From 8586e44de92c827d6a19a7700121c8b21d3687b1 Mon Sep 17 00:00:00 2001
From: default <nobody@localhost>
Date: Fri, 23 Aug 2024 17:22:10 +0200
Subject: [PATCH] Some optimizations.

---
 xs_json.h    |   7 +++-
 xs_unicode.h | 116 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 xs_version.h |   2 +-
 3 files changed, 120 insertions(+), 5 deletions(-)

diff --git a/xs_json.h b/xs_json.h
index a4112b0..3a91de9 100644
--- a/xs_json.h
+++ b/xs_json.h
@@ -208,6 +208,7 @@ static xs_val *_xs_json_load_lexer(FILE *f, js_type *t)
 {
     int c;
     xs_val *v = NULL;
+    int offset;
 
     *t = JS_ERROR;
 
@@ -236,6 +237,7 @@ static xs_val *_xs_json_load_lexer(FILE *f, js_type *t)
         *t = JS_STRING;
 
         v = xs_str_new(NULL);
+        offset = 0;
 
         while ((c = fgetc(f)) != '"' && c != EOF && *t != JS_ERROR) {
             if (c == '\\') {
@@ -274,11 +276,12 @@ static xs_val *_xs_json_load_lexer(FILE *f, js_type *t)
                     break;
                 }
 
-                v = xs_utf8_cat(v, cp);
+                v = xs_utf8_insert(v, cp, &offset);
             }
             else {
                 char cc = c;
-                v = xs_append_m(v, &cc, 1);
+                v = xs_insert_m(v, offset, &cc, 1);
+                offset++;
             }
         }
 
diff --git a/xs_unicode.h b/xs_unicode.h
index 2e9a754..a5a1dcb 100644
--- a/xs_unicode.h
+++ b/xs_unicode.h
@@ -9,6 +9,7 @@
  unsigned int xs_utf8_dec(const char **str);
  int xs_unicode_width(unsigned int cpoint);
  int xs_is_surrogate(unsigned int cpoint);
+ int xs_is_diacritic(unsigned int cpoint);
  unsigned int xs_surrogate_dec(unsigned int p1, unsigned int p2);
  unsigned int xs_surrogate_enc(unsigned int cpoint);
  unsigned int *_xs_unicode_upper_search(unsigned int cpoint);
@@ -22,7 +23,12 @@
  int xs_unicode_is_alpha(unsigned int cpoint);
 
 #ifdef _XS_H
+ xs_str *xs_utf8_insert(xs_str *str, unsigned int cpoint, int *offset);
  xs_str *xs_utf8_cat(xs_str *str, unsigned int cpoint);
+ xs_str *xs_utf8_to_upper(const char *str);
+ xs_str *xs_utf8_to_lower(const char *str);
+ xs_str *xs_utf8_to_nfd(const char *str);
+ xs_str *xs_utf8_to_nfc(const char *str);
 #endif
 
 #ifdef XS_IMPLEMENTATION
@@ -144,6 +150,12 @@ int xs_unicode_width(unsigned int cpoint)
 }
 
 
+int xs_is_diacritic(unsigned int cpoint)
+{
+    return cpoint >= 0x300 && cpoint <= 0x36f;
+}
+
+
 /** surrogate pairs **/
 
 int xs_is_surrogate(unsigned int cpoint)
@@ -172,14 +184,27 @@ unsigned int xs_surrogate_enc(unsigned int cpoint)
 
 #ifdef _XS_H
 
-xs_str *xs_utf8_cat(xs_str *str, unsigned int cpoint)
+xs_str *xs_utf8_insert(xs_str *str, unsigned int cpoint, int *offset)
 /* encodes an Unicode codepoint to utf-8 into str */
 {
     char tmp[4];
 
     int c = xs_utf8_enc(tmp, cpoint);
 
-    return xs_append_m(str, tmp, c);
+    str = xs_insert_m(str, *offset, tmp, c);
+
+    *offset += c;
+
+    return str;
+}
+
+
+xs_str *xs_utf8_cat(xs_str *str, unsigned int cpoint)
+/* encodes an Unicode codepoint to utf-8 into str */
+{
+    int offset = strlen(str);
+
+    return xs_utf8_insert(str, cpoint, &offset);
 }
 
 #endif /* _XS_H */
@@ -232,6 +257,9 @@ unsigned int *_xs_unicode_lower_search(unsigned int cpoint)
 unsigned int xs_unicode_to_lower(unsigned int cpoint)
 /* returns the cpoint to lowercase */
 {
+    if (cpoint < 0x80)
+        return tolower(cpoint);
+
     unsigned int *p = _xs_unicode_upper_search(cpoint);
 
     return p == NULL ? cpoint : p[1];
@@ -241,6 +269,9 @@ unsigned int xs_unicode_to_lower(unsigned int cpoint)
 unsigned int xs_unicode_to_upper(unsigned int cpoint)
 /* returns the cpoint to uppercase */
 {
+    if (cpoint < 0x80)
+        return toupper(cpoint);
+
     unsigned int *p = _xs_unicode_lower_search(cpoint);
 
     return p == NULL ? cpoint : p[0];
@@ -317,6 +348,87 @@ int xs_unicode_is_alpha(unsigned int cpoint)
 }
 
 
+#ifdef _XS_H
+
+xs_str *xs_utf8_to_upper(const char *str)
+{
+    xs_str *s = xs_str_new(NULL);
+    unsigned int cpoint;
+    int offset = 0;
+
+    while ((cpoint = xs_utf8_dec(&str))) {
+        cpoint = xs_unicode_to_upper(cpoint);
+        s = xs_utf8_insert(s, cpoint, &offset);
+    }
+
+    return s;
+}
+
+
+xs_str *xs_utf8_to_lower(const char *str)
+{
+    xs_str *s = xs_str_new(NULL);
+    unsigned int cpoint;
+    int offset = 0;
+
+    while ((cpoint = xs_utf8_dec(&str))) {
+        cpoint = xs_unicode_to_lower(cpoint);
+        s = xs_utf8_insert(s, cpoint, &offset);
+    }
+
+    return s;
+}
+
+
+xs_str *xs_utf8_to_nfd(const char *str)
+{
+    xs_str *s = xs_str_new(NULL);
+    unsigned int cpoint;
+    int offset = 0;
+
+    while ((cpoint = xs_utf8_dec(&str))) {
+        unsigned int base;
+        unsigned int diac;
+
+        if (xs_unicode_nfd(cpoint, &base, &diac)) {
+            s = xs_utf8_insert(s, base, &offset);
+            s = xs_utf8_insert(s, diac, &offset);
+        }
+        else
+            s = xs_utf8_insert(s, cpoint, &offset);
+    }
+
+    return s;
+}
+
+
+xs_str *xs_utf8_to_nfc(const char *str)
+{
+    xs_str *s = xs_str_new(NULL);
+    unsigned int cpoint;
+    unsigned int base = 0;
+    int offset = 0;
+
+    while ((cpoint = xs_utf8_dec(&str))) {
+        if (xs_is_diacritic(cpoint)) {
+            if (xs_unicode_nfc(base, cpoint, &base))
+                continue;
+        }
+
+        if (base)
+            s = xs_utf8_insert(s, base, &offset);
+
+        base = cpoint;
+    }
+
+    if (base)
+        s = xs_utf8_insert(s, base, &offset);
+
+    return s;
+}
+
+#endif /* _XS_H */
+
 #endif /* _XS_UNICODE_TBL_H */
 
 #endif /* XS_IMPLEMENTATION */
diff --git a/xs_version.h b/xs_version.h
index 4318c7e..ce88558 100644
--- a/xs_version.h
+++ b/xs_version.h
@@ -1 +1 @@
-/* c6eca9593f9b3d6791cba600e5950f682fdb36cf 2024-08-12T16:08:37+02:00 */
+/* cc9ebd36ae640e4701277327fbba9996143076f6 2024-08-23T17:17:08+02:00 */