Fix the text output of to_chars, snprintf, operator<< and std::format

doc/modules/ROOT/pages/charconv.adoc

@@ -168,6 +168,8 @@ Per IEEE 754 specifying both a precision and cohort preservation is an invalid o
 This follows because you are potentially rounding the number, or adding trailing zeros to the fraction which both destroy the cohort information.
 When using `cohort_preserving_fixed` the function will also return `{last, std::errc::invalid_argument}` for any value with a positive exponent, since rendering it in fixed notation would require appending trailing zeros.
 
+With the `general` format, a precision gives the same result as `printf` with `%.*g`: the precision is the number of significant digits (0 counts as 1), the fixed format is used when the decimal exponent is at least -4 and less than the precision, and trailing zeros are removed.
+
 IMPORTANT: Same as `from_chars`, `boost::decimal::to_chars` will return a `std::to_chars_result` if `std::chars_format` is used to specify the format; otherwise it returns a `boost::decimal::to_chars_result`.
 
 The library offers an additional feature for sizing buffers without specified precision and in general format

doc/modules/ROOT/pages/cstdio.adoc

@@ -31,6 +31,8 @@ int printf(const char* format, Dec... values) noexcept;
 ----
 
 These functions are limited in their support to only decimal floating point types, and are not a complete replacement for their `std::` equivalents.
+Like `std::snprintf`, `snprintf` writes as many characters as fit in the buffer, followed by a null terminator when `buf_size` is not 0, and returns the length of the full output.
+If a conversion has no value, the functions return -1 and set `errno` to `EINVAL`.
 
 == Type Modifiers
 
@@ -44,10 +46,14 @@ The remaining specifications of cstdio are the same as usual:
 
 - "g" or "G" for general format
 - "e" or "E" for scientific format
-- "f" for fixed format
+- "f" or "F" for fixed format
 - "a" or "A" for hex format
 
-NOTE: The uppercase format will return with all applicable values in uppercase (e.g. 3.14E+02 vs 3.14e+02)
+NOTE: The uppercase format will return with all applicable values in uppercase (e.g. 3.14E+02 vs 3.14e+02, and INF vs inf)
+
+A conversion can also have the flags "-", "+", " " and "0", and a minimum field width, before the precision (e.g. "%+010.3Df").
+"%%" writes a single "%".
+A conversion without a precision uses 6, but "a" and "A" without a precision print the exact value.
 
 == Examples
 

doc/modules/ROOT/pages/format.adoc

@@ -14,7 +14,8 @@ Format is supported when using C++20 and a compiler with appropriate support: GC
 
 == Locale Modifier
 
-If a format string begins with "L" the current global locale will be applied.
+If the format specifier has an "L" before the type, for example `{:.2Lf}`, the locale of the call is applied: the locale given to `std::format(loc, ...)` or `fmt::format(loc, ...)`, or else the current global locale.
+An "L" after the type, for example `{:.2fL}`, is also accepted.
 This can be set as so:
 
 [source, c++]
@@ -43,11 +44,13 @@ The following type modifiers are the same as those used by built-in floating poi
 | Modifier | Format
 | "g" or "G" | General
 | "e" or "E" | Scientific
-| "f" | Fixed
+| "f" or "F" | Fixed
 | "x" or "X" | Hex
 | "a" or "A" | Cohort Preserving Scientific
 |===
 
+The fast types do not keep a cohort, so "a" and "A" give a `format_error` for them.
+
 Example usage for scientific format would be: `{:e}`
 
 NOTE: The uppercase format will return with all applicable values in uppercase (e.g. 3.14E+02 vs 3.14e+02)
@@ -57,6 +60,9 @@ NOTE: The uppercase format will return with all applicable values in uppercase (
 Precision can be specified in the same way as built-in floating point values.
 For example a scientific format with 3 digits or precision would be: `{:.3e}`
 
+Without a precision, "f" and "e" give the shortest form that keeps the value (`{:f}` of 1.5 gives "1.5"), where a built-in floating point value gives 6 digits after the point.
+A width or a precision above `INT_MAX` gives a `format_error`.
+
 == Padding Modifiers
 
 If you want all values to be printed with a fixed width padding is allowed before the precision modifier.
@@ -67,6 +73,9 @@ For example with `{:10.3e}`:
 
 Note the space at the front of these string to keep with width at 10 characters
 
+A `0` before the width pads with zeros after the sign, for example `{:08.2f}` gives "-0003.14" for -3.14.
+An alignment overrides the `0`, and infinity and NaN are padded with spaces.
+
 == String Literal Support
 
 If you want the result to be a different string width than `char` you can specify this with the format string.
@@ -80,7 +89,7 @@ IMPORTANT: `std::format` only supports `char` and `wchar_t` types per the pass:[
 
 The appropriate order for the full format specifier is:
 
-String literal pass:["{Sign, Padding, Precision, Type, Locale}"]
+String literal pass:["{Fill and Alignment, Sign, 0, Width, Precision, Locale, Type}"]
 
 == Examples
 

doc/modules/ROOT/pages/generic_decimal.adoc

@@ -165,6 +165,10 @@ In the event of binary arithmetic between two decimal types the result will be t
 
 xref:design.adoc#non-finite-deviation[Non-finite values are supported]
 
+== 3.2.11 Note
+
+`operator<<` uses the stream flags as for a `double`: the float field (`fixed`, `scientific` or `hexfloat`), the precision, `showpos`, `showpoint`, `uppercase`, the width, fill and adjustment, and the grouping and decimal point of the `std::numpunct` of the stream character type. An exception during the output sets `badbit`, and goes to the caller only when `badbit` is in the exception mask.
+
 [#operator_behavior]
 == Operator Behaviors