This plays a tune through the PC speaker and is compatible with the
GRUB2 command of the same name.

Signed-off-by: Drew DeVault <sir@cmpwn.com>
---
 com32/modules/Makefile |  2 +-
 com32/modules/play.c   | 95 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 96 insertions(+), 1 deletion(-)
 create mode 100644 com32/modules/play.c

diff --git a/com32/modules/Makefile b/com32/modules/Makefile
index c01d6338..5e69a0d7 100644
--- a/com32/modules/Makefile
+++ b/com32/modules/Makefile
[message trimmed]

---
This should be the one. It addresses Nathaniel's feedback from IRC and
has been tested on new installs.

 setup-disk.in | 101 +++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 80 insertions(+), 21 deletions(-)

diff --git a/setup-disk.in b/setup-disk.in
index 5eb8638..71c2602 100644
--- a/setup-disk.in
+++ b/setup-disk.in
@@ -79,6 +79,26 @@ enumerate_fstab() {
 		done
 }
[message trimmed]

builds.sr.ht will automatically ensure that the correct sources URL is
added. By explicitly cloning the #website branch, we clone the repo
twice, which fails.
---
 .build.yml | 2 --
 1 file changed, 2 deletions(-)

diff --git a/.build.yml b/.build.yml
index 03148fa..e24746c 100644
--- a/.build.yml
+++ b/.build.yml
@@ -3,8 +3,6 @@ packages:
   - rsync
 secrets:
[message trimmed]

---
Note that this patch applies to the website branch.

 index.html | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/index.html b/index.html
index f3e0f2e..24f6db7 100644
--- a/index.html
+++ b/index.html
@@ -24,6 +24,12 @@
       portable shell scripts. Optional readline/libedit support is included to
       provide a comfortable interactive shell.
     </p>
[message trimmed]

Thanks!

To git.sr.ht:~sircmpwn/man.sr.ht
   1fd1ffb..1f83192  master -> master
 * [new tag]         0.14.2 -> 0.14.2

Thanks for investigating this. I'm pretty swamped at the moment so I
can't help much more.

I think I've found some records which hint at the problem, but I'm not
sure of the exact cause. Can you push some new commits and see if the
problem repeats itself?

On Mon Jan 20, 2020 at 3:10 PM, James Edwards-Jones wrote:
> Ah, my assumption was that making it zero size hides it until we
> re-size it later. It does work, but I'm guessing this approach will
> cause trouble later/elsewhere?

It actually doesn't work, or if it does, it'll only do so
non-deterministically.

> I attempted something along those lines, but saw a lot of
> configuration when the surfaces are first created. Obviously,
> attempting to destroy the surface without recreating seg-faulted.
> Should I extract surface creation to a method to repeat the
> configuration?

Thanks!

To git.sr.ht:~sircmpwn/scdoc
   8dfd30c..f092f83  master -> master

Can you clarify why this is wrong on Darwin? Do they not define
O_DIRECTORY? I'd rather define it to zero than add an ifdef with a
special Darwin case.