Revise string types

compiler/lib/src/main/scala/analysis/Semantics/Type.scala

@@ -384,18 +384,12 @@ object Type {
       case (Boolean -> Boolean) => true
       case _ => false
     }
-    def string = pair match {
-      case (String(None) -> String(None)) => true
-      case (String(Some(e1)) -> String(Some(e2))) => e1 == e2
-      case _ => false
-    }
     def sameDef = (t1.getDefNodeId, t2.getDefNodeId) match {
       case (Some(id1), Some(id2)) => id1 == id2
       case _ => false
     }
     numeric ||
     boolean ||
-    string ||
     sameDef
   }
 

compiler/lib/src/test/scala/semantics/TypeSpec.scala

@@ -25,7 +25,6 @@ class TypeSpec extends AnyWordSpec {
       duplicate(I64),
       duplicate(I8),
       duplicate(Integer),
-      duplicate(String(None)),
       duplicate(U16),
       duplicate(U32),
       duplicate(U64),
@@ -40,6 +39,7 @@ class TypeSpec extends AnyWordSpec {
       (defaultAbsType, defaultEnum),
       (defaultArray, defaultStruct),
       (Boolean,String(None)),
+      duplicate(String(None)),
       (F32,F64),
       (I8,U32),
       duplicate(AnonArray(None, I32)),

docs/fpp-spec.html

@@ -10661,14 +10661,28 @@ <h3 id="Types_String-Types">17.6. String Types</h3>
 A string type is written <code>string</code> or <code>string size</code> <em>n</em>,
 where <em>n</em> is an integer value in the range [1,2^31-1].
 There is one string type <code>string</code> and one string type <code>string size</code> <em>n</em>
-for each legal value of <em>n</em>.</p>
+for each legal value of <em>n</em>.
+The string type names map to the string types in the following way:</p>
+</div>
+<div class="ulist">
+<ul>
+<li>
+<p>The type name <code>string</code> maps to the type <code>string</code>.</p>
+</li>
+<li>
+<p>The type name <code>string</code> <code>size</code> <em>e</em> maps to the type <code>string</code> <code>size</code> <em>n</em>,
+after evaluating the expression <em>e</em> to the value <em>n</em>.</p>
+</li>
+</ul>
 </div>
 <div class="paragraph">
 <p>The value <em>n</em>, if present, is called the <strong>size</strong> associated with the string
 type.
-The size provides a maximum string length that the code generator can use
+The size provides a maximum string length that the code generator uses
 for allocating memory and serializing data.
-If no explicit size is present, then the code generator uses a default size.</p>
+If no size is present, then the code generator uses
+the <a href="#Definitions_Framework-Definitions_Constant-Definitions">framework definition</a> <code>FW_FIXED_LENGTH_STRING_SIZE</code> as the
+maximum string length.</p>
 </div>
 </div>
 <div class="sect2">
@@ -11337,11 +11351,6 @@ <h3 id="Type-Checking_Identical-Types">18.13. Identical Types</h3>
 <a href="#Types_The-Boolean-Type">the Boolean type</a>.</p>
 </li>
 <li>
-<p>\$T_1\$ and \$T_2\$
-are both  the same
-<a href="#Types_String-Types">string type</a>.</p>
-</li>
-<li>
 <p>Each of \$T_1\$ and \$T_2\$
 is an
 <a href="#Types_Abstract-Types">abstract type</a>,
@@ -11353,6 +11362,9 @@ <h3 id="Type-Checking_Identical-Types">18.13. Identical Types</h3>
 </li>
 </ol>
 </div>
+<div class="paragraph">
+<p>Note that string types are never identical.</p>
+</div>
 </div>
 <div class="sect2">
 <h3 id="Type-Checking_Type-Conversion">18.14. Type Conversion</h3>
@@ -11817,8 +11829,7 @@ <h2 id="Values">20. Values</h2>
 <div class="paragraph">
 <p>Every value <em>v</em> belongs to exactly one <a href="#Types_Canonical-Types">canonical type</a>
 <em>T</em>, except that
-every string value belongs to the type <code>string</code> and to
-all types <code>string size</code> <em>n</em>, for any <em>n</em>.</p>
+every string value belongs to every string type.</p>
 </div>
 <div class="paragraph">
 <p>Every value that belongs to type <em>T</em> also belongs to every
@@ -11914,20 +11925,8 @@ <h3 id="Values_String-Values">20.5. String Values</h3>
 characters in the sequence.</p>
 </div>
 <div class="paragraph">
-<p>A string value <em>v</em> belongs to the following canonical types:</p>
-</div>
-<div class="ulist">
-<ul>
-<li>
-<p><code>string</code></p>
-</li>
-<li>
-<p><code>string size</code> <em>n</em> for any <em>n</em>.</p>
-</li>
-</ul>
-</div>
-<div class="paragraph">
-<p>Note that <em>v</em> may have length <em>m</em> and type <code>string size</code> <em>n</em>,
+<p>A string value <em>v</em> belongs to every string type.
+Note that <em>v</em> may have length <em>m</em> and type <code>string size</code> <em>n</em>,
 with \$m &gt; n\$.
 In this case, the behavior is defined by the code generation strategy.
 In particular, when code is generated for the F Prime framework,
@@ -12471,7 +12470,7 @@ <h3 id="Analysis-and-Translation_Translation-Tools">22.4. Translation Tools</h3>
 </div>
 <div id="footer">
 <div id="footer-text">
-Last updated 2026-02-27 14:32:17 -0800
+Last updated 2026-03-03 08:30:06 -0800
 </div>
 </div>
 <script src="code-prettify/run_prettify.js"></script>

docs/fpp-users-guide.html

@@ -16543,7 +16543,7 @@ <h3 id="Writing-C-Plus-Plus-Implementations_Serialization-of-FPP-Values">17.4. S
 </div>
 <div id="footer">
 <div id="footer-text">
-Last updated 2026-02-26 16:15:38 -0800
+Last updated 2026-03-02 20:14:57 -0800
 </div>
 </div>
 <script src="code-prettify/run_prettify.js"></script>

docs/spec/Type-Checking.adoc

@@ -269,10 +269,6 @@ the type.
 are both
 <<Types_The-Boolean-Type,the Boolean type>>.
 
-. stem:[T_1] and stem:[T_2]
-are both  the same
-<<Types_String-Types,string type>>.
-
 . Each of stem:[T_1] and stem:[T_2]
 is an
 <<Types_Abstract-Types,abstract type>>,
@@ -282,6 +278,8 @@ is an
 <<Types_Struct-Types,struct type>>,
 and both types refer to the same definition.
 
+Note that string types are never identical.
+
 === Type Conversion
 
 We say that a type stem:[T_1] *may be converted to* another type stem:[T_2] if

docs/spec/Types.adoc

@@ -41,12 +41,21 @@ A string type is written `string` or `string size` _n_,
 where _n_ is an integer value in the range [1,2^31-1].
 There is one string type `string` and one string type `string size` _n_
 for each legal value of _n_.
+The string type names map to the string types in the following way:
+
+* The type name `string` maps to the type `string`.
+
+* The type name `string` `size` _e_ maps to the type `string` `size` _n_,
+after evaluating the expression _e_ to the value _n_.
 
 The value _n_, if present, is called the *size* associated with the string
 type.
-The size provides a maximum string length that the code generator can use
+The size provides a maximum string length that the code generator uses
 for allocating memory and serializing data.
-If no explicit size is present, then the code generator uses a default size.
+If no size is present, then the code generator uses
+the <<Definitions_Framework-Definitions_Constant-Definitions,
+framework definition>> `FW_FIXED_LENGTH_STRING_SIZE` as the
+maximum string length.
 
 === Array Types
 

docs/spec/Values.adoc

@@ -26,8 +26,7 @@ A *value* is one of the following:
 
 Every value _v_ belongs to exactly one <<Types_Canonical-Types,canonical type>>
 _T_, except that
-every string value belongs to the type `string` and to
-all types `string size` _n_, for any _n_.
+every string value belongs to every string type.
 
 Every value that belongs to type _T_ also belongs to every
 <<Types_Alias-Types,alias type>> whose
@@ -96,12 +95,7 @@ e.g., `"abc"`.
 Every string value _v_ has a *length*, which is the number of
 characters in the sequence.
 
-A string value _v_ belongs to the following canonical types:
-
-* `string`
-
-* `string size` _n_ for any _n_.
-
+A string value _v_ belongs to every string type.
 Note that _v_ may have length _m_ and type `string size` _n_,
 with stem:[m > n].
 In this case, the behavior is defined by the code generation strategy.