| /*********************************************************** |
| Copyright 1991 by Stichting Mathematisch Centrum, Amsterdam, The |
| Netherlands. |
| |
| All Rights Reserved |
| |
| Permission to use, copy, modify, and distribute this software and its |
| documentation for any purpose and without fee is hereby granted, |
| provided that the above copyright notice appear in all copies and that |
| both that copyright notice and this permission notice appear in |
| supporting documentation, and that the names of Stichting Mathematisch |
| Centrum or CWI not be used in advertising or publicity pertaining to |
| distribution of the software without specific, written prior permission. |
| |
| STICHTING MATHEMATISCH CENTRUM DISCLAIMS ALL WARRANTIES WITH REGARD TO |
| THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND |
| FITNESS, IN NO EVENT SHALL STICHTING MATHEMATISCH CENTRUM BE LIABLE |
| FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES |
| WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN |
| ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT |
| OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. |
| |
| ******************************************************************/ |
| |
| /* Tokenizer implementation */ |
| |
| /* XXX This is rather old, should be restructured perhaps */ |
| /* XXX Need a better interface to report errors than writing to stderr */ |
| /* XXX Should use editor resource to fetch true tab size on Macintosh */ |
| |
| #include "pgenheaders.h" |
| |
| #include <ctype.h> |
| #include "string.h" |
| |
| #include "fgetsintr.h" |
| #include "tokenizer.h" |
| #include "errcode.h" |
| |
| #ifdef macintosh |
| #define TABSIZE 4 |
| #endif |
| |
| #ifndef TABSIZE |
| #define TABSIZE 8 |
| #endif |
| |
| /* Forward */ |
| static struct tok_state *tok_new PROTO((void)); |
| static int tok_nextc PROTO((struct tok_state *tok)); |
| static void tok_backup PROTO((struct tok_state *tok, int c)); |
| |
| /* Token names */ |
| |
| char *tok_name[] = { |
| "ENDMARKER", |
| "NAME", |
| "NUMBER", |
| "STRING", |
| "NEWLINE", |
| "INDENT", |
| "DEDENT", |
| "LPAR", |
| "RPAR", |
| "LSQB", |
| "RSQB", |
| "COLON", |
| "COMMA", |
| "SEMI", |
| "PLUS", |
| "MINUS", |
| "STAR", |
| "SLASH", |
| "VBAR", |
| "AMPER", |
| "LESS", |
| "GREATER", |
| "EQUAL", |
| "DOT", |
| "PERCENT", |
| "BACKQUOTE", |
| "LBRACE", |
| "RBRACE", |
| "EQEQUAL", |
| "NOTEQUAL", |
| "LESSEQUAL", |
| "GREATEREQUAL", |
| "TILDE", |
| "CIRCUMFLEX", |
| "LEFTSHIFT", |
| "RIGHTSHIFT", |
| /* This table must match the #defines in token.h! */ |
| "OP", |
| "<ERRORTOKEN>", |
| "<N_TOKENS>" |
| }; |
| |
| |
| /* Create and initialize a new tok_state structure */ |
| |
| static struct tok_state * |
| tok_new() |
| { |
| struct tok_state *tok = NEW(struct tok_state, 1); |
| if (tok == NULL) |
| return NULL; |
| tok->buf = tok->cur = tok->end = tok->inp = NULL; |
| tok->done = E_OK; |
| tok->fp = NULL; |
| tok->tabsize = TABSIZE; |
| tok->indent = 0; |
| tok->indstack[0] = 0; |
| tok->atbol = 1; |
| tok->pendin = 0; |
| tok->prompt = tok->nextprompt = NULL; |
| tok->lineno = 0; |
| return tok; |
| } |
| |
| |
| /* Set up tokenizer for string */ |
| |
| struct tok_state * |
| tok_setups(str) |
| char *str; |
| { |
| struct tok_state *tok = tok_new(); |
| if (tok == NULL) |
| return NULL; |
| tok->buf = tok->cur = str; |
| tok->end = tok->inp = strchr(str, '\0'); |
| return tok; |
| } |
| |
| |
| /* Set up tokenizer for file */ |
| |
| struct tok_state * |
| tok_setupf(fp, ps1, ps2) |
| FILE *fp; |
| char *ps1, *ps2; |
| { |
| struct tok_state *tok = tok_new(); |
| if (tok == NULL) |
| return NULL; |
| if ((tok->buf = NEW(char, BUFSIZ)) == NULL) { |
| DEL(tok); |
| return NULL; |
| } |
| tok->cur = tok->inp = tok->buf; |
| tok->end = tok->buf + BUFSIZ; |
| tok->fp = fp; |
| tok->prompt = ps1; |
| tok->nextprompt = ps2; |
| return tok; |
| } |
| |
| |
| /* Free a tok_state structure */ |
| |
| void |
| tok_free(tok) |
| struct tok_state *tok; |
| { |
| /* XXX really need a separate flag to say 'my buffer' */ |
| if (tok->fp != NULL && tok->buf != NULL) |
| DEL(tok->buf); |
| DEL(tok); |
| } |
| |
| |
| /* Get next char, updating state; error code goes into tok->done */ |
| |
| static int |
| tok_nextc(tok) |
| register struct tok_state *tok; |
| { |
| if (tok->done != E_OK) |
| return EOF; |
| |
| for (;;) { |
| if (tok->cur < tok->inp) |
| return *tok->cur++; |
| if (tok->fp == NULL) { |
| tok->done = E_EOF; |
| return EOF; |
| } |
| if (tok->inp > tok->buf && tok->inp[-1] == '\n') |
| tok->inp = tok->buf; |
| if (tok->inp == tok->end) { |
| int n = tok->end - tok->buf; |
| char *new = tok->buf; |
| RESIZE(new, char, n+n); |
| if (new == NULL) { |
| fprintf(stderr, "tokenizer out of mem\n"); |
| tok->done = E_NOMEM; |
| return EOF; |
| } |
| tok->buf = new; |
| tok->inp = tok->buf + n; |
| tok->end = tok->inp + n; |
| } |
| #ifdef USE_READLINE |
| if (tok->prompt != NULL) { |
| extern char *readline PROTO((char *prompt)); |
| static int been_here; |
| if (!been_here) { |
| /* Force rebind of TAB to insert-tab */ |
| extern int rl_insert(); |
| rl_bind_key('\t', rl_insert); |
| been_here++; |
| } |
| if (tok->buf != NULL) |
| free(tok->buf); |
| tok->buf = readline(tok->prompt); |
| (void) intrcheck(); /* Clear pending interrupt */ |
| if (tok->nextprompt != NULL) |
| tok->prompt = tok->nextprompt; |
| /* XXX different semantics w/o readline()! */ |
| if (tok->buf == NULL) { |
| tok->done = E_EOF; |
| } |
| else { |
| unsigned int n = strlen(tok->buf); |
| if (n > 0) |
| add_history(tok->buf); |
| /* Append the '\n' that readline() |
| doesn't give us, for the tokenizer... */ |
| tok->buf = realloc(tok->buf, n+2); |
| if (tok->buf == NULL) |
| tok->done = E_NOMEM; |
| else { |
| tok->end = tok->buf + n; |
| *tok->end++ = '\n'; |
| *tok->end = '\0'; |
| tok->inp = tok->end; |
| tok->cur = tok->buf; |
| } |
| } |
| } |
| else |
| #endif |
| { |
| tok->cur = tok->inp; |
| if (tok->prompt != NULL && tok->inp == tok->buf) { |
| fprintf(stderr, "%s", tok->prompt); |
| tok->prompt = tok->nextprompt; |
| } |
| tok->done = fgets_intr(tok->inp, |
| (int)(tok->end - tok->inp), tok->fp); |
| } |
| if (tok->done != E_OK) { |
| if (tok->prompt != NULL) |
| fprintf(stderr, "\n"); |
| return EOF; |
| } |
| tok->inp = strchr(tok->inp, '\0'); |
| } |
| } |
| |
| |
| /* Back-up one character */ |
| |
| static void |
| tok_backup(tok, c) |
| register struct tok_state *tok; |
| register int c; |
| { |
| if (c != EOF) { |
| if (--tok->cur < tok->buf) { |
| fprintf(stderr, "tok_backup: begin of buffer\n"); |
| abort(); |
| } |
| if (*tok->cur != c) |
| *tok->cur = c; |
| } |
| } |
| |
| |
| /* Return the token corresponding to a single character */ |
| |
| int |
| tok_1char(c) |
| int c; |
| { |
| switch (c) { |
| case '(': return LPAR; |
| case ')': return RPAR; |
| case '[': return LSQB; |
| case ']': return RSQB; |
| case ':': return COLON; |
| case ',': return COMMA; |
| case ';': return SEMI; |
| case '+': return PLUS; |
| case '-': return MINUS; |
| case '*': return STAR; |
| case '/': return SLASH; |
| case '|': return VBAR; |
| case '&': return AMPER; |
| case '<': return LESS; |
| case '>': return GREATER; |
| case '=': return EQUAL; |
| case '.': return DOT; |
| case '%': return PERCENT; |
| case '`': return BACKQUOTE; |
| case '{': return LBRACE; |
| case '}': return RBRACE; |
| case '^': return CIRCUMFLEX; |
| case '~': return TILDE; |
| default: return OP; |
| } |
| } |
| |
| |
| int |
| tok_2char(c1, c2) |
| int c1, c2; |
| { |
| switch (c1) { |
| case '=': |
| switch (c2) { |
| case '=': return EQEQUAL; |
| } |
| break; |
| case '!': |
| switch (c2) { |
| case '=': return NOTEQUAL; |
| } |
| break; |
| case '<': |
| switch (c2) { |
| case '>': return NOTEQUAL; |
| case '=': return LESSEQUAL; |
| case '<': return LEFTSHIFT; |
| } |
| break; |
| case '>': |
| switch (c2) { |
| case '=': return GREATEREQUAL; |
| case '>': return RIGHTSHIFT; |
| } |
| break; |
| } |
| return OP; |
| } |
| |
| |
| /* Get next token, after space stripping etc. */ |
| |
| int |
| tok_get(tok, p_start, p_end) |
| register struct tok_state *tok; /* In/out: tokenizer state */ |
| char **p_start, **p_end; /* Out: point to start/end of token */ |
| { |
| register int c; |
| int blankline; |
| |
| nextline: |
| blankline = 0; |
| |
| /* Get indentation level */ |
| if (tok->atbol) { |
| register int col = 0; |
| tok->atbol = 0; |
| tok->lineno++; |
| for (;;) { |
| c = tok_nextc(tok); |
| if (c == ' ') |
| col++; |
| else if (c == '\t') |
| col = (col/tok->tabsize + 1) * tok->tabsize; |
| else |
| break; |
| } |
| tok_backup(tok, c); |
| if (c == '#' || c == '\n') { |
| /* Lines with only whitespace and/or comments |
| shouldn't affect the indentation and are |
| not passed to the parser as NEWLINE tokens, |
| except *totally* empty lines in interactive |
| mode, which signal the end of a command group. */ |
| if (col == 0 && c == '\n' && tok->prompt != NULL) |
| blankline = 0; /* Let it through */ |
| else |
| blankline = 1; /* Ignore completely */ |
| /* We can't jump back right here since we still |
| may need to skip to the end of a comment */ |
| } |
| if (!blankline) { |
| if (col == tok->indstack[tok->indent]) { |
| /* No change */ |
| } |
| else if (col > tok->indstack[tok->indent]) { |
| /* Indent -- always one */ |
| if (tok->indent+1 >= MAXINDENT) { |
| fprintf(stderr, "excessive indent\n"); |
| tok->done = E_TOKEN; |
| return ERRORTOKEN; |
| } |
| tok->pendin++; |
| tok->indstack[++tok->indent] = col; |
| } |
| else /* col < tok->indstack[tok->indent] */ { |
| /* Dedent -- any number, must be consistent */ |
| while (tok->indent > 0 && |
| col < tok->indstack[tok->indent]) { |
| tok->indent--; |
| tok->pendin--; |
| } |
| if (col != tok->indstack[tok->indent]) { |
| fprintf(stderr, "inconsistent dedent\n"); |
| tok->done = E_TOKEN; |
| return ERRORTOKEN; |
| } |
| } |
| } |
| } |
| |
| *p_start = *p_end = tok->cur; |
| |
| /* Return pending indents/dedents */ |
| if (tok->pendin != 0) { |
| if (tok->pendin < 0) { |
| tok->pendin++; |
| return DEDENT; |
| } |
| else { |
| tok->pendin--; |
| return INDENT; |
| } |
| } |
| |
| again: |
| /* Skip spaces */ |
| do { |
| c = tok_nextc(tok); |
| } while (c == ' ' || c == '\t'); |
| |
| /* Set start of current token */ |
| *p_start = tok->cur - 1; |
| |
| /* Skip comment */ |
| if (c == '#') { |
| /* Hack to allow overriding the tabsize in the file. |
| This is also recognized by vi, when it occurs near the |
| beginning or end of the file. (Will vi never die...?) |
| For Python it must be at the beginning of the file! */ |
| int x; |
| /* XXX The cast to (unsigned char *) is needed by THINK C 3.0 */ |
| if (sscanf(/*(unsigned char *)*/tok->cur, |
| " vi:set tabsize=%d:", &x) == 1 && |
| x >= 1 && x <= 40) { |
| /* fprintf(stderr, "# vi:set tabsize=%d:\n", x); */ |
| tok->tabsize = x; |
| } |
| do { |
| c = tok_nextc(tok); |
| } while (c != EOF && c != '\n'); |
| } |
| |
| /* Check for EOF and errors now */ |
| if (c == EOF) { |
| *p_start = *p_end = tok->cur; |
| return tok->done == E_EOF ? ENDMARKER : ERRORTOKEN; |
| } |
| |
| /* Identifier (most frequent token!) */ |
| if (isalpha(c) || c == '_') { |
| do { |
| c = tok_nextc(tok); |
| } while (isalnum(c) || c == '_'); |
| tok_backup(tok, c); |
| *p_end = tok->cur; |
| return NAME; |
| } |
| |
| /* Newline */ |
| if (c == '\n') { |
| tok->atbol = 1; |
| if (blankline) |
| goto nextline; |
| *p_end = tok->cur - 1; /* Leave '\n' out of the string */ |
| return NEWLINE; |
| } |
| |
| /* Period or number starting with period? */ |
| if (c == '.') { |
| c = tok_nextc(tok); |
| if (isdigit(c)) { |
| goto fraction; |
| } |
| else { |
| tok_backup(tok, c); |
| *p_end = tok->cur; |
| return DOT; |
| } |
| } |
| |
| /* Number */ |
| if (isdigit(c)) { |
| if (c == '0') { |
| /* Hex or octal */ |
| c = tok_nextc(tok); |
| if (c == '.') |
| goto fraction; |
| if (c == 'x' || c == 'X') { |
| /* Hex */ |
| do { |
| c = tok_nextc(tok); |
| } while (isxdigit(c)); |
| } |
| else { |
| /* XXX This is broken! E.g., |
| 09.9 should be accepted as float! */ |
| /* Octal; c is first char of it */ |
| /* There's no 'isoctdigit' macro, sigh */ |
| while ('0' <= c && c < '8') { |
| c = tok_nextc(tok); |
| } |
| } |
| if (c == 'l' || c == 'L') |
| c = tok_nextc(tok); |
| } |
| else { |
| /* Decimal */ |
| do { |
| c = tok_nextc(tok); |
| } while (isdigit(c)); |
| if (c == 'l' || c == 'L') |
| c = tok_nextc(tok); |
| else { |
| /* Accept floating point numbers. |
| XXX This accepts incomplete things like |
| XXX 12e or 1e+; worry run-time. |
| XXX Doesn't accept numbers |
| XXX starting with a dot */ |
| if (c == '.') { |
| fraction: |
| /* Fraction */ |
| do { |
| c = tok_nextc(tok); |
| } while (isdigit(c)); |
| } |
| if (c == 'e' || c == 'E') { |
| /* Exponent part */ |
| c = tok_nextc(tok); |
| if (c == '+' || c == '-') |
| c = tok_nextc(tok); |
| while (isdigit(c)) { |
| c = tok_nextc(tok); |
| } |
| } |
| } |
| } |
| tok_backup(tok, c); |
| *p_end = tok->cur; |
| return NUMBER; |
| } |
| |
| /* String */ |
| if (c == '\'') { |
| for (;;) { |
| c = tok_nextc(tok); |
| if (c == '\n' || c == EOF) { |
| tok->done = E_TOKEN; |
| return ERRORTOKEN; |
| } |
| if (c == '\\') { |
| c = tok_nextc(tok); |
| *p_end = tok->cur; |
| if (c == '\n' || c == EOF) { |
| tok->done = E_TOKEN; |
| return ERRORTOKEN; |
| } |
| continue; |
| } |
| if (c == '\'') |
| break; |
| } |
| *p_end = tok->cur; |
| return STRING; |
| } |
| |
| /* Line continuation */ |
| if (c == '\\') { |
| c = tok_nextc(tok); |
| if (c != '\n') { |
| tok->done = E_TOKEN; |
| return ERRORTOKEN; |
| } |
| tok->lineno++; |
| goto again; /* Read next line */ |
| } |
| |
| /* Check for two-character token */ |
| { |
| int c2 = tok_nextc(tok); |
| int token = tok_2char(c, c2); |
| if (token != OP) { |
| *p_end = tok->cur; |
| return token; |
| } |
| tok_backup(tok, c2); |
| } |
| |
| /* Punctuation character */ |
| *p_end = tok->cur; |
| return tok_1char(c); |
| } |
| |
| |
| #ifdef DEBUG |
| |
| void |
| tok_dump(type, start, end) |
| int type; |
| char *start, *end; |
| { |
| printf("%s", tok_name[type]); |
| if (type == NAME || type == NUMBER || type == STRING || type == OP) |
| printf("(%.*s)", (int)(end - start), start); |
| } |
| |
| #endif |