Regarding Lagrange, it's possibly SDL-on-wayland related.

I had the same issue with Wesnoth, but after I changed the UI scaling settings in dwm, it worked correctly in sway. Both are SDL-based, so I tried the trick with Lagrange. No success.

BTW, Wesnoth works almost perfectly on the PP, if you have tiny fingers. Almost like 2022 is the Year Of Linux Gaming On The Phone.
--
Bruno

Hey! just for my information... Is it possible to mark a patch as superseded myself? How so? It would remove the burden of cleaning my own mess from the shoulders of the maintainers...

Thanks!

--
Bruno Dupuis

------- Original Message -------

On Tuesday, March 29th, 2022 at 11:34, Stacy Harper <contact@stacyharper.net> wrote:

> Marked as superseded

This allow theming for wob, but the patch has an issue: to get the
width of the screen we must wait until sway is started. It's not
the case when `.config/sxmo/profile` is sourced (in `sxmo_winit.sh`),
so `profile` is maybe not the best place.

3 solutions:
1. We keep this configuration into `profile`, as we did
   with BEMENU_OPTS (BTW bemenu could also need info from sway
   to fine tune its options). This is less flexible for the
   user but more simple.
2. We advise to use the start hook, as I did in the patch
3. We create a new hook that is sourced in sxmo_hook_start.sh
   before we start the daemons. This removes the need to 
   override hook_start for a bunch of env vars
[message trimmed]

Another patch in the set didn't pass shellcheck. I splitted
this out.

---
 scripts/core/sxmo_status.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/core/sxmo_status.sh b/scripts/core/sxmo_status.sh
index 374983b..45bfae8 100755
--- a/scripts/core/sxmo_status.sh
+++ b/scripts/core/sxmo_status.sh
@@ -54,7 +54,7 @@ del() {
 	shift
 
[message trimmed]

---
 configs/default_hooks/sxmo_hook_start.sh | 4 ++++
 scripts/core/sxmo_wob.sh                 | 4 +++-
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/configs/default_hooks/sxmo_hook_start.sh b/configs/default_hooks/sxmo_hook_start.sh
index 5270284..7b93c54 100644
--- a/configs/default_hooks/sxmo_hook_start.sh
+++ b/configs/default_hooks/sxmo_hook_start.sh
@@ -15,6 +15,10 @@ echo "unlock" > "$SXMO_STATE"
 
 case "$SXMO_WM" in
 	sway)
+		# This env var requires sxmo to be started, hence we can't set
[message trimmed]

---
 scripts/core/sxmo_status.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/core/sxmo_status.sh b/scripts/core/sxmo_status.sh
index 374983b..45bfae8 100755
--- a/scripts/core/sxmo_status.sh
+++ b/scripts/core/sxmo_status.sh
@@ -54,7 +54,7 @@ del() {
 	shift
 
 	if [ -z "$id" ]; then
-		printf "usage: %s rm <id>\n" "$(basename "$0")" >&2
+		printf "usage: %s del <id>\n" "$(basename "$0")" >&2
[message trimmed]

After updating, sxmo_migrate.sh in a ssh session was broken, with
a strange error:

    usr/share/sxmo/migrations/1.9.0.0.sh: cd: line 8: can't cd to /home/user/.config/.config/.config/sxmo/hooks/: No such file or directory

This patch fix the issue introduced by c3991d4.

I'm not sure why phartman changed this in a second second version of
the patch, it was correct in the first one. Maybe I'm missing something.

---
 configs/profile.d/sxmo_init.sh | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)
[message trimmed]

This is more than cosmetic. In sxmo_appmenu.sh this line:

```
CMD="$(printf "%s\n" "$CHOICES" | grep -m1 -F "$PICKED" | cut -d '^' -f3)"
```

finds the command of the first entry that has `$PICKED` prefix.
Before this change it would pick `$icon_cls Save and Quit`.

The bug is realy sxmo_appmenu.sh, but it's hard to change the
`grep -F` into a `grep -e "^$PICKED\s^ [01] ^"` while guarantying
nothing breaks.

---
[message trimmed]

If you want to apply and test this, please, be very careful, it's in pre-alpha stage,
I've tested this only with dummy data on my laptop, with sxmo_* scripts stubs. It may eat your
data. At least, crate a backup of ~/.cache/share/sxmo/modem before running
sxmo_migrate.sh.

On Wed Mar 2, 2022 at 2:30 PM CET, Hartman, Peter wrote:
> On Wed, Mar 02, 2022 at 12:57:20PM +0000, Bruno Dupuis wrote:
> >Hi, Peter.
> >
> >I've got a patch implementing a new storage file format, a
> >a migration script and a tailtextlog hook for the new format.
> >
> >I just struggle to test it, I don't want to test this on
> >my daily driver and I've just received a new motherboard for
> >my second PP. Anyway, I'll propose the patch soon as a WIP,
> >for you folks to test it, and hopefully I can test it in the
> >week.
>
> I'd be happy to test it.