mirror of
https://github.com/python/cpython
synced 2024-09-16 02:40:00 +00:00
bpo-25643: Refactor the C tokenizer into smaller, logical units (GH-25050)
This commit is contained in:
parent
fb1d01b963
commit
261a452a13
|
@ -55,13 +55,13 @@ def test_line_continuation_EOF_from_file_bpo2180(self):
|
||||||
file_name = script_helper.make_script(temp_dir, 'foo', '\\')
|
file_name = script_helper.make_script(temp_dir, 'foo', '\\')
|
||||||
rc, out, err = script_helper.assert_python_failure(file_name)
|
rc, out, err = script_helper.assert_python_failure(file_name)
|
||||||
self.assertIn(b'unexpected EOF while parsing', err)
|
self.assertIn(b'unexpected EOF while parsing', err)
|
||||||
self.assertIn(b'line 2', err)
|
self.assertIn(b'line 1', err)
|
||||||
self.assertIn(b'\\', err)
|
self.assertIn(b'\\', err)
|
||||||
|
|
||||||
file_name = script_helper.make_script(temp_dir, 'foo', 'y = 6\\')
|
file_name = script_helper.make_script(temp_dir, 'foo', 'y = 6\\')
|
||||||
rc, out, err = script_helper.assert_python_failure(file_name)
|
rc, out, err = script_helper.assert_python_failure(file_name)
|
||||||
self.assertIn(b'unexpected EOF while parsing', err)
|
self.assertIn(b'unexpected EOF while parsing', err)
|
||||||
self.assertIn(b'line 2', err)
|
self.assertIn(b'line 1', err)
|
||||||
self.assertIn(b'y = 6\\', err)
|
self.assertIn(b'y = 6\\', err)
|
||||||
|
|
||||||
if __name__ == "__main__":
|
if __name__ == "__main__":
|
||||||
|
|
|
@ -205,6 +205,23 @@ def test_utf8_bom_and_utf8_coding_line(self):
|
||||||
b'print(ascii("\xc3\xa4"))\n')
|
b'print(ascii("\xc3\xa4"))\n')
|
||||||
self.check_script_output(src, br"'\xe4'")
|
self.check_script_output(src, br"'\xe4'")
|
||||||
|
|
||||||
|
def test_crlf(self):
|
||||||
|
src = (b'print(ascii("""\r\n"""))\n')
|
||||||
|
out = self.check_script_output(src, br"'\n'")
|
||||||
|
|
||||||
|
def test_crcrlf(self):
|
||||||
|
src = (b'print(ascii("""\r\r\n"""))\n')
|
||||||
|
out = self.check_script_output(src, br"'\n\n'")
|
||||||
|
|
||||||
|
def test_crcrcrlf(self):
|
||||||
|
src = (b'print(ascii("""\r\r\r\n"""))\n')
|
||||||
|
out = self.check_script_output(src, br"'\n\n\n'")
|
||||||
|
|
||||||
|
def test_crcrcrlf2(self):
|
||||||
|
src = (b'#coding:iso-8859-1\n'
|
||||||
|
b'print(ascii("""\r\r\r\n"""))\n')
|
||||||
|
out = self.check_script_output(src, br"'\n\n\n'")
|
||||||
|
|
||||||
|
|
||||||
class BytesSourceEncodingTest(AbstractSourceEncodingTest, unittest.TestCase):
|
class BytesSourceEncodingTest(AbstractSourceEncodingTest, unittest.TestCase):
|
||||||
|
|
||||||
|
|
|
@ -13,7 +13,6 @@
|
||||||
#include "unicodeobject.h"
|
#include "unicodeobject.h"
|
||||||
#include "bytesobject.h"
|
#include "bytesobject.h"
|
||||||
#include "fileobject.h"
|
#include "fileobject.h"
|
||||||
#include "codecs.h"
|
|
||||||
#include "abstract.h"
|
#include "abstract.h"
|
||||||
|
|
||||||
/* Alternate tab spacing */
|
/* Alternate tab spacing */
|
||||||
|
@ -75,7 +74,6 @@ tok_new(void)
|
||||||
tok->altindstack[0] = 0;
|
tok->altindstack[0] = 0;
|
||||||
tok->decoding_state = STATE_INIT;
|
tok->decoding_state = STATE_INIT;
|
||||||
tok->decoding_erred = 0;
|
tok->decoding_erred = 0;
|
||||||
tok->read_coding_spec = 0;
|
|
||||||
tok->enc = NULL;
|
tok->enc = NULL;
|
||||||
tok->encoding = NULL;
|
tok->encoding = NULL;
|
||||||
tok->cont_line = 0;
|
tok->cont_line = 0;
|
||||||
|
@ -164,14 +162,14 @@ get_coding_spec(const char *s, char **spec, Py_ssize_t size, struct tok_state *t
|
||||||
}
|
}
|
||||||
for (; i < size - 6; i++) { /* XXX inefficient search */
|
for (; i < size - 6; i++) { /* XXX inefficient search */
|
||||||
const char* t = s + i;
|
const char* t = s + i;
|
||||||
if (strncmp(t, "coding", 6) == 0) {
|
if (memcmp(t, "coding", 6) == 0) {
|
||||||
const char* begin = NULL;
|
const char* begin = NULL;
|
||||||
t += 6;
|
t += 6;
|
||||||
if (t[0] != ':' && t[0] != '=')
|
if (t[0] != ':' && t[0] != '=')
|
||||||
continue;
|
continue;
|
||||||
do {
|
do {
|
||||||
t++;
|
t++;
|
||||||
} while (t[0] == '\x20' || t[0] == '\t');
|
} while (t[0] == ' ' || t[0] == '\t');
|
||||||
|
|
||||||
begin = t;
|
begin = t;
|
||||||
while (Py_ISALNUM(t[0]) ||
|
while (Py_ISALNUM(t[0]) ||
|
||||||
|
@ -208,15 +206,14 @@ check_coding_spec(const char* line, Py_ssize_t size, struct tok_state *tok,
|
||||||
int set_readline(struct tok_state *, const char *))
|
int set_readline(struct tok_state *, const char *))
|
||||||
{
|
{
|
||||||
char *cs;
|
char *cs;
|
||||||
int r = 1;
|
|
||||||
|
|
||||||
if (tok->cont_line) {
|
if (tok->cont_line) {
|
||||||
/* It's a continuation line, so it can't be a coding spec. */
|
/* It's a continuation line, so it can't be a coding spec. */
|
||||||
tok->read_coding_spec = 1;
|
tok->decoding_state = STATE_NORMAL;
|
||||||
return 1;
|
return 1;
|
||||||
}
|
}
|
||||||
if (!get_coding_spec(line, &cs, size, tok))
|
if (!get_coding_spec(line, &cs, size, tok)) {
|
||||||
return 0;
|
return 0;
|
||||||
|
}
|
||||||
if (!cs) {
|
if (!cs) {
|
||||||
Py_ssize_t i;
|
Py_ssize_t i;
|
||||||
for (i = 0; i < size; i++) {
|
for (i = 0; i < size; i++) {
|
||||||
|
@ -225,37 +222,33 @@ check_coding_spec(const char* line, Py_ssize_t size, struct tok_state *tok,
|
||||||
if (line[i] != ' ' && line[i] != '\t' && line[i] != '\014') {
|
if (line[i] != ' ' && line[i] != '\t' && line[i] != '\014') {
|
||||||
/* Stop checking coding spec after a line containing
|
/* Stop checking coding spec after a line containing
|
||||||
* anything except a comment. */
|
* anything except a comment. */
|
||||||
tok->read_coding_spec = 1;
|
tok->decoding_state = STATE_NORMAL;
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
return 1;
|
return 1;
|
||||||
}
|
}
|
||||||
tok->read_coding_spec = 1;
|
|
||||||
if (tok->encoding == NULL) {
|
|
||||||
assert(tok->decoding_state == STATE_RAW);
|
|
||||||
if (strcmp(cs, "utf-8") == 0) {
|
|
||||||
tok->encoding = cs;
|
|
||||||
} else {
|
|
||||||
r = set_readline(tok, cs);
|
|
||||||
if (r) {
|
|
||||||
tok->encoding = cs;
|
|
||||||
tok->decoding_state = STATE_NORMAL;
|
tok->decoding_state = STATE_NORMAL;
|
||||||
}
|
if (tok->encoding == NULL) {
|
||||||
else {
|
assert(tok->decoding_readline == NULL);
|
||||||
PyErr_Format(PyExc_SyntaxError,
|
if (strcmp(cs, "utf-8") != 0 && !set_readline(tok, cs)) {
|
||||||
"encoding problem: %s", cs);
|
error_ret(tok);
|
||||||
|
PyErr_Format(PyExc_SyntaxError, "encoding problem: %s", cs);
|
||||||
PyMem_Free(cs);
|
PyMem_Free(cs);
|
||||||
|
return 0;
|
||||||
}
|
}
|
||||||
}
|
tok->encoding = cs;
|
||||||
} else { /* then, compare cs with BOM */
|
} else { /* then, compare cs with BOM */
|
||||||
r = (strcmp(tok->encoding, cs) == 0);
|
if (strcmp(tok->encoding, cs) != 0) {
|
||||||
if (!r)
|
error_ret(tok);
|
||||||
PyErr_Format(PyExc_SyntaxError,
|
PyErr_Format(PyExc_SyntaxError,
|
||||||
"encoding problem: %s with BOM", cs);
|
"encoding problem: %s with BOM", cs);
|
||||||
PyMem_Free(cs);
|
PyMem_Free(cs);
|
||||||
|
return 0;
|
||||||
}
|
}
|
||||||
return r;
|
PyMem_Free(cs);
|
||||||
|
}
|
||||||
|
return 1;
|
||||||
}
|
}
|
||||||
|
|
||||||
/* See whether the file starts with a BOM. If it does,
|
/* See whether the file starts with a BOM. If it does,
|
||||||
|
@ -270,7 +263,7 @@ check_bom(int get_char(struct tok_state *),
|
||||||
{
|
{
|
||||||
int ch1, ch2, ch3;
|
int ch1, ch2, ch3;
|
||||||
ch1 = get_char(tok);
|
ch1 = get_char(tok);
|
||||||
tok->decoding_state = STATE_RAW;
|
tok->decoding_state = STATE_SEEK_CODING;
|
||||||
if (ch1 == EOF) {
|
if (ch1 == EOF) {
|
||||||
return 1;
|
return 1;
|
||||||
} else if (ch1 == 0xEF) {
|
} else if (ch1 == 0xEF) {
|
||||||
|
@ -324,7 +317,8 @@ check_bom(int get_char(struct tok_state *),
|
||||||
return 1;
|
return 1;
|
||||||
}
|
}
|
||||||
|
|
||||||
static int tok_concatenate_interactive_new_line(struct tok_state* tok, char* line) {
|
static int
|
||||||
|
tok_concatenate_interactive_new_line(struct tok_state *tok, const char *line) {
|
||||||
assert(tok->fp_interactive);
|
assert(tok->fp_interactive);
|
||||||
|
|
||||||
if (!line) {
|
if (!line) {
|
||||||
|
@ -360,73 +354,73 @@ static int tok_concatenate_interactive_new_line(struct tok_state* tok, char* lin
|
||||||
1) NULL: need to call tok->decoding_readline to get a new line
|
1) NULL: need to call tok->decoding_readline to get a new line
|
||||||
2) PyUnicodeObject *: decoding_feof has called tok->decoding_readline and
|
2) PyUnicodeObject *: decoding_feof has called tok->decoding_readline and
|
||||||
stored the result in tok->decoding_buffer
|
stored the result in tok->decoding_buffer
|
||||||
3) PyByteArrayObject *: previous call to fp_readl did not have enough room
|
3) PyByteArrayObject *: previous call to tok_readline_recode did not have enough room
|
||||||
(in the s buffer) to copy entire contents of the line read
|
(in the s buffer) to copy entire contents of the line read
|
||||||
by tok->decoding_readline. tok->decoding_buffer has the overflow.
|
by tok->decoding_readline. tok->decoding_buffer has the overflow.
|
||||||
In this case, fp_readl is called in a loop (with an expanded buffer)
|
In this case, tok_readline_recode is called in a loop (with an expanded buffer)
|
||||||
until the buffer ends with a '\n' (or until the end of the file is
|
until the buffer ends with a '\n' (or until the end of the file is
|
||||||
reached): see tok_nextc and its calls to decoding_fgets.
|
reached): see tok_nextc and its calls to tok_reserve_buf.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
static char *
|
static int
|
||||||
fp_readl(char *s, int size, struct tok_state *tok)
|
tok_reserve_buf(struct tok_state *tok, Py_ssize_t size)
|
||||||
{
|
{
|
||||||
PyObject* bufobj;
|
Py_ssize_t cur = tok->cur - tok->buf;
|
||||||
|
Py_ssize_t oldsize = tok->inp - tok->buf;
|
||||||
|
Py_ssize_t newsize = oldsize + Py_MAX(size, oldsize >> 1);
|
||||||
|
if (newsize > tok->end - tok->buf) {
|
||||||
|
char *newbuf = tok->buf;
|
||||||
|
Py_ssize_t start = tok->start == NULL ? -1 : tok->start - tok->buf;
|
||||||
|
newbuf = (char *)PyMem_Realloc(newbuf, newsize);
|
||||||
|
if (newbuf == NULL) {
|
||||||
|
tok->done = E_NOMEM;
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
tok->buf = newbuf;
|
||||||
|
tok->cur = tok->buf + cur;
|
||||||
|
tok->inp = tok->buf + oldsize;
|
||||||
|
tok->end = tok->buf + newsize;
|
||||||
|
tok->start = start < 0 ? NULL : tok->buf + start;
|
||||||
|
}
|
||||||
|
return 1;
|
||||||
|
}
|
||||||
|
|
||||||
|
static int
|
||||||
|
tok_readline_recode(struct tok_state *tok) {
|
||||||
|
PyObject *line;
|
||||||
const char *buf;
|
const char *buf;
|
||||||
Py_ssize_t buflen;
|
Py_ssize_t buflen;
|
||||||
|
line = tok->decoding_buffer;
|
||||||
/* Ask for one less byte so we can terminate it */
|
if (line == NULL) {
|
||||||
assert(size > 0);
|
line = PyObject_CallNoArgs(tok->decoding_readline);
|
||||||
size--;
|
if (line == NULL) {
|
||||||
|
error_ret(tok);
|
||||||
if (tok->decoding_buffer) {
|
|
||||||
bufobj = tok->decoding_buffer;
|
|
||||||
Py_INCREF(bufobj);
|
|
||||||
}
|
|
||||||
else
|
|
||||||
{
|
|
||||||
bufobj = _PyObject_CallNoArg(tok->decoding_readline);
|
|
||||||
if (bufobj == NULL)
|
|
||||||
goto error;
|
|
||||||
}
|
|
||||||
if (PyUnicode_CheckExact(bufobj))
|
|
||||||
{
|
|
||||||
buf = PyUnicode_AsUTF8AndSize(bufobj, &buflen);
|
|
||||||
if (buf == NULL) {
|
|
||||||
goto error;
|
goto error;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
else
|
else {
|
||||||
{
|
|
||||||
buf = PyByteArray_AsString(bufobj);
|
|
||||||
if (buf == NULL) {
|
|
||||||
goto error;
|
|
||||||
}
|
|
||||||
buflen = PyByteArray_GET_SIZE(bufobj);
|
|
||||||
}
|
|
||||||
|
|
||||||
Py_XDECREF(tok->decoding_buffer);
|
|
||||||
if (buflen > size) {
|
|
||||||
/* Too many chars, the rest goes into tok->decoding_buffer */
|
|
||||||
tok->decoding_buffer = PyByteArray_FromStringAndSize(buf+size,
|
|
||||||
buflen-size);
|
|
||||||
if (tok->decoding_buffer == NULL)
|
|
||||||
goto error;
|
|
||||||
buflen = size;
|
|
||||||
}
|
|
||||||
else
|
|
||||||
tok->decoding_buffer = NULL;
|
tok->decoding_buffer = NULL;
|
||||||
|
}
|
||||||
memcpy(s, buf, buflen);
|
buf = PyUnicode_AsUTF8AndSize(line, &buflen);
|
||||||
s[buflen] = '\0';
|
if (buf == NULL) {
|
||||||
if (buflen == 0) /* EOF */
|
error_ret(tok);
|
||||||
s = NULL;
|
goto error;
|
||||||
Py_DECREF(bufobj);
|
}
|
||||||
return s;
|
if (!tok_reserve_buf(tok, buflen + 1)) {
|
||||||
|
goto error;
|
||||||
|
}
|
||||||
|
memcpy(tok->inp, buf, buflen);
|
||||||
|
tok->inp += buflen;
|
||||||
|
*tok->inp = '\0';
|
||||||
|
if (tok->fp_interactive &&
|
||||||
|
tok_concatenate_interactive_new_line(tok, buf) == -1) {
|
||||||
|
goto error;
|
||||||
|
}
|
||||||
|
Py_DECREF(line);
|
||||||
|
return 1;
|
||||||
error:
|
error:
|
||||||
Py_XDECREF(bufobj);
|
Py_XDECREF(line);
|
||||||
return error_ret(tok);
|
return 0;
|
||||||
}
|
}
|
||||||
|
|
||||||
/* Set the readline function for TOK to a StreamReader's
|
/* Set the readline function for TOK to a StreamReader's
|
||||||
|
@ -527,45 +521,13 @@ static int valid_utf8(const unsigned char* s)
|
||||||
return length;
|
return length;
|
||||||
}
|
}
|
||||||
|
|
||||||
/* Read a line of input from TOK. Determine encoding
|
static int
|
||||||
if necessary. */
|
ensure_utf8(char *line, struct tok_state *tok)
|
||||||
|
|
||||||
static char *
|
|
||||||
decoding_fgets(char *s, int size, struct tok_state *tok)
|
|
||||||
{
|
{
|
||||||
char *line = NULL;
|
|
||||||
int badchar = 0;
|
int badchar = 0;
|
||||||
for (;;) {
|
|
||||||
if (tok->decoding_state == STATE_NORMAL) {
|
|
||||||
/* We already have a codec associated with
|
|
||||||
this input. */
|
|
||||||
line = fp_readl(s, size, tok);
|
|
||||||
break;
|
|
||||||
} else if (tok->decoding_state == STATE_RAW) {
|
|
||||||
/* We want a 'raw' read. */
|
|
||||||
line = Py_UniversalNewlineFgets(s, size,
|
|
||||||
tok->fp, NULL);
|
|
||||||
break;
|
|
||||||
} else {
|
|
||||||
/* We have not yet determined the encoding.
|
|
||||||
If an encoding is found, use the file-pointer
|
|
||||||
reader functions from now on. */
|
|
||||||
if (!check_bom(fp_getc, fp_ungetc, fp_setreadl, tok))
|
|
||||||
return error_ret(tok);
|
|
||||||
assert(tok->decoding_state != STATE_INIT);
|
|
||||||
}
|
|
||||||
}
|
|
||||||
if (line != NULL && tok->lineno < 2 && !tok->read_coding_spec) {
|
|
||||||
if (!check_coding_spec(line, strlen(line), tok, fp_setreadl)) {
|
|
||||||
return error_ret(tok);
|
|
||||||
}
|
|
||||||
}
|
|
||||||
/* The default encoding is UTF-8, so make sure we don't have any
|
|
||||||
non-UTF-8 sequences in it. */
|
|
||||||
if (line && !tok->encoding) {
|
|
||||||
unsigned char *c;
|
unsigned char *c;
|
||||||
int length;
|
int length;
|
||||||
for (c = (unsigned char *)line; *c; c += length)
|
for (c = (unsigned char *)line; *c; c += length) {
|
||||||
if (!(length = valid_utf8(c))) {
|
if (!(length = valid_utf8(c))) {
|
||||||
badchar = *c;
|
badchar = *c;
|
||||||
break;
|
break;
|
||||||
|
@ -580,35 +542,9 @@ decoding_fgets(char *s, int size, struct tok_state *tok)
|
||||||
"but no encoding declared; "
|
"but no encoding declared; "
|
||||||
"see http://python.org/dev/peps/pep-0263/ for details",
|
"see http://python.org/dev/peps/pep-0263/ for details",
|
||||||
badchar, tok->filename, tok->lineno + 1);
|
badchar, tok->filename, tok->lineno + 1);
|
||||||
return error_ret(tok);
|
return 0;
|
||||||
}
|
}
|
||||||
|
|
||||||
if (tok->fp_interactive &&
|
|
||||||
tok_concatenate_interactive_new_line(tok, line) == -1) {
|
|
||||||
return NULL;
|
|
||||||
}
|
|
||||||
|
|
||||||
return line;
|
|
||||||
}
|
|
||||||
|
|
||||||
static int
|
|
||||||
decoding_feof(struct tok_state *tok)
|
|
||||||
{
|
|
||||||
if (tok->decoding_state != STATE_NORMAL) {
|
|
||||||
return feof(tok->fp);
|
|
||||||
} else {
|
|
||||||
PyObject* buf = tok->decoding_buffer;
|
|
||||||
if (buf == NULL) {
|
|
||||||
buf = _PyObject_CallNoArg(tok->decoding_readline);
|
|
||||||
if (buf == NULL) {
|
|
||||||
error_ret(tok);
|
|
||||||
return 1;
|
return 1;
|
||||||
} else {
|
|
||||||
tok->decoding_buffer = buf;
|
|
||||||
}
|
|
||||||
}
|
|
||||||
return PyObject_Length(buf) == 0;
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
|
||||||
/* Fetch a byte from TOK, using the string buffer. */
|
/* Fetch a byte from TOK, using the string buffer. */
|
||||||
|
@ -736,12 +672,13 @@ decode_str(const char *input, int single, struct tok_state *tok)
|
||||||
/* need to check line 1 and 2 separately since check_coding_spec
|
/* need to check line 1 and 2 separately since check_coding_spec
|
||||||
assumes a single line as input */
|
assumes a single line as input */
|
||||||
if (newl[0]) {
|
if (newl[0]) {
|
||||||
if (!check_coding_spec(str, newl[0] - str, tok, buf_setreadl))
|
if (!check_coding_spec(str, newl[0] - str, tok, buf_setreadl)) {
|
||||||
return error_ret(tok);
|
return NULL;
|
||||||
if (tok->enc == NULL && !tok->read_coding_spec && newl[1]) {
|
}
|
||||||
|
if (tok->enc == NULL && tok->decoding_state != STATE_NORMAL && newl[1]) {
|
||||||
if (!check_coding_spec(newl[0]+1, newl[1] - newl[0],
|
if (!check_coding_spec(newl[0]+1, newl[1] - newl[0],
|
||||||
tok, buf_setreadl))
|
tok, buf_setreadl))
|
||||||
return error_ret(tok);
|
return NULL;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
if (tok->enc != NULL) {
|
if (tok->enc != NULL) {
|
||||||
|
@ -777,6 +714,8 @@ PyTokenizer_FromString(const char *str, int exec_input)
|
||||||
return tok;
|
return tok;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
/* Set up tokenizer for UTF-8 string */
|
||||||
|
|
||||||
struct tok_state *
|
struct tok_state *
|
||||||
PyTokenizer_FromUTF8(const char *str, int exec_input)
|
PyTokenizer_FromUTF8(const char *str, int exec_input)
|
||||||
{
|
{
|
||||||
|
@ -789,16 +728,14 @@ PyTokenizer_FromUTF8(const char *str, int exec_input)
|
||||||
PyTokenizer_Free(tok);
|
PyTokenizer_Free(tok);
|
||||||
return NULL;
|
return NULL;
|
||||||
}
|
}
|
||||||
tok->decoding_state = STATE_RAW;
|
tok->decoding_state = STATE_NORMAL;
|
||||||
tok->read_coding_spec = 1;
|
|
||||||
tok->enc = NULL;
|
tok->enc = NULL;
|
||||||
tok->str = translated;
|
tok->str = translated;
|
||||||
tok->encoding = (char *)PyMem_Malloc(6);
|
tok->encoding = new_string("utf-8", 5, tok);
|
||||||
if (!tok->encoding) {
|
if (!tok->encoding) {
|
||||||
PyTokenizer_Free(tok);
|
PyTokenizer_Free(tok);
|
||||||
return NULL;
|
return NULL;
|
||||||
}
|
}
|
||||||
strcpy(tok->encoding, "utf-8");
|
|
||||||
|
|
||||||
tok->buf = tok->cur = tok->inp = translated;
|
tok->buf = tok->cur = tok->inp = translated;
|
||||||
tok->end = translated;
|
tok->end = translated;
|
||||||
|
@ -826,18 +763,16 @@ PyTokenizer_FromFile(FILE *fp, const char* enc,
|
||||||
if (enc != NULL) {
|
if (enc != NULL) {
|
||||||
/* Must copy encoding declaration since it
|
/* Must copy encoding declaration since it
|
||||||
gets copied into the parse tree. */
|
gets copied into the parse tree. */
|
||||||
tok->encoding = PyMem_Malloc(strlen(enc)+1);
|
tok->encoding = new_string(enc, strlen(enc), tok);
|
||||||
if (!tok->encoding) {
|
if (!tok->encoding) {
|
||||||
PyTokenizer_Free(tok);
|
PyTokenizer_Free(tok);
|
||||||
return NULL;
|
return NULL;
|
||||||
}
|
}
|
||||||
strcpy(tok->encoding, enc);
|
|
||||||
tok->decoding_state = STATE_NORMAL;
|
tok->decoding_state = STATE_NORMAL;
|
||||||
}
|
}
|
||||||
return tok;
|
return tok;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
/* Free a tok_state structure */
|
/* Free a tok_state structure */
|
||||||
|
|
||||||
void
|
void
|
||||||
|
@ -861,42 +796,59 @@ PyTokenizer_Free(struct tok_state *tok)
|
||||||
PyMem_Free(tok);
|
PyMem_Free(tok);
|
||||||
}
|
}
|
||||||
|
|
||||||
/* Get next char, updating state; error code goes into tok->done */
|
static int
|
||||||
|
tok_readline_raw(struct tok_state *tok)
|
||||||
|
{
|
||||||
|
do {
|
||||||
|
if (!tok_reserve_buf(tok, BUFSIZ)) {
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
char *line = Py_UniversalNewlineFgets(tok->inp,
|
||||||
|
(int)(tok->end - tok->inp),
|
||||||
|
tok->fp, NULL);
|
||||||
|
if (line == NULL) {
|
||||||
|
return 1;
|
||||||
|
}
|
||||||
|
if (tok->fp_interactive &&
|
||||||
|
tok_concatenate_interactive_new_line(tok, line) == -1) {
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
tok->inp = strchr(tok->inp, '\0');
|
||||||
|
} while (tok->inp[-1] != '\n');
|
||||||
|
return 1;
|
||||||
|
}
|
||||||
|
|
||||||
static int
|
static int
|
||||||
tok_nextc(struct tok_state *tok)
|
tok_underflow_string(struct tok_state *tok) {
|
||||||
{
|
|
||||||
for (;;) {
|
|
||||||
if (tok->cur != tok->inp) {
|
|
||||||
return Py_CHARMASK(*tok->cur++); /* Fast path */
|
|
||||||
}
|
|
||||||
if (tok->done != E_OK)
|
|
||||||
return EOF;
|
|
||||||
if (tok->fp == NULL) {
|
|
||||||
char *end = strchr(tok->inp, '\n');
|
char *end = strchr(tok->inp, '\n');
|
||||||
if (end != NULL)
|
if (end != NULL) {
|
||||||
end++;
|
end++;
|
||||||
|
}
|
||||||
else {
|
else {
|
||||||
end = strchr(tok->inp, '\0');
|
end = strchr(tok->inp, '\0');
|
||||||
if (end == tok->inp) {
|
if (end == tok->inp) {
|
||||||
tok->done = E_EOF;
|
tok->done = E_EOF;
|
||||||
return EOF;
|
return 0;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
if (tok->start == NULL)
|
if (tok->start == NULL) {
|
||||||
tok->buf = tok->cur;
|
tok->buf = tok->cur;
|
||||||
|
}
|
||||||
tok->line_start = tok->cur;
|
tok->line_start = tok->cur;
|
||||||
tok->lineno++;
|
tok->lineno++;
|
||||||
tok->inp = end;
|
tok->inp = end;
|
||||||
return Py_CHARMASK(*tok->cur++);
|
return 1;
|
||||||
}
|
}
|
||||||
if (tok->prompt != NULL) {
|
|
||||||
|
static int
|
||||||
|
tok_underflow_interactive(struct tok_state *tok) {
|
||||||
char *newtok = PyOS_Readline(stdin, stdout, tok->prompt);
|
char *newtok = PyOS_Readline(stdin, stdout, tok->prompt);
|
||||||
if (newtok != NULL) {
|
if (newtok != NULL) {
|
||||||
char *translated = translate_newlines(newtok, 0, tok);
|
char *translated = translate_newlines(newtok, 0, tok);
|
||||||
PyMem_Free(newtok);
|
PyMem_Free(newtok);
|
||||||
if (translated == NULL)
|
if (translated == NULL) {
|
||||||
return EOF;
|
return 0;
|
||||||
|
}
|
||||||
newtok = translated;
|
newtok = translated;
|
||||||
}
|
}
|
||||||
if (tok->encoding && newtok && *newtok) {
|
if (tok->encoding && newtok && *newtok) {
|
||||||
|
@ -905,9 +857,9 @@ tok_nextc(struct tok_state *tok)
|
||||||
const char* buf;
|
const char* buf;
|
||||||
PyObject *u = translate_into_utf8(newtok, tok->encoding);
|
PyObject *u = translate_into_utf8(newtok, tok->encoding);
|
||||||
PyMem_Free(newtok);
|
PyMem_Free(newtok);
|
||||||
if (!u) {
|
if (u == NULL) {
|
||||||
tok->done = E_DECODE;
|
tok->done = E_DECODE;
|
||||||
return EOF;
|
return 0;
|
||||||
}
|
}
|
||||||
buflen = PyBytes_GET_SIZE(u);
|
buflen = PyBytes_GET_SIZE(u);
|
||||||
buf = PyBytes_AS_STRING(u);
|
buf = PyBytes_AS_STRING(u);
|
||||||
|
@ -915,51 +867,43 @@ tok_nextc(struct tok_state *tok)
|
||||||
if (newtok == NULL) {
|
if (newtok == NULL) {
|
||||||
Py_DECREF(u);
|
Py_DECREF(u);
|
||||||
tok->done = E_NOMEM;
|
tok->done = E_NOMEM;
|
||||||
return EOF;
|
return 0;
|
||||||
}
|
}
|
||||||
strcpy(newtok, buf);
|
strcpy(newtok, buf);
|
||||||
Py_DECREF(u);
|
Py_DECREF(u);
|
||||||
}
|
}
|
||||||
if (tok->fp_interactive &&
|
if (tok->fp_interactive &&
|
||||||
tok_concatenate_interactive_new_line(tok, newtok) == -1) {
|
tok_concatenate_interactive_new_line(tok, newtok) == -1) {
|
||||||
return EOF;
|
PyMem_Free(newtok);
|
||||||
|
return 0;
|
||||||
}
|
}
|
||||||
if (tok->nextprompt != NULL)
|
if (tok->nextprompt != NULL) {
|
||||||
tok->prompt = tok->nextprompt;
|
tok->prompt = tok->nextprompt;
|
||||||
if (newtok == NULL)
|
}
|
||||||
|
if (newtok == NULL) {
|
||||||
tok->done = E_INTR;
|
tok->done = E_INTR;
|
||||||
|
}
|
||||||
else if (*newtok == '\0') {
|
else if (*newtok == '\0') {
|
||||||
PyMem_Free(newtok);
|
PyMem_Free(newtok);
|
||||||
tok->done = E_EOF;
|
tok->done = E_EOF;
|
||||||
}
|
}
|
||||||
else if (tok->start != NULL) {
|
else if (tok->start != NULL) {
|
||||||
size_t start = tok->start - tok->buf;
|
|
||||||
size_t oldlen = tok->cur - tok->buf;
|
|
||||||
size_t newlen = oldlen + strlen(newtok);
|
|
||||||
Py_ssize_t cur_multi_line_start = tok->multi_line_start - tok->buf;
|
Py_ssize_t cur_multi_line_start = tok->multi_line_start - tok->buf;
|
||||||
char *buf = tok->buf;
|
size_t size = strlen(newtok);
|
||||||
buf = (char *)PyMem_Realloc(buf, newlen+1);
|
|
||||||
tok->lineno++;
|
tok->lineno++;
|
||||||
if (buf == NULL) {
|
if (!tok_reserve_buf(tok, size + 1)) {
|
||||||
PyMem_Free(tok->buf);
|
PyMem_Free(tok->buf);
|
||||||
tok->buf = NULL;
|
tok->buf = NULL;
|
||||||
PyMem_Free(newtok);
|
PyMem_Free(newtok);
|
||||||
tok->done = E_NOMEM;
|
return 0;
|
||||||
return EOF;
|
|
||||||
}
|
}
|
||||||
tok->buf = buf;
|
memcpy(tok->cur, newtok, size + 1);
|
||||||
tok->cur = tok->buf + oldlen;
|
|
||||||
tok->multi_line_start = tok->buf + cur_multi_line_start;
|
|
||||||
tok->line_start = tok->cur;
|
|
||||||
strcpy(tok->buf + oldlen, newtok);
|
|
||||||
PyMem_Free(newtok);
|
PyMem_Free(newtok);
|
||||||
tok->inp = tok->buf + newlen;
|
tok->inp += size;
|
||||||
tok->end = tok->inp + 1;
|
tok->multi_line_start = tok->buf + cur_multi_line_start;
|
||||||
tok->start = tok->buf + start;
|
|
||||||
}
|
}
|
||||||
else {
|
else {
|
||||||
tok->lineno++;
|
tok->lineno++;
|
||||||
if (tok->buf != NULL)
|
|
||||||
PyMem_Free(tok->buf);
|
PyMem_Free(tok->buf);
|
||||||
tok->buf = newtok;
|
tok->buf = newtok;
|
||||||
tok->cur = tok->buf;
|
tok->cur = tok->buf;
|
||||||
|
@ -967,105 +911,138 @@ tok_nextc(struct tok_state *tok)
|
||||||
tok->inp = strchr(tok->buf, '\0');
|
tok->inp = strchr(tok->buf, '\0');
|
||||||
tok->end = tok->inp + 1;
|
tok->end = tok->inp + 1;
|
||||||
}
|
}
|
||||||
}
|
|
||||||
else {
|
|
||||||
int done = 0;
|
|
||||||
Py_ssize_t cur = 0;
|
|
||||||
char *pt;
|
|
||||||
if (tok->start == NULL) {
|
|
||||||
if (tok->buf == NULL) {
|
|
||||||
tok->buf = (char *)
|
|
||||||
PyMem_Malloc(BUFSIZ);
|
|
||||||
if (tok->buf == NULL) {
|
|
||||||
tok->done = E_NOMEM;
|
|
||||||
return EOF;
|
|
||||||
}
|
|
||||||
tok->end = tok->buf + BUFSIZ;
|
|
||||||
}
|
|
||||||
if (decoding_fgets(tok->buf, (int)(tok->end - tok->buf),
|
|
||||||
tok) == NULL) {
|
|
||||||
if (!tok->decoding_erred && !(tok->done == E_NOMEM))
|
|
||||||
tok->done = E_EOF;
|
|
||||||
done = 1;
|
|
||||||
}
|
|
||||||
else {
|
|
||||||
tok->done = E_OK;
|
|
||||||
tok->inp = strchr(tok->buf, '\0');
|
|
||||||
done = tok->inp == tok->buf || tok->inp[-1] == '\n';
|
|
||||||
}
|
|
||||||
}
|
|
||||||
else {
|
|
||||||
cur = tok->cur - tok->buf;
|
|
||||||
if (decoding_feof(tok)) {
|
|
||||||
tok->done = E_EOF;
|
|
||||||
done = 1;
|
|
||||||
}
|
|
||||||
else
|
|
||||||
tok->done = E_OK;
|
|
||||||
}
|
|
||||||
tok->lineno++;
|
|
||||||
/* Read until '\n' or EOF */
|
|
||||||
while (!done) {
|
|
||||||
Py_ssize_t curstart = tok->start == NULL ? -1 :
|
|
||||||
tok->start - tok->buf;
|
|
||||||
Py_ssize_t cur_multi_line_start = tok->multi_line_start - tok->buf;
|
|
||||||
Py_ssize_t curvalid = tok->inp - tok->buf;
|
|
||||||
Py_ssize_t newsize = curvalid + BUFSIZ;
|
|
||||||
char *newbuf = tok->buf;
|
|
||||||
newbuf = (char *)PyMem_Realloc(newbuf,
|
|
||||||
newsize);
|
|
||||||
if (newbuf == NULL) {
|
|
||||||
tok->done = E_NOMEM;
|
|
||||||
tok->cur = tok->inp;
|
|
||||||
return EOF;
|
|
||||||
}
|
|
||||||
tok->buf = newbuf;
|
|
||||||
tok->cur = tok->buf + cur;
|
|
||||||
tok->multi_line_start = tok->buf + cur_multi_line_start;
|
|
||||||
tok->line_start = tok->cur;
|
|
||||||
tok->inp = tok->buf + curvalid;
|
|
||||||
tok->end = tok->buf + newsize;
|
|
||||||
tok->start = curstart < 0 ? NULL :
|
|
||||||
tok->buf + curstart;
|
|
||||||
if (decoding_fgets(tok->inp,
|
|
||||||
(int)(tok->end - tok->inp),
|
|
||||||
tok) == NULL) {
|
|
||||||
/* Break out early on decoding
|
|
||||||
errors, as tok->buf will be NULL
|
|
||||||
*/
|
|
||||||
if (tok->decoding_erred)
|
|
||||||
return EOF;
|
|
||||||
/* Last line does not end in \n,
|
|
||||||
fake one */
|
|
||||||
if (tok->inp[-1] != '\n')
|
|
||||||
strcpy(tok->inp, "\n");
|
|
||||||
}
|
|
||||||
tok->inp = strchr(tok->inp, '\0');
|
|
||||||
done = tok->inp[-1] == '\n';
|
|
||||||
}
|
|
||||||
if (tok->buf != NULL) {
|
|
||||||
tok->cur = tok->buf + cur;
|
|
||||||
tok->line_start = tok->cur;
|
|
||||||
/* replace "\r\n" with "\n" */
|
|
||||||
/* For Mac leave the \r, giving a syntax error */
|
|
||||||
pt = tok->inp - 2;
|
|
||||||
if (pt >= tok->buf && *pt == '\r') {
|
|
||||||
*pt++ = '\n';
|
|
||||||
*pt = '\0';
|
|
||||||
tok->inp = pt;
|
|
||||||
}
|
|
||||||
}
|
|
||||||
}
|
|
||||||
if (tok->done != E_OK) {
|
if (tok->done != E_OK) {
|
||||||
if (tok->prompt != NULL)
|
if (tok->prompt != NULL) {
|
||||||
PySys_WriteStderr("\n");
|
PySys_WriteStderr("\n");
|
||||||
tok->cur = tok->inp;
|
|
||||||
return EOF;
|
|
||||||
}
|
}
|
||||||
|
return 0;
|
||||||
}
|
}
|
||||||
/*NOTREACHED*/
|
return 1;
|
||||||
}
|
}
|
||||||
|
|
||||||
|
static int
|
||||||
|
tok_underflow_file(struct tok_state *tok) {
|
||||||
|
if (tok->start == NULL) {
|
||||||
|
tok->cur = tok->inp = tok->buf;
|
||||||
|
}
|
||||||
|
if (tok->decoding_state == STATE_INIT) {
|
||||||
|
/* We have not yet determined the encoding.
|
||||||
|
If an encoding is found, use the file-pointer
|
||||||
|
reader functions from now on. */
|
||||||
|
if (!check_bom(fp_getc, fp_ungetc, fp_setreadl, tok)) {
|
||||||
|
error_ret(tok);
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
assert(tok->decoding_state != STATE_INIT);
|
||||||
|
}
|
||||||
|
/* Read until '\n' or EOF */
|
||||||
|
if (tok->decoding_readline != NULL) {
|
||||||
|
/* We already have a codec associated with this input. */
|
||||||
|
if (!tok_readline_recode(tok)) {
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
else {
|
||||||
|
/* We want a 'raw' read. */
|
||||||
|
if (!tok_readline_raw(tok)) {
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
if (tok->inp == tok->cur) {
|
||||||
|
tok->done = E_EOF;
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
if (tok->inp[-1] != '\n') {
|
||||||
|
/* Last line does not end in \n, fake one */
|
||||||
|
*tok->inp++ = '\n';
|
||||||
|
*tok->inp = '\0';
|
||||||
|
}
|
||||||
|
|
||||||
|
tok->lineno++;
|
||||||
|
if (tok->decoding_state != STATE_NORMAL) {
|
||||||
|
if (tok->lineno > 2) {
|
||||||
|
tok->decoding_state = STATE_NORMAL;
|
||||||
|
}
|
||||||
|
else if (!check_coding_spec(tok->cur, tok->end - tok->cur,
|
||||||
|
tok, fp_setreadl))
|
||||||
|
{
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
/* The default encoding is UTF-8, so make sure we don't have any
|
||||||
|
non-UTF-8 sequences in it. */
|
||||||
|
if (!tok->encoding
|
||||||
|
&& (tok->decoding_state != STATE_NORMAL || tok->lineno >= 2)) {
|
||||||
|
if (!ensure_utf8(tok->cur, tok)) {
|
||||||
|
error_ret(tok);
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
assert(tok->done == E_OK);
|
||||||
|
return tok->done == E_OK;
|
||||||
|
}
|
||||||
|
|
||||||
|
static void
|
||||||
|
print_escape(FILE *f, const char *s, Py_ssize_t size)
|
||||||
|
{
|
||||||
|
if (s == NULL) {
|
||||||
|
fputs("NULL", f);
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
putc('"', f);
|
||||||
|
while (size-- > 0) {
|
||||||
|
unsigned char c = *s++;
|
||||||
|
switch (c) {
|
||||||
|
case '\n': fputs("\\n", f); break;
|
||||||
|
case '\r': fputs("\\r", f); break;
|
||||||
|
case '\t': fputs("\\t", f); break;
|
||||||
|
case '\f': fputs("\\f", f); break;
|
||||||
|
case '\'': fputs("\\'", f); break;
|
||||||
|
case '"': fputs("\\\"", f); break;
|
||||||
|
default:
|
||||||
|
if (0x20 <= c && c <= 0x7f)
|
||||||
|
putc(c, f);
|
||||||
|
else
|
||||||
|
fprintf(f, "\\x%02x", c);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
putc('"', f);
|
||||||
|
}
|
||||||
|
|
||||||
|
/* Get next char, updating state; error code goes into tok->done */
|
||||||
|
|
||||||
|
static int
|
||||||
|
tok_nextc(struct tok_state *tok)
|
||||||
|
{
|
||||||
|
int rc;
|
||||||
|
for (;;) {
|
||||||
|
if (tok->cur != tok->inp) {
|
||||||
|
return Py_CHARMASK(*tok->cur++); /* Fast path */
|
||||||
|
}
|
||||||
|
if (tok->done != E_OK)
|
||||||
|
return EOF;
|
||||||
|
if (tok->fp == NULL) {
|
||||||
|
rc = tok_underflow_string(tok);
|
||||||
|
}
|
||||||
|
else if (tok->prompt != NULL) {
|
||||||
|
rc = tok_underflow_interactive(tok);
|
||||||
|
}
|
||||||
|
else {
|
||||||
|
rc = tok_underflow_file(tok);
|
||||||
|
}
|
||||||
|
if (Py_DebugFlag) {
|
||||||
|
printf("line[%d] = ", tok->lineno);
|
||||||
|
print_escape(stdout, tok->cur, tok->inp - tok->cur);
|
||||||
|
printf(" tok->done = %d\n", tok->done);
|
||||||
|
}
|
||||||
|
if (!rc) {
|
||||||
|
tok->cur = tok->inp;
|
||||||
|
return EOF;
|
||||||
|
}
|
||||||
|
tok->line_start = tok->cur;
|
||||||
|
}
|
||||||
|
Py_UNREACHABLE();
|
||||||
|
}
|
||||||
|
|
||||||
/* Back-up one character */
|
/* Back-up one character */
|
||||||
|
|
||||||
|
@ -1076,8 +1053,8 @@ tok_backup(struct tok_state *tok, int c)
|
||||||
if (--tok->cur < tok->buf) {
|
if (--tok->cur < tok->buf) {
|
||||||
Py_FatalError("tokenizer beginning of buffer");
|
Py_FatalError("tokenizer beginning of buffer");
|
||||||
}
|
}
|
||||||
if (*tok->cur != c) {
|
if ((int)(unsigned char)*tok->cur != c) {
|
||||||
*tok->cur = c;
|
Py_FatalError("tok_backup: wrong character");
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
@ -1951,9 +1928,10 @@ PyTokenizer_FindEncodingFilename(int fd, PyObject *filename)
|
||||||
fclose(fp);
|
fclose(fp);
|
||||||
if (tok->encoding) {
|
if (tok->encoding) {
|
||||||
encoding = (char *)PyMem_Malloc(strlen(tok->encoding) + 1);
|
encoding = (char *)PyMem_Malloc(strlen(tok->encoding) + 1);
|
||||||
if (encoding)
|
if (encoding) {
|
||||||
strcpy(encoding, tok->encoding);
|
strcpy(encoding, tok->encoding);
|
||||||
}
|
}
|
||||||
|
}
|
||||||
PyTokenizer_Free(tok);
|
PyTokenizer_Free(tok);
|
||||||
return encoding;
|
return encoding;
|
||||||
}
|
}
|
||||||
|
|
|
@ -15,8 +15,8 @@ extern "C" {
|
||||||
|
|
||||||
enum decoding_state {
|
enum decoding_state {
|
||||||
STATE_INIT,
|
STATE_INIT,
|
||||||
STATE_RAW,
|
STATE_SEEK_CODING,
|
||||||
STATE_NORMAL /* have a codec associated with input */
|
STATE_NORMAL
|
||||||
};
|
};
|
||||||
|
|
||||||
/* Tokenizer state */
|
/* Tokenizer state */
|
||||||
|
@ -54,7 +54,6 @@ struct tok_state {
|
||||||
/* Stuff for PEP 0263 */
|
/* Stuff for PEP 0263 */
|
||||||
enum decoding_state decoding_state;
|
enum decoding_state decoding_state;
|
||||||
int decoding_erred; /* whether erred in decoding */
|
int decoding_erred; /* whether erred in decoding */
|
||||||
int read_coding_spec; /* whether 'coding:...' has been read */
|
|
||||||
char *encoding; /* Source encoding. */
|
char *encoding; /* Source encoding. */
|
||||||
int cont_line; /* whether we are in a continuation line. */
|
int cont_line; /* whether we are in a continuation line. */
|
||||||
const char* line_start; /* pointer to start of current line */
|
const char* line_start; /* pointer to start of current line */
|
||||||
|
|
Loading…
Reference in a new issue