3.6.9 is the latest security fix release of Python 3.6.
- Fixes: CVE-2018-20852, CVE-2019-9948, CVE-2019-9740, CVE-2019-5010
- Refreshed patches
- Removed a few backports patches
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com>
# Note: keep in sync with setuptools & pip
PYTHON3_VERSION_MAJOR:=3
PYTHON3_VERSION_MINOR:=6
-PYTHON3_VERSION_MICRO:=5
+PYTHON3_VERSION_MICRO:=9
PYTHON3_VERSION:=$(PYTHON3_VERSION_MAJOR).$(PYTHON3_VERSION_MINOR)
-PYTHON3_SETUPTOOLS_PKG_RELEASE:=2
-PYTHON3_PIP_PKG_RELEASE:=2
+PYTHON3_SETUPTOOLS_PKG_RELEASE:=1
+PYTHON3_PIP_PKG_RELEASE:=1
-PYTHON3_SETUPTOOLS_VERSION:=39.0.1
-PYTHON3_PIP_VERSION:=9.0.3
+PYTHON3_SETUPTOOLS_VERSION:=40.6.2
+PYTHON3_PIP_VERSION:=18.1
PYTHON_VERSION_MICRO:=$(PYTHON3_VERSION_MICRO)
PKG_NAME:=python3
-PKG_RELEASE:=5
+PKG_RELEASE:=1
PKG_VERSION:=$(PYTHON_VERSION).$(PYTHON_VERSION_MICRO)
PKG_SOURCE:=Python-$(PKG_VERSION).tar.xz
PKG_SOURCE_URL:=https://www.python.org/ftp/python/$(PKG_VERSION)
-PKG_HASH:=f434053ba1b5c8a5cc597e966ead3c5143012af827fd3f0697d21450bb8d87a6
+PKG_HASH:=5e2f5f554e3f8f7f0296f7e73d8600c4e9acbaee6b2555b83206edf5153870da
PKG_LICENSE:=PSF
PKG_LICENSE_FILES:=LICENSE Modules/_ctypes/libffi_msvc/LICENSE Modules/_ctypes/darwin/LICENSE Modules/_ctypes/libffi/LICENSE Modules/_ctypes/libffi_osx/LICENSE Tools/pybench/LICENSE
+zlib zlibmodule.c -lz
# Interface to the Expat XML parser
- #
+ # More information on Expat can be found at www.libexpat.org.
--- a/Makefile.pre.in
+++ b/Makefile.pre.in
-@@ -1341,6 +1341,7 @@ libinstall: build_all $(srcdir)/Modules/
+@@ -1344,6 +1344,7 @@ libinstall: build_all $(srcdir)/Modules/
$(INSTALL_DATA) `cat pybuilddir.txt`/_sysconfigdata_$(ABIFLAGS)_$(MACHDEP)_$(MULTIARCH).py \
$(DESTDIR)$(LIBDEST); \
$(INSTALL_DATA) $(srcdir)/LICENSE $(DESTDIR)$(LIBDEST)/LICENSE.txt
if test -d $(DESTDIR)$(LIBDEST)/distutils/tests; then \
$(INSTALL_DATA) $(srcdir)/Modules/xxmodule.c \
$(DESTDIR)$(LIBDEST)/distutils/tests ; \
-@@ -1376,6 +1377,7 @@ libinstall: build_all $(srcdir)/Modules/
+@@ -1379,6 +1380,7 @@ libinstall: build_all $(srcdir)/Modules/
$(PYTHON_FOR_BUILD) -m lib2to3.pgen2.driver $(DESTDIR)$(LIBDEST)/lib2to3/Grammar.txt
-PYTHONPATH=$(DESTDIR)$(LIBDEST) $(RUNSHARED) \
$(PYTHON_FOR_BUILD) -m lib2to3.pgen2.driver $(DESTDIR)$(LIBDEST)/lib2to3/PatternGrammar.txt
--- a/setup.py
+++ b/setup.py
-@@ -512,16 +512,9 @@ class PyBuildExt(build_ext):
+@@ -517,16 +517,9 @@ class PyBuildExt(build_ext):
return ['m']
def detect_modules(self):
+++ /dev/null
-From f7666e828cc3d5873136473ea36ba2013d624fa1 Mon Sep 17 00:00:00 2001
-From: "Miss Islington (bot)"
- <31488909+miss-islington@users.noreply.github.com>
-Date: Tue, 18 Sep 2018 06:14:13 -0700
-Subject: [PATCH] bpo-34623: Use XML_SetHashSalt in _elementtree (GH-9146)
-
-The C accelerated _elementtree module now initializes hash randomization
-salt from _Py_HashSecret instead of libexpat's default CPRNG.
-
-Signed-off-by: Christian Heimes <christian@python.org>
-
-https://bugs.python.org/issue34623
-(cherry picked from commit cb5778f00ce48631c7140f33ba242496aaf7102b)
-
-Co-authored-by: Christian Heimes <christian@python.org>
----
- Include/pyexpat.h | 4 +++-
- .../next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst | 2 ++
- Modules/_elementtree.c | 5 +++++
- Modules/pyexpat.c | 5 +++++
- 4 files changed, 15 insertions(+), 1 deletion(-)
- create mode 100644 Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst
-
-diff --git a/Include/pyexpat.h b/Include/pyexpat.h
-index 44259bf6d7..07020b5dc9 100644
---- a/Include/pyexpat.h
-+++ b/Include/pyexpat.h
-@@ -3,7 +3,7 @@
-
- /* note: you must import expat.h before importing this module! */
-
--#define PyExpat_CAPI_MAGIC "pyexpat.expat_CAPI 1.0"
-+#define PyExpat_CAPI_MAGIC "pyexpat.expat_CAPI 1.1"
- #define PyExpat_CAPSULE_NAME "pyexpat.expat_CAPI"
-
- struct PyExpat_CAPI
-@@ -48,6 +48,8 @@ struct PyExpat_CAPI
- enum XML_Status (*SetEncoding)(XML_Parser parser, const XML_Char *encoding);
- int (*DefaultUnknownEncodingHandler)(
- void *encodingHandlerData, const XML_Char *name, XML_Encoding *info);
-+ /* might be none for expat < 2.1.0 */
-+ int (*SetHashSalt)(XML_Parser parser, unsigned long hash_salt);
- /* always add new stuff to the end! */
- };
-
-diff --git a/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst b/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst
-new file mode 100644
-index 0000000000..31ad92ef85
---- /dev/null
-+++ b/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst
-@@ -0,0 +1,2 @@
-+The C accelerated _elementtree module now initializes hash randomization
-+salt from _Py_HashSecret instead of libexpat's default CSPRNG.
-diff --git a/Modules/_elementtree.c b/Modules/_elementtree.c
-index 707ab2912b..53f05f937f 100644
---- a/Modules/_elementtree.c
-+++ b/Modules/_elementtree.c
-@@ -3261,6 +3261,11 @@ _elementtree_XMLParser___init___impl(XMLParserObject *self, PyObject *html,
- PyErr_NoMemory();
- return -1;
- }
-+ /* expat < 2.1.0 has no XML_SetHashSalt() */
-+ if (EXPAT(SetHashSalt) != NULL) {
-+ EXPAT(SetHashSalt)(self->parser,
-+ (unsigned long)_Py_HashSecret.expat.hashsalt);
-+ }
-
- if (target) {
- Py_INCREF(target);
-diff --git a/Modules/pyexpat.c b/Modules/pyexpat.c
-index 47c3e86c20..aa21d93c11 100644
---- a/Modules/pyexpat.c
-+++ b/Modules/pyexpat.c
-@@ -1887,6 +1887,11 @@ MODULE_INITFUNC(void)
- capi.SetStartDoctypeDeclHandler = XML_SetStartDoctypeDeclHandler;
- capi.SetEncoding = XML_SetEncoding;
- capi.DefaultUnknownEncodingHandler = PyUnknownEncodingHandler;
-+#if XML_COMBINED_VERSION >= 20100
-+ capi.SetHashSalt = XML_SetHashSalt;
-+#else
-+ capi.SetHashSalt = NULL;
-+#endif
-
- /* export using capsule */
- capi_object = PyCapsule_New(&capi, PyExpat_CAPSULE_NAME, NULL);
---
-2.19.1
-
+++ /dev/null
-From d1b336e530472f316b1d164d04626724c83b16d7 Mon Sep 17 00:00:00 2001
-From: "Miss Islington (bot)"
- <31488909+miss-islington@users.noreply.github.com>
-Date: Fri, 21 Sep 2018 21:57:28 -0700
-Subject: [PATCH] [3.6] bpo-34623: Mention CVE-2018-14647 in news entry
- (GH-9482) (GH-9489)
-
-https://bugs.python.org/issue34623
-(cherry picked from commit 026337a7101369297c8083047d2f3c6fc9dd1e2b)
-
-
-Co-authored-by: Christian Heimes <christian@python.org>
-
-
-https://bugs.python.org/issue34623
----
- .../next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst b/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst
-index 31ad92ef85..cbaa4b7506 100644
---- a/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst
-+++ b/Misc/NEWS.d/next/Security/2018-09-10-16-05-39.bpo-34623.Ua9jMv.rst
-@@ -1,2 +1,2 @@
--The C accelerated _elementtree module now initializes hash randomization
--salt from _Py_HashSecret instead of libexpat's default CSPRNG.
-+CVE-2018-14647: The C accelerated _elementtree module now initializes hash
-+randomization salt from _Py_HashSecret instead of libexpat's default CSPRNG.
---
-2.19.1
-
--- a/setup.py
+++ b/setup.py
-@@ -522,8 +522,9 @@ class PyBuildExt(build_ext):
+@@ -527,8 +527,9 @@ class PyBuildExt(build_ext):
# directly since an inconsistently reproducible issue comes up where
# the environment variable is not set even though the value were passed
# into configure and stored in the Makefile (issue found on OS X 10.3).
--- a/Makefile.pre.in
+++ b/Makefile.pre.in
-@@ -725,6 +725,16 @@ regen-all: regen-opcode regen-opcode-tar
+@@ -728,6 +728,16 @@ regen-all: regen-opcode regen-opcode-tar
############################################################################
# Special rules for object files
Modules/getbuildinfo.o: $(PARSER_OBJS) \
$(OBJECT_OBJS) \
$(PYTHON_OBJS) \
-@@ -732,6 +742,8 @@ Modules/getbuildinfo.o: $(PARSER_OBJS) \
+@@ -735,6 +745,8 @@ Modules/getbuildinfo.o: $(PARSER_OBJS) \
$(MODOBJS) \
$(srcdir)/Modules/getbuildinfo.c
$(CC) -c $(PY_CORE_CFLAGS) \
--- a/configure
+++ b/configure
-@@ -14995,7 +14995,7 @@ $as_echo_n "checking ABIFLAGS... " >&6;
+@@ -15171,7 +15171,7 @@ $as_echo_n "checking ABIFLAGS... " >&6;
$as_echo "$ABIFLAGS" >&6; }
{ $as_echo "$as_me:${as_lineno-$LINENO}: checking SOABI" >&5
$as_echo_n "checking SOABI... " >&6; }
--- a/configure.ac
+++ b/configure.ac
-@@ -4647,7 +4647,7 @@ AC_SUBST(SOABI)
+@@ -4753,7 +4753,7 @@ AC_SUBST(SOABI)
AC_MSG_CHECKING(ABIFLAGS)
AC_MSG_RESULT($ABIFLAGS)
AC_MSG_CHECKING(SOABI)
--- a/setup.py
+++ b/setup.py
-@@ -333,6 +333,7 @@ class PyBuildExt(build_ext):
+@@ -338,6 +338,7 @@ class PyBuildExt(build_ext):
print("Failed to build these modules:")
print_three_column(failed)
print()
abi=sys.abiflags,
--- a/Makefile.pre.in
+++ b/Makefile.pre.in
-@@ -1350,7 +1350,7 @@ libinstall: build_all $(srcdir)/Modules/
+@@ -1353,7 +1353,7 @@ libinstall: build_all $(srcdir)/Modules/
esac; \
done; \
done
$(DESTDIR)$(LIBDEST); \
$(INSTALL_DATA) $(srcdir)/LICENSE $(DESTDIR)$(LIBDEST)/LICENSE.txt
ifeq (@COMPILE_ALL_TESTS@,yes)
-@@ -1487,7 +1487,7 @@ sharedinstall: sharedmods
+@@ -1489,7 +1489,7 @@ sharedinstall: sharedmods
--install-scripts=$(BINDIR) \
--install-platlib=$(DESTSHARED) \
--root=$(DESTDIR)/
# Here are a couple of targets for MacOSX again, to install a full
--- a/configure
+++ b/configure
-@@ -15014,7 +15014,7 @@ LDVERSION='$(VERSION)$(ABIFLAGS)'
+@@ -15190,7 +15190,7 @@ LDVERSION='$(VERSION)$(ABIFLAGS)'
$as_echo "$LDVERSION" >&6; }
LIBPL='$(prefix)'"/lib/python${VERSION}/config-${LDVERSION}-${PLATFORM_TRIPLET}"
--- a/configure.ac
+++ b/configure.ac
-@@ -4664,7 +4664,7 @@ AC_MSG_RESULT($LDVERSION)
+@@ -4770,7 +4770,7 @@ AC_MSG_RESULT($LDVERSION)
dnl define LIBPL after ABIFLAGS and LDVERSION is defined.
AC_SUBST(PY_ENABLE_SHARED)
-diff --git a/Lib/lib2to3/refactor.py b/Lib/lib2to3/refactor.py
-index 7841b99..1e0d3b3 100644
--- a/Lib/lib2to3/refactor.py
+++ b/Lib/lib2to3/refactor.py
-@@ -37,6 +37,12 @@ def get_all_fix_names(fixer_pkg, remove_prefix=True):
+@@ -37,6 +37,12 @@ def get_all_fix_names(fixer_pkg, remove_
if remove_prefix:
name = name[4:]
fix_names.append(name[:-3])
+++ /dev/null
-From 71a9c65e74a70b6ed39adc4ba81d311ac1aa2acc Mon Sep 17 00:00:00 2001
-From: "Miss Islington (bot)"
- <31488909+miss-islington@users.noreply.github.com>
-Date: Thu, 20 Sep 2018 19:00:37 -0700
-Subject: [PATCH] closes bpo-34656: Avoid relying on signed overflow in _pickle
- memos. (GH-9261)
-
-(cherry picked from commit a4ae828ee416a66d8c7bf5ee71d653c2cc6a26dd)
-
-Co-authored-by: Benjamin Peterson <benjamin@python.org>
----
- Modules/_pickle.c | 62 +++++++++++++++++++++++------------------------
- 1 file changed, 31 insertions(+), 31 deletions(-)
-
-diff --git a/Modules/_pickle.c b/Modules/_pickle.c
-index f5202f50c5..93bc1c6fee 100644
---- a/Modules/_pickle.c
-+++ b/Modules/_pickle.c
-@@ -596,9 +596,9 @@ typedef struct {
- } PyMemoEntry;
-
- typedef struct {
-- Py_ssize_t mt_mask;
-- Py_ssize_t mt_used;
-- Py_ssize_t mt_allocated;
-+ size_t mt_mask;
-+ size_t mt_used;
-+ size_t mt_allocated;
- PyMemoEntry *mt_table;
- } PyMemoTable;
-
-@@ -644,8 +644,8 @@ typedef struct UnpicklerObject {
- /* The unpickler memo is just an array of PyObject *s. Using a dict
- is unnecessary, since the keys are contiguous ints. */
- PyObject **memo;
-- Py_ssize_t memo_size; /* Capacity of the memo array */
-- Py_ssize_t memo_len; /* Number of objects in the memo */
-+ size_t memo_size; /* Capacity of the memo array */
-+ size_t memo_len; /* Number of objects in the memo */
-
- PyObject *pers_func; /* persistent_load() method, can be NULL. */
- PyObject *pers_func_self; /* borrowed reference to self if pers_func
-@@ -731,7 +731,6 @@ PyMemoTable_New(void)
- static PyMemoTable *
- PyMemoTable_Copy(PyMemoTable *self)
- {
-- Py_ssize_t i;
- PyMemoTable *new = PyMemoTable_New();
- if (new == NULL)
- return NULL;
-@@ -748,7 +747,7 @@ PyMemoTable_Copy(PyMemoTable *self)
- PyErr_NoMemory();
- return NULL;
- }
-- for (i = 0; i < self->mt_allocated; i++) {
-+ for (size_t i = 0; i < self->mt_allocated; i++) {
- Py_XINCREF(self->mt_table[i].me_key);
- }
- memcpy(new->mt_table, self->mt_table,
-@@ -794,7 +793,7 @@ _PyMemoTable_Lookup(PyMemoTable *self, PyObject *key)
- {
- size_t i;
- size_t perturb;
-- size_t mask = (size_t)self->mt_mask;
-+ size_t mask = self->mt_mask;
- PyMemoEntry *table = self->mt_table;
- PyMemoEntry *entry;
- Py_hash_t hash = (Py_hash_t)key >> 3;
-@@ -816,22 +815,24 @@ _PyMemoTable_Lookup(PyMemoTable *self, PyObject *key)
-
- /* Returns -1 on failure, 0 on success. */
- static int
--_PyMemoTable_ResizeTable(PyMemoTable *self, Py_ssize_t min_size)
-+_PyMemoTable_ResizeTable(PyMemoTable *self, size_t min_size)
- {
- PyMemoEntry *oldtable = NULL;
- PyMemoEntry *oldentry, *newentry;
-- Py_ssize_t new_size = MT_MINSIZE;
-- Py_ssize_t to_process;
-+ size_t new_size = MT_MINSIZE;
-+ size_t to_process;
-
- assert(min_size > 0);
-
-- /* Find the smallest valid table size >= min_size. */
-- while (new_size < min_size && new_size > 0)
-- new_size <<= 1;
-- if (new_size <= 0) {
-+ if (min_size > PY_SSIZE_T_MAX) {
- PyErr_NoMemory();
- return -1;
- }
-+
-+ /* Find the smallest valid table size >= min_size. */
-+ while (new_size < min_size) {
-+ new_size <<= 1;
-+ }
- /* new_size needs to be a power of two. */
- assert((new_size & (new_size - 1)) == 0);
-
-@@ -904,10 +905,12 @@ PyMemoTable_Set(PyMemoTable *self, PyObject *key, Py_ssize_t value)
- * Very large memo tables (over 50K items) use doubling instead.
- * This may help applications with severe memory constraints.
- */
-- if (!(self->mt_used * 3 >= (self->mt_mask + 1) * 2))
-+ if (SIZE_MAX / 3 >= self->mt_used && self->mt_used * 3 < self->mt_allocated * 2) {
- return 0;
-- return _PyMemoTable_ResizeTable(self,
-- (self->mt_used > 50000 ? 2 : 4) * self->mt_used);
-+ }
-+ // self->mt_used is always < PY_SSIZE_T_MAX, so this can't overflow.
-+ size_t desired_size = (self->mt_used > 50000 ? 2 : 4) * self->mt_used;
-+ return _PyMemoTable_ResizeTable(self, desired_size);
- }
-
- #undef MT_MINSIZE
-@@ -1352,9 +1355,9 @@ _Unpickler_Readline(UnpicklerObject *self, char **result)
- /* Returns -1 (with an exception set) on failure, 0 on success. The memo array
- will be modified in place. */
- static int
--_Unpickler_ResizeMemoList(UnpicklerObject *self, Py_ssize_t new_size)
-+_Unpickler_ResizeMemoList(UnpicklerObject *self, size_t new_size)
- {
-- Py_ssize_t i;
-+ size_t i;
-
- assert(new_size > self->memo_size);
-
-@@ -1373,9 +1376,9 @@ _Unpickler_ResizeMemoList(UnpicklerObject *self, Py_ssize_t new_size)
-
- /* Returns NULL if idx is out of bounds. */
- static PyObject *
--_Unpickler_MemoGet(UnpicklerObject *self, Py_ssize_t idx)
-+_Unpickler_MemoGet(UnpicklerObject *self, size_t idx)
- {
-- if (idx < 0 || idx >= self->memo_size)
-+ if (idx >= self->memo_size)
- return NULL;
-
- return self->memo[idx];
-@@ -1384,7 +1387,7 @@ _Unpickler_MemoGet(UnpicklerObject *self, Py_ssize_t idx)
- /* Returns -1 (with an exception set) on failure, 0 on success.
- This takes its own reference to `value`. */
- static int
--_Unpickler_MemoPut(UnpicklerObject *self, Py_ssize_t idx, PyObject *value)
-+_Unpickler_MemoPut(UnpicklerObject *self, size_t idx, PyObject *value)
- {
- PyObject *old_item;
-
-@@ -4328,14 +4331,13 @@ static PyObject *
- _pickle_PicklerMemoProxy_copy_impl(PicklerMemoProxyObject *self)
- /*[clinic end generated code: output=bb83a919d29225ef input=b73043485ac30b36]*/
- {
-- Py_ssize_t i;
- PyMemoTable *memo;
- PyObject *new_memo = PyDict_New();
- if (new_memo == NULL)
- return NULL;
-
- memo = self->pickler->memo;
-- for (i = 0; i < memo->mt_allocated; ++i) {
-+ for (size_t i = 0; i < memo->mt_allocated; ++i) {
- PyMemoEntry entry = memo->mt_table[i];
- if (entry.me_key != NULL) {
- int status;
-@@ -6764,7 +6766,7 @@ static PyObject *
- _pickle_UnpicklerMemoProxy_copy_impl(UnpicklerMemoProxyObject *self)
- /*[clinic end generated code: output=e12af7e9bc1e4c77 input=97769247ce032c1d]*/
- {
-- Py_ssize_t i;
-+ size_t i;
- PyObject *new_memo = PyDict_New();
- if (new_memo == NULL)
- return NULL;
-@@ -6915,8 +6917,7 @@ static int
- Unpickler_set_memo(UnpicklerObject *self, PyObject *obj)
- {
- PyObject **new_memo;
-- Py_ssize_t new_memo_size = 0;
-- Py_ssize_t i;
-+ size_t new_memo_size = 0;
-
- if (obj == NULL) {
- PyErr_SetString(PyExc_TypeError,
-@@ -6933,7 +6934,7 @@ Unpickler_set_memo(UnpicklerObject *self, PyObject *obj)
- if (new_memo == NULL)
- return -1;
-
-- for (i = 0; i < new_memo_size; i++) {
-+ for (size_t i = 0; i < new_memo_size; i++) {
- Py_XINCREF(unpickler->memo[i]);
- new_memo[i] = unpickler->memo[i];
- }
-@@ -6981,8 +6982,7 @@ Unpickler_set_memo(UnpicklerObject *self, PyObject *obj)
-
- error:
- if (new_memo_size) {
-- i = new_memo_size;
-- while (--i >= 0) {
-+ for (size_t i = new_memo_size - 1; i != SIZE_MAX; i--) {
- Py_XDECREF(new_memo[i]);
- }
- PyMem_FREE(new_memo);
---
-2.17.1
-
+++ /dev/null
-From 8f5285ac12bd185202fdaac56fd19dfb858ec947 Mon Sep 17 00:00:00 2001
-From: Steve Dower <steve.dower@python.org>
-Date: Thu, 7 Mar 2019 08:09:56 -0800
-Subject: [PATCH 1/2] bpo-36216: Add check for characters in netloc that
- normalize to separators (GH-12201)
-
----
- Doc/library/urllib.parse.rst | 18 +++++++++++++++
- Lib/test/test_urlparse.py | 23 +++++++++++++++++++
- Lib/urllib/parse.py | 17 ++++++++++++++
- .../2019-03-06-09-38-40.bpo-36216.6q1m4a.rst | 3 +++
- 4 files changed, 61 insertions(+)
- create mode 100644 Misc/NEWS.d/next/Security/2019-03-06-09-38-40.bpo-36216.6q1m4a.rst
-
-diff --git a/Doc/library/urllib.parse.rst b/Doc/library/urllib.parse.rst
-index d991254d5ca1..647af613a315 100644
---- a/Doc/library/urllib.parse.rst
-+++ b/Doc/library/urllib.parse.rst
-@@ -121,6 +121,11 @@ or on combining URL components into a URL string.
- Unmatched square brackets in the :attr:`netloc` attribute will raise a
- :exc:`ValueError`.
-
-+ Characters in the :attr:`netloc` attribute that decompose under NFKC
-+ normalization (as used by the IDNA encoding) into any of ``/``, ``?``,
-+ ``#``, ``@``, or ``:`` will raise a :exc:`ValueError`. If the URL is
-+ decomposed before parsing, no error will be raised.
-+
- .. versionchanged:: 3.2
- Added IPv6 URL parsing capabilities.
-
-@@ -133,6 +138,10 @@ or on combining URL components into a URL string.
- Out-of-range port numbers now raise :exc:`ValueError`, instead of
- returning :const:`None`.
-
-+ .. versionchanged:: 3.6.9
-+ Characters that affect netloc parsing under NFKC normalization will
-+ now raise :exc:`ValueError`.
-+
-
- .. function:: parse_qs(qs, keep_blank_values=False, strict_parsing=False, encoding='utf-8', errors='replace', max_num_fields=None)
-
-@@ -256,10 +265,19 @@ or on combining URL components into a URL string.
- Unmatched square brackets in the :attr:`netloc` attribute will raise a
- :exc:`ValueError`.
-
-+ Characters in the :attr:`netloc` attribute that decompose under NFKC
-+ normalization (as used by the IDNA encoding) into any of ``/``, ``?``,
-+ ``#``, ``@``, or ``:`` will raise a :exc:`ValueError`. If the URL is
-+ decomposed before parsing, no error will be raised.
-+
- .. versionchanged:: 3.6
- Out-of-range port numbers now raise :exc:`ValueError`, instead of
- returning :const:`None`.
-
-+ .. versionchanged:: 3.6.9
-+ Characters that affect netloc parsing under NFKC normalization will
-+ now raise :exc:`ValueError`.
-+
-
- .. function:: urlunsplit(parts)
-
-diff --git a/Lib/test/test_urlparse.py b/Lib/test/test_urlparse.py
-index be50b47603aa..e6638aee2244 100644
---- a/Lib/test/test_urlparse.py
-+++ b/Lib/test/test_urlparse.py
-@@ -1,3 +1,5 @@
-+import sys
-+import unicodedata
- import unittest
- import urllib.parse
-
-@@ -984,6 +986,27 @@ def test_all(self):
- expected.append(name)
- self.assertCountEqual(urllib.parse.__all__, expected)
-
-+ def test_urlsplit_normalization(self):
-+ # Certain characters should never occur in the netloc,
-+ # including under normalization.
-+ # Ensure that ALL of them are detected and cause an error
-+ illegal_chars = '/:#?@'
-+ hex_chars = {'{:04X}'.format(ord(c)) for c in illegal_chars}
-+ denorm_chars = [
-+ c for c in map(chr, range(128, sys.maxunicode))
-+ if (hex_chars & set(unicodedata.decomposition(c).split()))
-+ and c not in illegal_chars
-+ ]
-+ # Sanity check that we found at least one such character
-+ self.assertIn('\u2100', denorm_chars)
-+ self.assertIn('\uFF03', denorm_chars)
-+
-+ for scheme in ["http", "https", "ftp"]:
-+ for c in denorm_chars:
-+ url = "{}://netloc{}false.netloc/path".format(scheme, c)
-+ with self.subTest(url=url, char='{:04X}'.format(ord(c))):
-+ with self.assertRaises(ValueError):
-+ urllib.parse.urlsplit(url)
-
- class Utility_Tests(unittest.TestCase):
- """Testcase to test the various utility functions in the urllib."""
-diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
-index 85e68c8b42c7..24a628915e53 100644
---- a/Lib/urllib/parse.py
-+++ b/Lib/urllib/parse.py
-@@ -391,6 +391,21 @@ def _splitnetloc(url, start=0):
- delim = min(delim, wdelim) # use earliest delim position
- return url[start:delim], url[delim:] # return (domain, rest)
-
-+def _checknetloc(netloc):
-+ if not netloc or netloc.isascii():
-+ return
-+ # looking for characters like \u2100 that expand to 'a/c'
-+ # IDNA uses NFKC equivalence, so normalize for this check
-+ import unicodedata
-+ netloc2 = unicodedata.normalize('NFKC', netloc)
-+ if netloc == netloc2:
-+ return
-+ _, _, netloc = netloc.rpartition('@') # anything to the left of '@' is okay
-+ for c in '/?#@:':
-+ if c in netloc2:
-+ raise ValueError("netloc '" + netloc2 + "' contains invalid " +
-+ "characters under NFKC normalization")
-+
- def urlsplit(url, scheme='', allow_fragments=True):
- """Parse a URL into 5 components:
- <scheme>://<netloc>/<path>?<query>#<fragment>
-@@ -420,6 +435,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
- url, fragment = url.split('#', 1)
- if '?' in url:
- url, query = url.split('?', 1)
-+ _checknetloc(netloc)
- v = SplitResult(scheme, netloc, url, query, fragment)
- _parse_cache[key] = v
- return _coerce_result(v)
-@@ -443,6 +459,7 @@ def urlsplit(url, scheme='', allow_fragments=True):
- url, fragment = url.split('#', 1)
- if '?' in url:
- url, query = url.split('?', 1)
-+ _checknetloc(netloc)
- v = SplitResult(scheme, netloc, url, query, fragment)
- _parse_cache[key] = v
- return _coerce_result(v)
-diff --git a/Misc/NEWS.d/next/Security/2019-03-06-09-38-40.bpo-36216.6q1m4a.rst b/Misc/NEWS.d/next/Security/2019-03-06-09-38-40.bpo-36216.6q1m4a.rst
-new file mode 100644
-index 000000000000..5546394157f9
---- /dev/null
-+++ b/Misc/NEWS.d/next/Security/2019-03-06-09-38-40.bpo-36216.6q1m4a.rst
-@@ -0,0 +1,3 @@
-+Changes urlsplit() to raise ValueError when the URL contains characters that
-+decompose under IDNA encoding (NFKC-normalization) into characters that
-+affect how the URL is parsed.
-
-From 379c71551a3f005bbedccc3758568918612c1765 Mon Sep 17 00:00:00 2001
-From: Steve Dower <steve.dower@python.org>
-Date: Thu, 7 Mar 2019 08:28:39 -0800
-Subject: [PATCH 2/2] Replace isascii call with any() call
-
----
- Lib/urllib/parse.py | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
-index 24a628915e53..7b06f4d71d67 100644
---- a/Lib/urllib/parse.py
-+++ b/Lib/urllib/parse.py
-@@ -392,7 +392,7 @@ def _splitnetloc(url, start=0):
- return url[start:delim], url[delim:] # return (domain, rest)
-
- def _checknetloc(netloc):
-- if not netloc or netloc.isascii():
-+ if not netloc or not any(ord(c) > 127 for c in netloc):
- return
- # looking for characters like \u2100 that expand to 'a/c'
- # IDNA uses NFKC equivalence, so normalize for this check
projects / feed / packages.git / commitdiff