---
Should fix: https://todo.sr.ht/~sircmpwn/aerc2/449

Much simpler than using mutexes, and I think this should be sufficient.

By setting `term.closed = true` as first thing, there is no moment where
`termin.closed == false && term.cmd == nil` so we avoid crashes in pty.

 widgets/terminal.go | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/widgets/terminal.go b/widgets/terminal.go
index f47f671..6bee46d 100644
--- a/widgets/terminal.go
[message trimmed]

Co-authored-by: Kevin Kuehler <keur@xcf.berkeley.edu>
Co-authored-by: Reto Brunner <reto@labrat.space>
---
v5 also had issues, the likely culprit is the sr.ht email server. This
version should circumvent the problem for now. Sorry for the noise.

For future reference, pass this to git send-mail:
--transfer-encoding=quoted-printable

Repeating the change with v3:
* Rebase on master
* Remove custom subject matcher in favor of sortthread.GetBaseSubject
* Fix conrtols typo
* Fix sorting of threads with IMAP (now sorts by UID, i.e. thread with newest
[message trimmed]

---
 config/aerc.conf.in            |   6 +
 config/config.go               |   1 +
 doc/aerc-config.5.scd          |   6 +
 lib/format/format.go           |  12 +-
 lib/msgstore.go                |  55 +++++++-
 widgets/account.go             |   8 ++
 widgets/msglist.go             | 248 ++++++++++++++++++++++-----------
 worker/imap/open.go            |  63 +++++++++
 worker/imap/worker.go          |   9 +-
 worker/notmuch/lib/database.go | 121 ++++++++++++++--
 worker/notmuch/lib/thread.go   |  14 ++
 worker/notmuch/worker.go       |  36 ++++-
 worker/types/messages.go       |  10 ++
[message trimmed]

On Mon Jan 18, 2021 at 1:53 PM GMT, Drew DeVault wrote:
> On Mon Jan 18, 2021 at 8:24 AM EST, y0ast wrote:
> > Using the web interface of lists.sr.ht, I am hitting issues with CRLF
> > line endings when applying patches using the suggested command. For
> > example,
>
> I cannot apply this patch, but I also cannot apply it from aerc. It
> looks like the CRLF is present in the message body after decoding, which
> would not have been affected by lists.sr.ht processing (evidence: the
> DKIM signature is still valid).
>
> I suspect that "Apple Git-128" in the signature has more to do with it
> than anything else.

---
Testing a patch that's broken elsewhere on sr.ht

 config/aerc.conf.in            |   6 +
 config/config.go               |   1 +
 doc/aerc-config.5.scd          |   6 +
 lib/format/format.go           |  12 +-
 lib/msgstore.go                |  55 +++++++-
 widgets/account.go             |   8 ++
 widgets/msglist.go             | 248 ++++++++++++++++++++++-----------
 worker/imap/open.go            |  63 +++++++++
 worker/imap/worker.go          |   9 +-
 worker/notmuch/lib/database.go | 121 ++++++++++++++--
 worker/notmuch/lib/thread.go   |  14 ++
[message trimmed]

Hi,

Using the web interface of lists.sr.ht, I am hitting issues with CRLF
line endings when applying patches using the suggested command. For example,

https://lists.sr.ht/~sircmpwn/aerc/patches/19612

curl -s https://lists.sr.ht/~sircmpwn/aerc/patches/19612/mbox | git am -3

Leads to an error.

However directly applying the patch from an email (through aerc) works fine.

Could this have something to do with the updates to go-milter?

Co-authored-by: Kevin Kuehler <keur@xcf.berkeley.edu>
Co-authored-by: Reto Brunner <reto@labrat.space>
---
Somehow v4 had issues with whitespace, this updated version fixes that.

Repeating the change with v3:
* Rebase on master
* Remove custom subject matcher in favor of sortthread.GetBaseSubject
* Fix conrtols typo
* Fix sorting of threads with IMAP (now sorts by UID, i.e. thread with newest
 message on top)

 config/aerc.conf.in            |   6 +
 config/config.go               |   1 +
[message trimmed]

Hmm, you're right. Not sure what happened (used a standard git
send-mail), but I'll look into it and update the patch.

Joost

This patch is marked as applied on the patch web interface, but I don't see the commit in the repo.

The bug still exists on what's currently live on sr.ht too.

Was there an accidental mishap or is this as intended?

Joost

Co-authored-by: Kevin Kuehler <keur@xcf.berkeley.edu>
Co-authored-by: Reto Brunner <reto@labrat.space>
---
Compared to v3:
* Rebase on master
* Remove custom subject matcher in favor of sortthread.GetBaseSubject
* Fix conrtols typo
* Fix sorting of threads with IMAP (now sorts by UID, i.e. thread with newest
 message on top)

Thanks Daniel for testing and reporting :)
[message trimmed]