Update: add UTF-8 BOM detection

Add support for detecting UTF-8 BOM in the f_utf_is_space() function.
Add new function f_utf_is_bom().

diff --git a/level_0/f_utf/c/utf.c b/level_0/f_utf/c/utf.c
index b07f574ccf3a39685f5c21339d363c0166c8ca4f..85a803fe3fa6fa727046c4b51bc47150b4788827 100644 (file)
--- a/level_0/f_utf/c/utf.c
+++ b/level_0/f_utf/c/utf.c
@@ -4,6 +4,32 @@
 extern "C" {
 #endif
 
+#ifndef _di_f_utf_is_bom_
+  f_return_status f_utf_is_bom(const f_string character, const f_u_short maxWidth) {
+    #ifndef _di_level_0_parameter_checking_
+      if (maxWidth < 1) return f_error_set_error(f_invalid_parameter);
+    #endif // _di_level_0_parameter_checking_
+
+    f_u_short width = f_macro_utf_byte_width(*character);
+
+    if (width == 1) {
+      return f_false;
+    }
+
+    if (width > maxWidth) {
+      return f_error_set_error(f_maybe);
+    }
+
+    if (width == 3) {
+      if (!memcmp(character, f_utf_bom, width)) {
+        return f_true;
+      }
+    }
+
+    return f_false;
+  }
+#endif // _di_f_utf_is_bom_
+
 #ifndef _di_f_utf_is_space_
   f_return_status f_utf_is_space(const f_string character, const f_u_short maxWidth) {
     #ifndef _di_level_0_parameter_checking_
@@ -129,6 +155,10 @@ extern "C" {
         return f_true;
       }
 
+      if (!memcmp(character, f_utf_bom, width)) {
+        return f_true;
+      }
+
       return f_false;
     }
 

diff --git a/level_0/f_utf/c/utf.h b/level_0/f_utf/c/utf.h
index 2965bc0345e6cebbdd583a9290c0b0863ca20bae..aee00cc4c294fc446ec987a9cfc5d3725d25fdcb 100644 (file)
--- a/level_0/f_utf/c/utf.h
+++ b/level_0/f_utf/c/utf.h
@@ -183,8 +183,30 @@ extern "C" {
 #endif // _di_f_utf_substitute_
 
 /**
+ * Check to see if the entire byte block of the character is a UTF-8 BOM.
+ *
+ * @param character
+ *   The character to validate.
+ *   There must be enough space allocated to compare against, as limited by maxWidth.
+ * @param maxWidth
+ *   The maximum width available for checking.
+ *   Can be anything greater than 0.
+ *
+ * @return
+ *   f_true if a UTF-8 whitespace or substitute.
+ *   f_false if not a UTF-8 whitespace or substitute.
+ *   f_maybe (with error bit) if this could be a whitespace or substitute but width is not long enough.
+ *   f_invalid_parameter (with error bit) if a parameter is invalid.
+ */
+#ifndef _di_f_utf_is_bom_
+  extern f_return_status f_utf_is_bom(const f_string character, const f_u_short maxWidth);
+#endif // _di_f_utf_is_bom_
+
+/**
  * Check to see if the entire byte block of the character is a UTF-8 whitespace or substitute character.
  *
+ * This will also return TRUE for the UTF-8 BOM.
+ *
  * This does not check non-UTF-8 whitespace.
  *
  * @param character
@@ -195,7 +217,7 @@ extern "C" {
  *   Can be anything greater than 0.
  *
  * @return
- *   f_true if a UTF-8 whitespace or substitute.
+ *   f_true if a UTF-8 whitespace, substitute, or UTF-8 BOM.
  *   f_false if not a UTF-8 whitespace or substitute.
  *   f_maybe (with error bit) if this could be a whitespace or substitute but width is not long enough.
  *   f_invalid_parameter (with error bit) if a parameter is invalid.