cmlenz@541: /*
cmlenz@719:  * Copyright (C) 2006-2008 Edgewall Software
cmlenz@541:  * All rights reserved.
cmlenz@541:  *
cmlenz@541:  * This software is licensed as described in the file COPYING, which
cmlenz@541:  * you should have received as part of this distribution. The terms
cmlenz@541:  * are also available at http://genshi.edgewall.org/wiki/License.
cmlenz@541:  *
cmlenz@541:  * This software consists of voluntary contributions made by many
cmlenz@541:  * individuals. For the exact contribution history, see the revision
cmlenz@541:  * history and logs, available at http://genshi.edgewall.org/log/.
cmlenz@541:  */
cmlenz@541: 
cmlenz@541: #include <Python.h>
cmlenz@541: #include <structmember.h>
cmlenz@541: 
hodgestar@930: #if PY_MAJOR_VERSION > 2
hodgestar@930: #   define IS_PY3K
hodgestar@930: #elif PY_VERSION_HEX < 0x02050000 && !defined(PY_SSIZE_T_MIN)
hodgestar@930:     typedef int Py_ssize_t;
hodgestar@930: #   define PY_SSIZE_T_MAX INT_MAX
hodgestar@930: #   define PY_SSIZE_T_MIN INT_MIN
hodgestar@930: #endif
hodgestar@930: 
hodgestar@930: /* We only use Unicode Strings in this module */
hodgestar@930: #ifndef IS_PY3K
hodgestar@930: #   define PyObject_Str PyObject_Unicode
jruigrok@723: #endif
jruigrok@723: 
cmlenz@541: static PyObject *amp1, *amp2, *lt1, *lt2, *gt1, *gt2, *qt1, *qt2;
cmlenz@541: static PyObject *stripentities, *striptags;
cmlenz@541: 
cmlenz@541: static void
cmlenz@541: init_constants(void)
cmlenz@541: {
cmlenz@541:     PyObject *util = PyImport_ImportModule("genshi.util");
cmlenz@541:     stripentities = PyObject_GetAttrString(util, "stripentities");
cmlenz@541:     striptags = PyObject_GetAttrString(util, "striptags");
cmlenz@541:     Py_DECREF(util);
cmlenz@541: 
cmlenz@541:     amp1 = PyUnicode_DecodeASCII("&", 1, NULL);
cmlenz@541:     amp2 = PyUnicode_DecodeASCII("&amp;", 5, NULL);
cmlenz@541:     lt1 = PyUnicode_DecodeASCII("<", 1, NULL);
cmlenz@541:     lt2 = PyUnicode_DecodeASCII("&lt;", 4, NULL);
cmlenz@541:     gt1 = PyUnicode_DecodeASCII(">", 1, NULL);
cmlenz@541:     gt2 = PyUnicode_DecodeASCII("&gt;", 4, NULL);
cmlenz@541:     qt1 = PyUnicode_DecodeASCII("\"", 1, NULL);
cmlenz@541:     qt2 = PyUnicode_DecodeASCII("&#34;", 5, NULL);
cmlenz@541: }
cmlenz@541: 
cmlenz@541: /* Markup class */
cmlenz@541: 
cmlenz@795: PyTypeObject MarkupType; /* declared later */
cmlenz@541: 
cmlenz@541: PyDoc_STRVAR(Markup__doc__,
cmlenz@541: "Marks a string as being safe for inclusion in HTML/XML output without\n\
cmlenz@541: needing to be escaped.");
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: escape(PyObject *text, int quotes)
cmlenz@541: {
cmlenz@541:     PyObject *args, *ret;
cmlenz@541:     PyUnicodeObject *in, *out;
cmlenz@541:     Py_UNICODE *inp, *outp;
cmlenz@541:     int len, inn, outn;
cmlenz@541: 
cmlenz@541:     if (PyObject_TypeCheck(text, &MarkupType)) {
cmlenz@541:         Py_INCREF(text);
cmlenz@541:         return text;
cmlenz@541:     }
cmlenz@737:     if (PyObject_HasAttrString(text, "__html__")) {
cmlenz@737:         ret = PyObject_CallMethod(text, "__html__", NULL);
cmlenz@737:         args = PyTuple_New(1);
cmlenz@737:         if (args == NULL) {
cmlenz@737:             Py_DECREF(ret);
cmlenz@737:             return NULL;
cmlenz@737:         }
cmlenz@737:         PyTuple_SET_ITEM(args, 0, ret);
cmlenz@737:         ret = MarkupType.tp_new(&MarkupType, args, NULL);
cmlenz@737:         Py_DECREF(args);
cmlenz@737:         return ret;
cmlenz@737:     }
hodgestar@930:     in = (PyUnicodeObject *) PyObject_Str(text);
cmlenz@541:     if (in == NULL) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     /* First we need to figure out how long the escaped string will be */
cmlenz@541:     len = inn = 0;
cmlenz@541:     inp = in->str;
cmlenz@541:     while (*(inp) || in->length > inp - in->str) {
cmlenz@541:         switch (*inp++) {
cmlenz@541:             case '&': len += 5; inn++;                                 break;
cmlenz@541:             case '"': len += quotes ? 5 : 1; inn += quotes ? 1 : 0;    break;
cmlenz@541:             case '<':
cmlenz@541:             case '>': len += 4; inn++;                                 break;
cmlenz@541:             default:  len++;
cmlenz@541:         }
cmlenz@541:     }
cmlenz@541: 
cmlenz@541:     /* Do we need to escape anything at all? */
cmlenz@541:     if (!inn) {
cmlenz@541:         args = PyTuple_New(1);
cmlenz@541:         if (args == NULL) {
cmlenz@541:             Py_DECREF((PyObject *) in);
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@541:         PyTuple_SET_ITEM(args, 0, (PyObject *) in);
cmlenz@541:         ret = MarkupType.tp_new(&MarkupType, args, NULL);
cmlenz@541:         Py_DECREF(args);
cmlenz@541:         return ret;
cmlenz@541:     }
cmlenz@541: 
cmlenz@541:     out = (PyUnicodeObject*) PyUnicode_FromUnicode(NULL, len);
cmlenz@541:     if (out == NULL) {
cmlenz@665:         Py_DECREF((PyObject *) in);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541: 
cmlenz@541:     outn = 0;
cmlenz@541:     inp = in->str;
cmlenz@541:     outp = out->str;
cmlenz@541:     while (*(inp) || in->length > inp - in->str) {
cmlenz@541:         if (outn == inn) {
cmlenz@541:             /* copy rest of string if we have already replaced everything */
cmlenz@541:             Py_UNICODE_COPY(outp, inp, in->length - (inp - in->str));
cmlenz@541:             break;
cmlenz@541:         }
cmlenz@541:         switch (*inp) {
cmlenz@541:             case '&':
cmlenz@541:                 Py_UNICODE_COPY(outp, ((PyUnicodeObject *) amp2)->str, 5);
cmlenz@541:                 outp += 5;
cmlenz@541:                 outn++;
cmlenz@541:                 break;
cmlenz@541:             case '"':
cmlenz@541:                 if (quotes) {
cmlenz@541:                     Py_UNICODE_COPY(outp, ((PyUnicodeObject *) qt2)->str, 5);
cmlenz@541:                     outp += 5;
cmlenz@541:                     outn++;
cmlenz@541:                 } else {
cmlenz@541:                     *outp++ = *inp;
cmlenz@541:                 }
cmlenz@541:                 break;
cmlenz@541:             case '<':
cmlenz@541:                 Py_UNICODE_COPY(outp, ((PyUnicodeObject *) lt2)->str, 4);
cmlenz@541:                 outp += 4;
cmlenz@541:                 outn++;
cmlenz@541:                 break;
cmlenz@541:             case '>':
cmlenz@541:                 Py_UNICODE_COPY(outp, ((PyUnicodeObject *) gt2)->str, 4);
cmlenz@541:                 outp += 4;
cmlenz@541:                 outn++;
cmlenz@541:                 break;
cmlenz@541:             default:
cmlenz@541:                 *outp++ = *inp;
cmlenz@541:         }
cmlenz@541:         inp++;
cmlenz@541:     }
cmlenz@541: 
cmlenz@665:     Py_DECREF((PyObject *) in);
cmlenz@665: 
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@541:         Py_DECREF((PyObject *) out);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, (PyObject *) out);
cmlenz@541:     ret = MarkupType.tp_new(&MarkupType, args, NULL);
cmlenz@541:     Py_DECREF(args);
cmlenz@541:     return ret;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: PyDoc_STRVAR(escape__doc__,
cmlenz@541: "Create a Markup instance from a string and escape special characters\n\
cmlenz@541: it may contain (<, >, & and \").\n\
cmlenz@541: \n\
cmlenz@547: >>> escape('\"1 < 2\"')\n\
cmlenz@547: <Markup u'&#34;1 &lt; 2&#34;'>\n\
cmlenz@547: \n\
cmlenz@541: If the `quotes` parameter is set to `False`, the \" character is left\n\
cmlenz@541: as is. Escaping quotes is generally only required for strings that are\n\
cmlenz@547: to be used in attribute values.\n\
cmlenz@547: \n\
cmlenz@547: >>> escape('\"1 < 2\"', quotes=False)\n\
cmlenz@547: <Markup u'\"1 &lt; 2\"'>\n\
cmlenz@547: \n\
cmlenz@547: :param text: the text to escape\n\
cmlenz@547: :param quotes: if ``True``, double quote characters are escaped in\n\
cmlenz@547:                addition to the other special characters\n\
cmlenz@547: :return: the escaped `Markup` string\n\
cmlenz@547: :rtype: `Markup`\n\
cmlenz@547: ");
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_escape(PyTypeObject* type, PyObject *args, PyObject *kwds)
cmlenz@541: {
cmlenz@541:     static char *kwlist[] = {"text", "quotes", 0};
cmlenz@541:     PyObject *text = NULL;
cmlenz@541:     char quotes = 1;
cmlenz@541: 
cmlenz@541:     if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|b", kwlist, &text, &quotes)) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     if (PyObject_Not(text)) {
cmlenz@541:         return type->tp_new(type, args, NULL);
cmlenz@541:     }
cmlenz@541:     if (PyObject_TypeCheck(text, type)) {
cmlenz@541:         Py_INCREF(text);
cmlenz@541:         return text;
cmlenz@541:     }
cmlenz@541:     return escape(text, quotes);
cmlenz@541: }
cmlenz@541: 
cmlenz@737: static PyObject *
cmlenz@737: Markup_html(PyObject *self)
cmlenz@737: {
cmlenz@737:     Py_INCREF(self);
cmlenz@737:     return self;
cmlenz@737: }
cmlenz@737: 
cmlenz@547: PyDoc_STRVAR(join__doc__,
cmlenz@547: "Return a `Markup` object which is the concatenation of the strings\n\
cmlenz@547: in the given sequence, where this `Markup` object is the separator\n\
cmlenz@547: between the joined elements.\n\
cmlenz@547: \n\
cmlenz@547: Any element in the sequence that is not a `Markup` instance is\n\
cmlenz@547: automatically escaped.\n\
cmlenz@547: \n\
cmlenz@547: :param seq: the sequence of strings to join\n\
cmlenz@547: :param escape_quotes: whether double quote characters in the elements\n\
cmlenz@547:                       should be escaped\n\
cmlenz@547: :return: the joined `Markup` object\n\
cmlenz@547: :rtype: `Markup`\n\
cmlenz@547: :see: `escape`\n\
cmlenz@547: ");
cmlenz@547: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_join(PyObject *self, PyObject *args, PyObject *kwds)
cmlenz@541: {
cmlenz@541:     static char *kwlist[] = {"seq", "escape_quotes", 0};
cmlenz@665:     PyObject *seq = NULL, *seq2, *tmp, *tmp2;
cmlenz@541:     char quotes = 1;
cmlenz@795:     Py_ssize_t n;
cmlenz@795:     int i;
cmlenz@541: 
cmlenz@541:     if (!PyArg_ParseTupleAndKeywords(args, kwds, "O|b", kwlist, &seq, &quotes)) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     if (!PySequence_Check(seq)) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     n = PySequence_Size(seq);
cmlenz@541:     if (n < 0) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     seq2 = PyTuple_New(n);
cmlenz@541:     if (seq2 == NULL) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     for (i = 0; i < n; i++) {
cmlenz@541:         tmp = PySequence_GetItem(seq, i);
cmlenz@541:         if (tmp == NULL) {
cmlenz@541:             Py_DECREF(seq2);
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@665:         tmp2 = escape(tmp, quotes);
cmlenz@665:         if (tmp2 == NULL) {
cmlenz@541:             Py_DECREF(seq2);
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@665:         PyTuple_SET_ITEM(seq2, i, tmp2);
cmlenz@665:         Py_DECREF(tmp);
cmlenz@541:     }
cmlenz@541:     tmp = PyUnicode_Join(self, seq2);
cmlenz@541:     Py_DECREF(seq2);
cmlenz@541:     if (tmp == NULL)
cmlenz@541:         return NULL;
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@541:         Py_DECREF(tmp);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, tmp);
cmlenz@541:     tmp = MarkupType.tp_new(&MarkupType, args, NULL);
cmlenz@541:     Py_DECREF(args);
cmlenz@541:     return tmp;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_add(PyObject *self, PyObject *other)
cmlenz@541: {
cmlenz@541:     PyObject *tmp, *tmp2, *args, *ret;
cmlenz@541:     if (PyObject_TypeCheck(self, &MarkupType)) {
cmlenz@541:         tmp = escape(other, 1);
cmlenz@541:         if (tmp == NULL)
cmlenz@541:             return NULL;
cmlenz@541:         tmp2 = PyUnicode_Concat(self, tmp);
cmlenz@541:     } else { // __radd__
cmlenz@541:         tmp = escape(self, 1);
cmlenz@541:         if (tmp == NULL)
cmlenz@541:             return NULL;
cmlenz@541:         tmp2 = PyUnicode_Concat(tmp, other);
cmlenz@541:     }
cmlenz@665:     Py_DECREF(tmp);
cmlenz@665:     if (tmp2 == NULL)
cmlenz@541:         return NULL;
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@541:         Py_DECREF(tmp2);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, tmp2);
cmlenz@541:     ret = MarkupType.tp_new(&MarkupType, args, NULL);
cmlenz@541:     Py_DECREF(args);
cmlenz@541:     return ret;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_mod(PyObject *self, PyObject *args)
cmlenz@541: {
cmlenz@541:     PyObject *tmp, *tmp2, *ret, *args2;
cmlenz@795:     int i;
cmlenz@795:     Py_ssize_t nargs = 0;
cmlenz@713:     PyObject *kwds = NULL;
cmlenz@541: 
cmlenz@713:     if (PyDict_Check(args)) {
cmlenz@713:         kwds = args;
cmlenz@713:     }
cmlenz@713:     if (kwds && PyDict_Size(kwds)) {
cmlenz@713:         PyObject *kwcopy, *key, *value;
cmlenz@713:         Py_ssize_t pos = 0;
cmlenz@713: 
cmlenz@713:         kwcopy = PyDict_Copy( kwds );
cmlenz@713:         if (kwcopy == NULL) {
cmlenz@713:             return NULL;
cmlenz@713:         }
cmlenz@713:         while (PyDict_Next(kwcopy, &pos, &key, &value)) {
cmlenz@713:             tmp = escape(value, 1);
cmlenz@713:             if (tmp == NULL) {
cmlenz@713:                 Py_DECREF(kwcopy);
cmlenz@713:                 return NULL;
cmlenz@713:             }
cmlenz@713:             if (PyDict_SetItem(kwcopy, key, tmp) < 0) {
cmlenz@713:                 Py_DECREF(tmp);
cmlenz@713:                 Py_DECREF(kwcopy);
cmlenz@713:                 return NULL;
cmlenz@713:             }
cmlenz@713:         }
cmlenz@713:         tmp = PyUnicode_Format(self, kwcopy);
cmlenz@713:         Py_DECREF(kwcopy);
cmlenz@713:         if (tmp == NULL) {
cmlenz@713:             return NULL;
cmlenz@713:         }
cmlenz@713:     } else if (PyTuple_Check(args)) {
cmlenz@541:         nargs = PyTuple_GET_SIZE(args);
cmlenz@541:         args2 = PyTuple_New(nargs);
cmlenz@541:         if (args2 == NULL) {
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@541:         for (i = 0; i < nargs; i++) {
cmlenz@541:             tmp = escape(PyTuple_GET_ITEM(args, i), 1);
cmlenz@541:             if (tmp == NULL) {
cmlenz@541:                 Py_DECREF(args2);
cmlenz@541:                 return NULL;
cmlenz@541:             }
cmlenz@541:             PyTuple_SET_ITEM(args2, i, tmp);
cmlenz@541:         }
cmlenz@541:         tmp = PyUnicode_Format(self, args2);
cmlenz@541:         Py_DECREF(args2);
cmlenz@541:         if (tmp == NULL) {
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@541:     } else {
cmlenz@541:         tmp2 = escape(args, 1);
cmlenz@541:         if (tmp2 == NULL) {
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@541:         tmp = PyUnicode_Format(self, tmp2);
cmlenz@541:         Py_DECREF(tmp2);
cmlenz@541:         if (tmp == NULL) {
cmlenz@541:             return NULL;
cmlenz@541:         }
cmlenz@541:     }
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@541:         Py_DECREF(tmp);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, tmp);
cmlenz@541:     ret = PyUnicode_Type.tp_new(&MarkupType, args, NULL);
cmlenz@541:     Py_DECREF(args);
cmlenz@541:     return ret;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_mul(PyObject *self, PyObject *num)
cmlenz@541: {
cmlenz@541:     PyObject *unicode, *result, *args;
cmlenz@541: 
cmlenz@541:     if (PyObject_TypeCheck(self, &MarkupType)) {
hodgestar@930:         unicode = PyObject_Str(self);
cmlenz@541:         if (unicode == NULL) return NULL;
cmlenz@541:         result = PyNumber_Multiply(unicode, num);
cmlenz@541:     } else { // __rmul__
hodgestar@930:         unicode = PyObject_Str(num);
cmlenz@541:         if (unicode == NULL) return NULL;
cmlenz@541:         result = PyNumber_Multiply(unicode, self);
cmlenz@541:     }
cmlenz@665:     Py_DECREF(unicode);
cmlenz@541: 
cmlenz@541:     if (result == NULL) return NULL;
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@541:         Py_DECREF(result);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, result);
cmlenz@541:     result = PyUnicode_Type.tp_new(&MarkupType, args, NULL);
cmlenz@541:     Py_DECREF(args);
cmlenz@541: 
cmlenz@541:     return result;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_repr(PyObject *self)
cmlenz@541: {
cmlenz@541:     PyObject *format, *result, *args;
cmlenz@541: 
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     format = PyUnicode_FromString("<Markup %r>");
hodgestar@930: #else
cmlenz@541:     format = PyString_FromString("<Markup %r>");
hodgestar@930: #endif
cmlenz@541:     if (format == NULL) return NULL;
hodgestar@930:     result = PyObject_Str(self);
cmlenz@665:     if (result == NULL) {
cmlenz@665:         Py_DECREF(format);
cmlenz@665:         return NULL;
cmlenz@665:     }
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@665:         Py_DECREF(format);
cmlenz@541:         Py_DECREF(result);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, result);
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     result = PyUnicode_Format(format, args);
hodgestar@930: #else
cmlenz@541:     result = PyString_Format(format, args);
hodgestar@930: #endif
cmlenz@681:     Py_DECREF(format);
cmlenz@541:     Py_DECREF(args);
cmlenz@541:     return result;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: PyDoc_STRVAR(unescape__doc__,
cmlenz@547: "Reverse-escapes &, <, >, and \" and returns a `unicode` object.\n\
cmlenz@547: \n\
cmlenz@547: >>> Markup('1 &lt; 2').unescape()\n\
cmlenz@547: u'1 < 2'\n\
cmlenz@547: \n\
cmlenz@547: :return: the unescaped string\n\
cmlenz@547: :rtype: `unicode`\n\
cmlenz@547: :see: `genshi.core.unescape`\n\
cmlenz@547: ");
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_unescape(PyObject* self)
cmlenz@541: {
cmlenz@541:     PyObject *tmp, *tmp2;
cmlenz@541: 
cmlenz@541:     tmp = PyUnicode_Replace(self, qt2, qt1, -1);
cmlenz@541:     if (tmp == NULL) return NULL;
cmlenz@541:     tmp2 = PyUnicode_Replace(tmp, gt2, gt1, -1);
cmlenz@541:     Py_DECREF(tmp);
cmlenz@541:     if (tmp2 == NULL) return NULL;
cmlenz@541:     tmp = PyUnicode_Replace(tmp2, lt2, lt1, -1);
cmlenz@541:     Py_DECREF(tmp2);
cmlenz@541:     if (tmp == NULL) return NULL;
cmlenz@541:     tmp2 = PyUnicode_Replace(tmp, amp2, amp1, -1);
cmlenz@541:     Py_DECREF(tmp);
cmlenz@541:     return tmp2;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: PyDoc_STRVAR(stripentities__doc__,
cmlenz@541: "Return a copy of the text with any character or numeric entities\n\
cmlenz@541: replaced by the equivalent UTF-8 characters.\n\
cmlenz@541: \n\
cmlenz@541: If the `keepxmlentities` parameter is provided and evaluates to `True`,\n\
cmlenz@547: the core XML entities (``&amp;``, ``&apos;``, ``&gt;``, ``&lt;`` and\n\
cmlenz@547: ``&quot;``) are not stripped.\n\
cmlenz@547: \n\
cmlenz@547: :return: a `Markup` instance with entities removed\n\
cmlenz@547: :rtype: `Markup`\n\
cmlenz@547: :see: `genshi.util.stripentities`\n\
cmlenz@547: ");
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_stripentities(PyObject* self, PyObject *args, PyObject *kwds)
cmlenz@541: {
cmlenz@541:     static char *kwlist[] = {"keepxmlentities", 0};
cmlenz@541:     PyObject *result, *args2;
cmlenz@541:     char keepxml = 0;
cmlenz@541: 
cmlenz@541:     if (!PyArg_ParseTupleAndKeywords(args, kwds, "|b", kwlist, &keepxml)) {
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541: 
cmlenz@541:     if (stripentities == NULL) return NULL;
cmlenz@541:     result = PyObject_CallFunction(stripentities, "Ob", self, keepxml);
cmlenz@541:     if (result == NULL) return NULL;
cmlenz@541:     args2 = PyTuple_New(1);
cmlenz@541:     if (args2 == NULL) {
cmlenz@541:         Py_DECREF(result);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args2, 0, result);
cmlenz@541:     result = MarkupType.tp_new(&MarkupType, args2, NULL);
cmlenz@541:     Py_DECREF(args2);
cmlenz@541:     return result;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: PyDoc_STRVAR(striptags__doc__,
cmlenz@547: """Return a copy of the text with all XML/HTML tags removed.\n\
cmlenz@547: \n\
cmlenz@547: :return: a `Markup` instance with all tags removed\n\
cmlenz@547: :rtype: `Markup`\n\
cmlenz@547: :see: `genshi.util.striptags`\n\
cmlenz@547: ");
cmlenz@541: 
cmlenz@541: static PyObject *
cmlenz@541: Markup_striptags(PyObject* self)
cmlenz@541: {
cmlenz@541:     PyObject *result, *args;
cmlenz@541: 
cmlenz@541:     if (striptags == NULL) return NULL;
cmlenz@541:     result = PyObject_CallFunction(striptags, "O", self);
cmlenz@541:     if (result == NULL) return NULL;
cmlenz@541:     args = PyTuple_New(1);
cmlenz@541:     if (args == NULL) {
cmlenz@541:         Py_DECREF(result);
cmlenz@541:         return NULL;
cmlenz@541:     }
cmlenz@541:     PyTuple_SET_ITEM(args, 0, result);
cmlenz@541:     result = MarkupType.tp_new(&MarkupType, args, NULL);
cmlenz@541:     Py_DECREF(args);
cmlenz@541:     return result;
cmlenz@541: }
cmlenz@541: 
cmlenz@541: typedef struct {
cmlenz@541:     PyUnicodeObject HEAD;
cmlenz@541: } MarkupObject;
cmlenz@541: 
cmlenz@541: static PyMethodDef Markup_methods[] = {
cmlenz@737:     {"__html__", (PyCFunction) Markup_html, METH_NOARGS, NULL},
cmlenz@541:     {"escape", (PyCFunction) Markup_escape,
cmlenz@547:      METH_VARARGS|METH_CLASS|METH_KEYWORDS, escape__doc__},
cmlenz@547:     {"join", (PyCFunction)Markup_join, METH_VARARGS|METH_KEYWORDS, join__doc__},
cmlenz@541:     {"unescape", (PyCFunction)Markup_unescape, METH_NOARGS, unescape__doc__},
cmlenz@541:     {"stripentities", (PyCFunction) Markup_stripentities,
cmlenz@541:      METH_VARARGS|METH_KEYWORDS, stripentities__doc__},
cmlenz@541:     {"striptags", (PyCFunction) Markup_striptags, METH_NOARGS,
cmlenz@541:      striptags__doc__},
cmlenz@541:     {NULL}  /* Sentinel */
cmlenz@541: };
cmlenz@541: 
cmlenz@541: static PyNumberMethods Markup_as_number = {
cmlenz@547:     Markup_add, /*nb_add*/
cmlenz@547:     0, /*nb_subtract*/
cmlenz@547:     Markup_mul, /*nb_multiply*/
hodgestar@930: #ifndef IS_PY3K
cmlenz@547:     0, /*nb_divide*/
hodgestar@930: #endif
cmlenz@547:     Markup_mod, /*nb_remainder*/
cmlenz@541: };
cmlenz@541: 
cmlenz@541: PyTypeObject MarkupType = {
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     PyVarObject_HEAD_INIT(NULL, 0)
hodgestar@930: #else
cmlenz@541:     PyObject_HEAD_INIT(NULL)
cmlenz@541:     0,
hodgestar@930: #endif
cmlenz@541:     "genshi._speedups.Markup",
cmlenz@541:     sizeof(MarkupObject),
cmlenz@541:     0,
cmlenz@541:     0,          /*tp_dealloc*/
jruigrok@722:     0,          /*tp_print*/
cmlenz@541:     0,          /*tp_getattr*/
cmlenz@541:     0,          /*tp_setattr*/
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     0,          /*tp_reserved*/
hodgestar@930: #else
cmlenz@541:     0,          /*tp_compare*/
hodgestar@930: #endif
cmlenz@541:     Markup_repr, /*tp_repr*/
cmlenz@541:     &Markup_as_number, /*tp_as_number*/
cmlenz@541:     0,          /*tp_as_sequence*/
cmlenz@541:     0,          /*tp_as_mapping*/
cmlenz@541:     0,          /*tp_hash */
cmlenz@541: 
cmlenz@541:     0,          /*tp_call*/
cmlenz@541:     0,          /*tp_str*/
cmlenz@541:     0,          /*tp_getattro*/
cmlenz@541:     0,          /*tp_setattro*/
cmlenz@541:     0,          /*tp_as_buffer*/
cmlenz@541: 
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_UNICODE_SUBCLASS, /*tp_flags*/
hodgestar@930: #elif defined(Py_TPFLAGS_UNICODE_SUBCLASS)
hodgestar@930:     Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_CHECKTYPES | Py_TPFLAGS_UNICODE_SUBCLASS, /*tp_flags*/
hodgestar@930: #else
cmlenz@541:     Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_CHECKTYPES, /*tp_flags*/
hodgestar@930: #endif
hodgestar@930: 
cmlenz@541:     Markup__doc__,/*tp_doc*/
jruigrok@722: 
cmlenz@541:     0,          /*tp_traverse*/
cmlenz@541:     0,          /*tp_clear*/
cmlenz@541: 
cmlenz@541:     0,          /*tp_richcompare*/
cmlenz@541:     0,          /*tp_weaklistoffset*/
cmlenz@541: 
cmlenz@541:     0,          /*tp_iter*/
cmlenz@541:     0,          /*tp_iternext*/
cmlenz@541: 
cmlenz@541:     /* Attribute descriptor and subclassing stuff */
cmlenz@541: 
cmlenz@541:     Markup_methods,/*tp_methods*/
cmlenz@541:     0,          /*tp_members*/
cmlenz@541:     0,          /*tp_getset*/
cmlenz@541:     0,          /*tp_base*/
cmlenz@541:     0,          /*tp_dict*/
jruigrok@722: 
cmlenz@541:     0,          /*tp_descr_get*/
cmlenz@541:     0,          /*tp_descr_set*/
cmlenz@541:     0,          /*tp_dictoffset*/
jruigrok@722: 
cmlenz@541:     0,          /*tp_init*/
cmlenz@541:     0,          /*tp_alloc  will be set to PyType_GenericAlloc in module init*/
cmlenz@713:     0,          /*tp_new*/
cmlenz@541:     0,          /*tp_free  Low-level free-memory routine */
cmlenz@541:     0,          /*tp_is_gc For PyObject_IS_GC */
cmlenz@541:     0,          /*tp_bases*/
cmlenz@541:     0,          /*tp_mro method resolution order */
cmlenz@541:     0,          /*tp_cache*/
cmlenz@541:     0,          /*tp_subclasses*/
cmlenz@541:     0           /*tp_weaklist*/
cmlenz@541: };
cmlenz@541: 
hodgestar@930: #ifdef IS_PY3K
hodgestar@930: struct PyModuleDef module_def = {
hodgestar@930:     PyModuleDef_HEAD_INIT, /*m_base*/
hodgestar@930:     "_speedups",           /*m_name*/
hodgestar@930:     NULL,                  /*m_doc*/
hodgestar@930:     -1,                    /*m_size*/
hodgestar@930:     NULL,                  /*m_methods*/
hodgestar@930:     NULL,                  /*m_reload*/
hodgestar@930:     NULL,                  /*m_traverse*/
hodgestar@930:     NULL,                  /*m_clear*/
hodgestar@930:     NULL                   /*m_free*/
hodgestar@930: };
hodgestar@930: 
hodgestar@930: PyObject *
hodgestar@930: PyInit__speedups(void)
hodgestar@930: #else
cmlenz@541: PyMODINIT_FUNC
cmlenz@541: init_speedups(void)
hodgestar@930: #endif
cmlenz@541: {
cmlenz@541:     PyObject *module;
cmlenz@541: 
cmlenz@541:     /* Workaround for quirk in Visual Studio, see
cmlenz@541:         <http://www.python.it/faq/faq-3.html#3.24> */
cmlenz@541:     MarkupType.tp_base = &PyUnicode_Type;
cmlenz@541: 
cmlenz@541:     if (PyType_Ready(&MarkupType) < 0)
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:         return NULL;
hodgestar@930: #else
cmlenz@541:         return;
hodgestar@930: #endif
cmlenz@541: 
cmlenz@541:     init_constants();
cmlenz@541: 
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     module = PyModule_Create(&module_def);
hodgestar@930: #else
cmlenz@541:     module = Py_InitModule("_speedups", NULL);
hodgestar@930: #endif
cmlenz@541:     Py_INCREF(&MarkupType);
cmlenz@541:     PyModule_AddObject(module, "Markup", (PyObject *) &MarkupType);
hodgestar@930: 
hodgestar@930: #ifdef IS_PY3K
hodgestar@930:     return module;
hodgestar@930: #endif
cmlenz@541: }