1991-02-19 12:39:46 +00:00
|
|
|
|
1990-10-14 12:07:46 +00:00
|
|
|
/* Parser-tokenizer link implementation */
|
|
|
|
|
1990-12-20 15:06:42 +00:00
|
|
|
#include "pgenheaders.h"
|
1990-10-14 12:07:46 +00:00
|
|
|
#include "tokenizer.h"
|
|
|
|
#include "node.h"
|
|
|
|
#include "grammar.h"
|
|
|
|
#include "parser.h"
|
1990-12-20 15:06:42 +00:00
|
|
|
#include "parsetok.h"
|
1990-10-14 12:07:46 +00:00
|
|
|
#include "errcode.h"
|
2002-08-04 17:29:52 +00:00
|
|
|
#include "graminit.h"
|
1990-10-14 12:07:46 +00:00
|
|
|
|
1998-04-10 19:35:06 +00:00
|
|
|
int Py_TabcheckFlag;
|
|
|
|
|
1990-10-14 12:07:46 +00:00
|
|
|
|
1990-12-20 15:06:42 +00:00
|
|
|
/* Forward */
|
2001-07-16 05:37:24 +00:00
|
|
|
static node *parsetok(struct tok_state *, grammar *, int, perrdetail *, int);
|
2002-12-11 14:04:59 +00:00
|
|
|
static void initerr(perrdetail *err_ret, const char* filename);
|
1990-10-14 12:07:46 +00:00
|
|
|
|
1990-12-20 15:06:42 +00:00
|
|
|
/* Parse input coming from a string. Return error code, print some errors. */
|
1994-08-29 12:25:45 +00:00
|
|
|
node *
|
2002-12-11 14:04:59 +00:00
|
|
|
PyParser_ParseString(const char *s, grammar *g, int start, perrdetail *err_ret)
|
2001-07-16 05:37:24 +00:00
|
|
|
{
|
2005-10-20 19:59:25 +00:00
|
|
|
return PyParser_ParseStringFlagsFilename(s, NULL, g, start, err_ret, 0);
|
2001-07-16 05:37:24 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
node *
|
2002-12-11 14:04:59 +00:00
|
|
|
PyParser_ParseStringFlags(const char *s, grammar *g, int start,
|
2001-07-16 05:37:24 +00:00
|
|
|
perrdetail *err_ret, int flags)
|
2002-07-09 09:23:27 +00:00
|
|
|
{
|
|
|
|
return PyParser_ParseStringFlagsFilename(s, NULL,
|
2004-07-08 01:54:07 +00:00
|
|
|
g, start, err_ret, flags);
|
2002-07-09 09:23:27 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
node *
|
2002-12-11 14:04:59 +00:00
|
|
|
PyParser_ParseStringFlagsFilename(const char *s, const char *filename,
|
2002-07-09 09:23:27 +00:00
|
|
|
grammar *g, int start,
|
|
|
|
perrdetail *err_ret, int flags)
|
1990-10-14 12:07:46 +00:00
|
|
|
{
|
1994-08-29 12:25:45 +00:00
|
|
|
struct tok_state *tok;
|
|
|
|
|
2002-07-09 09:23:27 +00:00
|
|
|
initerr(err_ret, filename);
|
1994-08-29 12:25:45 +00:00
|
|
|
|
1997-04-29 21:03:06 +00:00
|
|
|
if ((tok = PyTokenizer_FromString(s)) == NULL) {
|
2005-10-02 01:48:49 +00:00
|
|
|
err_ret->error = PyErr_Occurred() ? E_DECODE : E_NOMEM;
|
1994-08-29 12:25:45 +00:00
|
|
|
return NULL;
|
1990-10-14 12:07:46 +00:00
|
|
|
}
|
1994-08-29 12:25:45 +00:00
|
|
|
|
2002-08-04 17:29:52 +00:00
|
|
|
tok->filename = filename ? filename : "<string>";
|
1998-12-21 18:32:40 +00:00
|
|
|
if (Py_TabcheckFlag || Py_VerboseFlag) {
|
|
|
|
tok->altwarning = (tok->filename != NULL);
|
|
|
|
if (Py_TabcheckFlag >= 2)
|
|
|
|
tok->alterror++;
|
|
|
|
}
|
|
|
|
|
2001-07-16 05:37:24 +00:00
|
|
|
return parsetok(tok, g, start, err_ret, flags);
|
1990-10-14 12:07:46 +00:00
|
|
|
}
|
|
|
|
|
1990-12-20 15:06:42 +00:00
|
|
|
/* Parse input coming from a file. Return error code, print some errors. */
|
1990-10-14 12:07:46 +00:00
|
|
|
|
1994-08-29 12:25:45 +00:00
|
|
|
node *
|
2002-12-11 14:04:59 +00:00
|
|
|
PyParser_ParseFile(FILE *fp, const char *filename, grammar *g, int start,
|
2000-07-22 19:20:54 +00:00
|
|
|
char *ps1, char *ps2, perrdetail *err_ret)
|
2001-07-16 05:37:24 +00:00
|
|
|
{
|
|
|
|
return PyParser_ParseFileFlags(fp, filename, g, start, ps1, ps2,
|
|
|
|
err_ret, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
node *
|
2002-12-11 14:04:59 +00:00
|
|
|
PyParser_ParseFileFlags(FILE *fp, const char *filename, grammar *g, int start,
|
2001-07-16 05:37:24 +00:00
|
|
|
char *ps1, char *ps2, perrdetail *err_ret, int flags)
|
1990-10-14 12:07:46 +00:00
|
|
|
{
|
1994-08-29 12:25:45 +00:00
|
|
|
struct tok_state *tok;
|
|
|
|
|
2002-01-05 21:40:08 +00:00
|
|
|
initerr(err_ret, filename);
|
1994-08-29 12:25:45 +00:00
|
|
|
|
1997-04-29 21:03:06 +00:00
|
|
|
if ((tok = PyTokenizer_FromFile(fp, ps1, ps2)) == NULL) {
|
1994-08-29 12:25:45 +00:00
|
|
|
err_ret->error = E_NOMEM;
|
|
|
|
return NULL;
|
1990-10-14 12:07:46 +00:00
|
|
|
}
|
2002-08-04 17:29:52 +00:00
|
|
|
tok->filename = filename;
|
1998-04-10 19:35:06 +00:00
|
|
|
if (Py_TabcheckFlag || Py_VerboseFlag) {
|
|
|
|
tok->altwarning = (filename != NULL);
|
|
|
|
if (Py_TabcheckFlag >= 2)
|
|
|
|
tok->alterror++;
|
|
|
|
}
|
1994-08-29 12:25:45 +00:00
|
|
|
|
1990-12-20 15:06:42 +00:00
|
|
|
|
2001-07-16 05:37:24 +00:00
|
|
|
return parsetok(tok, g, start, err_ret, flags);
|
1994-08-29 12:25:45 +00:00
|
|
|
}
|
1990-12-20 15:06:42 +00:00
|
|
|
|
|
|
|
/* Parse input coming from the given tokenizer structure.
|
|
|
|
Return error code. */
|
|
|
|
|
2006-02-28 19:02:24 +00:00
|
|
|
static char with_msg[] =
|
|
|
|
"%s:%d: Warning: 'with' will become a reserved keyword in Python 2.6\n";
|
|
|
|
|
|
|
|
static char as_msg[] =
|
|
|
|
"%s:%d: Warning: 'as' will become a reserved keyword in Python 2.6\n";
|
|
|
|
|
|
|
|
static void
|
|
|
|
warn(const char *msg, const char *filename, int lineno)
|
|
|
|
{
|
|
|
|
if (filename == NULL)
|
|
|
|
filename = "<string>";
|
|
|
|
PySys_WriteStderr(msg, filename, lineno);
|
|
|
|
}
|
2001-07-17 16:53:11 +00:00
|
|
|
|
1994-08-29 12:25:45 +00:00
|
|
|
static node *
|
2001-07-16 05:37:24 +00:00
|
|
|
parsetok(struct tok_state *tok, grammar *g, int start, perrdetail *err_ret,
|
|
|
|
int flags)
|
1990-12-20 15:06:42 +00:00
|
|
|
{
|
|
|
|
parser_state *ps;
|
1994-08-29 12:25:45 +00:00
|
|
|
node *n;
|
2006-02-28 19:02:24 +00:00
|
|
|
int started = 0, handling_import = 0, handling_with = 0;
|
1994-08-29 12:25:45 +00:00
|
|
|
|
1997-04-29 21:03:06 +00:00
|
|
|
if ((ps = PyParser_New(g, start)) == NULL) {
|
1990-12-20 15:06:42 +00:00
|
|
|
fprintf(stderr, "no mem for new parser\n");
|
1994-08-29 12:25:45 +00:00
|
|
|
err_ret->error = E_NOMEM;
|
2005-11-15 04:54:23 +00:00
|
|
|
PyTokenizer_Free(tok);
|
1994-08-29 12:25:45 +00:00
|
|
|
return NULL;
|
1990-12-20 15:06:42 +00:00
|
|
|
}
|
2006-02-28 19:02:24 +00:00
|
|
|
#ifdef PY_PARSER_REQUIRES_FUTURE_KEYWORD
|
|
|
|
if (flags & PyPARSE_WITH_IS_KEYWORD)
|
|
|
|
ps->p_flags |= CO_FUTURE_WITH_STATEMENT;
|
2002-03-22 23:53:36 +00:00
|
|
|
#endif
|
1994-08-29 12:25:45 +00:00
|
|
|
|
1990-12-20 15:06:42 +00:00
|
|
|
for (;;) {
|
|
|
|
char *a, *b;
|
|
|
|
int type;
|
2000-06-28 22:00:02 +00:00
|
|
|
size_t len;
|
1990-12-20 15:06:42 +00:00
|
|
|
char *str;
|
1994-08-29 12:25:45 +00:00
|
|
|
|
1997-04-29 21:03:06 +00:00
|
|
|
type = PyTokenizer_Get(tok, &a, &b);
|
1990-12-20 15:06:42 +00:00
|
|
|
if (type == ERRORTOKEN) {
|
1994-08-29 12:25:45 +00:00
|
|
|
err_ret->error = tok->done;
|
1990-12-20 15:06:42 +00:00
|
|
|
break;
|
|
|
|
}
|
1992-03-04 16:40:44 +00:00
|
|
|
if (type == ENDMARKER && started) {
|
|
|
|
type = NEWLINE; /* Add an extra newline */
|
2006-02-28 19:02:24 +00:00
|
|
|
handling_with = handling_import = 0;
|
1992-03-04 16:40:44 +00:00
|
|
|
started = 0;
|
2003-02-13 22:07:59 +00:00
|
|
|
/* Add the right number of dedent tokens,
|
|
|
|
except if a certain flag is given --
|
|
|
|
codeop.py uses this. */
|
|
|
|
if (tok->indent &&
|
|
|
|
!(flags & PyPARSE_DONT_IMPLY_DEDENT))
|
|
|
|
{
|
|
|
|
tok->pendin = -tok->indent;
|
|
|
|
tok->indent = 0;
|
|
|
|
}
|
1992-03-04 16:40:44 +00:00
|
|
|
}
|
|
|
|
else
|
|
|
|
started = 1;
|
1994-08-29 12:25:45 +00:00
|
|
|
len = b - a; /* XXX this may compute NULL - NULL */
|
2002-08-04 06:28:21 +00:00
|
|
|
str = (char *) PyObject_MALLOC(len + 1);
|
1990-12-20 15:06:42 +00:00
|
|
|
if (str == NULL) {
|
|
|
|
fprintf(stderr, "no mem for next token\n");
|
1994-08-29 12:25:45 +00:00
|
|
|
err_ret->error = E_NOMEM;
|
1990-12-20 15:06:42 +00:00
|
|
|
break;
|
|
|
|
}
|
1994-08-29 12:25:45 +00:00
|
|
|
if (len > 0)
|
|
|
|
strncpy(str, a, len);
|
1990-12-20 15:06:42 +00:00
|
|
|
str[len] = '\0';
|
2001-07-17 16:53:11 +00:00
|
|
|
|
2006-02-28 19:02:24 +00:00
|
|
|
#ifdef PY_PARSER_REQUIRES_FUTURE_KEYWORD
|
|
|
|
/* This is only necessary to support the "as" warning, but
|
|
|
|
we don't want to warn about "as" in import statements. */
|
|
|
|
if (type == NAME &&
|
|
|
|
len == 6 && str[0] == 'i' && strcmp(str, "import") == 0)
|
|
|
|
handling_import = 1;
|
|
|
|
|
|
|
|
/* Warn about with as NAME */
|
|
|
|
if (type == NAME &&
|
|
|
|
!(ps->p_flags & CO_FUTURE_WITH_STATEMENT)) {
|
|
|
|
if (len == 4 && str[0] == 'w' && strcmp(str, "with") == 0)
|
|
|
|
warn(with_msg, err_ret->filename, tok->lineno);
|
|
|
|
else if (!(handling_import || handling_with) &&
|
2006-02-28 22:42:15 +00:00
|
|
|
len == 2 && str[0] == 'a' &&
|
|
|
|
strcmp(str, "as") == 0)
|
2006-02-28 19:02:24 +00:00
|
|
|
warn(as_msg, err_ret->filename, tok->lineno);
|
|
|
|
}
|
|
|
|
else if (type == NAME &&
|
|
|
|
(ps->p_flags & CO_FUTURE_WITH_STATEMENT) &&
|
|
|
|
len == 4 && str[0] == 'w' && strcmp(str, "with") == 0)
|
|
|
|
handling_with = 1;
|
2002-03-22 23:53:36 +00:00
|
|
|
#endif
|
2001-07-17 16:53:11 +00:00
|
|
|
|
1994-08-29 12:25:45 +00:00
|
|
|
if ((err_ret->error =
|
2000-07-11 17:53:00 +00:00
|
|
|
PyParser_AddToken(ps, (int)type, str, tok->lineno,
|
|
|
|
&(err_ret->expected))) != E_OK) {
|
1997-07-27 01:52:50 +00:00
|
|
|
if (err_ret->error != E_DONE)
|
2002-08-04 06:28:21 +00:00
|
|
|
PyObject_FREE(str);
|
1990-12-20 15:06:42 +00:00
|
|
|
break;
|
1997-07-27 01:52:50 +00:00
|
|
|
}
|
1990-12-20 15:06:42 +00:00
|
|
|
}
|
1994-08-29 12:25:45 +00:00
|
|
|
|
|
|
|
if (err_ret->error == E_DONE) {
|
|
|
|
n = ps->p_tree;
|
|
|
|
ps->p_tree = NULL;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
n = NULL;
|
|
|
|
|
1997-04-29 21:03:06 +00:00
|
|
|
PyParser_Delete(ps);
|
1994-08-29 12:25:45 +00:00
|
|
|
|
|
|
|
if (n == NULL) {
|
|
|
|
if (tok->lineno <= 1 && tok->done == E_EOF)
|
|
|
|
err_ret->error = E_EOF;
|
|
|
|
err_ret->lineno = tok->lineno;
|
|
|
|
if (tok->buf != NULL) {
|
2006-02-15 19:56:18 +00:00
|
|
|
size_t len;
|
2006-02-15 17:27:45 +00:00
|
|
|
assert(tok->cur - tok->buf < INT_MAX);
|
|
|
|
err_ret->offset = (int)(tok->cur - tok->buf);
|
2006-02-15 19:56:18 +00:00
|
|
|
len = tok->inp - tok->buf;
|
2002-08-04 06:28:21 +00:00
|
|
|
err_ret->text = (char *) PyObject_MALLOC(len + 1);
|
1994-08-29 12:25:45 +00:00
|
|
|
if (err_ret->text != NULL) {
|
1995-01-20 16:59:12 +00:00
|
|
|
if (len > 0)
|
|
|
|
strncpy(err_ret->text, tok->buf, len);
|
1994-08-29 12:25:45 +00:00
|
|
|
err_ret->text[len] = '\0';
|
|
|
|
}
|
|
|
|
}
|
2002-08-04 17:29:52 +00:00
|
|
|
} else if (tok->encoding != NULL) {
|
|
|
|
node* r = PyNode_New(encoding_decl);
|
|
|
|
r->n_str = tok->encoding;
|
|
|
|
r->n_nchildren = 1;
|
|
|
|
r->n_child = n;
|
|
|
|
tok->encoding = NULL;
|
|
|
|
n = r;
|
1994-08-29 12:25:45 +00:00
|
|
|
}
|
|
|
|
|
1997-04-29 21:03:06 +00:00
|
|
|
PyTokenizer_Free(tok);
|
1994-08-29 12:25:45 +00:00
|
|
|
|
|
|
|
return n;
|
1990-12-20 15:06:42 +00:00
|
|
|
}
|
2001-10-20 14:27:56 +00:00
|
|
|
|
|
|
|
static void
|
2005-10-20 19:59:25 +00:00
|
|
|
initerr(perrdetail *err_ret, const char *filename)
|
2001-10-20 14:27:56 +00:00
|
|
|
{
|
|
|
|
err_ret->error = E_OK;
|
2002-01-05 21:40:08 +00:00
|
|
|
err_ret->filename = filename;
|
2001-10-20 14:27:56 +00:00
|
|
|
err_ret->lineno = 0;
|
|
|
|
err_ret->offset = 0;
|
|
|
|
err_ret->text = NULL;
|
|
|
|
err_ret->token = -1;
|
|
|
|
err_ret->expected = -1;
|
|
|
|
}
|