mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
Merge branch 'ft-py3-tokenizer' of https://github.com/flaviut/calibre
This commit is contained in:
commit
36647d4762
1
.gitignore
vendored
1
.gitignore
vendored
@ -52,3 +52,4 @@ recipes/debug
|
||||
/.metadata/
|
||||
.idea
|
||||
/*env*/
|
||||
cmake-build-*
|
||||
|
@ -386,6 +386,18 @@ html_init(PyObject *self, PyObject *args) {
|
||||
Py_RETURN_NONE;
|
||||
}
|
||||
|
||||
static inline long number_to_long(PyObject *number) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
return PyLong_AsLong(number);
|
||||
#else
|
||||
if(PyInt_Check(number)) {
|
||||
return PyInt_AS_LONG(number);
|
||||
} else {
|
||||
return PyLong_AsLong(number);
|
||||
}
|
||||
#endif
|
||||
}
|
||||
|
||||
static PyObject*
|
||||
html_check_spelling(PyObject *self, PyObject *args) {
|
||||
PyObject *ans = NULL, *temp = NULL, *items = NULL, *text = NULL, *fmt = NULL, *locale = NULL, *sfmt = NULL, *_store_locale = NULL, *t = NULL, *utmp = NULL;
|
||||
@ -410,9 +422,9 @@ html_check_spelling(PyObject *self, PyObject *args) {
|
||||
|
||||
for (i = 0, j = 0; i < PyList_GET_SIZE(items); i++) {
|
||||
temp = PyList_GET_ITEM(items, i);
|
||||
start = PyLong_AsLong(PyTuple_GET_ITEM(temp, 0));
|
||||
start = number_to_long(PyTuple_GET_ITEM(temp, 0));
|
||||
if(start == -1 && PyErr_Occurred() != NULL) goto error;
|
||||
length = PyLong_AsLong(PyTuple_GET_ITEM(temp, 1));
|
||||
length = number_to_long(PyTuple_GET_ITEM(temp, 1));
|
||||
if(length == -1 && PyErr_Occurred() != NULL) goto error;
|
||||
temp = NULL;
|
||||
|
||||
|
@ -34,7 +34,7 @@ tokenizer_Token_dealloc(tokenizer_Token* self)
|
||||
Py_XDECREF(self->unit); self->unit = NULL;
|
||||
Py_XDECREF(self->line); self->line = NULL;
|
||||
Py_XDECREF(self->column); self->column = NULL;
|
||||
self->ob_type->tp_free((PyObject*)self);
|
||||
Py_TYPE(self)->tp_free((PyObject*)self);
|
||||
}
|
||||
|
||||
|
||||
@ -46,7 +46,8 @@ tokenizer_Token_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
|
||||
if (self == NULL) return PyErr_NoMemory();
|
||||
|
||||
if (!PyArg_ParseTuple(args, "OOOOOO", &(self->type), &(self->_as_css), &(self->value), &(self->unit), &(self->line), &(self->column))) {
|
||||
self->ob_type->tp_free((PyObject*)self); return NULL;
|
||||
Py_TYPE(self)->tp_free((PyObject *) self);
|
||||
return NULL;
|
||||
}
|
||||
Py_INCREF(self->type); Py_INCREF(self->_as_css); Py_INCREF(self->value); Py_INCREF(self->unit); Py_INCREF(self->line); Py_INCREF(self->column);
|
||||
self->is_container = Py_False; Py_INCREF(self->is_container);
|
||||
@ -54,16 +55,25 @@ tokenizer_Token_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
|
||||
return (PyObject *)self;
|
||||
}
|
||||
|
||||
#if PY_MAJOR_VERSION >= 3
|
||||
#define PyObject_Unicode_Compat(arg) PyObject_Str(arg)
|
||||
#else
|
||||
#define PyObject_Unicode_Compat(arg) PyObject_Unicode(arg)
|
||||
#endif
|
||||
|
||||
static PyObject *
|
||||
tokenizer_Token_repr(tokenizer_Token *self) {
|
||||
PyObject *type = NULL, *line = NULL, *column = NULL, *value = NULL, *ans = NULL, *unit = NULL;
|
||||
if (!self->type || !self->line || !self->column || !self->value)
|
||||
if (!self->type || !self->line || !self->column || !self->value)
|
||||
return PyBytes_FromString("<Token NULL fields>");
|
||||
type = PyObject_Unicode(self->type); line = PyObject_Unicode(self->line); column = PyObject_Unicode(self->column); value = PyObject_Unicode(self->value);
|
||||
type = PyObject_Unicode_Compat(self->type);
|
||||
line = PyObject_Unicode_Compat(self->line);
|
||||
column = PyObject_Unicode_Compat(self->column);
|
||||
value = PyObject_Unicode_Compat(self->value);
|
||||
if (type && line && column && value) {
|
||||
if (self->unit != NULL && PyObject_IsTrue(self->unit)) {
|
||||
unit = PyObject_Unicode(self->unit);
|
||||
if (unit != NULL)
|
||||
unit = PyObject_Unicode_Compat(self->unit);
|
||||
if (unit != NULL)
|
||||
ans = PyUnicode_FromFormat("<Token %U at %U:%U %U%U>", type, line, column, value, unit);
|
||||
else
|
||||
PyErr_NoMemory();
|
||||
@ -103,58 +113,57 @@ static PyMethodDef tokenizer_Token_methods[] = {
|
||||
};
|
||||
|
||||
static PyTypeObject tokenizer_TokenType = { // {{{
|
||||
PyObject_HEAD_INIT(NULL)
|
||||
0, /*ob_size*/
|
||||
"tokenizer.Token", /*tp_name*/
|
||||
sizeof(tokenizer_Token), /*tp_basicsize*/
|
||||
0, /*tp_itemsize*/
|
||||
(destructor)tokenizer_Token_dealloc, /*tp_dealloc*/
|
||||
0, /*tp_print*/
|
||||
0, /*tp_getattr*/
|
||||
0, /*tp_setattr*/
|
||||
0, /*tp_compare*/
|
||||
(reprfunc)tokenizer_Token_repr, /*tp_repr*/
|
||||
0, /*tp_as_number*/
|
||||
0, /*tp_as_sequence*/
|
||||
0, /*tp_as_mapping*/
|
||||
0, /*tp_hash */
|
||||
0, /*tp_call*/
|
||||
0, /*tp_str*/
|
||||
0, /*tp_getattro*/
|
||||
0, /*tp_setattro*/
|
||||
0, /*tp_as_buffer*/
|
||||
Py_TPFLAGS_DEFAULT|Py_TPFLAGS_BASETYPE, /*tp_flags*/
|
||||
"Token", /* tp_doc */
|
||||
0, /* tp_traverse */
|
||||
0, /* tp_clear */
|
||||
0, /* tp_richcompare */
|
||||
0, /* tp_weaklistoffset */
|
||||
0, /* tp_iter */
|
||||
0, /* tp_iternext */
|
||||
tokenizer_Token_methods, /* tp_methods */
|
||||
tokenizer_Token_members, /* tp_members */
|
||||
0, /* tp_getset */
|
||||
0, /* tp_base */
|
||||
0, /* tp_dict */
|
||||
0, /* tp_descr_get */
|
||||
0, /* tp_descr_set */
|
||||
0, /* tp_dictoffset */
|
||||
0, /* tp_init */
|
||||
0, /* tp_alloc */
|
||||
tokenizer_Token_new, /* tp_new */
|
||||
PyVarObject_HEAD_INIT(NULL, 0)
|
||||
/* tp_name */ "tokenizer.Token",
|
||||
/* tp_basicsize */ sizeof(tokenizer_Token),
|
||||
/* tp_itemsize */ 0,
|
||||
/* tp_dealloc */ (destructor) tokenizer_Token_dealloc,
|
||||
/* tp_print */ 0,
|
||||
/* tp_getattr */ 0,
|
||||
/* tp_setattr */ 0,
|
||||
/* tp_compare */ 0,
|
||||
/* tp_repr */ (reprfunc) tokenizer_Token_repr,
|
||||
/* tp_as_number */ 0,
|
||||
/* tp_as_sequence */ 0,
|
||||
/* tp_as_mapping */ 0,
|
||||
/* tp_hash */ 0,
|
||||
/* tp_call */ 0,
|
||||
/* tp_str */ 0,
|
||||
/* tp_getattro */ 0,
|
||||
/* tp_setattro */ 0,
|
||||
/* tp_as_buffer */ 0,
|
||||
/* tp_flags */ Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,
|
||||
/* tp_doc */ "Token",
|
||||
/* tp_traverse */ 0,
|
||||
/* tp_clear */ 0,
|
||||
/* tp_richcompare */ 0,
|
||||
/* tp_weaklistoffset */ 0,
|
||||
/* tp_iter */ 0,
|
||||
/* tp_iternext */ 0,
|
||||
/* tp_methods */ tokenizer_Token_methods,
|
||||
/* tp_members */ tokenizer_Token_members,
|
||||
/* tp_getset */ 0,
|
||||
/* tp_base */ 0,
|
||||
/* tp_dict */ 0,
|
||||
/* tp_descr_get */ 0,
|
||||
/* tp_descr_set */ 0,
|
||||
/* tp_dictoffset */ 0,
|
||||
/* tp_init */ 0,
|
||||
/* tp_alloc */ 0,
|
||||
/* tp_new */ tokenizer_Token_new,
|
||||
}; // }}}
|
||||
// }}}
|
||||
|
||||
static PyObject *COMPILED_TOKEN_REGEXPS = NULL, *UNICODE_UNESCAPE = NULL, *NEWLINE_UNESCAPE = NULL, *SIMPLE_UNESCAPE = NULL, *FIND_NEWLINES = NULL, *TOKEN_DISPATCH = NULL;
|
||||
static PyObject *COMPILED_TOKEN_REGEXPS = NULL, *UNICODE_UNESCAPE = NULL, *NEWLINE_UNESCAPE = NULL, *SIMPLE_UNESCAPE = NULL, *FIND_NEWLINES = NULL, *TOKEN_DISPATCH = NULL;
|
||||
static PyObject *COLON = NULL, *SCOLON = NULL, *LPAR = NULL, *RPAR = NULL, *LBRACE = NULL, *RBRACE = NULL, *LBOX = NULL, *RBOX = NULL, *DELIM_TOK = NULL, *INTEGER = NULL, *STRING_TOK = NULL;
|
||||
|
||||
static Py_ssize_t BAD_COMMENT, BAD_STRING, PERCENTAGE, DIMENSION, ATKEYWORD, FUNCTION, COMMENT, NUMBER, STRING, IDENT, HASH, URI, DELIM = -1;
|
||||
|
||||
#define CLEANUP(x) Py_XDECREF((x)); x = NULL;
|
||||
|
||||
#define CLEANUP(x) Py_XDECREF((x)); x = NULL;
|
||||
|
||||
static PyObject*
|
||||
tokenize_cleanup(PyObject *self, PyObject *args) {
|
||||
CLEANUP(COMPILED_TOKEN_REGEXPS); CLEANUP(UNICODE_UNESCAPE); CLEANUP(NEWLINE_UNESCAPE); CLEANUP(SIMPLE_UNESCAPE); CLEANUP(FIND_NEWLINES); CLEANUP(TOKEN_DISPATCH);
|
||||
CLEANUP(COMPILED_TOKEN_REGEXPS); CLEANUP(UNICODE_UNESCAPE); CLEANUP(NEWLINE_UNESCAPE); CLEANUP(SIMPLE_UNESCAPE); CLEANUP(FIND_NEWLINES); CLEANUP(TOKEN_DISPATCH);
|
||||
CLEANUP(COLON); CLEANUP(SCOLON); CLEANUP(LPAR); CLEANUP(RPAR); CLEANUP(LBRACE); CLEANUP(RBRACE); CLEANUP(LBOX); CLEANUP(RBOX); CLEANUP(DELIM_TOK); CLEANUP(INTEGER); CLEANUP(STRING_TOK);
|
||||
Py_RETURN_NONE;
|
||||
}
|
||||
@ -170,67 +179,103 @@ tokenize_init(PyObject *self, PyObject *args) {
|
||||
Py_INCREF(COMPILED_TOKEN_REGEXPS); Py_INCREF(UNICODE_UNESCAPE); Py_INCREF(NEWLINE_UNESCAPE); Py_INCREF(SIMPLE_UNESCAPE); Py_INCREF(FIND_NEWLINES); Py_INCREF(TOKEN_DISPATCH);
|
||||
Py_INCREF(COLON); Py_INCREF(SCOLON); Py_INCREF(LPAR); Py_INCREF(RPAR); Py_INCREF(LBRACE); Py_INCREF(RBRACE); Py_INCREF(LBOX); Py_INCREF(RBOX); Py_INCREF(DELIM_TOK); Py_INCREF(INTEGER); Py_INCREF(STRING_TOK);
|
||||
|
||||
#define SETCONST(x) x = PyInt_AsSsize_t(PyDict_GetItemString(cti, #x))
|
||||
#define SETCONST(x) do { (x) = PyNumber_AsSsize_t(PyDict_GetItemString(cti, #x), PyExc_OverflowError); \
|
||||
if((x) == -1 && PyErr_Occurred() != NULL) { return NULL; } \
|
||||
} while(0)
|
||||
SETCONST(BAD_COMMENT); SETCONST(BAD_STRING); SETCONST(PERCENTAGE); SETCONST(DIMENSION); SETCONST(ATKEYWORD); SETCONST(FUNCTION); SETCONST(COMMENT); SETCONST(NUMBER); SETCONST(STRING); SETCONST(IDENT); SETCONST(HASH); SETCONST(URI);
|
||||
|
||||
Py_RETURN_NONE;
|
||||
}
|
||||
|
||||
static int
|
||||
contains_char(PyObject *haystack, Py_UNICODE c) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
#error Not implemented for python >= 3.3
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
#define ITER_CODE_PTS(unicode_object) { \
|
||||
int _kind = PyUnicode_KIND(unicode_object); \
|
||||
void *_data = PyUnicode_DATA(unicode_object); \
|
||||
for (Py_ssize_t iteridx = 0; iteridx < PyUnicode_GET_LENGTH(unicode_object); iteridx++) { \
|
||||
Py_UCS4 ch = PyUnicode_READ(kind, data, i);
|
||||
#else
|
||||
#define PyUnicode_GET_LENGTH PyUnicode_GET_SIZE
|
||||
#define ITER_CODE_PTS(unicode_object) { \
|
||||
Py_UNICODE *_data = PyUnicode_AS_UNICODE(unicode_object); \
|
||||
for (Py_ssize_t iteridx = 0; iteridx < PyUnicode_GET_LENGTH(unicode_object); iteridx++) { \
|
||||
Py_UNICODE ch = _data[iteridx];
|
||||
#endif
|
||||
Py_ssize_t i = 0;
|
||||
Py_UNICODE *data = PyUnicode_AS_UNICODE(haystack);
|
||||
for (i = 0; i < PyUnicode_GET_SIZE(haystack); i++) {
|
||||
if (data[i] == c) return 1;
|
||||
}
|
||||
|
||||
#define END_ITER_CODE_PTS }}
|
||||
|
||||
static int
|
||||
contains_char(PyObject *haystack, const char c) {
|
||||
ITER_CODE_PTS(haystack)
|
||||
if (ch == c) return 1;
|
||||
END_ITER_CODE_PTS
|
||||
return 0;
|
||||
}
|
||||
|
||||
static PyObject *unicode_to_number(PyObject *src) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
#error Not implemented for python >= 3.3
|
||||
#endif
|
||||
PyObject *raw = NULL, *ans = NULL;
|
||||
raw = PyUnicode_AsASCIIString(src);
|
||||
if (raw == NULL) { return NULL; }
|
||||
if (contains_char(src, '.')) {
|
||||
#if PY_MAJOR_VERSION >= 3
|
||||
ans = PyFloat_FromString(raw);
|
||||
#else
|
||||
ans = PyFloat_FromString(raw, NULL);
|
||||
#endif
|
||||
} else {
|
||||
#if PY_MAJOR_VERSION >= 3
|
||||
ans = PyLong_FromUnicodeObject(raw, 10);
|
||||
#else
|
||||
ans = PyInt_FromString(PyString_AS_STRING(raw), NULL, 10);
|
||||
#endif
|
||||
}
|
||||
Py_DECREF(raw);
|
||||
return ans;
|
||||
}
|
||||
|
||||
|
||||
static void lowercase(PyObject *x) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
#error Not implemented for python >= 3.3
|
||||
ITER_CODE_PTS(x)
|
||||
if ('A' <= ch && ch <= 'Z') {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
PyUnicode_WRITE(_kind, _data, iteridx, ch + 32);
|
||||
#else
|
||||
_data[iteridx] += 32;
|
||||
#endif
|
||||
Py_ssize_t i = 0;
|
||||
Py_UNICODE *data = PyUnicode_AS_UNICODE(x);
|
||||
for (i = 0; i < PyUnicode_GET_SIZE(x); i++)
|
||||
data[i] = Py_UNICODE_TOLOWER(data[i]);
|
||||
}
|
||||
END_ITER_CODE_PTS
|
||||
}
|
||||
|
||||
static PyObject* clone_unicode(Py_UNICODE *x, Py_ssize_t sz) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
#error Not implemented for python >= 3.3
|
||||
static PyObject*
|
||||
clone_unicode(const PyObject* src, Py_ssize_t start_offset, Py_ssize_t end_offset) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
int kind = PyUnicode_KIND(src);
|
||||
void *data;
|
||||
switch(kind) {
|
||||
case PyUnicode_1BYTE_KIND:
|
||||
data = PyUnicode_1BYTE_DATA(src) + start_offset; break;
|
||||
case PyUnicode_2BYTE_KIND:
|
||||
data = PyUnicode_2BYTE_DATA(src) + start_offset; break;
|
||||
case PyUnicode_4BYTE_KIND:
|
||||
data = PyUnicode_4BYTE_DATA(src) + start_offset; break;
|
||||
|
||||
}
|
||||
return PyUnicode_FromKindAndData(kind, data, PyUnicode_GET_LENGTH(src) - start_offset - end_offset)
|
||||
#else
|
||||
return PyUnicode_FromUnicode(PyUnicode_AS_UNICODE(src) + start_offset, PyUnicode_GET_LENGTH(src) - start_offset - end_offset);
|
||||
#endif
|
||||
PyObject *ans = PyUnicode_FromUnicode(NULL, sz);
|
||||
if (ans == NULL) return PyErr_NoMemory();
|
||||
memcpy(PyUnicode_AS_UNICODE(ans), x, sz * sizeof(Py_UNICODE));
|
||||
return ans;
|
||||
}
|
||||
|
||||
static PyObject*
|
||||
tokenize_flat(PyObject *self, PyObject *args) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
#error Not implemented for python >= 3.3
|
||||
#endif
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
void *css_source = NULL; int css_kind; Py_UCS4 c = 0, codepoint = 0;
|
||||
#define first_char(string) PyUnicode_READ_CHAR(string, 0)
|
||||
#define unicode_from_data(data, sz) PyUnicode_FromKindAndData(css_kind, data, sz)
|
||||
#else
|
||||
Py_UNICODE *css_source = NULL, c = 0, codepoint = 0;
|
||||
#define first_char(string) PyUnicode_AS_UNICODE(string)[0]
|
||||
#define unicode_from_data(data, sz) PyUnicode_FromUnicode(data, sz)
|
||||
#endif
|
||||
PyObject *ic = NULL, *token = NULL, *tokens = NULL, *type_name = NULL, *css_value = NULL, *value = NULL, *unit = NULL, *tries = NULL, *match = NULL, *match_func = NULL, *py_source = NULL, *item = NULL, *newlines = NULL;
|
||||
int ignore_comments = 0;
|
||||
Py_ssize_t pos = 0, line = 1, column = 1, i = 0;
|
||||
@ -243,8 +288,13 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
|
||||
if (!PyArg_ParseTuple(args, "UO", &py_source, &ic)) return NULL;
|
||||
if (PyObject_IsTrue(ic)) ignore_comments = 1;
|
||||
source_len = PyUnicode_GET_SIZE(py_source);
|
||||
source_len = PyUnicode_GET_LENGTH(py_source);
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
if (PyUnicode_READY(py_source) != 0) return NULL;
|
||||
css_source = PyUnicode_DATA(py_source); css_kind = PyUnicode_KIND(py_source);
|
||||
#else
|
||||
css_source = PyUnicode_AS_UNICODE(py_source);
|
||||
#endif
|
||||
|
||||
tokens = PyList_New(0);
|
||||
if (tokens == NULL) return PyErr_NoMemory();
|
||||
@ -256,39 +306,44 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
#define SINGLE(x) { type_ = -1; type_name = x; Py_INCREF(type_name); css_value = x; Py_INCREF(css_value); }
|
||||
|
||||
while (pos < source_len) {
|
||||
#if PY_VERSION_HEX >= 0x03030000
|
||||
c = PyUnicode_READ(css_kind, css_data, pos);
|
||||
#else
|
||||
c = css_source[pos];
|
||||
#endif
|
||||
|
||||
css_value = NULL; type_name = NULL; value = NULL; unit = NULL; match = NULL;
|
||||
|
||||
if (c == ':') SINGLE(COLON) else if (c == ';') SINGLE(SCOLON) else if (c == '(') SINGLE(LPAR) else if (c == ')') SINGLE(RPAR) else if (c == '{') SINGLE(LBRACE) else if (c == '}') SINGLE(RBRACE) else if (c == '[') SINGLE(LBOX) else if (c == ']') SINGLE(RBOX) else
|
||||
if (c == ':') SINGLE(COLON) else if (c == ';') SINGLE(SCOLON) else if (c == '(') SINGLE(LPAR) else if (c == ')') SINGLE(RPAR) else if (c == '{') SINGLE(LBRACE) else if (c == '}') SINGLE(RBRACE) else if (c == '[') SINGLE(LBOX) else if (c == ']') SINGLE(RBOX) else
|
||||
{
|
||||
codepoint = (c > 160) ? 160: c;
|
||||
tries = PyList_GET_ITEM(TOKEN_DISPATCH, codepoint);
|
||||
for (i = 0; i < PyList_Size(tries); i++) {
|
||||
item = PyList_GET_ITEM(tries, i);
|
||||
match_func = PyTuple_GET_ITEM(item, 2);
|
||||
match = PyObject_CallFunction(match_func, "On", py_source, pos);
|
||||
match = PyObject_CallFunction(match_func, "On", py_source, pos);
|
||||
if (match == NULL) { goto error; }
|
||||
if (match != Py_None) {
|
||||
css_value = PyObject_CallMethod(match, "group", NULL);
|
||||
if (css_value == NULL) { goto error; }
|
||||
type_ = PyInt_AsSsize_t(PyTuple_GET_ITEM(item, 0));
|
||||
type_ = PyNumber_AsSsize_t(PyTuple_GET_ITEM(item, 0), PyExc_OverflowError);
|
||||
if(type_ == -1 && PyErr_Occurred() != NULL) { goto error; }
|
||||
type_name = PyTuple_GET_ITEM(item, 1);
|
||||
Py_INCREF(type_name);
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (css_value == NULL) { // No match
|
||||
type_ = DELIM; type_name = DELIM_TOK; Py_INCREF(type_name); css_value = clone_unicode(&c, 1);
|
||||
type_ = DELIM; type_name = DELIM_TOK; Py_INCREF(type_name); css_value = unicode_from_data(&c, 1);
|
||||
if (css_value == NULL) { goto error; }
|
||||
}
|
||||
}
|
||||
|
||||
length = PyUnicode_GET_SIZE(css_value);
|
||||
length = PyUnicode_GET_LENGTH(css_value);
|
||||
next_pos = pos + length;
|
||||
|
||||
// Now calculate the value and unit for this token (if any)
|
||||
if (! (ignore_comments && (type_ == COMMENT || type_ == BAD_COMMENT))) {
|
||||
if (! (ignore_comments && (type_ == COMMENT || type_ == BAD_COMMENT))) {
|
||||
if (type_ == DIMENSION) {
|
||||
value = PyObject_CallMethod(match, "group", "I", 1);
|
||||
if (value == NULL) { goto error; }
|
||||
@ -298,11 +353,11 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
UNESCAPE(unit, SIMPLE_UNESCAPE);
|
||||
UNESCAPE(unit, UNICODE_UNESCAPE);
|
||||
lowercase(unit);
|
||||
} else
|
||||
} else
|
||||
|
||||
if (type_ == PERCENTAGE) {
|
||||
if (PyUnicode_GET_SIZE(css_value) > 0) {
|
||||
value = clone_unicode(PyUnicode_AS_UNICODE(css_value), PyUnicode_GET_SIZE(css_value) - 1);
|
||||
if (PyUnicode_GET_LENGTH(css_value) > 0) {
|
||||
value = clone_unicode(css_value, 0, 1);
|
||||
if (value == NULL) goto error;
|
||||
} else { value = css_value; Py_INCREF(value); }
|
||||
if (value == NULL) goto error;
|
||||
@ -330,8 +385,8 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
if (type_ == URI) {
|
||||
value = PyObject_CallMethod(match, "group", "I", 1);
|
||||
if (value == NULL) { goto error; }
|
||||
if (PyObject_IsTrue(value) && PyUnicode_GET_SIZE(value) > 1 && (PyUnicode_AS_UNICODE(value)[0] == '"' || PyUnicode_AS_UNICODE(value)[0] == '\'')) {
|
||||
item = clone_unicode(PyUnicode_AS_UNICODE(value) + 1, PyUnicode_GET_SIZE(value) - 2);
|
||||
if (PyObject_IsTrue(value) && PyUnicode_GET_LENGTH(value) > 1 && (first_char(value) == '"' || first_char(value) == '\'')) {
|
||||
item = clone_unicode(value, 1, 1);
|
||||
if (item == NULL) goto error;
|
||||
Py_DECREF(value); value = item; item = NULL;
|
||||
UNESCAPE(value, NEWLINE_UNESCAPE);
|
||||
@ -341,8 +396,8 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
} else
|
||||
|
||||
if (type_ == STRING) {
|
||||
if (PyObject_IsTrue(css_value) && PyUnicode_GET_SIZE(css_value) > 1) { // remove quotes
|
||||
value = clone_unicode(PyUnicode_AS_UNICODE(css_value) + 1, PyUnicode_GET_SIZE(css_value) - 2);
|
||||
if (PyObject_IsTrue(css_value) && PyUnicode_GET_LENGTH(css_value) > 1) { // remove quotes
|
||||
value = clone_unicode(css_value, 1, 1);
|
||||
} else {
|
||||
value = css_value; Py_INCREF(value);
|
||||
}
|
||||
@ -353,8 +408,8 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
|
||||
if (type_ == BAD_STRING && next_pos == source_len) {
|
||||
Py_XDECREF(type_name); type_name = STRING_TOK; Py_INCREF(type_name);
|
||||
if (PyObject_IsTrue(css_value) && PyUnicode_GET_SIZE(css_value) > 0) { // remove quote
|
||||
value = clone_unicode(PyUnicode_AS_UNICODE(css_value) + 1, PyUnicode_GET_SIZE(css_value) - 1);
|
||||
if (PyObject_IsTrue(css_value) && PyUnicode_GET_LENGTH(css_value) > 0) { // remove quote
|
||||
value = clone_unicode(css_value, 1, 0);
|
||||
} else {
|
||||
value = css_value; Py_INCREF(value);
|
||||
}
|
||||
@ -386,8 +441,10 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
line += PyList_Size(newlines);
|
||||
item = PyObject_CallMethod(PyList_GET_ITEM(newlines, PyList_Size(newlines) - 1), "end", NULL);
|
||||
if (item == NULL) { Py_DECREF(newlines); newlines = NULL; goto error; }
|
||||
column = length - PyInt_AsSsize_t(item) + 1;
|
||||
Py_DECREF(item); item = NULL;
|
||||
column = PyNumber_AsSsize_t(item, PyExc_OverflowError);
|
||||
if(column == -1 && PyErr_Occurred()) { Py_DECREF(newlines); newlines = NULL; goto error; }
|
||||
column = length - column + 1;
|
||||
Py_DECREF(item); item = NULL;
|
||||
} else column += length;
|
||||
Py_DECREF(newlines); newlines = NULL;
|
||||
|
||||
@ -397,6 +454,8 @@ tokenize_flat(PyObject *self, PyObject *args) {
|
||||
error:
|
||||
Py_XDECREF(tokens); Py_XDECREF(css_value); Py_XDECREF(type_name); Py_XDECREF(value); Py_XDECREF(unit); Py_XDECREF(match);
|
||||
return NULL;
|
||||
#undef unicode_from_data
|
||||
#undef first_char
|
||||
}
|
||||
|
||||
static PyMethodDef tokenizer_methods[] = {
|
||||
@ -415,17 +474,41 @@ static PyMethodDef tokenizer_methods[] = {
|
||||
{NULL, NULL, 0, NULL}
|
||||
};
|
||||
|
||||
#if PY_MAJOR_VERSION >= 3
|
||||
#define INITERROR return NULL
|
||||
static struct PyModuleDef tokenizer_module = {
|
||||
/* m_base */ PyModuleDef_HEAD_INIT,
|
||||
/* m_name */ "tokenizer",
|
||||
/* m_doc */ "Implementation of tokenizer in C for speed.",
|
||||
/* m_size */ -1,
|
||||
/* m_methods */ tokenizer_methods,
|
||||
/* m_slots */ 0,
|
||||
/* m_traverse */ 0,
|
||||
/* m_clear */ 0,
|
||||
/* m_free */ 0,
|
||||
};
|
||||
|
||||
CALIBRE_MODINIT_FUNC
|
||||
inittokenizer(void) {
|
||||
PyObject *m;
|
||||
CALIBRE_MODINIT_FUNC PyInit_tokenizer(void) {
|
||||
if (PyType_Ready(&tokenizer_TokenType) < 0)
|
||||
return;
|
||||
INITERROR;
|
||||
|
||||
m = Py_InitModule3("tokenizer", tokenizer_methods,
|
||||
"Implementation of tokenizer in C for speed."
|
||||
);
|
||||
if (m == NULL) return;
|
||||
PyObject *mod = PyModule_Create(&tokenizer_module);
|
||||
#else
|
||||
#define INITERROR return
|
||||
CALIBRE_MODINIT_FUNC inittokenizer(void) {
|
||||
if (PyType_Ready(&tokenizer_TokenType) < 0)
|
||||
INITERROR;
|
||||
|
||||
PyObject *mod = Py_InitModule3("tokenizer", tokenizer_methods,
|
||||
"Implementation of tokenizer in C for speed.");
|
||||
#endif
|
||||
|
||||
if (mod == NULL) INITERROR;
|
||||
Py_INCREF(&tokenizer_TokenType);
|
||||
PyModule_AddObject(m, "Token", (PyObject *)&tokenizer_TokenType);
|
||||
PyModule_AddObject(mod, "Token", (PyObject *) &tokenizer_TokenType);
|
||||
|
||||
|
||||
#if PY_MAJOR_VERSION >= 3
|
||||
return mod;
|
||||
#endif
|
||||
}
|
||||
|
Loading…
x
Reference in New Issue
Block a user