Converting cartesian to polar according to the manual

[teepeemm]

I'm confused by this example from the siunitx manual in section 4.7: \complexnum[complex-mode = polar]{1 + i} has the output $1\angle 0\degree$. In a sample file:

\listfiles
\documentclass{article}
\usepackage{siunitx}
\begin{document}
\complexnum[complex-mode = polar]{1 + i} \\
\complexnum[complex-mode = polar]{1 + 1i}
\end{document}

We have the output:
$1\angle 0\degree$
$1.414 213 562 373 095\angle 45\degree$
Whereas I would expect both to match the second output. Am I overlooking something?

The file versions:

 article.cls    2023/05/17 v1.4n Standard LaTeX document class
  size10.clo    2023/05/17 v1.4n Standard LaTeX file (size option)
 siunitx.sty    2024-02-15 v3.3.12 A comprehensive (SI) units package
translations.sty    2022/02/05 v1.12 internationalization of LaTeX2e packages (CN)
etoolbox.sty    2020/10/05 v2.5k e-TeX tools for LaTeX (JAW)
pdftexcmds.sty    2020-06-27 v0.33 Utility functions of pdfTeX for LuaTeX (HO)
infwarerr.sty    2019/12/03 v1.5 Providing info/warning/error messages (HO)
   iftex.sty    2022/02/03 v1.0f TeX engine tests
 ltxcmds.sty    2023-12-04 v1.26 LaTeX kernel commands for general use (HO)
 amstext.sty    2021/08/26 v2.01 AMS text
  amsgen.sty    1999/11/30 v2.0 generic functions
   array.sty    2023/10/16 v2.5g Tabular extension package (FMi)
   color.sty    2022/01/06 v1.3d Standard LaTeX Color (DPC)
   color.cfg    2016/01/02 v1.6 sample color configuration
  pdftex.def    2022/09/22 v1.2b Graphics/color driver for pdftex
mathcolor.ltx
l3backend-pdftex.def    2024-03-14 L3 backend support: PDF output (pdfTeX)
translations-basic-dictionary-english.trsl    (english translation file `translations-basic-dictionary')
supp-pdf.mkii

[teepeemm]

Possibly related: \complexnum[complex-mode=cartesian, complex-angle-unit=radians]{1:3.14159265359} becomes $−1−2.067 615 373 566 167i$, where I would expect $-1$ (or at least something within a rounding error of that).

[josephwright]

You have a nearly year-old release: for me your example gives the correct output for both forms for the initial issue.

[josephwright]

\complexnum[complex-mode=cartesian, complex-angle-unit=radians]{1:3.14159265359}

This one is an issue: I had not allowed for the exponent part in conversion, so the very small residual from angle conversion is being misinterpreted.