[SystemZ][z/OS] ASCII/EBCDIC support with no coexistence
The aim of this patch is to break up the larger patch (https://reviews.llvm.org/D111323) to be more upstream friendly. In particular, this patch adds the char encoding sensitive changes but does not use inline namespaces as before. The use of namespaces to build both versions of the library, and localization of error messages will follow in a subsequent patch.
Differential Revision: https://reviews.llvm.org/D114813
NOKEYCHECK=True
GitOrigin-RevId: a1da73961d291c6a205150caa6ebda71757b9add
diff --git a/include/__config b/include/__config
index 654816b..85f334f 100644
--- a/include/__config
+++ b/include/__config
@@ -257,6 +257,10 @@
# endif // defined(__GLIBC_PREREQ)
#endif // defined(__linux__)
+#if defined(__MVS__)
+# include <features.h> // for __NATIVE_ASCII_F
+#endif
+
#ifdef __LITTLE_ENDIAN__
# if __LITTLE_ENDIAN__
# define _LIBCPP_LITTLE_ENDIAN
@@ -1220,8 +1224,8 @@
#endif
#if defined(__BIONIC__) || defined(__NuttX__) || \
- defined(__Fuchsia__) || defined(__wasi__) || defined(_LIBCPP_HAS_MUSL_LIBC) || \
- defined(__MVS__) || defined(__OpenBSD__)
+ defined(__Fuchsia__) || defined(__wasi__) || \
+ defined(_LIBCPP_HAS_MUSL_LIBC) || defined(__OpenBSD__)
#define _LIBCPP_PROVIDES_DEFAULT_RUNE_TABLE
#endif
diff --git a/include/__locale b/include/__locale
index 3bddbc8..6181f25 100644
--- a/include/__locale
+++ b/include/__locale
@@ -511,6 +511,33 @@
# define _LIBCPP_CTYPE_MASK_IS_COMPOSITE_PRINT
# define _LIBCPP_CTYPE_MASK_IS_COMPOSITE_ALPHA
# define _LIBCPP_CTYPE_MASK_IS_COMPOSITE_XDIGIT
+#elif defined(__MVS__)
+ static const mask __regex_word = 0x8000;
+# if defined(__NATIVE_ASCII_F)`
+ typedef unsigned int mask;
+ static const mask space = _ISSPACE_A;
+ static const mask print = _ISPRINT_A;
+ static const mask cntrl = _ISCNTRL_A;
+ static const mask upper = _ISUPPER_A;
+ static const mask lower = _ISLOWER_A;
+ static const mask alpha = _ISALPHA_A;
+ static const mask digit = _ISDIGIT_A;
+ static const mask punct = _ISPUNCT_A;
+ static const mask xdigit = _ISXDIGIT_A;
+ static const mask blank = _ISBLANK_A;
+# else
+ typedef unsigned short mask;
+ static const mask space = __ISSPACE;
+ static const mask print = __ISPRINT;
+ static const mask cntrl = __ISCNTRL;
+ static const mask upper = __ISUPPER;
+ static const mask lower = __ISLOWER;
+ static const mask alpha = __ISALPHA;
+ static const mask digit = __ISDIGIT;
+ static const mask punct = __ISPUNCT;
+ static const mask xdigit = __ISXDIGIT;
+ static const mask blank = __ISBLANK;
+# endif
#else
# error unknown rune table for this platform -- do you mean to define _LIBCPP_PROVIDES_DEFAULT_RUNE_TABLE?
#endif
@@ -734,6 +761,10 @@
static const short* __classic_upper_table() _NOEXCEPT;
static const short* __classic_lower_table() _NOEXCEPT;
#endif
+#if defined(__MVS__)
+ static const unsigned short* __classic_upper_table() _NOEXCEPT;
+ static const unsigned short* __classic_lower_table() _NOEXCEPT;
+#endif
protected:
~ctype();
diff --git a/include/regex b/include/regex
index 8203c81..59b2c9a 100644
--- a/include/regex
+++ b/include/regex
@@ -1310,19 +1310,51 @@
return (__c == '_' && (__m & __regex_word));
}
+inline _LIBCPP_INLINE_VISIBILITY
+bool __is_07(unsigned char c)
+{
+ return (c & 0xF8u) ==
+#if defined(__MVS__) && !defined(__NATIVE_ASCII_F)
+ 0xF0;
+#else
+ 0x30;
+#endif
+}
+
+inline _LIBCPP_INLINE_VISIBILITY
+bool __is_89(unsigned char c)
+{
+ return (c & 0xFEu) ==
+#if defined(__MVS__) && !defined(__NATIVE_ASCII_F)
+ 0xF8;
+#else
+ 0x38;
+#endif
+}
+
+inline _LIBCPP_INLINE_VISIBILITY
+unsigned char __to_lower(unsigned char c)
+{
+#if defined(__MVS__) && !defined(__NATIVE_ASCII_F)
+ return c & 0xBF;
+#else
+ return c | 0x20;
+#endif
+}
+
template <class _CharT>
int
regex_traits<_CharT>::__regex_traits_value(unsigned char __ch, int __radix)
{
- if ((__ch & 0xF8u) == 0x30) // '0' <= __ch && __ch <= '7'
+ if (__is_07(__ch)) // '0' <= __ch && __ch <= '7'
return __ch - '0';
if (__radix != 8)
{
- if ((__ch & 0xFEu) == 0x38) // '8' <= __ch && __ch <= '9'
+ if (__is_89(__ch)) // '8' <= __ch && __ch <= '9'
return __ch - '0';
if (__radix == 16)
{
- __ch |= 0x20; // tolower
+ __ch = __to_lower(__ch); // tolower
if ('a' <= __ch && __ch <= 'f')
return __ch - ('a' - 10);
}