From 5a0c79ef46aeac775a93ab1e69a2e31477c05d82 Mon Sep 17 00:00:00 2001 From: Sebastian Berg Date: Mon, 4 May 2026 11:56:19 +0200 Subject: [PATCH] PEP 3118: Clarify format string "unimplemented" note In order to wrap up the discussion around the choice of ``Zd`` vs. ``D`` a bit, I propose changing the "not implemented" note a bit. Not sure if it comes quite across, but I think: 1. There is no good reason to break with any of the "additions" here (I.e. don't use `&` to mean something different). 2. Python should feel free to diverge, but I think the note pointing to NumPy (which is "just" an example) hopefully makes it clear that this isn't a void. --- peps/pep-3118.rst | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/peps/pep-3118.rst b/peps/pep-3118.rst index 84710bb1dbd..fa08849df6a 100644 --- a/peps/pep-3118.rst +++ b/peps/pep-3118.rst @@ -14,7 +14,8 @@ Post-History: `09-Apr-2007 ' complex (whatever the next specifier is, e.g. ``'Zd'``) +'&' specific pointer (e.g. ``'&i'`` for a pointer to an int) 'T{}' structure (detailed layout inside {}) '(k1,k2,...,kn)' multi-dimensional array of whatever follows ':name:' optional name of the preceding element