Re: [PATCH] Fix assigning/typesetting of unset references (and related bugs)

X-seq: zsh-workers 54239

From: Philippe Altherr <philippe.altherr@xxxxxxxxx>

To: Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx>

Cc: Zsh hackers list <zsh-workers@xxxxxxx>

Subject: Re: [PATCH] Fix assigning/typesetting of unset references (and related bugs)

Date: Mon, 23 Mar 2026 01:25:20 +0000

Arc-authentication-results: i=1; mx.google.com; arc=none

Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20240605; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:dkim-signature; bh=w+ps41Cpj23qq7OTXu1AyMC6pJ+nPYoURYCR79nNJkE=; fh=vuDAGjptCPc/P1/HEZ3j98yPJEW1XjuEoEAmmwDS/+U=; b=d/1MOx11i/XtLrGsRQTr6FWySWn43AUKqnrYOUgMv8UDSmGQIi0x4H76Oc0FbA5mks 7HncdR1GeEAwG27GDGZ56n+KZVgG4kNukHNmT+C4EgErI4754OfZI5dwcuC5yNV7mAnx QXIKnmkoYYwPrqexCvb0SRtaztTELHoW4OJ8F7nsm7oKdAibOtiqniRP8N3+f9pzyLd0 +Oy+ju5T9N0ncCy8BtBEEZ+QmpmUV7NF0DjVxROl6hfKCQZ6Ahrxcw3QvT9CXU87wNcH 6UI+ww83PglmyCDdwfUFGPSCffw1WEIyKmEcTTYXxY3tFSDqNzqkYB+PdGx/hgaLOkEC 3QuA==; darn=zsh.org

Arc-seal: i=1; a=rsa-sha256; t=1774229132; cv=none; d=google.com; s=arc-20240605; b=TE9lON1q+atOMIB0xe86QsB4bUxwMX3vN3/VLGIGtBeCMURw6bDzrg2J0x62FCkPtp a7TbkWb6OOQNr2awiADctZZJwvcoakrG9tOZlQ8pJpjv0OSaY097NZCvu4ogPIO1Bmuh 8kFg00nkXsWi31JCo79TLBzCjWJ5B2e+4WwYO2dHPv0Qhi8ESseKo1XN4L2d3d8XeN9E E9B3h7wMnDlsiEE0vzKXkJSSDrINTS5alhf6zqzlvn1SMD5/TfBHmlTyZCu1A+/eG5Dj t+I+siI2I6tXA1w75epUMGO5qdc1cXpjkTrIzuUhNl85PwbnBaO6Q4aRdCVFzpmflMfq J7WQ==

Archived-at: <https://zsh.org/workers/54239>

In-reply-to: <CAH+w=7YqX_c==yQY4g=T9G-Z1zosTcbACtY9XSQtAu66pdO=YA@mail.gmail.com>

List-id: <zsh-workers.zsh.org>

References: <CAGdYchtYnyn-rrgD7mRhDg7qiwOp0GEgNy+R4ft39Kwf1RYYYA@mail.gmail.com> <CAH+w=7YqX_c==yQY4g=T9G-Z1zosTcbACtY9XSQtAu66pdO=YA@mail.gmail.com>

I suppose I have no problem with this, but it does contradict behavior
mentioned in workers/54224.

I'm not sure it contradicts any behavior mentioned in workers/54224. It would rather enable some of the corner cases I had in mind. As I (have now) replied there, I will try to write some tests for these corner cases. The proposed patch should in principle already handle them correctly.

> + # The only purpose of this test is to define the helper functions
> + # needed by the tests below about assigning and typesetting named
> + # references in different contexts.

Is there a reason these shouldn't go in the %prep section?

Mainly locality. I find it easier/more convenient to define the functions in the context where they are used. But, if you prefer that we stick with the %prep section, I can move them there.

Philippe

On Mon, Mar 23, 2026 at 12:19 AM Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx> wrote:

On Sun, Mar 22, 2026 at 4:41 PM Philippe Altherr
<philippe.altherr@xxxxxxxxx> wrote:
>
> Assigning an unset local reference R, wrongly resurrects the parameter R as a new reference instead of resurrecting it as a new string parameter R:

I suppose I have no problem with this, but it does contradict behavior
mentioned in workers/54224.

> + # The only purpose of this test is to define the helper functions
> + # needed by the tests below about assigning and typesetting named
> + # references in different contexts.

Is there a reason these shouldn't go in the %prep section?

Reverse Date

Date

Thread

Author