Removed cstr iter; => csview iter is now utf8 iter. See utf8.h example.

3 files changed, 23 insertions, 24 deletions

diff --git a/include/stc/cstr.h b/include/stc/cstr.h
index 7ba20d10..04592b69 100644
--- a/include/stc/cstr.h
+++ b/include/stc/cstr.h
@@ -31,7 +31,6 @@
 #include <ctype.h>

 

 typedef                 struct cstr { char* str; } cstr;

-typedef                 struct cstr_iter { char *ref; } cstr_iter;

 typedef                 char cstr_value;

 #define                 cstr_npos (SIZE_MAX >> 1)

 

@@ -110,11 +109,6 @@ STC_INLINE void         cstr_erase(cstr* self, const size_t pos)
 STC_INLINE char*        cstr_front(cstr* self) { return self->str; }

 STC_INLINE char*        cstr_back(cstr* self)

                             { return self->str + _cstr_rep(self)->size - 1; }

-STC_INLINE cstr_iter    cstr_begin(cstr* self)

-                            { return c_make(cstr_iter){self->str}; }

-STC_INLINE cstr_iter    cstr_end(cstr* self)

-                            { return c_make(cstr_iter){self->str + _cstr_rep(self)->size}; }

-STC_INLINE void         cstr_next(cstr_iter* it) {++it->ref; }

 STC_INLINE bool         cstr_equals(cstr s, const char* str)

                             { return strcmp(s.str, str) == 0; }

 STC_INLINE bool         cstr_equals_s(cstr s1, cstr s2)

diff --git a/include/stc/csview.h b/include/stc/csview.h
index 58a7f4f1..d8bb567e 100644
--- a/include/stc/csview.h
+++ b/include/stc/csview.h
@@ -27,7 +27,7 @@
 #include "utf8.h"

 

 typedef                 struct csview { const char* str; size_t size; } csview;

-typedef                 struct csview_iter { const char *ref; } csview_iter;

+typedef                 union csview_iter { const char *ref; csview cp; } csview_iter;

 typedef                 char csview_value;

 

 #define                 csview_null  c_make(csview){"", 0}

@@ -69,10 +69,11 @@ STC_INLINE bool         csview_ends_with(csview sv, csview sub)
                             { if (sub.size > sv.size) return false;

                               return !memcmp(sv.str + sv.size - sub.size, sub.str, sub.size); }

 STC_INLINE csview_iter  csview_begin(const csview* self)

-                            { return c_make(csview_iter){self->str}; }

+                            { return c_make(csview_iter){.cp = {self->str, utf8_codepoint_size(*self->str)}}; }

 STC_INLINE csview_iter  csview_end(const csview* self)

                             { return c_make(csview_iter){self->str + self->size}; }

-STC_INLINE void         csview_next(csview_iter* it) { ++it->ref; }

+STC_INLINE void         csview_next(csview_iter* it)

+                            { it->ref += it->cp.size; it->cp.size = utf8_codepoint_size(*it->ref); }

 

 /* utf8 */

 STC_INLINE bool csview_valid_utf8(csview sv)

diff --git a/include/stc/utf8.h b/include/stc/utf8.h
index 9a65906d..a064b906 100644
--- a/include/stc/utf8.h
+++ b/include/stc/utf8.h
@@ -8,13 +8,18 @@
 int main()

 {

     c_auto (cstr, s1) {

-        s1 = cstr_new("hell😀 world");

-        cstr_replace_v(&s1, utf8_substr(s1.str, 4, 1), c_sv("x"));

+        s1 = cstr_new("hell😀 w😀rld");

+        cstr_replace_v(&s1, utf8_substr(s1.str, 7, 1), c_sv("x"));

         printf("%s\n", s1.str);

+

+        csview sv = csview_from_s(s1);

+        c_foreach (i, csview, sv)

+            printf(c_PRIsv ",", c_ARGsv(i.cp));

     }

 }

 // Output:

-// Hellx world

+// hell😀 wxrld

+// h,e,l,l,😀, ,w,x,r,l,d,

 */

 #include "ccommon.h"

 #include <ctype.h>

@@ -32,20 +37,21 @@ STC_API const char* utf8_at(const char *s, size_t index);
 /* decode next utf8 codepoint. */

 STC_API uint32_t utf8_decode(uint32_t *state, uint32_t *codep, const uint32_t byte);

 

-STC_INLINE bool utf8_valid(const char* str)

-    { return utf8_size(str) != SIZE_MAX; }

+STC_INLINE size_t utf8_pos(const char* s, size_t index) 

+    { return utf8_at(s, index) - s; }

 

-STC_INLINE uint32_t utf8_peek(const char *s)

-{

+STC_INLINE bool utf8_valid(const char* s)

+    { return utf8_size(s) != SIZE_MAX; }

+

+STC_INLINE uint32_t utf8_peek(const char *s) {

     uint32_t state = 0, codepoint;

-    utf8_decode(&state, &codepoint, (uint8_t)s[0]);

+    utf8_decode(&state, &codepoint, (uint8_t)*s);

     return codepoint;

 }

 

-STC_INLINE int utf8_codepoint_size(char c)

-{

+STC_INLINE size_t utf8_codepoint_size(char c) {

     uint8_t u = (uint8_t)c;

-    int ret = (u & 0xF0) == 0xE0;

+    size_t ret = (u & 0xF0) == 0xE0;

     ret += ret << 1;                       // 3

     ret |= u < 0x80;                       // 1

     ret |= ((0xC1 < u) & (u < 0xE0)) << 1; // 2

@@ -53,9 +59,8 @@ STC_INLINE int utf8_codepoint_size(char c)
     return ret;

 }

 

-STC_INLINE const char *utf8_next(const char *s)

-{

-    const char* t = s + utf8_codepoint_size(s[0]);

+STC_INLINE const char *utf8_next(const char *s) {

+    const char* t = s + utf8_codepoint_size(*s);

     

     uintptr_t p = (uintptr_t)t;

     p &= (uintptr_t) -(*s != 0);

@@ -128,7 +133,6 @@ STC_DEF const char* utf8_at(const char *s, size_t index)
     return s;

 }

 

-

 #endif

 #endif

 #undef i_opt