format: fix precision field leading zero problems.

* stream.c (formatv): Do not recognize multiple leading zeros
as a single one; once the zero flag is set, if another zero is
seen, it must be treated as one of the digits specifying the
precision value. New requirement: before processing a format
specifier, check for the situation that the leading zero
has been specified, but no precision. Convert this situation
to that of a precision of zero being given, with no leading
zero.

* txr.1: Document the ambiguity around the leading zero and
how it is being handled when only the leading zero flag is
given, and no actual precision. Add a note about what happens
when zero precision is specified in ~a in conjunction with
a floating-point value. Misspelled "pas" word fixed.

2 files changed, 27 insertions, 8 deletions

diff --git a/stream.c b/stream.c
index 50f07495..248a51b2 100644
--- a/stream.c
+++ b/stream.c
@@ -3337,8 +3337,11 @@ val formatv(val stream_in, val fmtstr, struct args *al)
       case vf_precision:
         switch (ch) {
         case '0':
-          zeropad = 1;
-          continue;
+          if (!zeropad) {
+            zeropad = 1;
+            continue;
+          }
+          /* fallthrough */
         case '1': case '2': case '3': case '4': case '5':
         case '6': case '7': case '8': case '9':
           saved_state = state;
@@ -3404,6 +3407,10 @@ val formatv(val stream_in, val fmtstr, struct args *al)
         break;
       case vf_spec:
         state = vf_init;
+        if (zeropad && !precision_p) {
+          zeropad = precision = 0;
+          precision_p = 1;
+        }
         switch (ch) {
         case 'x': case 'X':
           obj = args_get_checked(self, al, &arg_ix);
diff --git a/txr.1 b/txr.1
index 32a38b4a..e2fd0f3a 100644
--- a/txr.1
+++ b/txr.1
@@ -54517,17 +54517,26 @@ is being omitted; there is only a precision field.
 The precision specifier may begin with these optional characters:
 .RS
 .coIP 0
-(the "leading zero flag"),
+(the "leading zero indicator"),
 .coIP +
 (print a sign for positive values")
 .IP space
 (print a space in place of a positive sign).
 .RE
 
-The precision specifier itself is either a decimal integer that does not
-begin with a zero digit, or the
+The precision specifier itself follows: it must be either a decimal integer
+or the
 .code *
-character.
+character indicating that the precision value comes from an integer argument.
+
+The leading zero indicator is only active if a precision specifier is
+present. If no precision specifier is present, then the leading zero indicator
+is interpreted as a specifier indicating a precision value of zero, rather
+than requesting leading zeros.  To request zero padding together with zero
+precision, either two or more zero digits are required, or else the leading
+zero indicator must be given together with the
+.code *
+specifier.
 
 The precision field's components have a meaning which depends on the type of
 object printed and the conversion specifier.
@@ -54615,6 +54624,7 @@ the
 special variable, whose default value is the same as that of the
 .code flo-dig
 variable.
+
 Floating point values which are integers are
 printed without a trailing
 .code .0
@@ -54627,7 +54637,9 @@ sign on nonnegative 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.
+sign.  A precision value of zero imposed on floating-point values is
+equivalent to a value of one; it is not possible to request zero significant
+figures.
 
 .coIP s
 Prints any object in a standard way, as if by the
@@ -54709,7 +54721,7 @@ precision of the number or its type.  If it is omitted, then the value
 is obtained from the special variable
 .codn *print-flo-digits* ,
 whose default value is three: three digits past the decimal point. A precision
-of zero means no digits pas the decimal point, and in this case the decimal
+of zero means no digits past the decimal point, and in this case the decimal
 point is suppressed (regardless of whether the numeric argument is
 floating-point or integer).