From 0ec2e3534f9b7751be484bd2f1344e24c49bb24f Mon Sep 17 00:00:00 2001 From: folz Date: Thu, 28 Apr 2016 15:08:28 +0200 Subject: fix problems associated with packing memoryviews fix wrong length when packing multibyte memoryviews in fallback add tests for memoryviews of different types and sizes and check contents of packed data --- msgpack/_packer.pyx | 23 ++++++++++++----------- msgpack/fallback.py | 26 ++++++++++++++++++++++++-- 2 files changed, 36 insertions(+), 13 deletions(-) (limited to 'msgpack') diff --git a/msgpack/_packer.pyx b/msgpack/_packer.pyx index b1a912b..e07b194 100644 --- a/msgpack/_packer.pyx +++ b/msgpack/_packer.pyx @@ -39,6 +39,7 @@ cdef extern from "pack.h": int msgpack_pack_ext(msgpack_packer* pk, char typecode, size_t l) cdef int DEFAULT_RECURSE_LIMIT=511 +cdef size_t ITEM_LIMIT = (2**32)-1 cdef class Packer(object): @@ -178,7 +179,7 @@ cdef class Packer(object): ret = msgpack_pack_double(&self.pk, dval) elif PyBytes_CheckExact(o) if strict_types else PyBytes_Check(o): L = len(o) - if L > (2**32)-1: + if L > ITEM_LIMIT: raise PackValueError("bytes is too large") rawval = o ret = msgpack_pack_bin(&self.pk, L) @@ -189,7 +190,7 @@ cdef class Packer(object): raise TypeError("Can't encode unicode string: no encoding is specified") o = PyUnicode_AsEncodedString(o, self.encoding, self.unicode_errors) L = len(o) - if L > (2**32)-1: + if L > ITEM_LIMIT: raise PackValueError("unicode string is too large") rawval = o ret = msgpack_pack_raw(&self.pk, L) @@ -198,7 +199,7 @@ cdef class Packer(object): elif PyDict_CheckExact(o): d = o L = len(d) - if L > (2**32)-1: + if L > ITEM_LIMIT: raise PackValueError("dict is too large") ret = msgpack_pack_map(&self.pk, L) if ret == 0: @@ -209,7 +210,7 @@ cdef class Packer(object): if ret != 0: break elif not strict_types and PyDict_Check(o): L = len(o) - if L > (2**32)-1: + if L > ITEM_LIMIT: raise PackValueError("dict is too large") ret = msgpack_pack_map(&self.pk, L) if ret == 0: @@ -223,13 +224,13 @@ cdef class Packer(object): longval = o.code rawval = o.data L = len(o.data) - if L > (2**32)-1: + if L > ITEM_LIMIT: raise PackValueError("EXT data is too large") ret = msgpack_pack_ext(&self.pk, longval, L) ret = msgpack_pack_raw_body(&self.pk, rawval, L) elif PyList_CheckExact(o) if strict_types else (PyTuple_Check(o) or PyList_Check(o)): L = len(o) - if L > (2**32)-1: + if L > ITEM_LIMIT: raise PackValueError("list is too large") ret = msgpack_pack_array(&self.pk, L) if ret == 0: @@ -240,7 +241,7 @@ cdef class Packer(object): if PyObject_GetBuffer(o, &view, PyBUF_SIMPLE) != 0: raise PackValueError("could not get buffer for memoryview") L = view.len - if L > (2**32)-1: + if L > ITEM_LIMIT: PyBuffer_Release(&view); raise PackValueError("memoryview is too large") ret = msgpack_pack_bin(&self.pk, L) @@ -271,8 +272,8 @@ cdef class Packer(object): msgpack_pack_ext(&self.pk, typecode, len(data)) msgpack_pack_raw_body(&self.pk, data, len(data)) - def pack_array_header(self, long long size): - if size > (2**32-1): + def pack_array_header(self, size_t size): + if size > ITEM_LIMIT: raise PackValueError cdef int ret = msgpack_pack_array(&self.pk, size) if ret == -1: @@ -284,8 +285,8 @@ cdef class Packer(object): self.pk.length = 0 return buf - def pack_map_header(self, long long size): - if size > (2**32-1): + def pack_map_header(self, size_t size): + if size > ITEM_LIMIT: raise PackValueError cdef int ret = msgpack_pack_map(&self.pk, size) if ret == -1: diff --git a/msgpack/fallback.py b/msgpack/fallback.py index d8c5d73..db47d5c 100644 --- a/msgpack/fallback.py +++ b/msgpack/fallback.py @@ -685,7 +685,7 @@ class Packer(object): default_used = True continue raise PackOverflowError("Integer value out of range") - if self._use_bin_type and check(obj, (bytes, memoryview)): + if self._use_bin_type and check(obj, bytes): n = len(obj) if n <= 0xff: self._buffer.write(struct.pack('>BB', 0xc4, n)) @@ -696,7 +696,7 @@ class Packer(object): else: raise PackValueError("Bytes is too large") return self._buffer.write(obj) - if check(obj, (Unicode, bytes, memoryview)): + if check(obj, (Unicode, bytes)): if check(obj, Unicode): if self._encoding is None: raise TypeError( @@ -715,6 +715,28 @@ class Packer(object): else: raise PackValueError("String is too large") return self._buffer.write(obj) + if check(obj, memoryview): + n = len(obj) * obj.itemsize + if self._use_bin_type: + if n <= 0xff: + self._buffer.write(struct.pack('>BB', 0xc4, n)) + elif n <= 0xffff: + self._buffer.write(struct.pack(">BH", 0xc5, n)) + elif n <= 0xffffffff: + self._buffer.write(struct.pack(">BI", 0xc6, n)) + else: + raise PackValueError("memoryview is too large") + return self._buffer.write(obj) + else: + if n <= 0x1f: + self._buffer.write(struct.pack('B', 0xa0 + n)) + elif n <= 0xffff: + self._buffer.write(struct.pack(">BH", 0xda, n)) + elif n <= 0xffffffff: + self._buffer.write(struct.pack(">BI", 0xdb, n)) + else: + raise PackValueError("memoryview is too large") + return self._buffer.write(obj) if check(obj, float): if self._use_float: return self._buffer.write(struct.pack(">Bf", 0xca, obj)) -- cgit v1.2.1 From 5860af953ae1c3f459ddc589cd815ec195db46a9 Mon Sep 17 00:00:00 2001 From: folz Date: Wed, 4 May 2016 11:01:27 +0200 Subject: refactor header packing for str and bin types --- msgpack/fallback.py | 83 +++++++++++++++++++++++++---------------------------- 1 file changed, 39 insertions(+), 44 deletions(-) (limited to 'msgpack') diff --git a/msgpack/fallback.py b/msgpack/fallback.py index db47d5c..abed3d9 100644 --- a/msgpack/fallback.py +++ b/msgpack/fallback.py @@ -685,58 +685,29 @@ class Packer(object): default_used = True continue raise PackOverflowError("Integer value out of range") - if self._use_bin_type and check(obj, bytes): + if check(obj, bytes): n = len(obj) - if n <= 0xff: - self._buffer.write(struct.pack('>BB', 0xc4, n)) - elif n <= 0xffff: - self._buffer.write(struct.pack(">BH", 0xc5, n)) - elif n <= 0xffffffff: - self._buffer.write(struct.pack(">BI", 0xc6, n)) - else: + if n >= 2**32: raise PackValueError("Bytes is too large") + self._fb_pack_bin_header(n) return self._buffer.write(obj) - if check(obj, (Unicode, bytes)): - if check(obj, Unicode): - if self._encoding is None: - raise TypeError( - "Can't encode unicode string: " - "no encoding is specified") - obj = obj.encode(self._encoding, self._unicode_errors) + if check(obj, Unicode): + if self._encoding is None: + raise TypeError( + "Can't encode unicode string: " + "no encoding is specified") + obj = obj.encode(self._encoding, self._unicode_errors) n = len(obj) - if n <= 0x1f: - self._buffer.write(struct.pack('B', 0xa0 + n)) - elif self._use_bin_type and n <= 0xff: - self._buffer.write(struct.pack('>BB', 0xd9, n)) - elif n <= 0xffff: - self._buffer.write(struct.pack(">BH", 0xda, n)) - elif n <= 0xffffffff: - self._buffer.write(struct.pack(">BI", 0xdb, n)) - else: + if n >= 2**32: raise PackValueError("String is too large") + self._fb_pack_raw_header(n) return self._buffer.write(obj) if check(obj, memoryview): n = len(obj) * obj.itemsize - if self._use_bin_type: - if n <= 0xff: - self._buffer.write(struct.pack('>BB', 0xc4, n)) - elif n <= 0xffff: - self._buffer.write(struct.pack(">BH", 0xc5, n)) - elif n <= 0xffffffff: - self._buffer.write(struct.pack(">BI", 0xc6, n)) - else: - raise PackValueError("memoryview is too large") - return self._buffer.write(obj) - else: - if n <= 0x1f: - self._buffer.write(struct.pack('B', 0xa0 + n)) - elif n <= 0xffff: - self._buffer.write(struct.pack(">BH", 0xda, n)) - elif n <= 0xffffffff: - self._buffer.write(struct.pack(">BI", 0xdb, n)) - else: - raise PackValueError("memoryview is too large") - return self._buffer.write(obj) + if n >= 2**32: + raise PackValueError("Memoryview is too large") + self._fb_pack_bin_header(n) + return self._buffer.write(obj) if check(obj, float): if self._use_float: return self._buffer.write(struct.pack(">Bf", 0xca, obj)) @@ -874,6 +845,30 @@ class Packer(object): self._pack(k, nest_limit - 1) self._pack(v, nest_limit - 1) + def _fb_pack_raw_header(self, n): + if n <= 0x1f: + self._buffer.write(struct.pack('B', 0xa0 + n)) + elif self._use_bin_type and n <= 0xff: + self._buffer.write(struct.pack('>BB', 0xd9, n)) + elif n <= 0xffff: + self._buffer.write(struct.pack(">BH", 0xda, n)) + elif n <= 0xffffffff: + self._buffer.write(struct.pack(">BI", 0xdb, n)) + else: + raise PackValueError('Raw is too large') + + def _fb_pack_bin_header(self, n): + if not self._use_bin_type: + return self._fb_pack_raw_header(n) + elif n <= 0xff: + return self._buffer.write(struct.pack('>BB', 0xc4, n)) + elif n <= 0xffff: + return self._buffer.write(struct.pack(">BH", 0xc5, n)) + elif n <= 0xffffffff: + return self._buffer.write(struct.pack(">BI", 0xc6, n)) + else: + raise PackValueError('Bin is too large') + def bytes(self): return self._buffer.getvalue() -- cgit v1.2.1