format: support leading zero printing for floats.

For a floating-point argument, the following behavior
is supported by the ~a, ~s and ~f format directives:
if the precision has a leading zero, then leading zeros
are added, up to one less than the field width.

* stream.c (formatv): Set the precision to one less than
the field width when rendering floats, rather than to
zero. The output_num code takes the precision as the cue
for padding zeros.

* txr.1: Updated format documentation, and clarified
a few things also.

2 files changed, 16 insertions, 3 deletions

diff --git a/stream.c b/stream.c
index 48738bb7..d5886336 100644
--- a/stream.c
+++ b/stream.c
@@ -3259,7 +3259,7 @@ val formatv(val stream_in, val fmtstr, struct args *al)
               vformat_str(stream, lit("#<bad-float>"), width, align, 0);
               continue;
             }
-            precision = 0;
+            precision = (width ? width - 1 : 0);
             goto output_num;
           }
         case 'd':
@@ -3348,7 +3348,7 @@ val formatv(val stream_in, val fmtstr, struct args *al)
               continue;
             }
 
-            precision = 0;
+            precision = (width ? width - 1 : 0);
             goto output_num;
           default:
             if (width != 0 || precision_p) {
diff --git a/txr.1 b/txr.1
index 22d0b3cb..3dd53160 100644
--- a/txr.1
+++ b/txr.1
@@ -40493,7 +40493,11 @@ or
 .codn s )
 is used. The details are
 documented in the description of each of these, below. The leading zero flag is
-ignored for floating-point values regardless of the conversion specifier.
+also taken into account for floating-point values, and treated uniformly by
+these directives.  If the flag is present, then the printed value's integer
+part will be padded with leading zeros up to the width of the field such that
+one character of unused space remains in the field, in case a positive or
+negative sign needs also to be rendered.
 
 For integer or floating-point arguments, if the precision specifier has a
 .code +
@@ -40560,6 +40564,15 @@ Floating point values which are integers are
 printed without a trailing
 .code .0
 (point zero).
+The
+.code +
+flag in the precision is honored for rendering an explicit
+.code +
+sign on non-negative values.
+If a leading zero is specified in the precision, and a nonzero width is
+specified, then the printed value's integer part will be padded with leading
+zeros up to one less than the field width. These zeros are placed before the
+sign.
 
 .coIP s
 Prints any object in a standard way, as if by the