mingw: allow git.exe to be used instead of the "Git wrapper"

Git for Windows wants to add `git.exe` to the users' `PATH`, without
cluttering the latter with unnecessary executables such as `wish.exe`.
To that end, it invented the concept of its "Git wrapper", i.e. a tiny
executable located in `C:\Program Files\Git\cmd\git.exe` (originally a
CMD script) whose sole purpose is to set up a couple of environment
variables and then spawn the _actual_ `git.exe` (which nowadays lives in
`C:\Program Files\Git\mingw64\bin\git.exe` for 64-bit, and the obvious
equivalent for 32-bit installations).

Currently, the following environment variables are set unless already
initialized:

- `MSYSTEM`, to make sure that the MSYS2 Bash and the MSYS2 Perl
  interpreter behave as expected, and

- `PLINK_PROTOCOL`, to force PuTTY's `plink.exe` to use the SSH
  protocol instead of Telnet,

- `PATH`, to make sure that the `bin` folder in the user's home
  directory, as well as the `/mingw64/bin` and the `/usr/bin`
  directories are included. The trick here is that the `/mingw64/bin/`
  and `/usr/bin/` directories are relative to the top-level installation
  directory of Git for Windows (which the included Bash interprets as
  `/`, i.e. as the MSYS pseudo root directory).

Using the absence of `MSYSTEM` as a tell-tale, we can detect in
`git.exe` whether these environment variables have been initialized
properly. Therefore we can call `C:\Program Files\Git\mingw64\bin\git`
in-place after this change, without having to call Git through the Git
wrapper.

Obviously, above-mentioned directories must be _prepended_ to the `PATH`
variable, otherwise we risk picking up executables from unrelated Git
installations. We do that by constructing the new `PATH` value from
scratch, appending `$HOME/bin` (if `HOME` is set), then the MSYS2 system
directories, and then appending the original `PATH`.

Side note: this modification of the `PATH` variable is independent of
the modification necessary to reach the executables and scripts in
`/mingw64/libexec/git-core/`, i.e. the `GIT_EXEC_PATH`. That
modification is still performed by Git, elsewhere, long after making the
changes described above.

While we _still_ cannot simply hard-link `mingw64\bin\git.exe` to `cmd`
(because the former depends on a couple of `.dll` files that are only in
`mingw64\bin`, i.e. calling `...\cmd\git.exe` would fail to load due to
missing dependencies), at least we can now avoid that extra process of
running the Git wrapper (which then has to wait for the spawned
`git.exe` to finish) by calling `...\mingw64\bin\git.exe` directly, via
its absolute path.

Testing this is in Git's test suite tricky: we set up a "new" MSYS
pseudo-root and copy the `git.exe` file into the appropriate location,
then verify that `MSYSTEM` is set properly, and also that the `PATH` is
modified so that scripts can be found in `$HOME/bin`, `/mingw64/bin/`
and `/usr/bin/`.

This addresses #2283

Signed-off-by: Johannes Schindelin <[email protected]>

Author: dscho

compat/mingw.c

@@ -2798,6 +2798,54 @@ int xwcstoutf(char *utf, const wchar_t *wcs, size_t utflen)
 	return -1;
 }
 
+#ifdef ENSURE_MSYSTEM_IS_SET
+#if !defined(RUNTIME_PREFIX) || !defined(HAVE_WPGMPTR)
+static size_t append_system_bin_dirs(char *path UNUSED, size_t size UNUSED)
+{
+	return 0;
+}
+#else
+static size_t append_system_bin_dirs(char *path, size_t size)
+{
+	char prefix[32768];
+	const char *slash;
+	size_t len = xwcstoutf(prefix, _wpgmptr, sizeof(prefix)), off = 0;
+
+	if (len == 0 || len >= sizeof(prefix) ||
+	    !(slash = find_last_dir_sep(prefix)))
+		return 0;
+	/* strip trailing `git.exe` */
+	len = slash - prefix;
+
+	/* strip trailing `cmd` or `mingw64\bin` or `mingw32\bin` or `bin` or `libexec\git-core` */
+	if (strip_suffix_mem(prefix, &len, "\\mingw64\\libexec\\git-core") ||
+	    strip_suffix_mem(prefix, &len, "\\mingw64\\bin"))
+		off += xsnprintf(path + off, size - off,
+				 "%.*s\\mingw64\\bin;", (int)len, prefix);
+	else if (strip_suffix_mem(prefix, &len, "\\clangarm64\\libexec\\git-core") ||
+	    strip_suffix_mem(prefix, &len, "\\clangarm64\\bin"))
+		off += xsnprintf(path + off, size - off,
+				 "%.*s\\clangarm64\\bin;", (int)len, prefix);
+	else if (strip_suffix_mem(prefix, &len, "\\mingw32\\libexec\\git-core") ||
+		 strip_suffix_mem(prefix, &len, "\\mingw32\\bin"))
+		off += xsnprintf(path + off, size - off,
+				 "%.*s\\mingw32\\bin;", (int)len, prefix);
+	else if (strip_suffix_mem(prefix, &len, "\\cmd") ||
+		 strip_suffix_mem(prefix, &len, "\\bin") ||
+		 strip_suffix_mem(prefix, &len, "\\libexec\\git-core"))
+		off += xsnprintf(path + off, size - off,
+				 "%.*s\\mingw%d\\bin;", (int)len, prefix,
+				 (int)(sizeof(void *) * 8));
+	else
+		return 0;
+
+	off += xsnprintf(path + off, size - off,
+			 "%.*s\\usr\\bin;", (int)len, prefix);
+	return off;
+}
+#endif
+#endif
+
 static void setup_windows_environment(void)
 {
 	char *tmp = getenv("TMPDIR");
@@ -2850,6 +2898,38 @@ static void setup_windows_environment(void)
 			setenv("HOME", tmp, 1);
 	}
 
+	if (!getenv("PLINK_PROTOCOL"))
+		setenv("PLINK_PROTOCOL", "ssh", 0);
+
+#ifdef ENSURE_MSYSTEM_IS_SET
+	if (!(tmp = getenv("MSYSTEM")) || !tmp[0]) {
+		const char *home = getenv("HOME"), *path = getenv("PATH");
+		char buf[32768];
+		size_t off = 0;
+
+#if defined(__aarch64__) || defined(_M_ARM64) || defined(_M_ARM64EC)
+		setenv("MSYSTEM", "CLANGARM64", 1);
+#elif defined(__MINGW64__) || defined(_M_AMD64)
+		setenv("MSYSTEM", "MINGW64", 1);
+#else
+		setenv("MSYSTEM", "MINGW32", 1);
+#endif
+
+		if (home)
+			off += xsnprintf(buf + off, sizeof(buf) - off,
+					 "%s\\bin;", home);
+		off += append_system_bin_dirs(buf + off, sizeof(buf) - off);
+		if (path)
+			off += xsnprintf(buf + off, sizeof(buf) - off,
+					 "%s", path);
+		else if (off > 0)
+			buf[off - 1] = '\0';
+		else
+			buf[0] = '\0';
+		setenv("PATH", buf, 1);
+	}
+#endif
+
 	if (!getenv("LC_ALL") && !getenv("LC_CTYPE") && !getenv("LANG"))
 		setenv("LC_CTYPE", "C.UTF-8", 1);
 }

config.mak.uname

@@ -528,7 +528,7 @@ endif
 		compat/win32/pthread.o compat/win32/syslog.o \
 		compat/win32/trace2_win32_process_info.o \
 		compat/win32/dirent.o
-	COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\"
+	COMPAT_CFLAGS = -D__USE_MINGW_ACCESS -DDETECT_MSYS_TTY -DENSURE_MSYSTEM_IS_SET -DNOGDI -DHAVE_STRING_H -Icompat -Icompat/regex -Icompat/win32 -DSTRIP_EXTENSION=\".exe\"
 	BASIC_LDFLAGS = -IGNORE:4217 -IGNORE:4049 -NOLOGO -ENTRY:wmainCRTStartup -SUBSYSTEM:CONSOLE
 	# invalidcontinue.obj allows Git's source code to close the same file
 	# handle twice, or to access the osfhandle of an already-closed stdout
@@ -761,7 +761,7 @@ ifeq ($(uname_S),MINGW)
         endif
 	CC = gcc
 	COMPAT_CFLAGS += -D__USE_MINGW_ANSI_STDIO=0 -DDETECT_MSYS_TTY \
-		-fstack-protector-strong
+		-DENSURE_MSYSTEM_IS_SET -fstack-protector-strong
 	EXTLIBS += -lntdll
 	EXTRA_PROGRAMS += headless-git$X
 	INSTALL = /bin/install

t/t0060-path-utils.sh

@@ -602,7 +602,8 @@ test_expect_success !VALGRIND,RUNTIME_PREFIX,CAN_EXEC_IN_PWD 'RUNTIME_PREFIX wor
 	echo "echo HERE" | write_script pretend/libexec/git-core/git-here &&
 	GIT_EXEC_PATH= ./pretend/bin/git here >actual &&
 	echo HERE >expect &&
-	test_cmp expect actual'
+	test_cmp expect actual
+'
 
 test_expect_success !VALGRIND,RUNTIME_PREFIX,CAN_EXEC_IN_PWD '%(prefix)/ works' '
 	git config yes.path "%(prefix)/yes" &&
@@ -611,4 +612,34 @@ test_expect_success !VALGRIND,RUNTIME_PREFIX,CAN_EXEC_IN_PWD '%(prefix)/ works'
 	test_cmp expect actual
 '
 
+test_expect_success MINGW,RUNTIME_PREFIX 'MSYSTEM/PATH is adjusted if necessary' '
+	if test -z "$MINGW_PREFIX"
+	then
+		MINGW_PREFIX="/$(echo "${MSYSTEM:-MINGW64}" | tr A-Z a-z)"
+	fi &&
+	mkdir -p "$HOME"/bin pretend"$MINGW_PREFIX"/bin \
+		pretend"$MINGW_PREFIX"/libexec/git-core pretend/usr/bin &&
+	cp "$GIT_EXEC_PATH"/git.exe pretend"$MINGW_PREFIX"/bin/ &&
+	cp "$GIT_EXEC_PATH"/git.exe pretend"$MINGW_PREFIX"/libexec/git-core/ &&
+	# copy the .dll files, if any (happens when building via CMake)
+	if test -n "$(ls "$GIT_EXEC_PATH"/*.dll 2>/dev/null)"
+	then
+		cp "$GIT_EXEC_PATH"/*.dll pretend"$MINGW_PREFIX"/bin/ &&
+		cp "$GIT_EXEC_PATH"/*.dll pretend"$MINGW_PREFIX"/libexec/git-core/
+	fi &&
+	echo "env | grep MSYSTEM=" | write_script "$HOME"/bin/git-test-home &&
+	echo "echo ${MINGW_PREFIX#/}" | write_script pretend"$MINGW_PREFIX"/bin/git-test-bin &&
+	echo "echo usr" | write_script pretend/usr/bin/git-test-bin2 &&
+
+	(
+		MSYSTEM= &&
+		GIT_EXEC_PATH= &&
+		pretend"$MINGW_PREFIX"/libexec/git-core/git.exe test-home >actual &&
+		pretend"$MINGW_PREFIX"/libexec/git-core/git.exe test-bin >>actual &&
+		pretend"$MINGW_PREFIX"/bin/git.exe test-bin2 >>actual
+	) &&
+	test_write_lines MSYSTEM=$MSYSTEM "${MINGW_PREFIX#/}" usr >expect &&
+	test_cmp expect actual
+'
+
 test_done