i have a banana


Recent activity

Re: panic in worker/imap/fetch.go 2 days ago

From ( to ~rjarry/aerc-discuss

Update: I seem to have gotten the same crash when pressing K twice
inside the Sent folder to switch back to the Inbox.

    -- (

Re: aerc panic when decrypting message 2 days ago

From ( to ~rjarry/aerc-devel

Oh, also, pinentry-ncurses is noticeably sluggish when typing in the
password, and the crash seems to happen as soon as pinentry-ncurses is
invoked (pinentry-ncurses obscures the log, which makes it seem like it
happens after the password is typed, though if you deliberately enter
the wrong password you may briefly see the log before pinentry-ncurses

    -- (

Re: aerc panic when decrypting message 2 days ago

From ( to ~rjarry/aerc-devel

I can confirm that it works when pinentry-gtk-2 is used. Perhaps aerc
and pinentry-ncurses try to fight over control of the terminal or

    -- (

Re: aerc panic when decrypting message 2 days ago

From ( to ~rjarry/aerc-devel

On Thu Jun 30, 2022 at 5:56 PM BST, Moritz Poldrack wrote:
> How reliably can you reproduce your sadness?

Seems like it only happens when pinentry-ncurses is invoked, as when the
password is cached by the GPG agent the crash doesn't occur. I'm not
sure whether it happens with pinentry-gtk-2. I'll test that in a

    -- (

aerc panic when decrypting message 2 days ago

From ( to ~rjarry/aerc-devel

After I typed the password into pinentry, aerc panicked:

panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x0 pc=0x884d6f]

goroutine 1 [running]:
panic({0x934ec0, 0xdc3300})

Re: [PATCH guixrus] guixrus: packages: Add iredis. 5 days ago

From ( to ~whereiseveryone/guixrus

Hi singpolyma,

On Mon Jun 27, 2022 at 4:01 PM BST, Stephen Paul Weber wrote:
> +    (arguments
> +     '(#:tests? #f
Could you please use the (list #:foo #~(...) #:bar #~(...)) gexps form
here instead?

> +           (lambda args
`(lambda _` would be better here, since we don't use the arguments.

> +               (lambda _ (display "from setuptools import setup\nif __name__ == '__main__':\n    setup()"))))))))
This line is a bit long :)

panic in worker/imap/fetch.go 21 days ago

From ( to ~rjarry/aerc-discuss

I hit this panic after having sent this email, which had been sitting in drafts for a few
days: <https://issues.guix.gnu.org/55903> and returned to the inbox, but i have no idea
what exactly i did that triggered it.

Error: runtime error: index out of range [82] with length 5

goroutine 356 [running]:
	/gnu/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-go-1.17.8/lib/go/src/runtime/debug/stack.go:24 +0x65
	/tmp/guix-build-aerc-0.0.0-20220531123251-2551dd1bfa2c.drv-0/src/git.sr.ht/~rjarry/aerc/logging/panic-logger.go:45 +0x414
panic({0x992c60, 0xc000529b48})
	/gnu/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-go-1.17.8/lib/go/src/runtime/panic.go:1038 +0x215

Re: aerc may effectively be GPL3.0-or-later licensed 29 days ago

From ( to ~rjarry/aerc-devel

Ah, I missed the part in the readme where it said that notmuch was optional. (In other news, my Guix package
works, and I will soon be sending it to upstream. Thanks for aerc!)

aerc may effectively be GPL3.0-or-later licensed 30 days ago

From to ~rjarry/aerc-devel

I noticed while packaging notmuch for GNU Guix <https://guix.gnu.org> that aerc depends on GPL-3.0-or-later software (notmuch and go-notmuch.) If my suspicion is correct, aerc is effectively GPL-3.0-or-later because the GPL propagates to any software that incorporates GPL-licensed software. Someone on #guix half-confirmed this (they said they were pretty sure I was right, but not certain.)

(Disclaimer: I am not a lawyer, nor even good at licensing things.)

This means that at commit 2485d509838013351672db50b2cb57880b7c3e1c ("add github.com/zenhack/go.notmuch"), aerc was effectively made GPL without anyone giving permission for their code to be relicensed. I suspect nobody would _really_ care about that, but the LICENSE file should probably be changed to reflect that aerc is actually GPLed. I do not, however, know whether this would require asking everyone who ever contributed (since it is already GPL, this would just be acknowledging it.)

I may be wrong about all this. Thoughts, anyone?

Re: It seems to be happening again when I edit the message a month ago

From to ~sircmpwn/sr.ht-discuss

May 27, 2022, 1:53 PM, paren@disroot.org paren@disroot.org wrote:

> This may be unrelated, but 500 is happening again, now when I click "Submit" after editing a message.

Oops, looks like I messed up sending the email, sorry. (Cause: having to use mail.disroot.org's
terrible web client.) It was meant to be a reply to <https://lists.sr.ht/~sircmpwn/sr.ht-discuss/%3C5af1e998f13b682847fd4c4e2f5b6a55fb7eea39%40disroot.org%3E>.

Also, s/message/ticket reply/.