implement mo file string lookup for translations

the core is based on a binary search; hash table is not used. both
native and reverse-endian mo files are supported. all offsets read
from the mapped mo file are checked against the mapping size to
prevent the possibility of reads outside the mapping.

this commit has no observable effects since there are not yet any
callers to the message translation code.

diff --git a/src/internal/locale_impl.h b/src/internal/locale_impl.h
index 0ee72d3..9142f0c 100644 (file)
--- a/src/internal/locale_impl.h
+++ b/src/internal/locale_impl.h
@@ -13,6 +13,12 @@ struct __locale_map {
 };
 
 int __setlocalecat(locale_t, int, const char *);
+const char *__mo_lookup(const void *, size_t, const char *);
+const char *__lctrans(const char *, const struct __locale_map *);
+const char *__lctrans_cur(const char *);
+
+#define LCTRANS(msg, lc, loc) __lctrans(msg, (loc)->cat[(lc)-2])
+#define LCTRANS_CUR(msg) __lctrans_cur(msg)
 
 #define CURRENT_LOCALE \
        (libc.uselocale_cnt ? __pthread_self()->locale : &libc.global_locale)

diff --git a/src/locale/__lctrans.c b/src/locale/__lctrans.c
new file mode 100644 (file)
index 0000000..2769c08
--- /dev/null
+++ b/src/locale/__lctrans.c
@@ -0,0 +1,20 @@
+#include <locale.h>
+#include "locale_impl.h"
+#include "libc.h"
+
+const char *dummy(const char *msg, const struct __locale_map *lm)
+{
+       return msg;
+}
+
+weak_alias(dummy, __lctrans_impl);
+
+const char *__lctrans(const char *msg, const struct __locale_map *lm)
+{
+       return __lctrans_impl(msg, lm);
+}
+
+const char *__lctrans_cur(const char *msg)
+{
+       return __lctrans_impl(msg, CURRENT_LOCALE->cat[LC_MESSAGES-2]);
+}

diff --git a/src/locale/__mo_lookup.c b/src/locale/__mo_lookup.c
new file mode 100644 (file)
index 0000000..8112d91
--- /dev/null
+++ b/src/locale/__mo_lookup.c
@@ -0,0 +1,38 @@
+#include <stdint.h>
+#include <string.h>
+
+static inline uint32_t swapc(uint32_t x, int c)
+{
+       return c ? x>>24 | x>>8&0xff00 | x<<8&0xff0000 | x<<24 : x;
+}
+
+const char *__mo_lookup(const void *p, size_t size, const char *s)
+{
+       const uint32_t *mo = p;
+       int sw = *mo - 0x950412de;
+       uint32_t b = 0, n = swapc(mo[2], sw);
+       uint32_t o = swapc(mo[3], sw);
+       uint32_t t = swapc(mo[4], sw);
+       if (n>=size/4 || o>=size-4*n || t>=size-4*n || ((o|t)%4))
+               return 0;
+       o/=4;
+       t/=4;
+       for (;;) {
+               uint32_t os = swapc(mo[o+2*(b+n/2)+1], sw);
+               if (os >= size) return 0;
+               int sign = strcmp(s, (char *)p + os);
+               if (!sign) {
+                       uint32_t ts = swapc(mo[t+2*(b+n/2)+1], sw);
+                       if (ts >= size) return 0;
+                       return (char *)p + ts;
+               }
+               else if (n == 1) return 0;
+               else if (sign < 0)
+                       n /= 2;
+               else {
+                       b += n/2;
+                       n -= n/2;
+               }
+       }
+       return 0;
+}

diff --git a/src/locale/__setlocalecat.c b/src/locale/__setlocalecat.c
index bbecde4..44385e0 100644 (file)
--- a/src/locale/__setlocalecat.c
+++ b/src/locale/__setlocalecat.c
@@ -4,6 +4,13 @@
 #include "libc.h"
 #include "atomic.h"
 
+const char *__lctrans_impl(const char *msg, const struct __locale_map *lm)
+{
+       const char *trans = 0;
+       if (lm) trans = __mo_lookup(lm->map, lm->map_size, msg);
+       return trans ? trans : msg;
+}
+
 const unsigned char *__map_file(const char *, size_t *);
 int __munmap(void *, size_t);
 char *__strchrnul(const char *, int);