Compose String — Control Systems/Strings

Compose String

Control Systems / Strings

Builds text from a format and its inputs, as C's and MATLAB's sprintf do: y = sprintf(Format, u1, u2, …). With the Format "x=%d, v=%.2f" and the inputs 7 and 3.14159, the output is x=7, v=3.14.

Ports

• Input 1 (i32, ICoreInt32, for the default Format's %d) – the value for the first conversion, one value [1,1].
• Input 2 – the value for the default Format's %f, one number [1,1].
• Output – the text (str, ICoreString), one value, cut at 255 bytes.

There is one input per conversion, in order; the number of inputs is yours to set and must equal the number of conversions in the Format. What each input carries is decided by its conversion, exactly as in Simulink: %d and %i take a signed integer of at most 32 bits (i8, i16, i32) or a bool; %u, %x, %X and %o an unsigned one (u8, u16, u32) or a bool; %f, %e, %E, %g and %G a floating-point number (f64, f32); %s text (str, ICoreString). Anything else is refused at the port. A double does not go into %d: convert it with Data Type Conversion first, as in Simulink.

Parameters

• Format – the format, written with its quotes as in Simulink: "%d %f" (the default). A doubled quote inside is one quote. Text is copied as it is; %% is a percent sign; the escapes \n, \t, \r, \a, \b, \f, \v and \\ are the control characters they name. A conversion is %, then optional flags (- left-justify, + always show the sign,   a space for a positive sign, 0 pad with zeros), an optional width, an optional .precision, and one of d i u x X o f e E g G s. Between 1 and 128 conversions.
• Sampling Time (s) – zero or less inherits the solver's rate; a positive value runs the block at that period.

What is not offered

Each of these stops the run with a message naming it: %c, %a, %p, %n and a * width or precision (Simulink refuses them too); length modifiers such as %ld and %hd, which Simulink accepts; the # flag; 0 together with -, with text, or with an integer precision; + or a space on an unsigned or text conversion; an integer precision of 0; and the escapes \x.. and octal \... These are left out because the six target languages do not agree on what they mean, and a block whose exported code disagreed with it would be worse than one that says no.

Not-a-number and infinity

A floating-point conversion writes NaN, Inf or -Inf – +Inf under the + flag – and applies no width, precision or zero padding to them. That is Simulink's own rule, measured; C's own printf would write nan and pad it.

Code export

C, C++, Python, MATLAB, Java and Rust all write the same text. The first four use their own printf for each conversion; Java and Rust have no equivalent that agrees with C on floating point, so their generated code rounds each value exactly and lays it out by the same rules. The Format is read when the code is generated, not when it runs.

VHDL, Verilog and SystemVerilog do not carry text, and PLC Structured Text is not supported either: the language has a string type, but the check that runs an exported core cannot read one back. An export to any of these four stops and names this block and the reason.

Simulink bridge

Import and export, mapped to simulink/String/Compose String. "Format" crosses unchanged, quotes and all; on import the number of inputs and what each one carries are read off it, since Simulink derives both from the Format as well. The output data type is always string and is written as a fixed setting. The Simulink block has no sampling-time setting of its own, so a positive Sampling Time (s) stays on this side and is reported rather than written.

Notes

• Algebraic, with no state: the output depends only on the current inputs.
• Width and precision on %s count bytes of UTF-8 here and in C, and characters in Simulink and in Java, Python and MATLAB – the same number for plain English text.