Merge from HEAD:

>Rename unpack_* functions to be consistent
>
>Decode CP850 strings when unpacking
>
>Add variable giving default encoding name

source/python/py_tdbpack.c

@@ -31,7 +31,7 @@ static PyObject * pytdbpack_number(char ch, PyObject *val_iter, PyObject *packed
 static PyObject * pytdbpack_str_850(PyObject *val_iter, PyObject *packed_list);
 static PyObject * pytdbpack_buffer(PyObject *val_iter, PyObject *packed_list);
 
-static PyObject *pytdbpack_unpack_item(char, char **pbuf, int *plen, PyObject *);
+static PyObject *pytdbunpack_item(char, char **pbuf, int *plen, PyObject *);
 
 static PyObject *pytdbpack_data(const char *format_str,
 				     PyObject *val_seq,
@@ -117,7 +117,7 @@ notes:
 ";
 
 
-static char const pytdbpack_unpack_doc[] =
+static char const pytdbunpack_doc[] =
 "unpack(format, buffer) -> (values, rest)
 Unpack Samba binary data according to format string.
 
@@ -145,6 +145,7 @@ notes:
 ";
 
 
+const char *pytdb_string_encoding = "cp850";
 
 
 /*
@@ -324,7 +325,7 @@ pytdbpack_str_850(PyObject *val_iter, PyObject *packed_list)
 		val_obj = NULL;
 	}
 
-	if (!(cp850_str = PyUnicode_AsEncodedString(unicode_obj, "cp850", NULL)))
+	if (!(cp850_str = PyUnicode_AsEncodedString(unicode_obj, pytdb_string_encoding, NULL)))
 		goto out;
 
 	if (!nul_str)
@@ -378,208 +379,6 @@ pytdbpack_buffer(PyObject *val_iter, PyObject *packed_list)
 }
 
 
-#if 0
-else if (ch == 'B') {
-			long size;
-			char *sval;
-
-			if (!PyNumber_Check(val_obj)) {
-				pytdbpack_bad_type(ch, "Number", val_obj);
-				return NULL;
-			}
-
-			if (!(val_obj = PyNumber_Long(val_obj)))
-				return NULL;
-
-			size = PyLong_AsLong(val_obj);
-			pack_le_uint32(size, &packed);
-
-			/* Release the new reference created by the cast */
-			Py_DECREF(val_obj);
-
-			val_obj = PySequence_GetItem(val_seq, val_i++);
-			if (!val_obj)
-				return NULL;
-			
-			sval = PyString_AsString(val_obj);
-			if (!sval)
-				return NULL;
-			
-			pack_bytes(size, sval, &packed); /* do not include nul */
-		}
-		else {
-		
-	}
-		
-	return Py_None;
-}
-#endif
-
-static PyObject *
-pytdbpack_unpack(PyObject *self,
-		 PyObject *args)
-{
-	char *format_str, *packed_str, *ppacked;
-	PyObject *val_list = NULL, *ret_tuple = NULL;
-	PyObject *rest_string = NULL;
-	int format_len, packed_len;
-	char last_format = '#';	/* invalid */
-	int i;
-	
-	/* get arguments */
-	if (!PyArg_ParseTuple(args, "ss#", &format_str, &packed_str, &packed_len))
-		return NULL;
-
-	format_len = strlen(format_str);
-	
-	/* Allocate list to hold results.  Initially empty, and we append
-	   results as we go along. */
-	val_list = PyList_New(0);
-	if (!val_list)
-		goto failed;
-	ret_tuple = PyTuple_New(2);
-	if (!ret_tuple)
-		goto failed;
-	
-	/* For every object, unpack.  */
-	for (ppacked = packed_str, i = 0; i < format_len && format_str[i] != '$'; i++) {
-		last_format = format_str[i];
-		/* packed_len is reduced in place */
-		if (!pytdbpack_unpack_item(format_str[i], &ppacked, &packed_len, val_list))
-			goto failed;
-	}
-
-	/* If the last character was '$', keep going until out of space */
-	if (format_str[i] == '$') {
-		if (i == 0) {
-			PyErr_Format(PyExc_ValueError,
-				     "%s: '$' may not be first character in format",
-				     __FUNCTION__);
-			return NULL;
-		} 
-		while (packed_len > 0)
-			if (!pytdbpack_unpack_item(last_format, &ppacked, &packed_len, val_list))
-				goto failed;
-	}
-	
-	/* save leftovers for next time */
-	rest_string = PyString_FromStringAndSize(ppacked, packed_len);
-	if (!rest_string)
-		goto failed;
-
-	/* return (values, rest) tuple; give up references to them */
-	PyTuple_SET_ITEM(ret_tuple, 0, val_list);
-	val_list = NULL;
-	PyTuple_SET_ITEM(ret_tuple, 1, rest_string);
-	val_list = NULL;
-	return ret_tuple;
-
-  failed:
-	/* handle failure: deallocate anything.  XDECREF forms handle NULL
-	   pointers for objects that haven't been allocated yet. */
-	Py_XDECREF(val_list);
-	Py_XDECREF(ret_tuple);
-	Py_XDECREF(rest_string);
-	return NULL;
-}
-
-
-
-#if 0
-/*
-  Internal routine that calculates how many bytes will be required to
-  encode the values in the format.
-
-  Also checks that the value list is the right size for the format list.
-
-  Returns number of bytes (may be 0), or -1 if there's something wrong, in
-  which case a Python exception has been raised.
-
-  Arguments:
-
-    val_seq: a Fast Sequence (list or tuple), being all the values
-*/
-static int
-pytdbpack_calc_reqd_len(char *format_str,
-			PyObject *val_seq)
-{
-	int len = 0;
-	char *p;
-	int val_i;
-	int val_len;
-
-	val_len = PySequence_Length(val_seq);
-	if (val_len == -1)
-		return -1;
-
-	for (p = format_str, val_i = 0; *p; p++, val_i++) {
-		char ch = *p;
-
-		if (val_i >= val_len) {
-			PyErr_Format(PyExc_IndexError,
-				     "%s: value list is too short for format string",
-				     __FUNCTION__);
-			return -1;
-		}
-
-		/* borrow a reference to the item */
-		if (ch == 'd' || ch == 'p') 
-			len += 4;
-		else if (ch == 'w')
-			len += 2;
-		else if (ch == 'f' || ch == 'P') {
-			/* nul-terminated 8-bit string */
-			int item_len;
-			PyObject *str_obj;
-
-			str_obj = PySequence_GetItem(val_seq, val_i);
-			if (!str_obj)
-				return -1;
-
-			if (!PyString_Check(str_obj) || ((item_len = PyString_Size(str_obj)) == -1)) {
-				pytdbpack_bad_type(ch, "String", str_obj);
-				return -1;
-			}
-			
-			len += 1 + item_len;
-		}
-		else if (ch == 'B') {
-			/* length-preceded byte buffer: n bytes, plus a preceding
-			 * word */
-			PyObject *len_obj;
-			long len_val;
-
-			len_obj = PySequence_GetItem(val_seq, val_i);
-			val_i++; /* skip over buffer */
-
-			if (!PyNumber_Check(len_obj)) {
-				pytdbpack_bad_type(ch, "Number", len_obj);
-				return -1;
-			}
-
-			len_val = PyInt_AsLong(len_obj);
-			if (len_val < 0) {
-				PyErr_Format(PyExc_ValueError,
-					     "%s: format 'B' requires positive integer", __FUNCTION__);
-				return -1;
-			}
-
-			len += 4 + len_val;
-		}
-		else {	
-			PyErr_Format(PyExc_ValueError,
-				     "%s: format character '%c' is not supported",
-				     __FUNCTION__, ch);
-		
-			return -1;
-		}
-	}
-
-	return len;
-}
-#endif
-
-
 static PyObject *pytdbpack_bad_type(char ch,
 				    const char *expected,
 				    PyObject *val_obj)
@@ -618,8 +417,78 @@ static void pack_bytes(long len, const char *from,
 }
 
 
+
+static PyObject *
+pytdbunpack(PyObject *self,
+		 PyObject *args)
+{
+	char *format_str, *packed_str, *ppacked;
+	PyObject *val_list = NULL, *ret_tuple = NULL;
+	PyObject *rest_string = NULL;
+	int format_len, packed_len;
+	char last_format = '#';	/* invalid */
+	int i;
+	
+	/* get arguments */
+	if (!PyArg_ParseTuple(args, "ss#", &format_str, &packed_str, &packed_len))
+		return NULL;
+
+	format_len = strlen(format_str);
+	
+	/* Allocate list to hold results.  Initially empty, and we append
+	   results as we go along. */
+	val_list = PyList_New(0);
+	if (!val_list)
+		goto failed;
+	ret_tuple = PyTuple_New(2);
+	if (!ret_tuple)
+		goto failed;
+	
+	/* For every object, unpack.  */
+	for (ppacked = packed_str, i = 0; i < format_len && format_str[i] != '$'; i++) {
+		last_format = format_str[i];
+		/* packed_len is reduced in place */
+		if (!pytdbunpack_item(format_str[i], &ppacked, &packed_len, val_list))
+			goto failed;
+	}
+
+	/* If the last character was '$', keep going until out of space */
+	if (format_str[i] == '$') {
+		if (i == 0) {
+			PyErr_Format(PyExc_ValueError,
+				     "%s: '$' may not be first character in format",
+				     __FUNCTION__);
+			return NULL;
+		} 
+		while (packed_len > 0)
+			if (!pytdbunpack_item(last_format, &ppacked, &packed_len, val_list))
+				goto failed;
+	}
+	
+	/* save leftovers for next time */
+	rest_string = PyString_FromStringAndSize(ppacked, packed_len);
+	if (!rest_string)
+		goto failed;
+
+	/* return (values, rest) tuple; give up references to them */
+	PyTuple_SET_ITEM(ret_tuple, 0, val_list);
+	val_list = NULL;
+	PyTuple_SET_ITEM(ret_tuple, 1, rest_string);
+	val_list = NULL;
+	return ret_tuple;
+
+  failed:
+	/* handle failure: deallocate anything.  XDECREF forms handle NULL
+	   pointers for objects that haven't been allocated yet. */
+	Py_XDECREF(val_list);
+	Py_XDECREF(ret_tuple);
+	Py_XDECREF(rest_string);
+	return NULL;
+}
+
+
 static void
-unpack_err_too_short(void)
+pytdbunpack_err_too_short(void)
 {
 	PyErr_Format(PyExc_IndexError,
 		     __FUNCTION__ ": data too short for unpack format");
@@ -627,13 +496,13 @@ unpack_err_too_short(void)
 
 
 static PyObject *
-unpack_uint32(char **pbuf, int *plen)
+pytdbunpack_uint32(char **pbuf, int *plen)
 {
 	unsigned long v;
 	unsigned char *b;
 	
 	if (*plen < 4) {
-		unpack_err_too_short();
+		pytdbunpack_err_too_short();
 		return NULL;
 	}
 
@@ -647,13 +516,13 @@ unpack_uint32(char **pbuf, int *plen)
 }
 
 
-static PyObject *unpack_int16(char **pbuf, int *plen)
+static PyObject *pytdbunpack_int16(char **pbuf, int *plen)
 {
 	long v;
 	unsigned char *b;
 	
 	if (*plen < 2) {
-		unpack_err_too_short();
+		pytdbunpack_err_too_short();
 		return NULL;
 	}
 
@@ -668,7 +537,7 @@ static PyObject *unpack_int16(char **pbuf, int *plen)
 
 
 static PyObject *
-unpack_string(char **pbuf, int *plen)
+pytdbunpack_string(char **pbuf, int *plen)
 {
 	int len;
 	char *nul_ptr, *start;
@@ -677,7 +546,7 @@ unpack_string(char **pbuf, int *plen)
 	
 	nul_ptr = memchr(start, '\0', *plen);
 	if (!nul_ptr) {
-		unpack_err_too_short();
+		pytdbunpack_err_too_short();
 		return NULL;
 	}
 
@@ -686,12 +555,12 @@ unpack_string(char **pbuf, int *plen)
 	*pbuf += len + 1;	/* skip \0 */
 	*plen -= len + 1;
 
-	return PyString_FromStringAndSize(start, len);
+	return PyString_Decode(start, len, pytdb_string_encoding, NULL);
 }
 
 
 static PyObject *
-unpack_buffer(char **pbuf, int *plen, PyObject *val_list)
+pytdbunpack_buffer(char **pbuf, int *plen, PyObject *val_list)
 {
 	/* first get 32-bit len */
 	long slen;
@@ -700,7 +569,7 @@ unpack_buffer(char **pbuf, int *plen, PyObject *val_list)
 	PyObject *str_obj = NULL, *len_obj = NULL;
 	
 	if (*plen < 4) {
-		unpack_err_too_short();
+		pytdbunpack_err_too_short();
 		return NULL;
 	}
 	
@@ -757,7 +626,7 @@ unpack_buffer(char **pbuf, int *plen, PyObject *val_list)
 
    Returns a reference to None, or NULL for failure.
 */
-static PyObject *pytdbpack_unpack_item(char ch,
+static PyObject *pytdbunpack_item(char ch,
 				       char **pbuf,
 				       int *plen,
 				       PyObject *val_list)
@@ -765,17 +634,17 @@ static PyObject *pytdbpack_unpack_item(char ch,
 	PyObject *result;
 	
 	if (ch == 'w') {	/* 16-bit int */
-		result = unpack_int16(pbuf, plen);
+		result = pytdbunpack_int16(pbuf, plen);
 	}
 	else if (ch == 'd' || ch == 'p') { /* 32-bit int */
 		/* pointers can just come through as integers */
-		result = unpack_uint32(pbuf, plen);
+		result = pytdbunpack_uint32(pbuf, plen);
 	}
 	else if (ch == 'f' || ch == 'P') { /* nul-term string  */
-		result = unpack_string(pbuf, plen);
+		result = pytdbunpack_string(pbuf, plen);
 	}
 	else if (ch == 'B') { /* length, buffer */
-		return unpack_buffer(pbuf, plen, val_list);
+		return pytdbunpack_buffer(pbuf, plen, val_list);
 	}
 	else {
 		PyErr_Format(PyExc_ValueError,
@@ -801,7 +670,7 @@ static PyObject *pytdbpack_unpack_item(char ch,
 
 static PyMethodDef pytdbpack_methods[] = {
 	{ "pack", pytdbpack, METH_VARARGS, (char *) pytdbpack_doc },
-	{ "unpack", pytdbpack_unpack, METH_VARARGS, (char *) pytdbpack_unpack_doc },
+	{ "unpack", pytdbunpack, METH_VARARGS, (char *) pytdbunpack_doc },
 };
 
 DL_EXPORT(void)