Translate OCAML_STDLIB_DIR correctly on Windows (#1448)

dra27 · gasche · commit 7f21a3fcdbf2 · 2017-10-27T17:25:28.000+02:00
The build generates invalid code in byterun/dynlink.c if LIBDIR contains non-1252 characters. Use iconv to translate LIBDIR to JAVA which, unlike C99 will produce only \u sequences with no \U sequences. The \u is then translated to \x which is the only form recognised pre-Visual Studio 2013. (cherry picked from commit 3b434eb)
diff --git a/Changes b/Changes
@@ -405,7 +405,7 @@ Release branch for 4.06:
 ### Runtime system:
 
 * MPR#3771, GPR#153, GPR#1200, GPR#1357, GPR#1362, GPR#1363, GPR#1369, GPR#1398,
-  GPR#1446: Unicode support for the Windows runtime.
+  GPR#1446, GPR#1448: Unicode support for the Windows runtime.
   (ygrek, Nicolas Ojeda Bar, review by Alain Frisch, David Allsopp, Damien
   Doligez)
 
diff --git a/byterun/Makefile b/byterun/Makefile
@@ -49,7 +49,17 @@ endif
 # On Windows, OCAML_STDLIB_DIR needs to be defined dynamically
 
 ifeq "$(UNIX_OR_WIN32)" "win32"
-CFLAGS += -DOCAML_STDLIB_DIR='"$(LIBDIR)"'
+# OCAML_STDLIB_DIR needs to arrive in dynlink.c as a string which both gcc and
+# msvc are willing parse without warning. This means we can't pass UTF-8
+# directly since, as far as I can tell, cl can cope, but the pre-processor
+# can't. So the string needs to be directly translated to L"" form. To do this,
+# we take advantage of the fact that Cygwin uses GNU libiconv which includes a
+# Java pseudo-encoding which translates any UTF-8 sequences to \uXXXX (and,
+# unlike the C99 pseudo-encoding, emits two surrogate values when needed, rather
+# than \UXXXXXXXX). The \u is then translated to \x in order to accommodate
+# pre-Visual Studio 2013 compilers where \x is a non-standard alias for \u.
+OCAML_STDLIB_DIR = $(shell echo $(LIBDIR)| iconv -t JAVA | sed -e 's/\\u/\\x/g')
+CFLAGS += -DOCAML_STDLIB_DIR='L"$(OCAML_STDLIB_DIR)"'
 endif
 
 CFLAGS += $(IFLEXDIR)
diff --git a/byterun/dynlink.c b/byterun/dynlink.c
@@ -77,7 +77,7 @@ static c_primitive lookup_primitive(char * name)
 
 static char_os * parse_ld_conf(void)
 {
-  char_os * stdlib, * ldconfname, * wconfig, * p, * q, * tofree = NULL;
+  char_os * stdlib, * ldconfname, * wconfig, * p, * q;
   char * config;
 #ifdef _WIN32
   struct _stati64 st;
@@ -88,9 +88,8 @@ static char_os * parse_ld_conf(void)
 
   stdlib = caml_secure_getenv(_T("OCAMLLIB"));
   if (stdlib == NULL) stdlib = caml_secure_getenv(_T("CAMLLIB"));
-  if (stdlib == NULL) stdlib = tofree = caml_stat_strdup_to_os(OCAML_STDLIB_DIR);
+  if (stdlib == NULL) stdlib = OCAML_STDLIB_DIR;
   ldconfname = caml_stat_strconcat_os(3, stdlib, _T("/"), LD_CONF_NAME);
-  if (tofree != NULL) caml_stat_free(tofree);
   if (stat_os(ldconfname, &st) == -1) {
     caml_stat_free(ldconfname);
     return NULL;
