On Tue, Aug 26, 2008 at 09:32:32AM +0200, Bruno Haible wrote:
> Alexander V. Lukyanov wrote:
> > Let's measure it.
> >
> > $ time ./wcwidth-solaris
> > wcwidth(0x2022)=2
> >
> > real 0m2.205s
> > user 0m2.200s
> > sys 0m0.000s
> >
> > $ time ./wcwidth-rpl
> > wcwidth(0x2022)=1
> >
> > real 0m55.477s
> > user 0m55.350s
> > sys 0m0.000s
> >
> > $ time ./wcwidth-mk
> > wcwidth(0x2022)=1
> >
> > real 0m1.944s
> > user 0m1.940s
> > sys 0m0.010s
>
> This is not a fair comparison: wcwidth-mk works only in UTF-8 locales,
> whereas wcwidth() from the system and from gnulib return the right result
> in all locales. The test whether the locale encoding is UTF-8 is precisely
> what takes up most time in the gnulib replacement.
Ok. Would you accept this patch, which caches charset properties?
BTW, it also fixes a bug - cjk case could not ever be executed before.
$ time ./wcwidth-rpl1
wcwidth(0x2022)=1
real 0m4.160s
user 0m4.150s
sys 0m0.010s
Worse than Solaris native wcwidth (nl_langinfo+strncmp costs), but still 13
times faster than before.
--
Alexander.
diff --git a/lib/uniwidth.h b/lib/uniwidth.h
index cdc7d96..3979187 100644
--- a/lib/uniwidth.h
+++ b/lib/uniwidth.h
@@ -37,25 +37,7 @@ extern "C" {
/* Determine number of column positions required for UC. */
extern int
- uc_width (ucs4_t uc, const char *encoding);
-
-/* Determine number of column positions required for first N units
- (or fewer if S ends before this) in S. */
-extern int
- u8_width (const uint8_t *s, size_t n, const char *encoding);
-extern int
- u16_width (const uint16_t *s, size_t n, const char *encoding);
-extern int
- u32_width (const uint32_t *s, size_t n, const char *encoding);
-
-/* Determine number of column positions required for S. */
-extern int
- u8_strwidth (const uint8_t *s, const char *encoding);
-extern int
- u16_strwidth (const uint16_t *s, const char *encoding);
-extern int
- u32_strwidth (const uint32_t *s, const char *encoding);
-
+ uc_width (ucs4_t uc, int is_cjk_encoding);
#ifdef __cplusplus
}
diff --git a/lib/uniwidth/width.c b/lib/uniwidth/width.c
index 4161c26..e43d789 100644
--- a/lib/uniwidth/width.c
+++ b/lib/uniwidth/width.c
@@ -20,8 +20,6 @@
/* Specification. */
#include "uniwidth.h"
-#include "cjk.h"
-
/*
* Non-spacing attribute table.
* Consists of:
@@ -267,7 +265,7 @@ static const signed char nonspacing_table_ind[240] = {
/* Determine number of column positions required for UC. */
int
-uc_width (ucs4_t uc, const char *encoding)
+uc_width (ucs4_t uc, int is_cjk_encoding)
{
/* Test for non-spacing or control character. */
if ((uc >> 9) < 240)
@@ -317,7 +315,7 @@ uc_width (ucs4_t uc, const char *encoding)
/* In ancient CJK encodings, Cyrillic and most other characters are
double-width as well. */
if (uc >= 0x00A1 && uc < 0xFF61 && uc != 0x20A9
- && is_cjk_encoding (encoding))
+ && is_cjk_encoding)
return 2;
return 1;
}
diff --git a/lib/wcwidth.c b/lib/wcwidth.c
index 4885071..c2de83b 100644
--- a/lib/wcwidth.c
+++ b/lib/wcwidth.c
@@ -22,22 +22,54 @@
/* Get iswprint. */
#include <wctype.h>
+#if HAVE_LANGINFO_CODESET
+# include <langinfo.h>
+#endif
+
#include "localcharset.h"
#include "streq.h"
#include "uniwidth.h"
#undef wcwidth
+#include "uniwidth/cjk.h"
+
+static char cached_encoding[32];
+static int cached_is_cjk_encoding;
+static int cached_is_utf8_encoding;
+
+static const char *locale_charset_simple ()
+{
+#if HAVE_LANGINFO_CODESET
+ /* Most systems support nl_langinfo (CODESET) nowadays. */
+ return nl_langinfo (CODESET);
+# else
+ /* Do the complex case */
+ return locale_charset ();
+# endif
+}
+
+static void cache_encoding ()
+{
+ const char *encoding = locale_charset_simple ();
+ if (!strncmp(encoding, cached_encoding, sizeof (cached_encoding)))
+ return;
+ strncpy (cached_encoding, encoding, sizeof (cached_encoding));
+ encoding = locale_charset ();
+ cached_is_utf8_encoding = STREQ (encoding, "UTF-8", 'U', 'T', 'F', '-', '8',
0, 0, 0 ,0);
+ cached_is_cjk_encoding = is_cjk_encoding (encoding);
+}
+
int
rpl_wcwidth (wchar_t wc)
{
+ cache_encoding ();
/* In UTF-8 locales, use a Unicode aware width function. */
- const char *encoding = locale_charset ();
- if (STREQ (encoding, "UTF-8", 'U', 'T', 'F', '-', '8', 0, 0, 0 ,0))
+ if (cached_is_utf8_encoding || cached_is_cjk_encoding)
{
/* We assume that in a UTF-8 locale, a wide character is the same as a
Unicode character. */
- return uc_width (wc, encoding);
+ return uc_width (wc, cached_is_cjk_encoding);
}
else
{