---
 contrib/crontab | 1 +
 1 file changed, 1 insertion(+)

diff --git a/contrib/crontab b/contrib/crontab
index a4acea47ceb6..080483e4ec96 100644
--- a/contrib/crontab
+++ b/contrib/crontab
@@ -22,6 +22,7 @@
  0 19 * * 5 chronic /usr/local/bin/submit_image_build -n "Weekly scheduled build image refresh" -a amd64 freebsd/13.x
  0 19 * * 6 chronic /usr/local/bin/submit_image_build -n "Weekly scheduled build image refresh" -a amd64 openbsd/latest
 30 19 * * 6 chronic /usr/local/bin/submit_image_build -n "Weekly scheduled build image refresh" -a x86_64 rockylinux/9
+ 0 19 * * 7 chronic /usr/local/bin/submit_image_build -n "Weekly scheduled build image refresh" -a amd64 freebsd/14.x
 
[message trimmed]

On Thursday, November 30th, 2023 at 22:18, Conrad Hoffmann <ch@bitfehler.net> wrote:

> Can you confirm that this is the same pattern and should be fixed also?
> 
> https://git.sr.ht/~sircmpwn/todo.sr.ht/tree/master/item/api/trackers/import.go#L190-206

Yup, looks like it.

This is not the first time I'm fixing one of these, and there may be more.

In case a panic happened, not doing anything will continue to
unwind the stack. Not doing anything is actually better because
that avoids loosing the original panic's stack trace.
---
 api/lists/middleware.go | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/api/lists/middleware.go b/api/lists/middleware.go
index f9da37da0b72..f7113a8a2b06 100644
--- a/api/lists/middleware.go
+++ b/api/lists/middleware.go
@@ -39,8 +39,6 @@ func ImportMailingListSpool(ctx context.Context, listID int, spool io.Reader) {
 		defer cancel()
 
[message trimmed]

On Thursday, November 30th, 2023 at 16:30, Conrad Hoffmann <ch@bitfehler.net> wrote:

> On 11/30/23 16:22, Simon Ser wrote:
> 
> > On Thursday, November 30th, 2023 at 16:18, Conrad Hoffmann ch@bitfehler.net wrote:
> > 
> > > 1. On export, we are modifying the patch email (adding the status
> > > header), breaking DKIM along the way, and sort of, I don't know,
> > > rewriting history?
> > 
> > Not really. Prepending extra header fields to an e-mail is a typical
> > thing to do to an email, the DKIM signature won't be broken because it
> > doesn't cover the new header field. For instance, lists.sr.ht will
> > prepend List-* and Sender, destination email servers will prepend

On Thursday, November 30th, 2023 at 16:18, Conrad Hoffmann <ch@bitfehler.net> wrote:

> 1. On export, we are modifying the patch email (adding the status
> header), breaking DKIM along the way, and sort of, I don't know,
> rewriting history?

Not really. Prepending extra header fields to an e-mail is a typical
thing to do to an email, the DKIM signature won't be broken because it
doesn't cover the new header field. For instance, lists.sr.ht will
prepend List-* and Sender, destination email servers will prepend
Received, etc.

> 2. If this change goes live as is, imports of data that was created
> before this change will not import any patchset status

Let's just use the toplevel go.mod instead.

This upgrades core-go for the worker, so a small adjustment of
JobContext.oauth2Token is required.
---
 go.mod          |  10 +
 go.sum          |  34 ++
 worker/go.mod   |  23 --
 worker/go.sum   | 837 ------------------------------------------------
 worker/tasks.go |   4 +-
 5 files changed, 46 insertions(+), 862 deletions(-)
 delete mode 100644 worker/go.mod
 delete mode 100644 worker/go.sum
[message trimmed]

v1 is not maintained anymore.
---
 go.mod           | 2 +-
 go.sum           | 4 ++--
 server/server.go | 4 ++--
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/go.mod b/go.mod
index 987ca5da69a6..875aef83b5c9 100644
--- a/go.mod
+++ b/go.mod
@@ -15,7 +15,7 @@ require (
 	github.com/emersion/go-sasl v0.0.0-20220912192320-0145f2c60ead
 	github.com/emersion/go-smtp v0.18.0
[message trimmed]

Checking context.Cancelled is not enough, I also see the same issue
with context.DeadlineExceeded:

    panic: Transaction error: driver: bad connection
    Closure error: context deadline exceeded

    goroutine 66664 [running]:
    git.sr.ht/~sircmpwn/core-go/database.WithTx({0x5578e1847a58, 0xc0002c44e0}, 0x50?, 0xc000285f38)
            git.sr.ht/~sircmpwn/core-go@v0.0.0-20231129165057-e7b8e02696a1/database/middleware.go:74 +0x2ce
    git.sr.ht/~sircmpwn/builds.sr.ht/api/loaders.fetchUsersByID.func1({0xc0002cf190, 0x1, 0x1})
            git.sr.ht/~sircmpwn/builds.sr.ht/api/loaders/middleware.go:34 +0x118
    git.sr.ht/~sircmpwn/builds.sr.ht/api/loaders.(*usersByIDLoaderBatch).end(...)
            git.sr.ht/~sircmpwn/builds.sr.ht/api/loaders/usersbyidloader_gen.go:222
    git.sr.ht/~sircmpwn/builds.sr.ht/api/loaders.(*usersByIDLoaderBatch).startTimer(0xc00044a360, 0xc00042d380)
[message trimmed]

---
 doc/hut.1.scd |  2 +-
 import.go     | 20 +++++++++++++-------
 2 files changed, 14 insertions(+), 8 deletions(-)

diff --git a/doc/hut.1.scd b/doc/hut.1.scd
index f8581c89473b..1b1ce0ee60dd 100644
--- a/doc/hut.1.scd
+++ b/doc/hut.1.scd
@@ -86,7 +86,7 @@ Additionally, mailing lists can be referred to by their email address.
 *export* <directory>
 	Export account data.
 
-*import* <directory>
[message trimmed]

I've addressed the remaining points, and pushed a version with these changes:

- The filepath.Walk was effectively iterating recursively over all files in the
  imported directory. This is unnecessary, no need to walk a dir if there is an
  info.json there.
- The PGP import was silently ignoring malformed files. A warning is now printed.
- Added an error in case we find nothing to import in the directory.

Thanks!