Support '+' format for floating-point types as well

Sources/Plasma/CoreLib/plFormat.cpp

@@ -121,10 +121,9 @@ namespace plFormat_Private
                 spec.fDigitClass = kDigitHexUpper;
                 break;
             case '+':
-                spec.fDigitClass = kDigitDecAlwaysSigned;
+                spec.fAlwaysSigned = true;
                 break;
             case 'd':
-                if (spec.fDigitClass != kDigitDecAlwaysSigned)
                 spec.fDigitClass = kDigitDec;
                 break;
             case 'o':
@@ -265,7 +264,7 @@ static void _formatDecimal(const plFormat_Private::FormatSpec &format,
     if (format_size == 0)
         format_size = 1;
 
-    if (value < 0 || format.fDigitClass == plFormat_Private::kDigitDecAlwaysSigned)
+    if (value < 0 || format.fAlwaysSigned)
         ++format_size;
 
     hsAssert(format_size < 24, "Format length too long");
@@ -275,7 +274,7 @@ static void _formatDecimal(const plFormat_Private::FormatSpec &format,
 
     if (value < 0)
         buffer[0] = '-';
-    else if (format.fDigitClass == plFormat_Private::kDigitDecAlwaysSigned)
+    else if (format.fAlwaysSigned)
         buffer[0] = '+';
 
     _formatString(format, output, buffer, format_size, plFormat_Private::kAlignRight);
@@ -340,7 +339,6 @@ static void _formatChar(const plFormat_Private::FormatSpec &format,
             _formatNumeric<_utype>(format, output, value, 16, true); \
             break; \
         case plFormat_Private::kDigitDec: \
-        case plFormat_Private::kDigitDecAlwaysSigned: \
         case plFormat_Private::kDigitDefault: \
             _formatDecimal<_stype>(format, output, value); \
             break; \
@@ -369,7 +367,6 @@ static void _formatChar(const plFormat_Private::FormatSpec &format,
             _formatNumeric<_utype>(format, output, value, 16, true); \
             break; \
         case plFormat_Private::kDigitDec: \
-        case plFormat_Private::kDigitDecAlwaysSigned: \
         case plFormat_Private::kDigitDefault: \
             _formatDecimal<_utype>(format, output, value); \
             break; \
@@ -402,6 +399,10 @@ PL_FORMAT_IMPL(double)
     size_t end = 0;
 
     format_buffer[end++] = '%';
+
+    if (format.fAlwaysSigned)
+        format_buffer[end++] = '+';
+
     if (format.fPrecision >= 0) {
         int count = snprintf(format_buffer + end, arrsize(format_buffer) - end,
                              ".%d", format.fPrecision);
@@ -459,7 +460,6 @@ PL_FORMAT_IMPL(char)
         _formatNumeric<unsigned char>(format, output, value, 16, true);
         break;
     case plFormat_Private::kDigitDec:
-    case plFormat_Private::kDigitDecAlwaysSigned:
         _formatDecimal<signed char>(format, output, value);
         break;
     case plFormat_Private::kDigitChar:
@@ -488,7 +488,6 @@ PL_FORMAT_IMPL(wchar_t)
         _formatNumeric<uint32_t>(format, output, value, 16, true);
         break;
     case plFormat_Private::kDigitDec:
-    case plFormat_Private::kDigitDecAlwaysSigned:
         _formatDecimal<uint32_t>(format, output, value);
         break;
     case plFormat_Private::kDigitChar:

Sources/Plasma/CoreLib/plFormat.h

@@ -65,7 +65,7 @@ Mead, WA   99021
  * `<`           | Align left
  * `>`           | Align right
  * `NNN`         | Pad to NNN characters (minimum - can be more)
- * `+`           | Show a '+' char for positive signed values (decimal only)
+ * `+`           | Show a '+' char for positive signed values (decimal / float only)
  * `_C`          | Use C as the pad character (only '\001'..'\177' supported for now)
  * `x`           | Hex (lower-case)
  * `X`           | Hex (upper-case)
@@ -95,7 +95,6 @@ namespace plFormat_Private
     {
         kDigitDefault,          /**< Default digit formatting */
         kDigitDec,              /**< Format as decimal integer */
-        kDigitDecAlwaysSigned,  /**< Same as `kDigitDec`, but include a '+' for positive numbers too */
         kDigitHex,              /**< Hex integer (assume unsigned) */
         kDigitHexUpper,         /**< Hex integer with upper-case digits */
         kDigitOct,              /**< Octal integer (assume unsigned) */
@@ -118,6 +117,7 @@ namespace plFormat_Private
         int fPrecision = -1;        /**< Requested precision for floating-point */
 
         char fPadChar = 0;          /**< Explicit padding char (default is space) */
+        bool fAlwaysSigned = false; /**< Show + for positive numbers (dec/float) */
         Alignment fAlignment = kAlignDefault;   /**< Requested pad alignment */
         DigitClass fDigitClass = kDigitDefault; /**< Requested int formatting */
         FloatClass fFloatClass = kFloatDefault; /**< Requested float formatting */