bpo-34589: Add -X coerce_c_locale command line option (GH-9378)

Add a new -X coerce_c_locale command line option to control C locale
coercion (PEP 538).

1 files changed, 79 insertions, 35 deletions

diff --git a/Python/coreconfig.c b/Python/coreconfig.c
index 131a043..b2459dc 100644
--- a/Python/coreconfig.c
+++ b/Python/coreconfig.c
@@ -705,6 +705,17 @@ config_init_utf8_mode(_PyCoreConfig *config)
         return _Py_INIT_OK();
     }
 
+#ifndef MS_WINDOWS
+    /* The C locale and the POSIX locale enable the UTF-8 Mode (PEP 540) */
+    const char *ctype_loc = setlocale(LC_CTYPE, NULL);
+    if (ctype_loc != NULL
+        && (strcmp(ctype_loc, "C") == 0 || strcmp(ctype_loc, "POSIX") == 0))
+    {
+        config->utf8_mode = 1;
+        return _Py_INIT_OK();
+    }
+#endif
+
     return _Py_INIT_OK();
 }
 
@@ -808,25 +819,6 @@ config_read_env_vars(_PyCoreConfig *config)
         config->malloc_stats = 1;
     }
 
-    const char *env = _PyCoreConfig_GetEnv(config, "PYTHONCOERCECLOCALE");
-    if (env) {
-        if (strcmp(env, "0") == 0) {
-            if (config->_coerce_c_locale < 0) {
-                config->_coerce_c_locale = 0;
-            }
-        }
-        else if (strcmp(env, "warn") == 0) {
-            if (config->_coerce_c_locale_warn < 0) {
-                config->_coerce_c_locale_warn = 1;
-            }
-        }
-        else {
-            if (config->_coerce_c_locale < 0) {
-                config->_coerce_c_locale = 1;
-            }
-        }
-    }
-
     wchar_t *path;
     int res = _PyCoreConfig_GetEnvDup(config, &path,
                                       L"PYTHONPATH", "PYTHONPATH");
@@ -966,28 +958,76 @@ config_read_complex_options(_PyCoreConfig *config)
 }
 
 
-static void
-config_init_locale(_PyCoreConfig *config)
+static _PyInitError
+config_init_coerce_c_locale(_PyCoreConfig *config)
 {
+    const wchar_t *xopt = config_get_xoption(config, L"coerce_c_locale");
+    if (xopt) {
+        wchar_t *sep = wcschr(xopt, L'=');
+        if (sep) {
+            xopt = sep + 1;
+            if (wcscmp(xopt, L"1") == 0) {
+                if (config->_coerce_c_locale < 0) {
+                    config->_coerce_c_locale = 1;
+                }
+            }
+            else if (wcscmp(xopt, L"0") == 0) {
+                if (config->_coerce_c_locale < 0) {
+                    config->_coerce_c_locale = 0;
+                }
+            }
+            else if (wcscmp(xopt, L"warn") == 0) {
+                if (config->_coerce_c_locale_warn < 0) {
+                    config->_coerce_c_locale_warn = 1;
+                }
+            }
+            else {
+                return _Py_INIT_USER_ERR("invalid -X coerce_c_locale option value");
+            }
+        }
+        else {
+            if (config->_coerce_c_locale < 0) {
+                config->_coerce_c_locale = 1;
+            }
+        }
+
+        if (config->_coerce_c_locale_warn < 0) {
+            config->_coerce_c_locale_warn = 0;
+        }
+    }
+
+    const char *env = _PyCoreConfig_GetEnv(config, "PYTHONCOERCECLOCALE");
+    if (env) {
+        if (strcmp(env, "0") == 0) {
+            if (config->_coerce_c_locale < 0) {
+                config->_coerce_c_locale = 0;
+            }
+        }
+        else if (strcmp(env, "warn") == 0) {
+            if (config->_coerce_c_locale_warn < 0) {
+                config->_coerce_c_locale_warn = 1;
+            }
+        }
+        else {
+            if (config->_coerce_c_locale < 0) {
+                config->_coerce_c_locale = 1;
+            }
+        }
+
+        if (config->_coerce_c_locale_warn < 0) {
+            config->_coerce_c_locale_warn = 0;
+        }
+    }
+
     if (config->_coerce_c_locale < 0) {
         /* The C locale enables the C locale coercion (PEP 538) */
         if (_Py_LegacyLocaleDetected()) {
             config->_coerce_c_locale = 1;
+            return _Py_INIT_OK();
         }
     }
 
-#ifndef MS_WINDOWS
-    if (config->utf8_mode < 0) {
-        /* The C locale and the POSIX locale enable the UTF-8 Mode (PEP 540) */
-        const char *ctype_loc = setlocale(LC_CTYPE, NULL);
-        if (ctype_loc != NULL
-           && (strcmp(ctype_loc, "C") == 0
-               || strcmp(ctype_loc, "POSIX") == 0))
-        {
-            config->utf8_mode = 1;
-        }
-    }
-#endif
+    return _Py_INIT_OK();
 }
 
 
@@ -1293,8 +1333,11 @@ _PyCoreConfig_Read(_PyCoreConfig *config)
         }
     }
 
-    if (config->utf8_mode < 0 || config->_coerce_c_locale < 0) {
-        config_init_locale(config);
+    if (config->_coerce_c_locale < 0 || config->_coerce_c_locale_warn < 0) {
+        err = config_init_coerce_c_locale(config);
+        if (_Py_INIT_FAILED(err)) {
+            return err;
+        }
     }
 
     if (config->_install_importlib) {
@@ -1349,6 +1392,7 @@ _PyCoreConfig_Read(_PyCoreConfig *config)
     }
 
     assert(config->_coerce_c_locale >= 0);
+    assert(config->_coerce_c_locale_warn >= 0);
     assert(config->use_environment >= 0);
     assert(config->filesystem_encoding != NULL);
     assert(config->filesystem_errors != NULL);