bpo-42260: PyConfig_Read() only parses argv once (GH-23168)

The PyConfig_Read() function now only parses PyConfig.argv arguments
once: PyConfig.parse_argv is set to 2 after arguments are parsed.
Since Python arguments are strippped from PyConfig.argv, parsing
arguments twice would parse the application options as Python
options.

* Rework the PyConfig documentation.
* Fix _testinternalcapi.set_config() error handling.
* SetConfigTests no longer needs parse_argv=0 when restoring the old
  configuration.

6 files changed, 134 insertions, 105 deletions

diff --git a/Doc/c-api/init_config.rst b/Doc/c-api/init_config.rst
index c957d6c..edfeba5 100644
--- a/Doc/c-api/init_config.rst
+++ b/Doc/c-api/init_config.rst
@@ -8,55 +8,68 @@ Python Initialization Configuration
 
 .. versionadded:: 3.8
 
-Structures:
-
-* :c:type:`PyConfig`
-* :c:type:`PyPreConfig`
-* :c:type:`PyStatus`
-* :c:type:`PyWideStringList`
-
-Functions:
-
-* :c:func:`PyConfig_Clear`
-* :c:func:`PyConfig_InitIsolatedConfig`
-* :c:func:`PyConfig_InitPythonConfig`
-* :c:func:`PyConfig_Read`
-* :c:func:`PyConfig_SetArgv`
-* :c:func:`PyConfig_SetBytesArgv`
-* :c:func:`PyConfig_SetBytesString`
-* :c:func:`PyConfig_SetString`
-* :c:func:`PyConfig_SetWideStringList`
-* :c:func:`PyPreConfig_InitIsolatedConfig`
-* :c:func:`PyPreConfig_InitPythonConfig`
-* :c:func:`PyStatus_Error`
-* :c:func:`PyStatus_Exception`
-* :c:func:`PyStatus_Exit`
-* :c:func:`PyStatus_IsError`
-* :c:func:`PyStatus_IsExit`
-* :c:func:`PyStatus_NoMemory`
-* :c:func:`PyStatus_Ok`
-* :c:func:`PyWideStringList_Append`
-* :c:func:`PyWideStringList_Insert`
-* :c:func:`Py_ExitStatusException`
-* :c:func:`Py_InitializeFromConfig`
-* :c:func:`Py_PreInitialize`
-* :c:func:`Py_PreInitializeFromArgs`
-* :c:func:`Py_PreInitializeFromBytesArgs`
-* :c:func:`Py_RunMain`
-* :c:func:`Py_GetArgcArgv`
-
-The preconfiguration (``PyPreConfig`` type) is stored in
-``_PyRuntime.preconfig`` and the configuration (``PyConfig`` type) is stored in
-``PyInterpreterState.config``.
+Python can be initialized with :c:func:`Py_InitializeFromConfig` and the
+:c:type:`PyConfig` structure. It can be preinitialized with
+:c:func:`Py_PreInitialize` and the :c:type:`PyPreConfig` structure.
+
+There are two kinds of configuration:
+
+* The :ref:`Python Configuration <init-python-config>` can be used to build a
+  customized Python which behaves as the regular Python. For example,
+  environments variables and command line arguments are used to configure
+  Python.
+
+* The :ref:`Isolated Configuration <init-isolated-conf>` can be used to embed
+  Python into an application. It isolates Python from the system. For example,
+  environments variables are ignored, the LC_CTYPE locale is left unchanged and
+  no signal handler is registred.
 
 See also :ref:`Initialization, Finalization, and Threads <initialization>`.
 
 .. seealso::
    :pep:`587` "Python Initialization Configuration".
 
+Example
+=======
+
+Example of customized Python always running in isolated mode::
+
+    int main(int argc, char **argv)
+    {
+        PyStatus status;
+
+        PyConfig config;
+        PyConfig_InitPythonConfig(&config);
+        config.isolated = 1;
+
+        /* Decode command line arguments.
+           Implicitly preinitialize Python (in isolated mode). */
+        status = PyConfig_SetBytesArgv(&config, argc, argv);
+        if (PyStatus_Exception(status)) {
+            goto exception;
+        }
+
+        status = Py_InitializeFromConfig(&config);
+        if (PyStatus_Exception(status)) {
+            goto exception;
+        }
+        PyConfig_Clear(&config);
+
+        return Py_RunMain();
+
+    exception:
+        PyConfig_Clear(&config);
+        if (PyStatus_IsExit(status)) {
+            return status.exitcode;
+        }
+        /* Display the error message and exit the process with
+           non-zero exit code */
+        Py_ExitStatusException(status);
+    }
+
 
 PyWideStringList
-----------------
+================
 
 .. c:type:: PyWideStringList
 
@@ -95,7 +108,7 @@ PyWideStringList
       List items.
 
 PyStatus
---------
+========
 
 .. c:type:: PyStatus
 
@@ -187,7 +200,7 @@ Example::
 
 
 PyPreConfig
------------
+===========
 
 .. c:type:: PyPreConfig
 
@@ -317,7 +330,7 @@ PyPreConfig
 .. _c-preinit:
 
 Preinitialize Python with PyPreConfig
--------------------------------------
+=====================================
 
 The preinitialization of Python:
 
@@ -326,12 +339,17 @@ The preinitialization of Python:
 * Set the :ref:`Python UTF-8 Mode <utf8-mode>`
   (:c:member:`PyPreConfig.utf8_mode`)
 
+The current preconfiguration (``PyPreConfig`` type) is stored in
+``_PyRuntime.preconfig``.
+
 Functions to preinitialize Python:
 
 .. c:function:: PyStatus Py_PreInitialize(const PyPreConfig *preconfig)
 
    Preinitialize Python from *preconfig* preconfiguration.
 
+   *preconfig* must not be ``NULL``.
+
 .. c:function:: PyStatus Py_PreInitializeFromBytesArgs(const PyPreConfig *preconfig, int argc, char * const *argv)
 
    Preinitialize Python from *preconfig* preconfiguration.
@@ -339,6 +357,8 @@ Functions to preinitialize Python:
    Parse *argv* command line arguments (bytes strings) if
    :c:member:`~PyPreConfig.parse_argv` of *preconfig* is non-zero.
 
+   *preconfig* must not be ``NULL``.
+
 .. c:function:: PyStatus Py_PreInitializeFromArgs(const PyPreConfig *preconfig, int argc, wchar_t * const * argv)
 
    Preinitialize Python from *preconfig* preconfiguration.
@@ -346,6 +366,8 @@ Functions to preinitialize Python:
    Parse *argv* command line arguments (wide strings) if
    :c:member:`~PyPreConfig.parse_argv` of *preconfig* is non-zero.
 
+   *preconfig* must not be ``NULL``.
+
 The caller is responsible to handle exceptions (error or exit) using
 :c:func:`PyStatus_Exception` and :c:func:`Py_ExitStatusException`.
 
@@ -388,7 +410,7 @@ the :ref:`Python UTF-8 Mode <utf8-mode>`::
 
 
 PyConfig
---------
+========
 
 .. c:type:: PyConfig
 
@@ -449,8 +471,20 @@ PyConfig
 
       Fields which are already initialized are left unchanged.
 
+      The :c:func:`PyConfig_Read` function only parses
+      :c:member:`PyConfig.argv` arguments once: :c:member:`PyConfig.parse_argv`
+      is set to ``2`` after arguments are parsed. Since Python arguments are
+      strippped from :c:member:`PyConfig.argv`, parsing arguments twice would
+      parse the application options as Python options.
+
       :ref:`Preinitialize Python <c-preinit>` if needed.
 
+      .. versionchanged:: 3.10
+         The :c:member:`PyConfig.argv` arguments are now only parsed once,
+         :c:member:`PyConfig.parse_argv` is set to ``2`` after arguments are
+         parsed, and arguments are only parsed if
+         :c:member:`PyConfig.parse_argv` equals ``1``.
+
    .. c:function:: void PyConfig_Clear(PyConfig *config)
 
       Release configuration memory.
@@ -833,7 +867,7 @@ PyConfig
 
       If :c:member:`~PyConfig.orig_argv` list is empty and
       :c:member:`~PyConfig.argv` is not a list only containing an empty
-      string, :c:func:`PyConfig_Read()` copies :c:member:`~PyConfig.argv` into
+      string, :c:func:`PyConfig_Read` copies :c:member:`~PyConfig.argv` into
       :c:member:`~PyConfig.orig_argv` before modifying
       :c:member:`~PyConfig.argv` (if :c:member:`~PyConfig.parse_argv` is
       non-zero).
@@ -849,12 +883,22 @@ PyConfig
 
       Parse command line arguments?
 
-      If non-zero, parse :c:member:`~PyConfig.argv` the same way the regular
+      If equals to ``1``, parse :c:member:`~PyConfig.argv` the same way the regular
       Python parses :ref:`command line arguments <using-on-cmdline>`, and strip
       Python arguments from :c:member:`~PyConfig.argv`.
 
+      The :c:func:`PyConfig_Read` function only parses
+      :c:member:`PyConfig.argv` arguments once: :c:member:`PyConfig.parse_argv`
+      is set to ``2`` after arguments are parsed. Since Python arguments are
+      strippped from :c:member:`PyConfig.argv`, parsing arguments twice would
+      parse the application options as Python options.
+
       Default: ``1`` in Python mode, ``0`` in isolated mode.
 
+      .. versionchanged:: 3.10
+         The :c:member:`PyConfig.argv` arguments are now only parsed if
+         :c:member:`PyConfig.parse_argv` equals to ``1``.
+
    .. c:member:: int parser_debug
 
       Parser debug mode. If greater than 0, turn on parser debugging output (for expert only, depending
@@ -1108,7 +1152,7 @@ the :option:`-X` command line option.
 
 
 Initialization with PyConfig
-----------------------------
+============================
 
 Function to initialize Python:
 
@@ -1123,6 +1167,9 @@ If :c:func:`PyImport_FrozenModules`, :c:func:`PyImport_AppendInittab` or
 :c:func:`PyImport_ExtendInittab` are used, they must be set or called after
 Python preinitialization and before the Python initialization.
 
+The current configuration (``PyConfig`` type) is stored in
+``PyInterpreterState.config``.
+
 Example setting the program name::
 
     void init_python(void)
@@ -1136,17 +1183,17 @@ Example setting the program name::
         status = PyConfig_SetString(&config, &config.program_name,
                                     L"/path/to/my_program");
         if (PyStatus_Exception(status)) {
-            goto fail;
+            goto exception;
         }
 
         status = Py_InitializeFromConfig(&config);
         if (PyStatus_Exception(status)) {
-            goto fail;
+            goto exception;
         }
         PyConfig_Clear(&config);
         return;
 
-    fail:
+    exception:
         PyConfig_Clear(&config);
         Py_ExitStatusException(status);
     }
@@ -1202,7 +1249,7 @@ configuration, and then override some parameters::
 .. _init-isolated-conf:
 
 Isolated Configuration
-----------------------
+======================
 
 :c:func:`PyPreConfig_InitIsolatedConfig` and
 :c:func:`PyConfig_InitIsolatedConfig` functions create a configuration to
@@ -1223,7 +1270,7 @@ configuration.
 .. _init-python-config:
 
 Python Configuration
---------------------
+====================
 
 :c:func:`PyPreConfig_InitPythonConfig` and :c:func:`PyConfig_InitPythonConfig`
 functions create a configuration to build a customized Python which behaves as
@@ -1237,46 +1284,11 @@ and :ref:`Python UTF-8 Mode <utf8-mode>`
 (:pep:`540`) depending on the LC_CTYPE locale, :envvar:`PYTHONUTF8` and
 :envvar:`PYTHONCOERCECLOCALE` environment variables.
 
-Example of customized Python always running in isolated mode::
-
-    int main(int argc, char **argv)
-    {
-        PyStatus status;
-
-        PyConfig config;
-        PyConfig_InitPythonConfig(&config);
-        config.isolated = 1;
-
-        /* Decode command line arguments.
-           Implicitly preinitialize Python (in isolated mode). */
-        status = PyConfig_SetBytesArgv(&config, argc, argv);
-        if (PyStatus_Exception(status)) {
-            goto fail;
-        }
-
-        status = Py_InitializeFromConfig(&config);
-        if (PyStatus_Exception(status)) {
-            goto fail;
-        }
-        PyConfig_Clear(&config);
-
-        return Py_RunMain();
-
-    fail:
-        PyConfig_Clear(&config);
-        if (PyStatus_IsExit(status)) {
-            return status.exitcode;
-        }
-        /* Display the error message and exit the process with
-           non-zero exit code */
-        Py_ExitStatusException(status);
-    }
-
 
 .. _init-path-config:
 
 Path Configuration
-------------------
+==================
 
 :c:type:`PyConfig` contains multiple fields for the path configuration:
 
@@ -1356,7 +1368,7 @@ The ``__PYVENV_LAUNCHER__`` environment variable is used to set
 
 
 Py_RunMain()
-------------
+============
 
 .. c:function:: int Py_RunMain(void)
 
@@ -1376,7 +1388,7 @@ customized Python always running in isolated mode using
 
 
 Py_GetArgcArgv()
-----------------
+================
 
 .. c:function:: void Py_GetArgcArgv(int *argc, wchar_t ***argv)
 
@@ -1386,7 +1398,7 @@ Py_GetArgcArgv()
 
 
 Multi-Phase Initialization Private Provisional API
---------------------------------------------------
+==================================================
 
 This section is a private provisional API introducing multi-phase
 initialization, the core feature of the :pep:`432`:
diff --git a/Lib/test/_test_embed_set_config.py b/Lib/test/_test_embed_set_config.py
index 7c91381..a19f8db 100644
--- a/Lib/test/_test_embed_set_config.py
+++ b/Lib/test/_test_embed_set_config.py
@@ -20,7 +20,7 @@ class SetConfigTests(unittest.TestCase):
         self.sys_copy = dict(sys.__dict__)
 
     def tearDown(self):
-        self.set_config(parse_argv=0)
+        _testinternalcapi.set_config(self.old_config)
         sys.__dict__.clear()
         sys.__dict__.update(self.sys_copy)
 
@@ -234,6 +234,12 @@ class SetConfigTests(unittest.TestCase):
         self.assertEqual(sys.argv, ['python_program', 'args'])
         self.assertEqual(sys.orig_argv, ['orig', 'orig_args'])
 
+        self.set_config(parse_argv=0,
+                        argv=[],
+                        orig_argv=[])
+        self.assertEqual(sys.argv, [''])
+        self.assertEqual(sys.orig_argv, [])
+
     def test_pycache_prefix(self):
         self.check(pycache_prefix=None)
         self.check(pycache_prefix="pycache_prefix")
diff --git a/Lib/test/test_embed.py b/Lib/test/test_embed.py
index 9182061..a7d9121 100644
--- a/Lib/test/test_embed.py
+++ b/Lib/test/test_embed.py
@@ -422,7 +422,7 @@ class InitConfigTests(EmbeddingTestsMixin, unittest.TestCase):
     CONFIG_PYTHON = dict(CONFIG_COMPAT,
         _config_init=API_PYTHON,
         configure_c_stdio=1,
-        parse_argv=1,
+        parse_argv=2,
     )
     CONFIG_ISOLATED = dict(CONFIG_COMPAT,
         _config_init=API_ISOLATED,
@@ -800,7 +800,7 @@ class InitConfigTests(EmbeddingTestsMixin, unittest.TestCase):
                           '-X', 'cmdline_xoption',
                           '-c', 'pass',
                           'arg2'],
-            'parse_argv': 1,
+            'parse_argv': 2,
             'xoptions': [
                 'config_xoption1=3',
                 'config_xoption2=',
@@ -1045,7 +1045,7 @@ class InitConfigTests(EmbeddingTestsMixin, unittest.TestCase):
             'orig_argv': ['python3', '-c', code, 'arg2'],
             'program_name': './python3',
             'run_command': code + '\n',
-            'parse_argv': 1,
+            'parse_argv': 2,
         }
         self.check_all_configs("test_init_run_main", config, api=API_PYTHON)
 
@@ -1059,7 +1059,7 @@ class InitConfigTests(EmbeddingTestsMixin, unittest.TestCase):
                           'arg2'],
             'program_name': './python3',
             'run_command': code + '\n',
-            'parse_argv': 1,
+            'parse_argv': 2,
             '_init_main': 0,
         }
         self.check_all_configs("test_init_main", config,
@@ -1068,7 +1068,7 @@ class InitConfigTests(EmbeddingTestsMixin, unittest.TestCase):
 
     def test_init_parse_argv(self):
         config = {
-            'parse_argv': 1,
+            'parse_argv': 2,
             'argv': ['-c', 'arg1', '-v', 'arg3'],
             'orig_argv': ['./argv0', '-E', '-c', 'pass', 'arg1', '-v', 'arg3'],
             'program_name': './argv0',
diff --git a/Misc/NEWS.d/next/C API/2020-11-05-18-02-07.bpo-42260.pAeaNR.rst b/Misc/NEWS.d/next/C API/2020-11-05-18-02-07.bpo-42260.pAeaNR.rst
new file mode 100644
index 0000000..0d6a277
--- /dev/null
+++ b/Misc/NEWS.d/next/C API/2020-11-05-18-02-07.bpo-42260.pAeaNR.rst
@@ -0,0 +1,5 @@
+The :c:func:`PyConfig_Read` function now only parses :c:member:`PyConfig.argv`
+arguments once: :c:member:`PyConfig.parse_argv` is set to ``2`` after arguments
+are parsed. Since Python arguments are strippped from
+:c:member:`PyConfig.argv`, parsing arguments twice would parse the application
+options as Python options.
diff --git a/Modules/_testinternalcapi.c b/Modules/_testinternalcapi.c
index be144bf..df4725e 100644
--- a/Modules/_testinternalcapi.c
+++ b/Modules/_testinternalcapi.c
@@ -253,14 +253,17 @@ test_set_config(PyObject *Py_UNUSED(self), PyObject *dict)
     PyConfig config;
     PyConfig_InitIsolatedConfig(&config);
     if (_PyConfig_FromDict(&config, dict) < 0) {
-        PyConfig_Clear(&config);
-        return NULL;
+        goto error;
     }
     if (_PyInterpreterState_SetConfig(&config) < 0) {
-        return NULL;
+        goto error;
     }
     PyConfig_Clear(&config);
     Py_RETURN_NONE;
+
+error:
+    PyConfig_Clear(&config);
+    return NULL;
 }
 
 
diff --git a/Python/initconfig.c b/Python/initconfig.c
index d54d5b7..e0811b5 100644
--- a/Python/initconfig.c
+++ b/Python/initconfig.c
@@ -1325,8 +1325,6 @@ _PyConfig_FromDict(PyConfig *config, PyObject *dict)
     GET_UINT(_init_main);
     GET_UINT(_isolated_interpreter);
 
-    assert(config_check_consistency(config));
-
 #undef CHECK_VALUE
 #undef GET_UINT
 #undef GET_WSTR
@@ -2145,6 +2143,11 @@ config_read(PyConfig *config)
         config->configure_c_stdio = 1;
     }
 
+    // Only parse arguments once.
+    if (config->parse_argv == 1) {
+        config->parse_argv = 2;
+    }
+
     return _PyStatus_OK();
 }
 
@@ -2635,7 +2638,7 @@ core_read_precmdline(PyConfig *config, _PyPreCmdline *precmdline)
 {
     PyStatus status;
 
-    if (config->parse_argv) {
+    if (config->parse_argv == 1) {
         if (_PyWideStringList_Copy(&precmdline->argv, &config->argv) < 0) {
             return _PyStatus_NO_MEMORY();
         }
@@ -2713,7 +2716,7 @@ config_read_cmdline(PyConfig *config)
         }
     }
 
-    if (config->parse_argv) {
+    if (config->parse_argv == 1) {
         Py_ssize_t opt_index;
         status = config_parse_cmdline(config, &cmdline_warnoptions, &opt_index);
         if (_PyStatus_EXCEPTION(status)) {