Update: improve IKI syntax rules.

This is a relaxation on the specification where previously only whitespace was required to exist before a valid variable name.

This allows punctuation to exist before the variable name.
Content such as 'A:"b"' may now be specified where A is the variable name (notice the single-quote before A which would previously prevent A from being recognized as a variable name).

diff --git a/level_0/f_iki/c/iki-common.h b/level_0/f_iki/c/iki-common.h
index 8f37731a636aaf9d471e16f3f1128775f5fe24c4..a07a1cf7d09ed70e4c9e545e2c2683c28fa45161 100644 (file)
--- a/level_0/f_iki/c/iki-common.h
+++ b/level_0/f_iki/c/iki-common.h
@@ -265,6 +265,7 @@ extern "C" {
  * status:    The return status to use.
  * buffer:    (A pointer) The buffer to seek through.
  * range:     (A pointer) The range within that buffer to seek through.
+ * width_max: The width_max variable to use fo calculating width_max.
  * condition: Set to TRUE to seek until whitespace is found and FALSE to seek until non-whitespace.
  */
 #ifndef _di_f_macro_iki_seek_whitespace_
@@ -283,6 +284,33 @@ extern "C" {
     }
 #endif // _di_f_macro_iki_seek_whitespace_
 
+/**
+ * Seek until a word, dash, or plus is found or not found.
+ *
+ * This will ignore the delimit placeholder.
+ *
+ * status:    The return status to use.
+ * buffer:    (A pointer) The buffer to seek through.
+ * range:     (A pointer) The range within that buffer to seek through.
+ * width_max: The width_max variable to use fo calculating width_max.
+ * condition: Set to TRUE to seek until a word character, dash character, or plus character is found and FALSE to seek until the opposite is found.
+ */
+#ifndef _di_f_macro_iki_seek_word_dash_plus_
+  #define f_macro_iki_seek_word_dash_plus(status, buffer, range, width_max, condition) \
+    while (range->start <= range->stop && range->start < buffer->used) { \
+      if (buffer->string[range->start] == f_iki_syntax_placeholder) { \
+        range->start++; \
+        continue; \
+      } \
+      f_macro_iki_determine_width_max(buffer, range, width_max); \
+      status = f_utf_is_word_dash_plus(buffer->string + range->start, width_max); \
+      if (status == condition) break; \
+      else if (F_status_is_error(status)) break; \
+      status = f_utf_buffer_increment(*buffer, range, 1); \
+      if (F_status_is_error(status)) break; \
+    }
+#endif // _di_f_macro_iki_seek_word_dash_plus_
+
 #ifdef __cplusplus
 } // extern "C"
 #endif

diff --git a/level_0/f_iki/c/iki.c b/level_0/f_iki/c/iki.c
index 6205aaed1d5c8027f5714e16f24c77ba17825329..698e5599c19663e62c4ee2bbe0d99db79297122a 100644 (file)
--- a/level_0/f_iki/c/iki.c
+++ b/level_0/f_iki/c/iki.c
@@ -26,27 +26,12 @@ extern "C" {
     }
 
     // skip past all initial non-word, non-dash, and non-plus.
-    while (range->start <= range->stop && range->start < buffer->used) {
-      if (buffer->string[range->start] == f_iki_syntax_placeholder) {
-        range->start++;
-        continue;
-      }
-
-      f_macro_iki_determine_width_max(buffer, range, width_max);
-
-      status = f_utf_is_word_dash_plus(buffer->string + range->start, width_max);
-      if (F_status_is_error(status)) return status;
-
-      if (status == F_true) break;
-
-      f_macro_iki_seek_whitespace(status, buffer, range, width_max, F_true);
-      if (F_status_is_error(status)) return status;
-
-      f_macro_iki_seek_whitespace(status, buffer, range, width_max, F_false);
-      if (F_status_is_error(status)) return status;
-    } // while
+    f_macro_iki_seek_word_dash_plus(status, buffer, range, width_max, F_true);
 
-    if (range->start > range->stop) {
+    if (F_status_is_error(status)) {
+      return status;
+    }
+    else if (range->start > range->stop) {
       return F_data_not_stop;
     }
     else if (range->start >= buffer->used) {
@@ -106,8 +91,8 @@ extern "C" {
             break;
           }
 
-          // this is not a valid vocabulary name so seek until a whitespace to prepare for the next main loop pass.
-          f_macro_iki_seek_whitespace(status, buffer, range, width_max, F_true);
+          // this is not a valid vocabulary name so seek until a non-word, non-dash, or non-plus character.
+          f_macro_iki_seek_word_dash_plus(status, buffer, range, width_max, F_false);
           if (F_status_is_error(status)) {
             f_macro_string_lengths_delete(status, delimits);
             return status;
@@ -326,28 +311,7 @@ extern "C" {
                     range->start++;
 
                     // skip past all initial non-word, non-dash, and non-plus.
-                    while (range->start <= range->stop && range->start < buffer->used) {
-                      if (buffer->string[range->start] == f_iki_syntax_placeholder) {
-                        range->start++;
-                        continue;
-                      }
-
-                      f_macro_iki_determine_width_max(buffer, range, width_max);
-
-                      status = f_utf_is_word_dash_plus(buffer->string + range->start, width_max);
-                      if (F_status_is_error(status)) {
-                        f_macro_string_lengths_delete(status, delimits);
-                        return status;
-                      }
-
-                      if (status == F_true) break;
-
-                      status = f_utf_buffer_increment(*buffer, range, 1);
-                      if (F_status_is_error(status)) {
-                        f_macro_string_lengths_delete(status, delimits);
-                        return status;
-                      }
-                    } // while
+                    f_macro_iki_seek_word_dash_plus(status, buffer, range, width_max, F_true);
 
                     found_vocabulary.start = range->start;
                   }
@@ -429,39 +393,7 @@ extern "C" {
       }
 
       if (find_next) {
-
-        while (range->start <= range->stop && range->start < buffer->used) {
-
-          f_macro_iki_seek_whitespace(status, buffer, range, width_max, F_true);
-          if (F_status_is_error(status)) {
-            f_macro_string_lengths_delete(status, delimits);
-            return status;
-          }
-
-          f_macro_iki_seek_whitespace(status, buffer, range, width_max, F_false);
-          if (F_status_is_error(status)) {
-            f_macro_string_lengths_delete(status, delimits);
-            return status;
-          }
-
-          if (range->start > range->stop || range->start >= buffer->used) break;
-
-          f_macro_iki_determine_width_max(buffer, range, width_max);
-
-          status = f_utf_is_word_dash_plus(buffer->string + range->start, width_max);
-          if (F_status_is_error(status)) {
-            f_macro_string_lengths_delete(status, delimits);
-            return status;
-          }
-
-          if (status == F_true) break;
-
-          status = f_utf_buffer_increment(*buffer, range, 1);
-          if (F_status_is_error(status)) {
-            f_macro_string_lengths_delete(status, delimits);
-            return status;
-          }
-        } // while
+        f_macro_iki_seek_word_dash_plus(status, buffer, range, width_max, F_true);
 
         found_vocabulary.start = range->start;
         find_next = F_false;

diff --git a/specifications/fss-000D.txt b/specifications/fss-000D.txt
index 1abcd430c17936973cab8d136160c2d6610e8f20..1e09b6576aa984c1434270558b3d79b273bebb54 100644 (file)
--- a/specifications/fss-000D.txt
+++ b/specifications/fss-000D.txt
@@ -16,22 +16,23 @@ Featureless Settings Specification: 000D - Iki Text:
 
   For compatibility with the FSS terminology, the Vocabulary Name is to be considered the Object and the Vocabulary value is to be considered the Content.
 
-  Whitespace (or start of file), must exist before any valid variable name.
+  Whitespace, non-word (and non "_", "-", "+") character punctuations, or the start of file must exist before any valid variable name.
+  Whitespace and non-word (and non "_", "-", "+") character punctuations may not exist as part of the variable name.
 
   The IKI format will use IKI-0000 to represent an IKI with no explicitly defined vocabulary.
   Whereas IKI-0001 and beyond represent a specific IKI vocabulary.
 
   Key\:
-    \s = whitespace.
-    \o = any printable word character, including "_", "-", "+".
+    \o = any printable word character, including "_", "-", "+" (and Unicode equivalents).
     \c = any character, including whitespace and non-printing, and any delimited quote (used as the opening quote) or a any quote (undelimited) not used as the opening quote.
     \q = either a single quote (') or a double quote (").
     \x = any character.
+    \W = any non-word character, discluding "_", "-", "+" (and Unicode equivalents).
     * = 0 or more occurrences.
     ~ = one or more occurrences, or 0 if at start of file.
 
   Structure\:
-    \x*\s~\o:\q\c\q
+    \x*\W~\o:\q\c\q
 
   Example\:
     # fss-000d iki-0000