Questions about pgfmath

Question

\pgfkeys{/pgf/fpu} 

\pgfmathsetlengthmacro\x{5^2}
\show\x -> 25.0pt % OK, thanks.

\pgfmathparse{5cm^2}

What is the meaning of ] in the following?
```
\show\pgfmathresult % ->1Y2.0238983e4].

\pgfkeys{/pgf/fpu/output format=sci}
```
gives the expected format, but what is the cryptic ]?
Why does this fail? Is unit not allowed in {<expr>}? The manual says \pgfmathsetlengthmacro "defines <macro> as the value of <expression> in points."
```
\pgfmathsetlengthmacro\y{5cm^2} % ! Illegal unit.
\pgfmathsetlengthmacro\y{5^2cm} % ! Illegal unit.
\show\y
```

EDIT

Here is a minimal example:

\documentclass{article}
\usepackage{pgfplots}

\begin{document}
\pgfkeys{/pgf/fpu}
\pgfkeys{/pgf/fpu/output format=sci}
\pgfmathparse{5cm^2}
%\show\pgfmathresult % -> 2.0238983e4.

% Both of these give error:
\pgfmathsetlengthmacro\x{5mm^2}
\pgfmathsetlengthmacro\x{5^2}

\pgfkeys{/pgf/fpu=false}

\end{document}

Error msg:

! Illegal unit of measure (pt inserted).
<to be read again>
                   e
l.10 \pgfmathsetlengthmacro\x{5mm^2}

Wouldn't a MWE help? You know which packages you're loading, we don't. I get an error from the first \pgfmathsetlengthmacro call. — egreg, Oct 01 '12 at 14:53
This is related to the internal number represantation of the fpu library and I think ] is doing something similar to \relax. You have to use \pgfmathprintnumber to convert it to the human-readable form. — percusse, Oct 01 '12 at 14:56
I get ! Dimension too large. for both of the examples in part 2, probably because 5 cm comes out 'too big' in pt. — Joseph Wright, Oct 01 '12 at 15:09
Related question on the topic of FPU internal representation: tikz pgf - pgfplots and calculations; without fpu 'dimension too large', with fpu 'Illegal unit of measure (pt inserted)' - TeX - LaTeX Stack Exchange — user202729, Nov 23 '22 at 02:27

score 9 · Accepted Answer · answered Oct 01 '12 at 19:09

The 1Y2.0238983e4] is the internal representation of the number. With fpu you have to explicitly specify the output format or use \pgfmathfloattofixed. The default format is float (with Y and ]), \pgfmathsetlengthmacro is not changed by the fpu library, thus the length is not a valid TeX length with format float. Examples:

\documentclass{article}
\usepackage{pgf}
\usepgflibrary{fpu}

\begingroup
  \pgfkeys{/pgf/fpu,/pgf/fpu/output format=fixed}
  \pgfmathsetlengthmacro\x{5^2}
  \typeout{\x}
\endgroup

\begingroup
  \pgfkeys{/pgf/fpu}
  \pgfmathparse{5^2}
  \pgfmathfloattofixed{\pgfmathresult}
  \edef\x{\pgfmathresult pt}
  \typeout{\x}
\endgroup

\begingroup
  \pgfkeys{/pgf/fpu,/pgf/fpu/output format=fixed}
  \pgfmathparse{5cm^2}
  \typeout{\pgfmathresult}
\endgroup

\begin{document}
\end{document}

Result:

25.0pt
25.000000000pt
20238.983000000000

More details in 36 Floating Point Unit Library.

Questions about pgfmath

1 Answers1

Linked