bug#74666: 31.0.50; Regression in replace-match with empty-adjacent groups

[Eli Zaretskii <eliz@gnu.org>]

> Date: Tue, 3 Dec 2024 21:56:02 +1100
> From: Campbell Barton <ideasman42@gmail.com>
> 
> 
> Run the following file as an executable script:
> 
> emacs -Q --batch -l ./emacs-test.el
> 
> ;; BEGIN SCRIPT:
> #!/usr/bin/env -S emacs --batch -l
> ;;; emacs-git-log --- Batch check emacs-lisp. -*- lexical-binding: t -*-
> 
> (defun printf (&rest args)
>    (princ (apply #'format args) #'external-debugging-output))
> 
> (defun test-me (is-forward)
>    (let ((result ""))
>      (with-temp-buffer
>        (insert "__B_\n")
>        (save-match-data
>          (set-match-data (list 2 4 2 2 2 4))
>          (cond
>           (is-forward
>            (replace-match "HELLO" t t nil 1)
>            (replace-match "WORLD" t t nil 2))
>           (t
>            (replace-match "WORLD" t t nil 2)
>            (replace-match "HELLO" t t nil 1))))
>        (setq result (buffer-substring-no-properties (point-min) 
> (point-max))))
>      result))
> 
> (printf "A: %s" (test-me t))
> (printf "B: %s" (test-me nil))
> ;; END SCRIPT:
> 
> In emacs 29.4 this prints:
> 
> A: _HELLOWORLD_
> B: _HELLOWORLD_
> 
> In emacs 31.0.50 this prints:
> 
> A: _WORLD_
> B: _HELLOWORLD_
> 
> This is a regression in 47b497b4dac91e5ea56102018223bdeb5e21a93b.

Stefan, could you please take a look?