Offline
Hi,
I just wrote a PDF reader based on Chromium PDF engine.
It's pretty dummy, written in 30 minutes. But quite useful (at least for me) given the limitations of Document viewer, and that sturmreader is currently broken on noble.
I think it will greatly benefit from Qt6 upgrade, when it comes, because of improvement of the engine in chromium.
It is published in the openstore:
https://github.com/pparent76/chromiumpdf
Regards,
Pierre.
I've succeeded in opening the keyboard when editing text in signal-desktop !!!! 
I could do it reusing the work of @ChromiumOS-Guy in uWolf!
I can officially send signal messages from a native Ubuntu Touch app, without Waydroid! 
Even though for now it's not convenient , since keyboard hides the text area, and the enter button of the keyboard somehow does not trigger sending the message, and the "send" button in signal is hidden until you expend the editing area. But still very usable.
For now everything is messy, but I will try to make it available soon (I think it should be possible to do a click package, but I'm not sure yet)
Ps: Note that I could do it with Mobian version only for now.
Hi,
Just wanted to share that VLC 4 (not yet released) has an interface very close to VLC for Android.
I will be a very good candidate to port to Ubuntu Touch. The interface would probably be usable as such, but with few minimal patches to the interface it could probably be near perfect on the phone screen.
I don't have time to do that now, but probably later (possibly when release officially)
Hi,
Just wanted to mention that I contacted the in-development French identity mobile app, to ask them for interoperability, and remind them their duty to make sure public service is available for everybody, and that their role was not to enforce the choice of mobile OS of their citizens.
https://france-identite.gouv.fr/
Surprisingly I got a pretty open and positive reply, apologizing that their app was not compatible with my OS (Ubuntu Touch), and saying they shared the goal of interoperability, and would study that with the development team what was possible. I did not expect such a positive answer, although I don't know if anything concrete will get out of this.
As a side note I wonder why for this kind of app, with a real need of interoperability, they don't use more Qt, because from what I understand Qt allows to develop apps for Android, iOS, Windows, Mac, Linux desktop, Linux mobile (Ubuntu Touch, Sailfishos, Mobian, postMarketOs, pureOs, ect...), all at once, with only few custom things to go for each platform. Any thoughts on that?
Hi,
Just wanted to mention that a nice and easy addition to Ubuntu touch would be wifi captive portal via RFC 8910 and RFC 8908.
It is actually very simple:
1- When you get a DHCP lease in the wifi interface look for option 114
2- If there is a URL in option 114 it's an API, just make a simple HTTP GET request on this URL (no parameters or header)
3- The result will be a json, if it contains "captive"="true" and has an "user-portal-url" field then we should just open this "user-portal-url" in the browser and it will allow the user to authenticate to the captive portal and use the wifi!
Really I think that's any easy addition, I may try to contribute this inside the OS when I have time.
Ps: Adding a simple script in /etc/dhcp/dhclient-exit-hooks.d/ to do that should do the trick I guess.
Hi,
Today I just figured why not try:
sudo snap install signal-desktop
And I realize it just works! This is crazy. With this simple command I can link my device and see my chats. Well it is certainly not a responsive interface (but not terrible either, there's just the left menu coming in the way) and the keyboard does not show for now. But I can consult messages, and this is major.
It means:
1°) We can certainly create quite easily a version of Signal-Desktop that is adapted to mobile display (with few patches to the interface to make it responsive), and that we can get a full-fledge signal client on Ubtunu touch, that way.
2°) That Mir2.0 will certainly help with that.
Ps: I guess the first easy thing to make it usable as a consul-only stuff is adapt the scale, which I guess must be easily doable.
Anyway after all the investigations that I've made I'm extremely confident that we'll be able to have a fully responsive client based on Signal-Desktop landed in the openstore within few month. I see exactly how to make it perfectly adapted to the phone Interface like whatsweb. (Funny enough it is possible to use a modified version of the js script of whatsweb to make the interface responsive, because the interface has an extremely similar structure. )
It's not anymore a question of "if" but a question of "when". But there are few tedious tasks in the way, and I need to go little by little, because I can't dedicate all my time to this.
Of course I'm open to any offer for help and testing!
I've been able to recompile Signal-Desktop from source and apply my responsive script.
It is available in the Openstore alongside these other improvements:
-Fix startup from OpenStore
-Fix open url externaly
-Fix battery problem
-Fix double window problem
There still are some keyboard issues, it's probably the next thing I will work on.
They have not reached out to me again, but I would say every french citizen that use UT is entitled to contact them about this concern and the more they receive requests about that, the more they are likely to take it seriously. And I guess it is also possible to send them a message in the name of the fundation with an offer for assistance on the technical side.
https://france-identite.gouv.fr/contact/
Ps: By the way an argument that seems efficient to draw their attention lately, in the current context, is stressing that by not being inter-operable (as they are legally obliged) they are effectively offering a duopoly to 2 American companies, and preventing any potential alternative from emerging.
Craziest thing:
The calls are partially working out of the box!!!!
Which means we can certainly get calls working one day!
From what I know all the gtk/XWayland apps are affected. Something was probably changed in the system concerning maliit. ( Chromium for UT, Min browser, Signal UT, uWolf, Rocket.chat, ect... )
For now I cannot test 24.04-2.x, maybe later. Probably system devs should know more about that problem, and what may have been changed in regard to OSK and maliit.
Well if I run the xev from the Min browser in the terminal from the phone screen, somehow it gets killed while I navigate in the Min Browser. So I indeed cannot see the same log, but the cause seems to be totally different, simply that xev is not running anymore.
Anyway it seems better to me to run xev directly from adb/ssh as I showed before.
Yes but did you try to reproduce with the method I posted above?
I can't reproduce yours because xev is not installed by default in the system. I guess you have installed it.
This is very strange, just tu make sure I explain the process:
1- I start The Min Browser.
2- I open an adb shell session and I do the following:
phablet@ubuntu-phablet:~$ if [ "$DISPLAY" = "" ]; then
i=0
while [ -e "/tmp/.X11-unix/X$i" ] ; do
i=$(( i + 1 ))
done
i=$(( i - 1 ))
display=":$i"
export DISPLAY=$display
fi
cd /opt/click.ubuntu.com/min.pparent/current
export LD_LIBRARY_PATH=lib/aarch64-linux-gnu/
bin/xev -root
3- I try to open the top left hand corner menu, and exactly at that time I see the following log appear in the shell
CreateNotify event, serial 18, synthetic NO, window 0x3ef,
parent 0x3ef, window 0x600013, (-53,66), width 1189, height 1433
border_width 0, override YES
MapNotify event, serial 18, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600013, override YES
ConfigureNotify event, serial 18, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600013, (-54,302), width 1189, height 1433,
border_width 0, above 0x800003, override YES
PropertyNotify event, serial 18, synthetic NO, window 0x3ef,
atom 0x111 (_NET_ACTIVE_WINDOW), time 9737872, state PropertyNewValue
FocusOut event, serial 18, synthetic NO, window 0x3ef,
mode NotifyNormal, detail NotifyNonlinearVirtual
FocusIn event, serial 18, synthetic NO, window 0x3ef,
mode NotifyNormal, detail NotifyDetailNone
KeymapNotify event, serial 18, synthetic NO, window 0x0,
keys: 130 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
ConfigureNotify event, serial 18, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600004, (0,0), width 1080, height 2277,
border_width 0, above 0x600013, override NO
PropertyNotify event, serial 18, synthetic NO, window 0x3ef,
atom 0x111 (_NET_ACTIVE_WINDOW), time 9737873, state PropertyNewValue
ConfigureNotify event, serial 18, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600013, (-54,302), width 1189, height 1433,
border_width 0, above 0x600004, override YES
ConfigureNotify event, serial 19, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600013, (-54,302), width 1080, height 2277,
border_width 0, above 0x600004, override YES
UnmapNotify event, serial 19, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600013, from_configure NO
UnmapNotify event, serial 19, synthetic YES, window 0x3ef,
event 0x3ef, window 0x600013, from_configure NO
DestroyNotify event, serial 19, synthetic NO, window 0x3ef,
event 0x3ef, window 0x600013
PropertyNotify event, serial 19, synthetic NO, window 0x3ef,
atom 0x111 (_NET_ACTIVE_WINDOW), time 9737927, state PropertyNewValue
FocusOut event, serial 19, synthetic NO, window 0x3ef,
mode NotifyNormal, detail NotifyDetailNone
FocusIn event, serial 19, synthetic NO, window 0x3ef,
mode NotifyNormal, detail NotifyNonlinearVirtual
KeymapNotify event, serial 19, synthetic NO, window 0x0,
keys: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
@gpatel-fr said in XWayland: Menu windows resized full screen and crash?:
are you sure that the messages you see are coming from the menu ? What I tried is to redirect the output of xev to a file that I examined by tailing it from ssh. When the menu opened and did its particular dance, I saw nothing on my screen.
From my understanding, the window created when opening a menu is not rooted, so it's not seen when using xev -root (err, warning here, I am decidedly no X window expert)
Yes I'm sure that the window corresponding to the menu appears in xev, and in xdotools for a second. are you sure you have the exported the correct DISPLAY ? Everything seems to show that the window is instantaneously resized which is not normal for a menu window.
the window created when opening a menu is not rooted
What do you mean not rooted?
Apart of that, I have no idea on the uWolf mystery and it's indeed baffling. The only far fetched hypothesis that I can think is that this difference is not coming from the packaging but from the underlying code base (ie, Firefox vs Chrome) is not using Gtk in the same way.
Or that they have found a workaround to make it work? @chromiumos-guy @fuseteam ?
Anyone? Any info about that would of tremendous use for all Wayland/gtk apps ( Chromium UT, Min browser, Signal UT, Rocket.Chat )
@o.o. For the record, with Signal UT it's now possible to send and receive messages, photos, and audios in Signal UT without Waydroid, and to get standard UT notifications when a new message arrives even when the phone is locked.
Thank you very much for proposing, I currently have on top of my main phone:
*An old vollaphone (with broken screen, and I have datas on my waydroid session there that for now I would not like to relfash it for that reason)
*A pine-phone, that I wonder if I could use to try in-development versions (24.04.2)
*My old Nexus 5 that was my first UT device but completely outdated! 
What are you proposing?
@domubpkm said in French ID app interoperability.:
@DJac For that, I don't believe it. Community too small. No point wasting development and maintenance resources.
I do not agree:
First because interoperability is a legal requirement for public services, especially those supposed to be universal like ID. And they replied to me that they agreed on the goal of interoperability, were sorry about the issue, and were looking for solutions. ( https://eur-lex.europa.eu/eli/reg/2024/903/oj/eng )
Second because the french gouvernement is currently actively trying to reduce its dependency to foreign company (see above), in this case Apple, and Google who would have monopoly over a strategic state function as vital as identity, if this were to become a central way to verify identity.
Third because they don't need to develop a solution specific for Ubuntu Touch: they can release an App based on Qt framework, that with principally the same code will work on Windows, Mac, Linux Desktop, Android, iOS, and all Linux Mobile distributions (Ubuntu Touch, Mobian, postmarkeOs, SailfishOs, ect...); or even they can make a progressive webapp that nowadays can be compatible with NFC and will be universally compatible.
If we feel defeated before we even start the fight, or even send a short email, we will probably not go very far.
Edit: "The first step for institutions to think about it would be for the OS to be distributed pre-installed on a large scale. " -> On the other hand they will probably not do that because the state does not like this role to have direct influance on markets and choices of consumers.
Well I guess this type of gouvernemental organization will not be interested in individual feedback of individual, but it is consulting some associations, so with a lot of luck fondations like Ubports might have a word to say.