Re: A few changes to the QML UI

[Christian Schneider]

Also after exchanging the battery, the bluelink interface is still dead.
I will get a replacement in 1-2 weeks and I will follow up on this then.
BR, Christian

Am 14.05.2018 um 15:41 schrieb Christian Schneider:
> When I tried again, it turned out, that the bluelink device is not working 
> anymore. I hope it's just the battery, but since I'm not at home, I'm not 
> sure if I find a matching battery somewhere.
> I will report back, when/if I can provide further logs.
> Br, Christian
> 
> Am 12. Mai 2018 07:07:23 OEZ schrieb Dirk Hohndel :
>>
>>> On May 11, 2018, at 9:53 PM, Christian Schneider 
>> wrote:
>>>
>>> I tried, but in the log, I can see 'not recognized as dive computer'
>> for the paired mares device.
>>
>> Outstanding. This is why we need good testers!
>>
>> The log shows:
>>
>> Found new device: "Mares bluelink pro" "LE:00:1A:85:E0:0B:FA"
>> Not recognized as dive computer
>>
>> Which means that we need to add this name to the "special handling"
>> section.
>>
>> But even with that, you should be able to pick Mares / Quad (or
>> whatever model
>> you have) and that not recognized LE:00... entry. And try to download.
>>
>> I'd love to see the log for THAT.
>>
>> Thanks
>>
>> /D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Long, Martin]

I've spotted the pattern

First of all, I kill the app (by swiping from the task list).

To start with, regardless of which dive I tap, it always opens the very top
dive (the last one I did).

Now, if I swipe that dive left, to move to the next one down, then tap the
back button to return to the dive list...

The behaviour now changes. With each tap of a dive (regardless of which
one), it will keep moving down the list, opening then next one in sequence.

If instead of using the back button, I swipe back through the dives, until
I reach the top, and then right again to get the list... then the original
behaviour is reverted (any tap always opens the most recent dive)

Sounds like it relates to the back stack on Android, and possibly some
confusion/difference between what happens when swiping versus using the
back button.


On 14 May 2018 at 21:56, Dirk Hohndel  wrote:

>
> > On May 14, 2018, at 1:40 PM, Long, Martin  wrote:
> >
> > Aside from the jerky/weird scrolling, I'm getting the old issue where
> whichever dive I "tap" on the list, a random one seems to be show. It
> doesn't *seem* to be deterministic. Tapping the same dive over and over can
> result in different ones showing. They're not always necessarily from the
> same trip. it's quite odd.
>
> Every time I think we have this one fixed it comes back. Not the slightest
> clue.
>
> /D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Long, Martin]

Foiled. I'm using this forum to drop encoded messages to the NSA (or
perhaps MI5) in Haitian Creole, disguised as English (that be correct
English English). We were one step away from catching the eBay Rambler. You
might be interested to know my wife is a native Creole speaker, but
Seychelles Kreol, not Haitian.

On 14 May 2018 at 21:58, Dirk Hohndel  wrote:

>
> On May 14, 2018, at 1:50 PM, Linus Torvalds 
> wrote:
>
> I have no answer for you, but I found it hilarious that apparently Google
> mail thinks you're speaking "Haitian Creole". Don't ask me why.
>
> Original:
>
> [image: orig.png]
>
> so obviously I had to ask google to "translate" your incomprehensible
> ramblings to me:
>
> [image: translated.png]
> Apparently you didn't write enough "Rambler eBay" in your original
> message, and apparently your capitalization needed some work too. Very odd.
>
> Sorry that I can't help you with your actual problem, but I felt I had to
> share this language issue with you.
>
>
> Well, Martin has a .uk domain, so it seems reasonable to assume that he
> doeth not speaka de English correctedly
>
> /D
>
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On May 14, 2018, at 1:50 PM, Linus Torvalds  
> wrote:
> 
> I have no answer for you, but I found it hilarious that apparently Google 
> mail thinks you're speaking "Haitian Creole". Don't ask me why.
> 
> Original:
> 
> 
> 
> so obviously I had to ask google to "translate" your incomprehensible 
> ramblings to me:
> 
> 
> Apparently you didn't write enough "Rambler eBay" in your original message, 
> and apparently your capitalization needed some work too. Very odd.
> 
> Sorry that I can't help you with your actual problem, but I felt I had to 
> share this language issue with you.

Well, Martin has a .uk domain, so it seems reasonable to assume that he doeth 
not speaka de English correctedly

/D___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On May 14, 2018, at 1:40 PM, Long, Martin  wrote:
> 
> Aside from the jerky/weird scrolling, I'm getting the old issue where 
> whichever dive I "tap" on the list, a random one seems to be show. It doesn't 
> *seem* to be deterministic. Tapping the same dive over and over can result in 
> different ones showing. They're not always necessarily from the same trip. 
> it's quite odd. 

Every time I think we have this one fixed it comes back. Not the slightest clue.

/D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Linus Torvalds]

On Mon, May 14, 2018 at 1:40 PM Long, Martin  wrote:
>
> Aside from the jerky/weird scrolling, I'm getting the old issue [..]

I have no answer for you, but I found it hilarious that apparently Google
mail thinks you're speaking "Haitian Creole". Don't ask me why.

Original:

[image: orig.png]

so obviously I had to ask google to "translate" your incomprehensible
ramblings to me:

[image: translated.png]
Apparently you didn't write enough "Rambler eBay" in your original message,
and apparently your capitalization needed some work too. Very odd.

Sorry that I can't help you with your actual problem, but I felt I had to
share this language issue with you.

   Linus
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Long, Martin]

Aside from the jerky/weird scrolling, I'm getting the old issue where
whichever dive I "tap" on the list, a random one seems to be show. It
doesn't *seem* to be deterministic. Tapping the same dive over and over can
result in different ones showing. They're not always necessarily from the
same trip. it's quite odd.

On 14 May 2018 at 14:41, Christian Schneider  wrote:

> When I tried again, it turned out, that the bluelink device is not working
> anymore. I hope it's just the battery, but since I'm not at home, I'm not
> sure if I find a matching battery somewhere.
> I will report back, when/if I can provide further logs.
> Br, Christian
>
> Am 12. Mai 2018 07:07:23 OEZ schrieb Dirk Hohndel :
> >
> >> On May 11, 2018, at 9:53 PM, Christian Schneider 
> >wrote:
> >>
> >> I tried, but in the log, I can see 'not recognized as dive computer'
> >for the paired mares device.
> >
> >Outstanding. This is why we need good testers!
> >
> >The log shows:
> >
> >Found new device: "Mares bluelink pro" "LE:00:1A:85:E0:0B:FA"
> >Not recognized as dive computer
> >
> >Which means that we need to add this name to the "special handling"
> >section.
> >
> >But even with that, you should be able to pick Mares / Quad (or
> >whatever model
> >you have) and that not recognized LE:00... entry. And try to download.
> >
> >I'd love to see the log for THAT.
> >
> >Thanks
> >
> >/D
> ___
> subsurface mailing list
> subsurface@subsurface-divelog.org
> http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
>
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Christian Schneider]

When I tried again, it turned out, that the bluelink device is not working 
anymore. I hope it's just the battery, but since I'm not at home, I'm not sure 
if I find a matching battery somewhere.
I will report back, when/if I can provide further logs.
Br, Christian

Am 12. Mai 2018 07:07:23 OEZ schrieb Dirk Hohndel :
>
>> On May 11, 2018, at 9:53 PM, Christian Schneider 
>wrote:
>> 
>> I tried, but in the log, I can see 'not recognized as dive computer'
>for the paired mares device.
>
>Outstanding. This is why we need good testers!
>
>The log shows:
>
>Found new device: "Mares bluelink pro" "LE:00:1A:85:E0:0B:FA"
>Not recognized as dive computer
>
>Which means that we need to add this name to the "special handling"
>section.
>
>But even with that, you should be able to pick Mares / Quad (or
>whatever model
>you have) and that not recognized LE:00... entry. And try to download.
>
>I'd love to see the log for THAT.
>
>Thanks
>
>/D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On May 11, 2018, at 9:53 PM, Christian Schneider  wrote:
> 
> I tried, but in the log, I can see 'not recognized as dive computer' for the 
> paired mares device.

Outstanding. This is why we need good testers!

The log shows:

Found new device: "Mares bluelink pro" "LE:00:1A:85:E0:0B:FA"
Not recognized as dive computer

Which means that we need to add this name to the "special handling" section.

But even with that, you should be able to pick Mares / Quad (or whatever model
you have) and that not recognized LE:00... entry. And try to download.

I'd love to see the log for THAT.

Thanks

/D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Christian Schneider]

I tried, but in the log, I can see 'not recognized as dive computer' for the 
paired mares device.
Br, Christian
Ps: I removed other paired BT devices from log

Am 15. April 2018 03:52:23 MESZ schrieb Dirk Hohndel :
>
>I have pushed new versions to both Android and iOS testers.
>
>- experimental, still completely untested Mares BLE support (I copied a
>few people who I believe have access to one of those)
>- ability to share the log file on iOS (easiest way is to use the Mail
>app and attach the log file and send it to yourself - then you can look
>at it...)
>- folding dive list is back (and this time it seems to work much
>better)
>- on Android I still see an occasional weird scrolling behavior - I
>don't see this on iOS
>- when editing dives "Use current GPS location" should now actually
>work (this has been broken since forever)
>
>(see what happens if I get to go diving? - Y'all need to make sure I go
>diving more frequently... just saying)
>
>For all of these changes I of course would love to see some testing.
>You can add yourself to the Android testing group here:
>https://play.google.com/apps/testing/org.subsurfacedivelog.mobile
>
>For iOS please send me an email so I can add you to the tester list (no
>self service there, sorry)
>
>/D 


subsurface.log.md
Description: Binary data
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Christian Schneider]

Thanks for notifying me. I'm happy to help with testing Mares BLE, but
unfortunately, I'm quite busy right now. Maybe I can give it a try next
week.
BR, CHristian

Am 15.04.2018 um 03:52 schrieb Dirk Hohndel:
> 
> I have pushed new versions to both Android and iOS testers.
> 
> - experimental, still completely untested Mares BLE support (I copied a
> few people who I believe have access to one of those)
> - ability to share the log file on iOS (easiest way is to use the Mail
> app and attach the log file and send it to yourself - then you can look
> at it...)
> - folding dive list is back (and this time it seems to work much better)
> - on Android I still see an occasional weird scrolling behavior - I
> don't see this on iOS
> - when editing dives "Use current GPS location" should now actually work
> (this has been broken since forever)
> 
> (see what happens if I get to go diving? - Y'all need to make sure I go
> diving more frequently... just saying)
> 
> For all of these changes I of course would love to see some testing. You
> can add yourself to the Android testing group
> here: https://play.google.com/apps/testing/org.subsurfacedivelog.mobile
> For iOS please send me an email so I can add you to the tester list (no
> self service there, sorry)
> 
> /D 
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Miika Turkia]

On Sun, Apr 15, 2018 at 10:20 PM, Dirk Hohndel  wrote:

> The latest build that I pushed this morning seems to behave much better,
> even on my old phone where I could "kind of" reproduce what you describe.
>
> Did you have a chance to try it?
>

Unfortunately I still have jerkynes when scrolling on iPhone 5S. The log
seems to have a couple of "QObject: startTimer: Timers cannot be started
from another thread" messages. No idea if this is related or not. I think
the scrolling has improved quite a bit but still very painful on this
device.

Scrolling the debug log works fine when going to one direction, but when
switching it has a brief moment of immobility. The location at the top of
the debug log is either shown or hidden when switching direction and this
might be related to the problem. Scrolling pauses once during the
transition and once it is completed the scrolling to this direction works
fine.

miika
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On Apr 15, 2018, at 1:24 PM, Linus Torvalds  
> wrote:
> 
> On Sun, Apr 15, 2018, 13:19 Dirk Hohndel  > wrote:
> 
> If you feel strongly about being able to have multiple trips open that should 
> not be hard. 
> 
> I don't feel strongly at all. I think it was more of a reaction from "this is 
> different from what I expected from the desktop version".
> 
> So as far as I'm concerned, leave it as is until somebody comes up with a 
> good reason to have multiple trips open.

Cool, will do.

BTW, I also fixed the "no SAC rate shown" bug that we noticed last weekend.
This now also works as expected - after you download, when you update the 
cylinder information, the derived info is updated as well.

/D

___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Linus Torvalds]

On Sun, Apr 15, 2018, 13:19 Dirk Hohndel  wrote:

>
> If you feel strongly about being able to have multiple trips open that
> should not be hard.
>

I don't feel strongly at all. I think it was more of a reaction from "this
is different from what I expected from the desktop version".

The mobile behavior might indeed be better for the small screen.

The only issue I see is if you would want to compare dives in two different
trips, but I'm not sure that's an actual issue at all. The mobile version
doesn't show the kinds of things that I normally look at in the dive list
anyway (ie SAC rate etc) so you have to go into the dives for that anyway.

So as far as I'm concerned, leave it as is until somebody comes up with a
good reason to have multiple trips open.

  Linus
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

Sorry, on the phone.
That was an intentional design decision to simplify things for the mobile UI.
If you feel strongly about being able to have multiple trips open that should 
not be hard. I somehow thought this was better for the small screen (as the 
goal was to make it easier for people to get a quick overview)
What do you think?

/D

On April 15, 2018 1:00:22 PM PDT, Linus Torvalds 
 wrote:
>On Sun, Apr 15, 2018, 12:53 Linus Torvalds
>
>wrote:
>
>>
>> I'll go check on my tablet too, but I'm assuming it's fixed there as
>well.
>> If you hear nothing, assume it's all good.
>>
>
>All good there too.
>
>But now that I can actually scroll around and test, I notice that I can
>only have at most one trip expanded.
>
>That works, and maybe it's the simplest model, but my UI expectations
>were
>that I can expand and collapse trips independently of each other.
>
>So not sure this is a problem or not. Perhaps more of a "that wasn't
>what I
>expected".
>
>Still very lovely, and makes it much easier to find a dive.
>
>  Linus

-- 
from my phone.___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Linus Torvalds]

On Sun, Apr 15, 2018, 12:53 Linus Torvalds 
wrote:

>
> I'll go check on my tablet too, but I'm assuming it's fixed there as well.
> If you hear nothing, assume it's all good.
>

All good there too.

But now that I can actually scroll around and test, I notice that I can
only have at most one trip expanded.

That works, and maybe it's the simplest model, but my UI expectations were
that I can expand and collapse trips independently of each other.

So not sure this is a problem or not. Perhaps more of a "that wasn't what I
expected".

Still very lovely, and makes it much easier to find a dive.

  Linus
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

The latest build that I pushed this morning seems to behave much better, even 
on my old phone where I could "kind of" reproduce what you describe.

Did you have a chance to try it?

/D

On April 14, 2018 10:11:31 PM PDT, Linus Torvalds 
 wrote:
>On Sat, Apr 14, 2018, 21:08 Linus Torvalds
>
>wrote:
>
>>
>> It's very clear, and happens both with all trips collapsed and with
>one or
>> more trips exposed. I can scroll down once, but when I try to scroll
>> further, at first nothing happens at all, then it resets to an
>earlier
>> state (maybe back to the top) and then it scrolls down some more.
>>
>
>Ok, so there is a pattern to the oddity.
>
>I tried to move my finger very slowly and keep track of where my finger
>was
>and which entry I was touching.
>
>I may not be great at explaining the behavior I see, but imagine that
>you
>have all trips collapsed, and the first scroll event is you touching a
>trip
>right in the middle of the screen, and slowly moving your finger up.
>Everything works fine.
>
>It at least *almost* fine. I think there is an off-by-one error, where
>when
>I put my finger down tp before I move it, the dive list jumps by one
>entry.
>So there is a small jerk event happening even there, but it's almost
>unnoticeable. The scrolling itself is fairly smooth once I start moving
>my
>finger.
>
>Then, release the finger, and start the next scroll event lower down
>(say,
>three quarters of to the bottom), and moved your finger up to scroll.
>
>During that second scroll event, *nothing* happens in the screen for me
>at
>first, until my finger hits the point where I started the *first*
>scroll
>(so I moved my finger from 3/4 down the screen up to the halfway point
>without anything moving at all). But when I hit that point where I
>started
>my previous scroll, the dive log screen resets to where it was the
>*last*
>scroll event, and then starts scrolling with my finger again.
>
>So I never and up making any actual progress, and it feels really jerky
>because part of the time the screen doesn't react to the finger at all,
>and
>then when you pass a certain point it jumps and starts reacting again,
>just
>at the wrong place.
>
>So there seems to be some "screen offset" thing going on, with some
>history.
>
>Except *occasionally* the second scroll event works fine. So I can make
>some progress every once in a while. I didn't figure out the pattern
>for
>that, though.
>
>   Linus
>
>>

-- 
from my phone.___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Joakim Bygdell]

I see a lot of jerkyness when scrolling, seems like it always restarts at
the top.
Cant' really scroll in small steps, to get beyond the top display I need to
flick really hard.
And then the scrolling will move in steps before it finally comes to a full
stop.

Can't get a cloud resync to happen by scrolling up from the top of the list
due to position reset issues.

/ Jocke
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Linus Torvalds]

On Sat, Apr 14, 2018, 21:08 Linus Torvalds 
wrote:

>
> It's very clear, and happens both with all trips collapsed and with one or
> more trips exposed. I can scroll down once, but when I try to scroll
> further, at first nothing happens at all, then it resets to an earlier
> state (maybe back to the top) and then it scrolls down some more.
>

Ok, so there is a pattern to the oddity.

I tried to move my finger very slowly and keep track of where my finger was
and which entry I was touching.

I may not be great at explaining the behavior I see, but imagine that you
have all trips collapsed, and the first scroll event is you touching a trip
right in the middle of the screen, and slowly moving your finger up.
Everything works fine.

It at least *almost* fine. I think there is an off-by-one error, where when
I put my finger down tp before I move it, the dive list jumps by one entry.
So there is a small jerk event happening even there, but it's almost
unnoticeable. The scrolling itself is fairly smooth once I start moving my
finger.

Then, release the finger, and start the next scroll event lower down (say,
three quarters of to the bottom), and moved your finger up to scroll.

During that second scroll event, *nothing* happens in the screen for me at
first, until my finger hits the point where I started the *first* scroll
(so I moved my finger from 3/4 down the screen up to the halfway point
without anything moving at all). But when I hit that point where I started
my previous scroll, the dive log screen resets to where it was the *last*
scroll event, and then starts scrolling with my finger again.

So I never and up making any actual progress, and it feels really jerky
because part of the time the screen doesn't react to the finger at all, and
then when you pass a certain point it jumps and starts reacting again, just
at the wrong place.

So there seems to be some "screen offset" thing going on, with some
history.

Except *occasionally* the second scroll event works fine. So I can make
some progress every once in a while. I didn't figure out the pattern for
that, though.

   Linus

>
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On Apr 14, 2018, at 9:27 PM, Miika Turkia  wrote:
> 
> I have the same scrolling jerkyness as Linus on iPhone 5S. iPhone 8 is a lot 
> better but as I know about the problem, I can see it there as well. Feels 
> like the problem might be related to scrolling to next trip. Maybe processing 
> all the dives within a trip?

We are simply making all the dives that we aren't showing (a) invisible and (b) 
have a height of 0.
I don't see how that would be MORE work than not doing that and actually having 
to render the dives.

/D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Miika Turkia]

On Sun, Apr 15, 2018 at 7:12 AM, Dirk Hohndel  wrote:

>
> On Apr 14, 2018, at 9:08 PM, Linus Torvalds 
> wrote:
>
>
>
> On Sat, Apr 14, 2018, 20:37 Dirk Hohndel  wrote:
>
>>
>> That was as easy as I thought. New test binary for Android is on its way
>> (didn't do a new iOS app)
>>
>
> Updated again, and now the trips close too. Lovely.
>
> Except the scrolling is still very broken.
>
> I tested on my old Samsung tablet too, and it shows the same behavior as
> my phone. So both a Pixel 2 and a Samsung Tab S2 show the problem for me.
>
>
> So I guess I'm just lucky that it works on my Pixel XL, iPad Pro and
> Samsung S3 without a flaw.
>

I have the same scrolling jerkyness as Linus on iPhone 5S. iPhone 8 is a
lot better but as I know about the problem, I can see it there as well.
Feels like the problem might be related to scrolling to next trip. Maybe
processing all the dives within a trip?

miika
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Linus Torvalds]

On Sat, Apr 14, 2018, 20:37 Dirk Hohndel  wrote:

>
> That was as easy as I thought. New test binary for Android is on its way
> (didn't do a new iOS app)
>

Updated again, and now the trips close too. Lovely.

Except the scrolling is still very broken.

I tested on my old Samsung tablet too, and it shows the same behavior as my
phone. So both a Pixel 2 and a Samsung Tab S2 show the problem for me.

It's very clear, and happens both with all trips collapsed and with one or
more trips exposed. I can scroll down once, but when I try to scroll
further, at first nothing happens at all, then it resets to an earlier
state (maybe back to the top) and then it scrolls down some more.

This was what used to happen for a while, and then I was going to show you
last weekend when diving, but you had somehow magically fixed it. So it
worked for a week, but then the first update today brought the problem
back, and the updates after haven't fixed it.

So it seems very reproducible, but the fact that it was magically fixed for
a week may mean that there is some really odd trigger, that just isn't
obvious. Was there something else you changed in the build?

But you definitely did make the trip collapsing work. The trip thing makes
me very excited about this, because if the scrolling just worked, it would
really be a very nice feature.

If it is somehow data dependent, feel free to try with my dives. We used to
have that shared git repository, now you'd have to clone my cloud repo.

 Linus

>
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On Apr 14, 2018, at 8:38 PM, Federico Masias  wrote:
> 
> A very quick test on my end (6S running 11.2.6). Dives that are not in a trip 
> are not shown at all. Also, opening one trip collapses the open trip- is this 
> expected behavior? Some jerkiness with a small log (26 dives... I should go 
> dive soon...) 

Thanks for that report. My log didn't have dives outside of trips. Silly me.
Should be easy to fix.

/D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Federico Masias]

On Sat, Apr 14, 2018 at 23:30 Dirk Hohndel  wrote:

> On Apr 14, 2018, at 7:44 PM, Linus Torvalds 
> wrote:
>
> On Sat, Apr 14, 2018, 18:52 Dirk Hohndel  wrote:
>
>>
>> I have pushed new versions to both Android and iOS testers.
>>
>> - folding dive list is back (and this time it seems to work much better)
>>
>
> I love the folding dive list, although it's seems to have performance
> problems.
>
>
> That surprises me - I use it with my reasonably large dive list on an
> older and slower phone than you.
> And I tried on an even older phone where it was tolerable.
>
> Or maybe scrolling up triggers "refresh" and another cloud sync? It's very
> jerky, and I seem to get the cloud sync message several times.
>
>
> The one situation that seems to trigger weirdness appears to be if you
> have a trip with a lot of dives and you try to scroll down really fast,
> flicking your finger. But slowing down seems to make it all work.
>
> I also seem unable to collapse a dive trip after I've exposed it. But it's
> *so* jerky that it's hard to tell.
>
>
> You can not close all trips again - that's something that would be easy to
> add, though.
>
> Maybe it's not performance, and it's just a symptom of the scrolling bug:
>
> - on Android I still see an occasional weird scrolling behavior - I don't
>> see this on iOS
>>
>
> It was fine for a while, but it's back with a vengeance now.
>
>
> The odd thing is that for the exact same binary some people seem to see it
> and others don't.
>
> The collapsing dive trip feature is very very promising, but there is
> definitely something seriously wrong with the dive list. I'm assuming it's
> all kirigami under the hood?
>
>
> Yes, all Kirigami, and all really straight forward stuff.
>
> I'll admit I'm disappointed that this is working so poorly for you. I
> played with this quite a bit on Android and iOS and it looked really nice...
>
> /D
>
> ___
> subsurface mailing list
> subsurface@subsurface-divelog.org
> http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface


A very quick test on my end (6S running 11.2.6). Dives that are not in a
trip are not shown at all. Also, opening one trip collapses the open trip-
is this expected behavior? Some jerkiness with a small log (26 dives... I
should go dive soon...)

Im in the back of an uber so I don’t have a ton of testing time but I’ll
take a better look tomorrow morning

-Federico
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On Apr 14, 2018, at 8:30 PM, Dirk Hohndel  wrote:
> 
> You can not close all trips again - that's something that would be easy to 
> add, though.

That was as easy as I thought. New test binary for Android is on its way 
(didn't do a new iOS app)

/D
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Dirk Hohndel]

> On Apr 14, 2018, at 7:44 PM, Linus Torvalds  
> wrote:
> 
> On Sat, Apr 14, 2018, 18:52 Dirk Hohndel  > wrote:
> 
> I have pushed new versions to both Android and iOS testers.
> 
> - folding dive list is back (and this time it seems to work much better)
> 
> I love the folding dive list, although it's seems to have performance 
> problems. 

That surprises me - I use it with my reasonably large dive list on an older and 
slower phone than you.
And I tried on an even older phone where it was tolerable.

> Or maybe scrolling up triggers "refresh" and another cloud sync? It's very 
> jerky, and I seem to get the cloud sync message several times.

The one situation that seems to trigger weirdness appears to be if you have a 
trip with a lot of dives and you try to scroll down really fast, flicking your 
finger. But slowing down seems to make it all work.

> I also seem unable to collapse a dive trip after I've exposed it. But it's 
> *so* jerky that it's hard to tell.

You can not close all trips again - that's something that would be easy to add, 
though.

> Maybe it's not performance, and it's just a symptom of the scrolling bug:
> 
> - on Android I still see an occasional weird scrolling behavior - I don't see 
> this on iOS
> 
> It was fine for a while, but it's back with a vengeance now. 

The odd thing is that for the exact same binary some people seem to see it and 
others don't.

> The collapsing dive trip feature is very very promising, but there is 
> definitely something seriously wrong with the dive list. I'm assuming it's 
> all kirigami under the hood?

Yes, all Kirigami, and all really straight forward stuff.

I'll admit I'm disappointed that this is working so poorly for you. I played 
with this quite a bit on Android and iOS and it looked really nice...

/D

___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

Re: A few changes to the QML UI

[Linus Torvalds]

On Sat, Apr 14, 2018, 18:52 Dirk Hohndel  wrote:

>
> I have pushed new versions to both Android and iOS testers.
>
> - folding dive list is back (and this time it seems to work much better)
>

I love the folding dive list, although it's seems to have performance
problems.

Or maybe scrolling up triggers "refresh" and another cloud sync? It's very
jerky, and I seem to get the cloud sync message several times.

I also seem unable to collapse a dive trip after I've exposed it. But it's
*so* jerky that it's hard to tell.

Maybe it's not performance, and it's just a symptom of the scrolling bug:

- on Android I still see an occasional weird scrolling behavior - I don't
> see this on iOS
>

It was fine for a while, but it's back with a vengeance now.

The collapsing dive trip feature is very very promising, but there is
definitely something seriously wrong with the dive list. I'm assuming it's
all kirigami under the hood?

 Linus
___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface

A few changes to the QML UI

[Dirk Hohndel]

I have pushed new versions to both Android and iOS testers.

- experimental, still completely untested Mares BLE support (I copied a few 
people who I believe have access to one of those)
- ability to share the log file on iOS (easiest way is to use the Mail app and 
attach the log file and send it to yourself - then you can look at it...)
- folding dive list is back (and this time it seems to work much better)
- on Android I still see an occasional weird scrolling behavior - I don't see 
this on iOS
- when editing dives "Use current GPS location" should now actually work (this 
has been broken since forever)

(see what happens if I get to go diving? - Y'all need to make sure I go diving 
more frequently... just saying)

For all of these changes I of course would love to see some testing. You can 
add yourself to the Android testing group here: 
https://play.google.com/apps/testing/org.subsurfacedivelog.mobile 

For iOS please send me an email so I can add you to the tester list (no self 
service there, sorry)

/D ___
subsurface mailing list
subsurface@subsurface-divelog.org
http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface