time: d,c,dd,ddd,dddd pattern support for parse_format()

[jeffangelion]

Allow weekday name to be parseable (doesn't affect Time struct)

[JalonSolov]

long_days.index(weekday)

should be

long_days[weekday]

[jeffangelion]

@JalonSolov yeah you're right

already changed fn output from !int to !string (don't see much difference since fn result is always ignoring)

[JalonSolov]

One possibility is that you declared the function as returning a Result, but there are no error returns. If either of the if statements "fail" (as in the check doesn't let the code go into the body of the if), you could wind up with no answer at all, which should probably be an error...

[spytheman]

One possibility is that you declared the function as returning a Result, but there are no error returns. If either of the if statements "fail" (as in the check doesn't let the code go into the body of the if), you could wind up with no answer at all, which should probably be an error...

The function in question ends with:
return error_invalid_time(0, 'invalid weekday, at: ${p.current_pos_datetime}')

If I comment that out, I get:

I think it works fine.

The only usage of that function is currently this one:

            'dddd' {
                p.must_be_valid_week_day() or { return err }
            }

[JalonSolov]

And of course, that can be shortened to

            'dddd' {
                p.must_be_valid_week_day()!
            }

[jeffangelion]

And of course, that can be shortened to

IMHO writing functions for handling weekdays is overengineering since we're dropping them right away