summaryrefslogtreecommitdiff
path: root/libguile/symbols.c
diff options
context:
space:
mode:
authorMichael Gran <spk121@yahoo.com>2009-08-21 08:57:35 -0700
committerMichael Gran <spk121@yahoo.com>2009-08-21 08:57:35 -0700
commite23106d53eb03d7cb4962282396269176ea7482e (patch)
treed7736a8f7f16af0c73365f688f75561aac81c5eb /libguile/symbols.c
parent90305ce9e429f0381ff79427e71287fdafd4d201 (diff)
downloadguile-e23106d53eb03d7cb4962282396269176ea7482e.tar.gz
Add initial support for wide symbols
* libguile/hash.c (scm_i_string_hash): new function (scm_hasher): don't unpack string: use scm_i_string_hash * libguile/hash.h: new declaration for scm_i_string_hash * libguile/print.c (quote_keywordish_symbol): use symbol accessors (scm_i_print_symbol_name): new function (scm_print_symbol_name): call scm_i_print_symbol_name (iprin1): use scm_i_print_symbol_name to print symbols * libguile/print.h: new declaration for scm_i_print_symbol_name * libguile/symbols.c (lookup_interned_symbol): now takes scheme string instead of c string; callers changed (lookup_interned_symbol): add wide symbol support (scm_i_c_mem2symbol): removed (scm_i_mem2symbol): removed and replaced with scm_i_str2symbol (scm_i_str2symbol): new function (scm_i_mem2uninterned_symbol): removed and replaced with scm_i_str2uninterned_symbol (scm_i_str2uninterned_symbol): new function (scm_make_symbol, scm_string_to_symbol, scm_from_locale_symbol) (scm_from_locale_symboln): use scm_i_str2symbol * test-suite/tests/symbols.test: new tests
Diffstat (limited to 'libguile/symbols.c')
-rw-r--r--libguile/symbols.c108
1 files changed, 41 insertions, 67 deletions
diff --git a/libguile/symbols.c b/libguile/symbols.c
index c0ba2a8b4..a9320163a 100644
--- a/libguile/symbols.c
+++ b/libguile/symbols.c
@@ -89,11 +89,11 @@ scm_i_hash_symbol (SCM obj, unsigned long n, void *closure)
}
static SCM
-lookup_interned_symbol (const char *name, size_t len,
- unsigned long raw_hash)
+lookup_interned_symbol (SCM name, unsigned long raw_hash)
{
/* Try to find the symbol in the symbols table */
SCM l;
+ size_t len = scm_i_string_length (name);
unsigned long hash = raw_hash % SCM_HASHTABLE_N_BUCKETS (symbols);
for (l = SCM_HASHTABLE_BUCKET (symbols, hash);
@@ -104,15 +104,32 @@ lookup_interned_symbol (const char *name, size_t len,
if (scm_i_symbol_hash (sym) == raw_hash
&& scm_i_symbol_length (sym) == len)
{
- const char *chrs = scm_i_symbol_chars (sym);
- size_t i = len;
-
- while (i != 0)
- {
- --i;
- if (name[i] != chrs[i])
- goto next_symbol;
- }
+ size_t i = len;
+
+ /* Slightly faster path for comparing narrow to narrow. */
+ if (scm_i_is_narrow_string (name) && scm_i_is_narrow_symbol (sym))
+ {
+ const char *chrs = scm_i_symbol_chars (sym);
+ const char *str = scm_i_string_chars (name);
+
+ while (i != 0)
+ {
+ --i;
+ if (str[i] != chrs[i])
+ goto next_symbol;
+ }
+ }
+ else
+ {
+ /* Somewhat slower path for comparing narrow to wide or
+ wide to wide. */
+ while (i != 0)
+ {
+ --i;
+ if (scm_i_string_ref (name, i) != scm_i_symbol_ref (sym, i))
+ goto next_symbol;
+ }
+ }
return sym;
}
@@ -142,32 +159,12 @@ intern_symbol (SCM symbol)
}
static SCM
-scm_i_c_mem2symbol (const char *name, size_t len)
-{
- SCM symbol;
- size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
-
- symbol = lookup_interned_symbol (name, len, raw_hash);
- if (scm_is_false (symbol))
- {
- /* The symbol was not found, create it. */
- symbol = scm_i_c_make_symbol (name, len, 0, raw_hash,
- scm_cons (SCM_BOOL_F, SCM_EOL));
- intern_symbol (symbol);
- }
-
- return symbol;
-}
-
-static SCM
-scm_i_mem2symbol (SCM str)
+scm_i_str2symbol (SCM str)
{
SCM symbol;
- const char *name = scm_i_string_chars (str);
- size_t len = scm_i_string_length (str);
- size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
+ size_t raw_hash = scm_i_string_hash (str);
- symbol = lookup_interned_symbol (name, len, raw_hash);
+ symbol = lookup_interned_symbol (str, raw_hash);
if (scm_is_false (symbol))
{
/* The symbol was not found, create it. */
@@ -181,11 +178,9 @@ scm_i_mem2symbol (SCM str)
static SCM
-scm_i_mem2uninterned_symbol (SCM str)
+scm_i_str2uninterned_symbol (SCM str)
{
- const char *name = scm_i_string_chars (str);
- size_t len = scm_i_string_length (str);
- size_t raw_hash = scm_string_hash ((const unsigned char *) name, len);
+ size_t raw_hash = scm_i_string_hash (str);
return scm_i_make_symbol (str, SCM_I_F_SYMBOL_UNINTERNED,
raw_hash, scm_cons (SCM_BOOL_F, SCM_EOL));
@@ -220,7 +215,7 @@ SCM_DEFINE (scm_make_symbol, "make-symbol", 1, 0, 0,
#define FUNC_NAME s_scm_make_symbol
{
SCM_VALIDATE_STRING (1, name);
- return scm_i_mem2uninterned_symbol (name);
+ return scm_i_str2uninterned_symbol (name);
}
#undef FUNC_NAME
@@ -282,7 +277,7 @@ SCM_DEFINE (scm_string_to_symbol, "string->symbol", 1, 0, 0,
#define FUNC_NAME s_scm_string_to_symbol
{
SCM_VALIDATE_STRING (1, string);
- return scm_i_mem2symbol (string);
+ return scm_i_str2symbol (string);
}
#undef FUNC_NAME
@@ -389,44 +384,23 @@ SCM_DEFINE (scm_symbol_pset_x, "symbol-pset!", 2, 0, 0,
SCM
scm_from_locale_symbol (const char *sym)
{
- return scm_i_c_mem2symbol (sym, strlen (sym));
+ return scm_from_locale_symboln (sym, -1);
}
SCM
scm_from_locale_symboln (const char *sym, size_t len)
{
- return scm_i_c_mem2symbol (sym, len);
+ SCM str = scm_from_locale_stringn (sym, len);
+ return scm_i_str2symbol (str);
}
SCM
scm_take_locale_symboln (char *sym, size_t len)
{
- SCM res;
- unsigned long raw_hash;
-
- if (len == (size_t)-1)
- len = strlen (sym);
- else
- {
- /* Ensure STR is null terminated. A realloc for 1 extra byte should
- often be satisfied from the alignment padding after the block, with
- no actual data movement. */
- sym = scm_realloc (sym, len+1);
- sym[len] = '\0';
- }
-
- raw_hash = scm_string_hash ((unsigned char *)sym, len);
- res = lookup_interned_symbol (sym, len, raw_hash);
- if (scm_is_false (res))
- {
- res = scm_i_c_take_symbol (sym, len, 0, raw_hash,
- scm_cons (SCM_BOOL_F, SCM_EOL));
- intern_symbol (res);
- }
- else
- free (sym);
+ SCM str;
- return res;
+ str = scm_take_locale_stringn (sym, len);
+ return scm_i_str2symbol (str);
}
SCM