msys mingw prefer unix sep if MSYSTEM

Co-authored-by: Алексей <alexey.pawlow@gmail.com>
Co-authored-by: Christoph Reiter <reiter.christoph@gmail.com>

Include/pylifecycle.h

@@ -21,6 +21,12 @@ PyAPI_FUNC(int) Py_IsInitialized(void);
 PyAPI_FUNC(PyThreadState *) Py_NewInterpreter(void);
 PyAPI_FUNC(void) Py_EndInterpreter(PyThreadState *);
 
+PyAPI_FUNC(wchar_t) Py_GetSepW(const wchar_t *);
+PyAPI_FUNC(char) Py_GetSepA(const char *);
+
+PyAPI_FUNC(void) Py_NormalizeSepsW(wchar_t *);
+PyAPI_FUNC(void) Py_NormalizeSepsA(char *);
+
 
 /* Py_PyAtExit is for the atexit module, Py_AtExit is for low-level
  * exit functions.

Lib/ntpath.py

@@ -11,9 +11,7 @@
 curdir = '.'
 pardir = '..'
 extsep = '.'
-sep = '\\'
 pathsep = ';'
-altsep = '/'
 defpath = '.;C:\\bin'
 devnull = 'nul'
 
@@ -23,6 +21,15 @@
 import genericpath
 from genericpath import *
 
+if sys.platform == "win32" and "MSYSTEM" in os.environ:
+    sep = '/'
+    altsep = '\\'
+else:
+    sep = '\\'
+    altsep = '/'
+bsep = str.encode(sep)
+baltsep = str.encode(altsep)
+
 __all__ = ["normcase","isabs","join","splitdrive","split","splitext",
            "basename","dirname","commonprefix","getsize","getmtime",
            "getatime","getctime", "islink","exists","lexists","isdir","isfile",
@@ -33,9 +40,27 @@
 
 def _get_bothseps(path):
     if isinstance(path, bytes):
-        return b'\\/'
+        return bsep+baltsep
+    else:
+        return sep+altsep
+
+def _get_sep(path):
+    if isinstance(path, bytes):
+        return bsep
+    else:
+        return sep
+
+def _get_altsep(path):
+    if isinstance(path, bytes):
+        return baltsep
     else:
-        return '\\/'
+        return altsep
+
+def _get_colon(path):
+    if isinstance(path, bytes):
+        return b':'
+    else:
+        return ':'
 
 # Normalize the case of a pathname and map slashes to backslashes.
 # Other normalizations (such as optimizing '../' away) are not done
@@ -47,9 +72,9 @@ def normcase(s):
     Makes all characters lowercase and all slashes into backslashes."""
     s = os.fspath(s)
     if isinstance(s, bytes):
-        return s.replace(b'/', b'\\').lower()
+        return s.replace(baltsep, bsep).lower()
     else:
-        return s.replace('/', '\\').lower()
+        return s.replace(altsep, sep).lower()
 
 
 # Return whether a path is absolute.
@@ -76,14 +101,9 @@ def isabs(s):
 # Join two (or more) paths.
 def join(path, *paths):
     path = os.fspath(path)
-    if isinstance(path, bytes):
-        sep = b'\\'
-        seps = b'\\/'
-        colon = b':'
-    else:
-        sep = '\\'
-        seps = '\\/'
-        colon = ':'
+    sep = _get_sep(path)
+    seps = _get_bothseps(path)
+    colon = _get_colon(path)
     try:
         if not paths:
             path[:0] + sep  #23780: Ensure compatible data type even if p is null.
@@ -142,14 +162,9 @@ def splitdrive(p):
     """
     p = os.fspath(p)
     if len(p) >= 2:
-        if isinstance(p, bytes):
-            sep = b'\\'
-            altsep = b'/'
-            colon = b':'
-        else:
-            sep = '\\'
-            altsep = '/'
-            colon = ':'
+        sep = _get_sep(p)
+        altsep = _get_altsep(p)
+        colon = _get_colon(p)
         normp = p.replace(altsep, sep)
         if (normp[0:2] == sep*2) and (normp[2:3] != sep):
             # is a UNC path:
@@ -203,9 +218,9 @@ def split(p):
 def splitext(p):
     p = os.fspath(p)
     if isinstance(p, bytes):
-        return genericpath._splitext(p, b'\\', b'/', b'.')
+        return genericpath._splitext(p, bsep, baltsep, b'.')
     else:
-        return genericpath._splitext(p, '\\', '/', '.')
+        return genericpath._splitext(p, sep, altsep, '.')
 splitext.__doc__ = genericpath._splitext.__doc__
 
 
@@ -450,15 +465,13 @@ def expandvars(path):
 def normpath(path):
     """Normalize path, eliminating double slashes, etc."""
     path = os.fspath(path)
+    sep = _get_sep(path)
+    altsep = _get_altsep(path)
     if isinstance(path, bytes):
-        sep = b'\\'
-        altsep = b'/'
         curdir = b'.'
         pardir = b'..'
         special_prefixes = (b'\\\\.\\', b'\\\\?\\')
     else:
-        sep = '\\'
-        altsep = '/'
         curdir = '.'
         pardir = '..'
         special_prefixes = ('\\\\.\\', '\\\\?\\')
@@ -668,6 +681,7 @@ def realpath(path):
                 # strip the prefix anyway.
                 if ex.winerror == initial_winerror:
                     path = spath
+        path = normpath(path)
         return path
 
 
@@ -678,12 +692,11 @@ def realpath(path):
 def relpath(path, start=None):
     """Return a relative version of a path"""
     path = os.fspath(path)
+    sep = _get_sep(path)
     if isinstance(path, bytes):
-        sep = b'\\'
         curdir = b'.'
         pardir = b'..'
     else:
-        sep = '\\'
         curdir = '.'
         pardir = '..'
 
@@ -738,13 +751,11 @@ def commonpath(paths):
         raise ValueError('commonpath() arg is an empty sequence')
 
     paths = tuple(map(os.fspath, paths))
+    sep = _get_sep(paths[0])
+    altsep = _get_altsep(paths[0])
     if isinstance(paths[0], bytes):
-        sep = b'\\'
-        altsep = b'/'
         curdir = b'.'
     else:
-        sep = '\\'
-        altsep = '/'
         curdir = '.'
 
     try:

Modules/posixmodule.c

@@ -3759,6 +3759,7 @@ posix_getcwd(int use_bytes)
         return PyErr_SetFromWindowsErr(0);
     }
 
+    Py_NormalizeSepsW(wbuf2);
     PyObject *resobj = PyUnicode_FromWideChar(wbuf2, len);
     if (wbuf2 != wbuf) {
         PyMem_RawFree(wbuf2);
@@ -4322,6 +4323,7 @@ os__getfinalpathname_impl(PyObject *module, path_t *path)
         target_path = tmp;
     }
 
+    Py_NormalizeSepsW(target_path);
     result = PyUnicode_FromWideChar(target_path, result_length);
     if (result && path->narrow) {
         Py_SETREF(result, PyUnicode_EncodeFSDefault(result));

Python/initconfig.c

@@ -136,7 +136,7 @@ static const char usage_6[] =
 "PYTHONDEVMODE: enable the development mode.\n"
 "PYTHONPYCACHEPREFIX: root directory for bytecode cache (pyc) files.\n";
 
-#if defined(MS_WINDOWS)
+#if defined(_MSC_VER)
 #  define PYTHONHOMEHELP "<prefix>\\python{major}{minor}"
 #else
 #  define PYTHONHOMEHELP "<prefix>/lib/pythonX.X"
@@ -1187,7 +1187,7 @@ config_init_program_name(PyConfig *config)
     }
 
     /* Last fall back: hardcoded name */
-#ifdef MS_WINDOWS
+#ifdef _MSC_VER
     const wchar_t *default_program_name = L"python";
 #else
     const wchar_t *default_program_name = L"python3";

Python/pathconfig.c

@@ -15,6 +15,114 @@
 extern "C" {
 #endif
 
+#ifdef __MINGW32__
+#define wcstok  wcstok_s
+#include <windows.h>
+#endif
+
+char
+Py_GetSepA(const char *name)
+{
+    char* msystem = (char*)2; /* So that non Windows use / as sep */
+    static char sep = '\0';
+#ifdef _WIN32
+    /* https://msdn.microsoft.com/en-gb/library/windows/desktop/aa365247%28v=vs.85%29.aspx
+     * The "\\?\" prefix .. indicate that the path should be passed to the system with minimal
+     * modification, which means that you cannot use forward slashes to represent path separators
+     */
+    if (name != NULL && memcmp(name, "\\\\?\\", sizeof("\\\\?\\") - sizeof(char)) == 0)
+    {
+        return '\\';
+    }
+#endif
+    if (sep != '\0')
+        return sep;
+#if defined(__MINGW32__)
+    msystem = Py_GETENV("MSYSTEM");
+#endif
+    if (msystem != NULL)
+        sep = '/';
+    else
+        sep = '\\';
+    return sep;
+}
+
+static char
+Py_GetAltSepA(const char *name)
+{
+    char sep = Py_GetSepA(name);
+    if (sep == '/')
+        return '\\';
+    return '/';
+}
+
+void
+Py_NormalizeSepsA(char *name)
+{
+    char sep = Py_GetSepA(name);
+    char altsep = Py_GetAltSepA(name);
+    char* seps;
+    if (strlen(name) > 1 && name[1] == ':') {
+        name[0] = toupper(name[0]);
+    }
+    seps = strchr(name, altsep);
+    while(seps) {
+        *seps = sep;
+        seps = strchr(seps, altsep);
+    }
+}
+
+wchar_t
+Py_GetSepW(const wchar_t *name)
+{
+    char* msystem = (char*)2; /* So that non Windows use / as sep */
+    static wchar_t sep = L'\0';
+#ifdef _WIN32
+    /* https://msdn.microsoft.com/en-gb/library/windows/desktop/aa365247%28v=vs.85%29.aspx
+     * The "\\?\" prefix .. indicate that the path should be passed to the system with minimal
+     * modification, which means that you cannot use forward slashes to represent path separators
+     */
+    if (name != NULL && memcmp(name, L"\\\\?\\", sizeof(L"\\\\?\\") - sizeof(wchar_t)) == 0)
+    {
+        return L'\\';
+    }
+#endif
+    if (sep != L'\0')
+        return sep;
+#if defined(__MINGW32__)
+    msystem = Py_GETENV("MSYSTEM");
+#endif
+    if (msystem != NULL)
+        sep = L'/';
+    else
+        sep = L'\\';
+    return sep;
+}
+
+static wchar_t
+Py_GetAltSepW(const wchar_t *name)
+{
+    char sep = Py_GetSepW(name);
+    if (sep == L'/')
+        return L'\\';
+    return L'/';
+}
+
+void
+Py_NormalizeSepsW(wchar_t *name)
+{
+    wchar_t sep = Py_GetSepW(name);
+    wchar_t altsep = Py_GetAltSepW(name);
+    wchar_t* seps;
+    if (wcslen(name) > 1 && name[1] == L':') {
+        name[0] = towupper(name[0]);
+    }
+    seps = wcschr(name, altsep);
+    while(seps) {
+        *seps = sep;
+        seps = wcschr(seps, altsep);
+    }
+}
 
 _PyPathConfig _Py_path_config = _PyPathConfig_INIT;
 
@@ -541,6 +649,7 @@ _Py_SetProgramFullPath(const wchar_t *program_full_path)
     if (_Py_path_config.program_full_path == NULL) {
         path_out_of_memory(__func__);
     }
+    Py_NormalizeSepsW(_Py_path_config.program_name);
 }
 
 

Python/traceback.c

@@ -314,7 +314,7 @@ _Py_FindSourceFile(PyObject *filename, char* namebuf, size_t namelen, PyObject *
     filepath = PyBytes_AS_STRING(filebytes);
 
     /* Search tail of filename in sys.path before giving up */
-    tail = strrchr(filepath, SEP);
+    tail = strrchr(filepath, Py_GetSepA(filepath));
     if (tail == NULL)
         tail = filepath;
     else