Zsh Mailing List Archive Messages sorted by: Reverse Date, Date, Thread, Author

Re: Bug: Searching through sufficiently large $historywords causes seg fault

X-seq: zsh-workers 51732
From: Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx>
To: Marlon Richert <marlon.richert@xxxxxxxxx>
Cc: Zsh hackers list <zsh-workers@xxxxxxx>
Subject: Re: Bug: Searching through sufficiently large $historywords causes seg fault
Date: Tue, 9 May 2023 09:53:12 -0700
Archived-at: <https://zsh.org/workers/51732>
In-reply-to: <CAH+w=7bp9hix9trxDq1PsHfRuq=nM9LQarrmzkg4j4uDwq+UDA@mail.gmail.com>
List-id: <zsh-workers.zsh.org>
References: <CAHLkEDuXELQ66nEW+D74q3GiMcoOvNYfBZSSxFKfvZOZopQLkw@mail.gmail.com> <CAH+w=7bp9hix9trxDq1PsHfRuq=nM9LQarrmzkg4j4uDwq+UDA@mail.gmail.com>

On Mon, May 8, 2023 at 10:01 AM Bart Schaefer <schaefer@xxxxxxxxxxxxxxxx> wrote:
>
> Does PWS's patch from https://zsh.org/workers/51722 make any difference?

To attempt to answer my own question:

I don't get the segfault on $historywords[(r)foo] in either case
(without, or with, 51722).  However, after [(r)foo] in the "without"
case I get an immediate segfault attempting $#historywords.

With 51722, $#historywords is incorrect (returns "51" where it should
be about 100,000) and I eventually get a segfault after scrolling back
through the history past the event that was added with "print -S'.

Follow-Ups:
- Re: Bug: Searching through sufficiently large $historywords causes seg fault
  - From: Bart Schaefer

References:
- Bug: Searching through sufficiently large $historywords causes seg fault
  - From: Marlon Richert
- Re: Bug: Searching through sufficiently large $historywords causes seg fault
  - From: Bart Schaefer

Messages sorted by: Reverse Date, Date, Thread, Author