Let us make Sage reals and rationals more friendly
to formatting commands that expect Python floats.

For example, with a Python float one can do

sage: a = 1.25r
sage: "{:0.3f}".format(a)
'1.250'

but a similar command fails for a Sage real or rational:

sage: a = 1.25
sage: "{:0.3f}".format(a)
Traceback (most recent call last)
...
TypeError: unsupported format string passed to
sage.rings.real_mpfr.RealLiteral.__format__

sage: a = 5/4
sage: "{:0.3f}".format(a)
Traceback (most recent call last)
...
TypeError: unsupported format string passed to
sage.rings.rational.Rational.__format__

This puzzles users especially when using Python packages with Sage.
As things stand one needs to know to work around Sage's preparser.

For example this came up in

• Ask Sage question 46408: tqdm's "total" parameter bug
• Ask Sage question 52617: f string float not working

The case of Sage integers was dealt with in #12788 and can provide inspiration.

We could also consider doing the same for number field elements, algebraic numbers,
complex floating-point numbers, continued fractions... and other numbers in Sage
for which it makes sense.

This is part of #29728 (Meta-ticket: improve compatibility with the Python library).

CC: @slel @jcamp0x2a

Component: numerical

Keywords: format

Issue created by migration from https://trac.sagemath.org/ticket/27788