Notices
Page 9 of 72 « First 78 9 101119 Last »
Reply
Thread Tools
ruff is offline ruff
2016-03-22 , 22:19
Posts: 207 | Thanked: 482 times | Joined on Mar 2016
#81
Sent the PR. So the selector is implemented using floating dynamic component. This is the only missing piece from pebbeld's app config. I.e. I still see glitches on JS rendered popups - like color selector. I can select the color but the popup is highly distorted. But list now works fine (to me at least).

Also noted strange thing - daemon is returning version 4.0-1 even though I've repackaged RPM and pushed it. Correct me if I'm wrong but the version string is provided as -DVERSION so it cannot be stale in some file. Will try to dig more in the morning but with such behaviour the implication is - when app starts it checks daemon version and if it's lower than app's version - it restarts the daemon.
To end user it looks like app started, connected, disconnected, reconnected - then it's steady till the client is running.
 

The Following 5 Users Say Thank You to ruff For This Useful Post:
abranson is offline abranson
2016-03-22 , 23:08
Posts: 287 | Thanked: 862 times | Joined on Dec 2015
#82
Originally Posted by ruff View Post
Sent the PR. So the selector is implemented using floating dynamic component. This is the only missing piece from pebbeld's app config. I.e. I still see glitches on JS rendered popups - like color selector. I can select the color but the popup is highly distorted. But list now works fine (to me at least).
The new selector looks a lot nicer than the pebbled one. Nicely done.

Originally Posted by ruff View Post
Also noted strange thing - daemon is returning version 4.0-1 even though I've repackaged RPM and pushed it. Correct me if I'm wrong but the version string is provided as -DVERSION so it cannot be stale in some file. Will try to dig more in the morning but with such behaviour the implication is - when app starts it checks daemon version and if it's lower than app's version - it restarts the daemon.
To end user it looks like app started, connected, disconnected, reconnected - then it's steady till the client is running.
I was getting the same until I did a clean build. That's a bit disappointing. The version.pri is from Rockwork though - maybe that particular way of doing it doesn't work in SF?

Edit: 0.7-1 on openrepos with these changes. Version number seems to be ok.

Edit2: I figured out what qmake is for
Last edited by abranson; 2016-03-25 at 16:45.
 

The Following 6 Users Say Thank You to abranson For This Useful Post:
kombipom is offline kombipom
2016-03-23 , 00:50
Posts: 15 | Thanked: 12 times | Joined on Jun 2012
#83
I just wanted to thank abranson and ruff for their great work and rapid progress. I spent way too long playing with watchfaces last night.

Is the firmware update button just a placeholder at the moment? I'm still on 3.6.1 and pressing the firmware update doesn't seem to do anything.
 

The Following 4 Users Say Thank You to kombipom For This Useful Post:
ruff is offline ruff
2016-03-23 , 07:30
Posts: 207 | Thanked: 482 times | Joined on Mar 2016
#84
Originally Posted by kombipom View Post
Is the firmware update button just a placeholder at the moment? I'm still on 3.6.1 and pressing the firmware update doesn't seem to do anything.
It is a real button which is calling firmware update page - but that page hasn't yet been cleaned up - so it fails to load.
I don't see any specific dependencies on this page so will clean it up today.
 

The Following 4 Users Say Thank You to ruff For This Useful Post:
abranson is offline abranson
2016-03-23 , 07:32
Posts: 287 | Thanked: 862 times | Joined on Dec 2015
#85
Originally Posted by kombipom View Post
I just wanted to thank abranson and ruff for their great work and rapid progress. I spent way too long playing with watchfaces last night.
You're very welcome, but don't forget mzanetti's refactoring of smokku's core and javispedro's interfacing with sailfish's notifications, calendar, media player and clock. Lots of contributors on the Rockwork side too. I'm trying to maintain a full list in the readme that we can include in the about page, because I found the rockwork on a little lean.

Originally Posted by kombipom View Post
Is the firmware update button just a placeholder at the moment? I'm still on 3.6.1 and pressing the firmware update doesn't seem to do anything.
Yes, I think that may be the last remaining page! The only other functionality I think we're missing is re-ordering the apps in the watch menu. I'm not sure how that was done in the Rockwork QML though.

I've got another bug too that reminds me of a Pebbled one - the Misfit configuration page doesn't load properly because its URL is built on a base URL initialized in a callback which isn't being called. On pebbled, it was because that callback wasn't supported, but I think it might be a race here.
 

The Following 4 Users Say Thank You to abranson For This Useful Post:
ruff is offline ruff
2016-03-23 , 08:00
Posts: 207 | Thanked: 482 times | Joined on Mar 2016
#86
Sent PR for firmware update page. I cannot test it thoroughly bcz my fmw is recent. But as far as I can see it's really straightforward with all the dirty job being done in the daemon.

Edit: Will check misfit, never used that so don't really understand what is it all about
 

The Following 3 Users Say Thank You to ruff For This Useful Post:
abranson is offline abranson
2016-03-23 , 13:33
Posts: 287 | Thanked: 862 times | Joined on Dec 2015
#87
Thanks, building it now. Confirmed the version thing btw - changing the version.pri file doesn't update the build files, so the -DVERSION param doesn't change. So far the only way I've found to get it to update is to clean first.
 

The Following 3 Users Say Thank You to abranson For This Useful Post:
ronny_b is offline ronny_b
2016-03-23 , 14:52
Posts: 14 | Thanked: 28 times | Joined on Feb 2014 @ Germany
#88
Hello,

I also have a problem with the firmware-update. I cannot finish the recovery mode of the Pebble Time Round. The firmware-check of rockpoold announces the following:

Code:
FirmwareDownloader::checkForNewFirmware:169 - fetching firmware info: "https://pebblefw.s3.amazonaws.com/pebble/spalding/release-v3.8/latest.json" 
[D] FirmwareDownloader::checkForNewFirmware:175 - firmware info reply: "{
    "normal": {
        "friendlyVersion": "v3.10.1",
        ...
        "timestamp": 1457631509,
        "url": "https://pebblefw.s3.amazonaws.com/pebble/spalding/release-v3.8/pbz/Pebble-3.10.1-spalding.pbz"
    }
}
" 
[D] FirmwareDownloader::checkForNewFirmware:186 - current: "v3.2-prf5" candidate: "v3.10.1" 
[D] FirmwareDownloader::checkForNewFirmware:200 - Watch firmware is up to date 
[D] Pebble::slotUpdateAvailableChanged:687 - update available false ""
Please Help.

Ronny
 

The Following 2 Users Say Thank You to ronny_b For This Useful Post:
abranson is offline abranson
2016-03-23 , 15:06
Posts: 287 | Thanked: 862 times | Joined on Dec 2015
#89
Ok this firmware checking is a bit silly - does anyone see the point of trying to compare the watch firmware with the current to see if it's more recent? Shouldn't we just offer a flash if it's different? I can't see how the pebble servers would offer anything but the latest.

Edit. For some context - the RockWork firmware checking broke when pebble updated to 2.10.1 because string matching thinks that 2.9.1 is greater than 2.10.1. Michael did a much more complex check that turns the version number into three numbers and compares them individually, but that broke for v3.4 which only has two. I fixed that to consider 1, 2 or 3 numbers, but the recovery firmware has a dash in instead.

In 0.7-2 I've dropped all that and just made it offer an upgrade whenever your version is different from the candidate offered by pebble. Hopefully it'll fix your problem, Ronny, and stop all the silliness. Check on openrepos in a bit.
Last edited by abranson; 2016-03-23 at 15:25.
 

The Following 3 Users Say Thank You to abranson For This Useful Post:
ruff is offline ruff
2016-03-23 , 15:36
Posts: 207 | Thanked: 482 times | Joined on Mar 2016
#90
Can you just flash normal over prf? if yes then indeed there's no point comparing, just flash the offered one if it's not yet flashed
Otherwise we may make something similar to linux kernel version - converting it to long integer - ((a?a:0)<<24 | (b?b:0) << 16 | (c?c:0) << 8 | (d?d:0))

btw the one thing which is itching me the most (however i didn't have time to look at it yet) - is developer connection. Is it something present in the libpebble2 and thus could be implemented in the rockworkd/rockpoold?
 

The Following 2 Users Say Thank You to ruff For This Useful Post:
Reply
Page 9 of 72 « First 78 9 101119 Last »

Tags
pebble, smartwatch

« Previous Thread | Next Thread »

 
Forum Jump


All times are GMT. The time now is 15:32.

Contact Us - Home - Archive - Top