rename XFactory.Regs to XFactory.Generators.
This commit is contained in:
+15
-1
@@ -6,7 +6,7 @@
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
||||
<meta name="title" content="Class NcExpressionEvaluator | HiAPI-C# 2025 ">
|
||||
|
||||
<meta name="description" content="Walks an AST and produces an . Resolves #nnn via an ; built-in function names are matched case-insensitively against a fixed table. Phase-1 supports: SIN COS TAN ASIN ACOS ATAN SQRT ABS ROUND FIX FUP LN EXP POW. Trigonometric arguments and results are in degrees, matching Fanuc Custom Macro B convention. Unknown function names surface as ; arity mismatches as ; division / MOD by zero and domain errors (e.g. SQRT[-1]) as ; vacant operands as .">
|
||||
<meta name="description" content="Walks an AST and produces an . Resolves #nnn via an ; built-in function names are matched case-insensitively against a fixed table. Phase-1 supports: SIN COS TAN ASIN ACOS ATAN SQRT ABS ROUND FIX FUP LN EXP POW. Trigonometric arguments and results are in degrees, matching Fanuc Custom Macro B convention. Unknown function names surface as ; arity mismatches as ; division / MOD by zero and domain errors (e.g. SQRT[-1]) as ; vacant operands as . Numeric domain &amp; type conventions. All values are IEEE 754 — there is no separate bool / int type at runtime. Comparison ops (EQ NE GT GE LT LE) yield 1.0 (true) or 0.0 (false), using strict double equality / ordering (NaN compares as IEEE specifies — NaN EQ NaN is 0.0). Logical ops (AND OR XOR) truncate each operand to a 64-bit signed integer ( then cast to ) before applying the bitwise operation; non-finite or out-of-range operands surface rather than silently wrapping. Truthiness at caller-side IF / WHILE gates is value != 0 — any non-zero value (bit, float, comparator result) is true.">
|
||||
<link rel="icon" href="../img/HiAPI.favicon.ico">
|
||||
<link rel="stylesheet" href="../public/docfx.min.css">
|
||||
<link rel="stylesheet" href="../public/main.css">
|
||||
@@ -109,6 +109,20 @@ Fanuc Custom Macro B convention. Unknown function names surface as
|
||||
errors (e.g. <code>SQRT[-1]</code>) as <a class="xref" href="Hi.NcParsers.EvaluationSyntaxs.Evaluation.NcExpressionEvaluator.html#Hi_NcParsers_EvaluationSyntaxs_Evaluation_NcExpressionEvaluator_MathErrorCode">MathErrorCode</a>; vacant
|
||||
operands as <a class="xref" href="Hi.NcParsers.EvaluationSyntaxs.Evaluation.NcExpressionEvaluator.html#Hi_NcParsers_EvaluationSyntaxs_Evaluation_NcExpressionEvaluator_VacantErrorCode">VacantErrorCode</a>.
|
||||
</p>
|
||||
<p>
|
||||
<b>Numeric domain & type conventions.</b> All values are IEEE 754
|
||||
<a class="xref" href="https://learn.microsoft.com/dotnet/api/system.double">double</a> — there is no separate bool / int type at runtime.
|
||||
Comparison ops (<code>EQ NE GT GE LT LE</code>) yield <code>1.0</code> (true) or
|
||||
<code>0.0</code> (false), using strict double equality / ordering (NaN
|
||||
compares as IEEE specifies — <code>NaN EQ NaN</code> is <code>0.0</code>). Logical
|
||||
ops (<code>AND OR XOR</code>) truncate each operand to a 64-bit signed
|
||||
integer (<a class="xref" href="https://learn.microsoft.com/dotnet/api/system.math.truncate#system-math-truncate(system-double)">Truncate(double)</a> then cast to
|
||||
<a class="xref" href="https://learn.microsoft.com/dotnet/api/system.int64">long</a>) before applying the bitwise operation; non-finite
|
||||
or out-of-range operands surface <a class="xref" href="Hi.NcParsers.EvaluationSyntaxs.Evaluation.NcExpressionEvaluator.html#Hi_NcParsers_EvaluationSyntaxs_Evaluation_NcExpressionEvaluator_MathErrorCode">MathErrorCode</a> rather
|
||||
than silently wrapping. Truthiness at caller-side <code>IF / WHILE</code>
|
||||
gates is <code>value != 0</code> — any non-zero value (bit, float, comparator
|
||||
result) is true.
|
||||
</p>
|
||||
</div>
|
||||
<div class="markdown conceptual"></div>
|
||||
|
||||
|
||||
Reference in New Issue
Block a user