
- token_buffer = (char *) xrealloc(token_buffer, maxtoken + 1);
+ token_buffer = (char *) xrealloc (token_buffer, maxtoken + 1);
return token_buffer + offset;
}
int
-skip_white_space()
+skip_white_space (void)
{
- register int c;
- register int inside;
+ int c;
+ int inside;
- c = getc(finput);
+ c = getc (finput);
for (;;)
{
switch (c)
{
case '/':
- c = getc(finput);
+ c = getc (finput);
if (c != '*' && c != '/')
- fatals("unexpected `/%c' found",c);
+ {
+ complain (_("unexpected `/' found and ignored"));
+ break;
+ }
cplus_comment = (c == '/');
- c = getc(finput);
+ c = getc (finput);
inside = 1;
while (inside)
if (!cplus_comment && c == '*')
{
while (c == '*')
- c = getc(finput);
+ c = getc (finput);
if (c == '/')
{
inside = 0;
- c = getc(finput);
+ c = getc (finput);
}
}
else if (c == '\n')
lineno++;
if (cplus_comment)
inside = 0;
- c = getc(finput);
+ c = getc (finput);
}
else if (c == EOF)
- fatal("unterminated comment");
+ fatal (_("unterminated comment"));
else
- c = getc(finput);
+ c = getc (finput);
}
break;
case ' ':
case '\t':
case '\f':
- c = getc(finput);
+ c = getc (finput);
break;
default:
- return (c);
+ return c;
+ }
+ }
+}
+
+/* do a getc, but give error message if EOF encountered */
+static int
+xgetc (FILE *f)
+{
+ int c = getc (f);
+ if (c == EOF)
+ fatal (_("unexpected end of file"));
+ return c;
+}
+
+
+/*------------------------------------------------------------------.
+| Read one literal character from finput. Process \ escapes. |
+| Append the normalized string version of the char to *PP. Assign |
+| the character code to *PCODE. Return 1 unless the character is an |
+| unescaped `term' or \n report error for \n |
+`------------------------------------------------------------------*/
+
+static int
+literalchar (char **pp, int *pcode, char term)
+{
+ int c;
+ char *p;
+ int code;
+ int wasquote = 0;
+
+ c = xgetc (finput);
+ if (c == '\n')
+ {
+ complain (_("unescaped newline in constant"));
+ ungetc (c, finput);
+ code = '?';
+ wasquote = 1;
+ }
+ else if (c != '\\')
+ {
+ code = c;
+ if (c == term)
+ wasquote = 1;
+ }
+ else
+ {
+ c = xgetc (finput);
+ if (c == 't')
+ code = '\t';
+ else if (c == 'n')
+ code = '\n';
+ else if (c == 'a')
+ code = '\007';
+ else if (c == 'r')
+ code = '\r';
+ else if (c == 'f')
+ code = '\f';
+ else if (c == 'b')
+ code = '\b';
+ else if (c == 'v')
+ code = '\013';
+ else if (c == '\\')
+ code = '\\';
+ else if (c == '\'')
+ code = '\'';
+ else if (c == '\"')
+ code = '\"';
+ else if (c <= '7' && c >= '0')
+ {
+ code = 0;
+ while (c <= '7' && c >= '0')
+ {
+ code = (code * 8) + (c - '0');
+ if (code >= 256 || code < 0)
+ {
+ complain (_("octal value outside range 0...255: `\\%o'"),
+ code);
+ code &= 0xFF;
+ break;
+ }
+ c = xgetc (finput);
+ }
+ ungetc (c, finput);
+ }
+ else if (c == 'x')
+ {
+ c = xgetc (finput);
+ code = 0;
+ while (1)
+ {
+ if (c >= '0' && c <= '9')
+ code *= 16, code += c - '0';
+ else if (c >= 'a' && c <= 'f')
+ code *= 16, code += c - 'a' + 10;
+ else if (c >= 'A' && c <= 'F')
+ code *= 16, code += c - 'A' + 10;
+ else
+ break;
+ if (code >= 256 || code < 0)
+ {
+ complain (_("hexadecimal value above 255: `\\x%x'"), code);
+ code &= 0xFF;
+ break;
+ }
+ c = xgetc (finput);
+ }
+ ungetc (c, finput);
+ }
+ else
+ {
+ complain (_("unknown escape sequence: `\\' followed by `%s'"),
+ printable_version (c));
+ code = '?';
}
+ } /* has \ */
+
+ /* now fill token_buffer with the canonical name for this character
+ as a literal token. Do not use what the user typed,
+ so that `\012' and `\n' can be interchangeable. */
+
+ p = *pp;
+ if (code == term && wasquote)
+ *p++ = code;
+ else if (code == '\\')
+ {
+ *p++ = '\\';
+ *p++ = '\\';
+ }
+ else if (code == '\'')
+ {
+ *p++ = '\\';
+ *p++ = '\'';
+ }
+ else if (code == '\"')
+ {
+ *p++ = '\\';
+ *p++ = '\"';
+ }
+ else if (code >= 040 && code < 0177)
+ *p++ = code;
+ else if (code == '\t')
+ {
+ *p++ = '\\';
+ *p++ = 't';
}
+ else if (code == '\n')
+ {
+ *p++ = '\\';
+ *p++ = 'n';
+ }
+ else if (code == '\r')
+ {
+ *p++ = '\\';
+ *p++ = 'r';
+ }
+ else if (code == '\v')
+ {
+ *p++ = '\\';
+ *p++ = 'v';
+ }
+ else if (code == '\b')
+ {
+ *p++ = '\\';
+ *p++ = 'b';
+ }
+ else if (code == '\f')
+ {
+ *p++ = '\\';
+ *p++ = 'f';
+ }
+ else
+ {
+ *p++ = '\\';
+ *p++ = code / 0100 + '0';
+ *p++ = ((code / 010) & 07) + '0';
+ *p++ = (code & 07) + '0';
+ }
+ *pp = p;
+ *pcode = code;
+ return !wasquote;
}
void
-unlex(token)
-int token;
+unlex (int token)
{
unlexed = token;
unlexed_symval = symval;
}
-
int
-lex()
+lex (void)
{
- register int c;
- register char *p;
+ int c;
+ char *p;
if (unlexed >= 0)
{
symval = unlexed_symval;
c = unlexed;
unlexed = -1;
- return (c);
+ return c;
}
- c = skip_white_space();
+ c = skip_white_space ();
+ *token_buffer = c; /* for error messages (token buffer always valid) */
+ token_buffer[1] = 0;
switch (c)
{
case EOF:
- return (ENDFILE);
-
- case 'A': case 'B': case 'C': case 'D': case 'E':
- case 'F': case 'G': case 'H': case 'I': case 'J':
- case 'K': case 'L': case 'M': case 'N': case 'O':
- case 'P': case 'Q': case 'R': case 'S': case 'T':
- case 'U': case 'V': case 'W': case 'X': case 'Y':
+ strcpy (token_buffer, "EOF");
+ return ENDFILE;
+
+ case 'A': case 'B': case 'C': case 'D': case 'E':
+ case 'F': case 'G': case 'H': case 'I': case 'J':
+ case 'K': case 'L': case 'M': case 'N': case 'O':
+ case 'P': case 'Q': case 'R': case 'S': case 'T':
+ case 'U': case 'V': case 'W': case 'X': case 'Y':
case 'Z':
- case 'a': case 'b': case 'c': case 'd': case 'e':
- case 'f': case 'g': case 'h': case 'i': case 'j':
- case 'k': case 'l': case 'm': case 'n': case 'o':
- case 'p': case 'q': case 'r': case 's': case 't':
- case 'u': case 'v': case 'w': case 'x': case 'y':
+ case 'a': case 'b': case 'c': case 'd': case 'e':
+ case 'f': case 'g': case 'h': case 'i': case 'j':
+ case 'k': case 'l': case 'm': case 'n': case 'o':
+ case 'p': case 'q': case 'r': case 's': case 't':
+ case 'u': case 'v': case 'w': case 'x': case 'y':
case 'z':
- case '.': case '_':
+ case '.': case '_':
+
p = token_buffer;
- while (isalnum(c) || c == '_' || c == '.')
+ while (isalnum (c) || c == '_' || c == '.')
{
if (p == token_buffer + maxtoken)
- p = grow_token_buffer(p);
+ p = grow_token_buffer (p);
*p++ = c;
- c = getc(finput);
+ c = getc (finput);
}
*p = 0;
- ungetc(c, finput);
- symval = getsym(token_buffer);
- return (IDENTIFIER);
+ ungetc (c, finput);
+ symval = getsym (token_buffer);
+ return IDENTIFIER;
- case '0': case '1': case '2': case '3': case '4':
- case '5': case '6': case '7': case '8': case '9':
+ case '0': case '1': case '2': case '3': case '4':
+ case '5': case '6': case '7': case '8': case '9':
{
numval = 0;
- while (isdigit(c))
+ p = token_buffer;
+ while (isdigit (c))
{
- numval = numval*10 + c - '0';
- c = getc(finput);
+ if (p == token_buffer + maxtoken)
+ p = grow_token_buffer (p);
+
+ *p++ = c;
+ numval = numval * 10 + c - '0';
+ c = getc (finput);
}
- ungetc(c, finput);
- return (NUMBER);
+ *p = 0;
+ ungetc (c, finput);
+ return NUMBER;
}
case '\'':
- translations = -1;
-
/* parse the literal token and compute character code in code */
- c = getc(finput);
+ translations = -1;
{
- register int code = 0;
+ int code, discode;
+ char discard[10], *dp;
- if (c == '\\')
- {
- c = getc(finput);
+ p = token_buffer;
+ *p++ = '\'';
+ literalchar (&p, &code, '\'');
- if (c <= '7' && c >= '0')
- {
- while (c <= '7' && c >= '0')
- {
- code = (code * 8) + (c - '0');
- c = getc(finput);
- if (code >= 256 || code < 0)
- fatals("malformatted literal token `\\%03o'", code);
- }
- }
- else
+ c = getc (finput);
+ if (c != '\'')
+ {
+ complain (_("use \"...\" for multi-character literal tokens"));
+ while (1)
{
- if (c == 't')
- code = '\t';
- else if (c == 'n')
- code = '\n';
- else if (c == 'a')
- code = '\007';
- else if (c == 'r')
- code = '\r';
- else if (c == 'f')
- code = '\f';
- else if (c == 'b')
- code = '\b';
- else if (c == 'v')
- code = 013;
- else if (c == 'x')
- {
- c = getc(finput);
- while ((c <= '9' && c >= '0')
- || (c >= 'a' && c <= 'z')
- || (c >= 'A' && c <= 'Z'))
- {
- code *= 16;
- if (c <= '9' && c >= '0')
- code += c - '0';
- else if (c >= 'a' && c <= 'z')
- code += c - 'a' + 10;
- else if (c >= 'A' && c <= 'Z')
- code += c - 'A' + 10;
- if (code >= 256 || code<0)/* JF this said if(c>=128) */
- fatals("malformatted literal token `\\x%x'",code);
- c = getc(finput);
- }
- ungetc(c, finput);
- }
- else if (c == '\\')
- code = '\\';
- else if (c == '\'')
- code = '\'';
- else if (c == '\"') /* JF this is a good idea */
- code = '\"';
- else
- {
- if (c >= 040 && c <= 0177)
- fatals ("unknown escape sequence `\\%c'", c);
- else
- fatals ("unknown escape sequence: `\\' followed by char code 0x%x", c);
- }
-
- c = getc(finput);
+ dp = discard;
+ if (!literalchar (&dp, &discode, '\''))
+ break;
}
}
- else
- {
- code = c;
- c = getc(finput);
- }
- if (c != '\'')
- fatal("multicharacter literal tokens not supported");
+ *p++ = '\'';
+ *p = 0;
+ symval = getsym (token_buffer);
+ symval->class = STOKEN;
+ if (!symval->user_token_number)
+ symval->user_token_number = code;
+ return IDENTIFIER;
+ }
- /* now fill token_buffer with the canonical name for this character
- as a literal token. Do not use what the user typed,
- so that '\012' and '\n' can be interchangeable. */
+ case '\"':
+ /* parse the literal string token and treat as an identifier */
+ translations = -1;
+ {
+ int code; /* ignored here */
p = token_buffer;
- *p++ = '\'';
- if (code == '\\')
- {
- *p++ = '\\';
- *p++ = '\\';
- }
- else if (code == '\'')
- {
- *p++ = '\\';
- *p++ = '\'';
- }
- else if (code >= 040 && code != 0177)
- *p++ = code;
- else if (code == '\t')
- {
- *p++ = '\\';
- *p++ = 't';
- }
- else if (code == '\n')
+ *p++ = '\"';
+ while (literalchar (&p, &code, '\"')) /* read up to and including " */
{
- *p++ = '\\';
- *p++ = 'n';
+ if (p >= token_buffer + maxtoken - 4)
+ p = grow_token_buffer (p);
}
- else if (code == '\r')
- {
- *p++ = '\\';
- *p++ = 'r';
- }
- else if (code == '\v')
- {
- *p++ = '\\';
- *p++ = 'v';
- }
- else if (code == '\b')
- {
- *p++ = '\\';
- *p++ = 'b';
- }
- else if (code == '\f')
- {
- *p++ = '\\';
- *p++ = 'f';
- }
- else
- {
- *p++ = code / 0100 + '0';
- *p++ = ((code / 010) & 07) + '0';
- *p++ = (code & 07) + '0';
- }
- *p++ = '\'';
*p = 0;
- symval = getsym(token_buffer);
+
+ symval = getsym (token_buffer);
symval->class = STOKEN;
- if (! symval->user_token_number)
- symval->user_token_number = code;
- return (IDENTIFIER);
+
+ return IDENTIFIER;
}
case ',':
- return (COMMA);
+ return COMMA;
case ':':
- return (COLON);
+ return COLON;
case ';':
- return (SEMICOLON);
+ return SEMICOLON;
case '|':
- return (BAR);
+ return BAR;
case '{':
- return (LEFT_CURLY);
+ return LEFT_CURLY;
case '=':
do
{
- c = getc(finput);
- if (c == '\n') lineno++;
+ c = getc (finput);
+ if (c == '\n')
+ lineno++;
}
- while(c==' ' || c=='\n' || c=='\t');
+ while (c == ' ' || c == '\n' || c == '\t');
if (c == '{')
- return(LEFT_CURLY);
+ {
+ strcpy (token_buffer, "={");
+ return LEFT_CURLY;
+ }
else
{
- ungetc(c, finput);
- return(ILLEGAL);
+ ungetc (c, finput);
+ return ILLEGAL;
}
case '<':
p = token_buffer;
- c = getc(finput);
+ c = getc (finput);
while (c != '>')
{
- if (c == '\n' || c == EOF)
- fatal("unterminated type name");
+ if (c == EOF)
+ fatal (_("unterminated type name at end of file"));
+ if (c == '\n')
+ {
+ complain (_("unterminated type name"));
+ ungetc (c, finput);
+ break;
+ }
if (p == token_buffer + maxtoken)
- p = grow_token_buffer(p);
+ p = grow_token_buffer (p);
*p++ = c;
- c = getc(finput);
+ c = getc (finput);
}
*p = 0;
- return (TYPENAME);
-
+ return TYPENAME;
+
case '%':
- return (parse_percent_token());
+ return parse_percent_token ();
default:
- return (ILLEGAL);
+ return ILLEGAL;
}
}
-
-/* parse a token which starts with %. Assumes the % has already been read and discarded. */
+/* the following table dictates the action taken for the various %
+ directives. A setflag value causes the named flag to be set. A
+ retval action returns the code. */
+struct percent_table_struct
+{
+ const char *name;
+ void *setflag;
+ int retval;
+}
+percent_table[] =
+{
+ { "token", NULL, TOKEN },
+ { "term", NULL, TOKEN },
+ { "nterm", NULL, NTERM },
+ { "type", NULL, TYPE },
+ { "guard", NULL, GUARD },
+ { "union", NULL, UNION },
+ { "expect", NULL, EXPECT },
+ { "thong", NULL, THONG },
+ { "start", NULL, START },
+ { "left", NULL, LEFT },
+ { "right", NULL, RIGHT },
+ { "nonassoc", NULL, NONASSOC },
+ { "binary", NULL, NONASSOC },
+ { "semantic_parser", NULL, SEMANTIC_PARSER },
+ { "pure_parser", NULL, PURE_PARSER },
+ { "prec", NULL, PREC },
+ { "no_lines", &nolinesflag, NOOP}, /* -l */
+ { "raw", &rawtoknumflag, NOOP }, /* -r */
+ { "token_table", &toknumflag, NOOP}, /* -k */
+#if 0
+ /* These can be utilized after main is reoganized so
+ open_files() is deferred 'til after read_declarations().
+ But %{ and %union both put information into files
+ that have to be opened before read_declarations().
+ */
+ { "yacc", &fixed_outfiles, NOOP}, /* -y */
+ { "fixed_output_files", &fixed_outfiles, NOOP}, /* -y */
+ { "defines", &definesflag, NOOP}, /* -d */
+ { "no_parser", &noparserflag, NOOP}, /* -n */
+ { "output_file", &spec_outfile, SETOPT}, /* -o */
+ { "file_prefix", &spec_file_prefix, SETOPT}, /* -b */
+ { "name_prefix", &spec_name_prefix, SETOPT}, /* -p */
+ /* These would be acceptable, but they do not affect processing */
+ { "verbose", &verboseflag, NOOP}, /* -v */
+ { "debug", &debugflag, NOOP}, /* -t */
+/* {"help", <print usage stmt>, NOOP}, *//* -h */
+/* {"version", <print version number> , NOOP}, *//* -V */
+#endif
+ { NULL, NULL, ILLEGAL}
+};
+
+/* Parse a token which starts with %.
+ Assumes the % has already been read and discarded. */
int
-parse_percent_token ()
+parse_percent_token (void)
{
- register int c;
- register char *p;
+ int c;
+ char *p;
+ struct percent_table_struct *tx;
p = token_buffer;
- c = getc(finput);
+ c = getc (finput);
+ *p++ = '%';
+ *p++ = c; /* for error msg */
+ *p = 0;
switch (c)
{
case '%':
- return (TWO_PERCENTS);
+ return TWO_PERCENTS;
case '{':
- return (PERCENT_LEFT_CURLY);
+ return PERCENT_LEFT_CURLY;
case '<':
- return (LEFT);
+ return LEFT;
case '>':
- return (RIGHT);
+ return RIGHT;
case '2':
- return (NONASSOC);
+ return NONASSOC;
case '0':
- return (TOKEN);
+ return TOKEN;
case '=':
- return (PREC);
+ return PREC;
}
- if (!isalpha(c))
- return (ILLEGAL);
+ if (!isalpha (c))
+ return ILLEGAL;
- while (isalpha(c) || c == '_')
+ p = token_buffer;
+ *p++ = '%';
+ while (isalpha (c) || c == '_' || c == '-')
{
if (p == token_buffer + maxtoken)
- p = grow_token_buffer(p);
+ p = grow_token_buffer (p);
+ if (c == '-')
+ c = '_';
*p++ = c;
- c = getc(finput);
+ c = getc (finput);
}
- ungetc(c, finput);
+ ungetc (c, finput);
*p = 0;
- if (strcmp(token_buffer, "token") == 0
- ||
- strcmp(token_buffer, "term") == 0)
- return (TOKEN);
- else if (strcmp(token_buffer, "nterm") == 0)
- return (NTERM);
- else if (strcmp(token_buffer, "type") == 0)
- return (TYPE);
- else if (strcmp(token_buffer, "guard") == 0)
- return (GUARD);
- else if (strcmp(token_buffer, "union") == 0)
- return (UNION);
- else if (strcmp(token_buffer, "expect") == 0)
- return (EXPECT);
- else if (strcmp(token_buffer, "start") == 0)
- return (START);
- else if (strcmp(token_buffer, "left") == 0)
- return (LEFT);
- else if (strcmp(token_buffer, "right") == 0)
- return (RIGHT);
- else if (strcmp(token_buffer, "nonassoc") == 0
- ||
- strcmp(token_buffer, "binary") == 0)
- return (NONASSOC);
- else if (strcmp(token_buffer, "semantic_parser") == 0)
- return (SEMANTIC_PARSER);
- else if (strcmp(token_buffer, "pure_parser") == 0)
- return (PURE_PARSER);
- else if (strcmp(token_buffer, "prec") == 0)
- return (PREC);
- else return (ILLEGAL);
+ /* table lookup % directive */
+ for (tx = percent_table; tx->name; tx++)
+ if (strcmp (token_buffer + 1, tx->name) == 0)
+ break;
+ if (tx->retval == SETOPT)
+ {
+ *((char **) (tx->setflag)) = optarg;
+ return NOOP;
+ }
+ if (tx->setflag)
+ {
+ *((int *) (tx->setflag)) = 1;
+ return NOOP;
+ }
+ return tx->retval;
}
projects / bison.git / blobdiff