v0.11.0 #146

loudnate · 2016-08-31T04:38:45Z

This version contains important safety fixes related to insulin interpretation:

Boluses of 10 U or greater on x23+ pumps are not interpreted as primes (x23 pumps can deliver bolus insulin as fast as 1U/10s LoopKit#66)
Empty reservoirs are not interpreted as no delivery (Loop seems to do the wrong thing when reservoir reports 0 #142)

Additional improvements:

Updating Amplitude SDK to 3.8.5
Updating LoopKit to 0.6.0 (Fixes Incorrect "since" time for total insulin #134)
Max bolus and recommended bolus are available when using the bolus screen on watchOS

…128) * Adding a current glucose HUD * DoseStore upload delegation * Updating interface; fixing watch * Reverting font weight change * Better display of history values

…ace (#137) * Hide the navigation bar on the status screen to give more vertical space * Updating the launch screen to match

Preventing a loop from starting if reservoir data is bad and history fetch fails. Fixes #142 Fixes #134 Fixes #24

* Getting insulin dosing data from pump event history * Updating LoopKit to 0.6.0. Preventing a loop from starting if reservoir data is bad and history fetch fails. Fixes #142 Fixes #134 Fixes #24

… bolus. Progress toward #30

… bolus. (#145) Progress toward #30

sarahspins · 2016-08-31T06:22:31Z

You are amazing @loudnate, thanks!

sarahspins · 2016-08-31T07:07:50Z

It looks like treatments and Loop info for the NS plugin are no longer being uploaded to NS with this update.. and I missing something? The option switch is no longer there and I am not sure if that broke it :(

loudnate · 2016-08-31T13:50:42Z

@sarahspins you're no longer seeing treatments or loop status? So, nothing is uploading to Nightscout at all?

sarahspins · 2016-08-31T13:58:55Z

Nothing was getting uploaded to NS - I went back to 0.10.1 for now.

loudnate · 2016-08-31T15:05:28Z

Nothing should have changed for loop status upload in this release. If you try installing again, check that you are seeing entries in the Event History tab in the insulin data screen. They should show checkmarks when uploaded.

sarahspins · 2016-08-31T16:26:04Z

Looks like they're showing up in history, and with the checkmark showing they've been uploaded, but they're not showing up on NS...

(chose this one because it was a high temp)

NS shows only the previous zero temp over that time... it's as if it's just updating non-zero values :(

loudnate · 2016-08-31T16:28:44Z

OK I think what I'm hearing is that they are uploading (and you can see
them in your treatments.json endpoint or whatever) but they're not
rendering in the graph correctly anymore?
On Wed, Aug 31, 2016 at 9:26 AM sarahspins [email protected] wrote:

Looks like they're showing up in history, and with the checkmark showing
they've been uploaded, but they're not showing up on NS...
[image: img_0301]
https://cloud.githubusercontent.com/assets/16582717/18137215/11d3eb56-6f6d-11e6-937d-094ae3c025d6.PNG
(chose this one because it was a high temp)
[image: img_0300]
https://cloud.githubusercontent.com/assets/16582717/18137209/0e097996-6f6d-11e6-8133-a3b3bcd2d2b8.PNG
NS shows only the previous zero temp over that time... it's as if it's
just updating non-zero values :(
[image: screen shot 2016-08-31 at 11 17 31 am]
https://cloud.githubusercontent.com/assets/16582717/18137303/6767f6ac-6f6d-11e6-808f-828accf37503.png

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#146 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAM_jCaMzOS81DXlrhQcvxxut3TgKLtyks5qlascgaJpZM4JxOUo
.

sarahspins · 2016-08-31T16:44:49Z

The non-0 values do not seem to be in treatments.json that I can see (and they're not in my mlab database either)... only the 0 temps are in there (plus the .5 bolus I did to see if that would show up). Everything that is older than these was from when I was still running 0.10.1 (prior to about 10:20 central time) so I didn't include it.

[  
   {  
      "_id":"57c706e9a0f1339b22e68a3c",
      "created_at":"2016-08-31T16:28:40Z",
      "eventType":"Correction Bolus",
      "unabsorbed":0,
      "timestamp":"2016-08-31T16:28:40Z",
      "enteredBy":"loop://Sarah's iPhone ",
      "insulin":0.5,
      "duration":0,
      "programmed":0.5,
      "type":"normal"
   },
   {  
      "_id":"57c6fea8a0f1339b22e686fd",
      "created_at":"2016-08-31T15:53:31Z",
      "temp":"absolute",
      "rate":0,
      "eventType":"Temp Basal",
      "timestamp":"2016-08-31T15:53:31Z",
      "enteredBy":"loop://Sarah's iPhone ",
      "duration":30,
      "absolute":0
   },
   {  
      "_id":"57c6f2fda0f1339b22e6823d",
      "created_at":"2016-08-31T15:03:41Z",
      "temp":"absolute",
      "rate":0,
      "eventType":"Temp Basal",
      "timestamp":"2016-08-31T15:03:41Z",
      "enteredBy":"loop://Sarah's iPhone ",
      "duration":30,
      "absolute":0
   },

sarahspins · 2016-08-31T17:33:25Z

I went back to 0.10.1 again and it updated all of the missing temp basal data. Very concerning is that that what was being displayed on NS didn't even represent what was actually happening at all... which I suspected, based on what I was watching within Loop :(

on 0.11.0:

back to 0.10.1:

TempBasal and TempBasalDuration must appear in the same order as in the pump

loudnate · 2016-09-01T03:32:02Z

@sarahspins if you get a chance, please give it another try.

sarahspins · 2016-09-01T04:03:47Z

Updated, built and running as of about 20 minutes ago.... but I am still not convinced that it's working as it should. I will give it a little while longer and report back.

sarahspins · 2016-09-01T04:46:56Z

I'm neither convinced it's working as it should, or that it's not, but I've gone back to 0.10.1 for the night and I'll give it a better evaluation in the morning.

sarahspins · 2016-09-01T23:29:19Z

Seems to be working, though it seems to be not particularly reliable for real-time monitoring of what Loop is actually doing.. it seems that with the increased RF comms, many of them fail (I'm seeing this in the logs - but Loop is still running fine), in effect delaying the visibility of temp basals and such on NS. Eventually it does seem to catch up.

ps2 · 2016-09-02T04:32:50Z

I have been running this today, and have been watching the temp basal records in nightscout update, and they are updating reliably for me.

They are lagged a bit (as they always have been, and as they are in OpenAPS), since they get fetched from history 5 minutes after they're enacted.

Another thing I saw is that Riley has an emoji in her phone name. This build puts the phone name in a url in the 'enteredBy' attribute of treatments, which gives NS some troubles (it was sometimes displaying, sometimes not). I had to clean those entries out of my mongo db before NS would work properly. If you think you are seeing this issue, open up the JS console on your NS site, and look for Uncaught URIError: URI malformed errors.

@sarahspins if you are seeing issues, could you note precisely what the inconsistency is and create a new ticket? I.E. note what you see in NS, and in treatments.json, vs what you expect to see.

sarahspins · 2016-09-02T04:38:20Z

@ps2 since Nate's update yesterday it is in fact working well enough (it's just more delayed than I had become accustomed to) - prior to that it really wasn't working at all to upload any temp basals to NS, but that has been fixed.

ps2 · 2016-09-02T04:41:11Z

I don't think we are expecting any additional delays for getting temp basals uploaded with this update. If you are seeing delays, could you file an issue?

channemann · 2016-09-02T04:48:57Z

The sharp drop at about 1pm on the basal chart corresponds to a zero-duration, zero temp being set to cancel the previous temp. From discussion with @loudnate, I believe this was fixed at some point previously, so perhaps it's either a reversion in Loopkit or building against the wrong branch.

loudnate added 13 commits August 27, 2016 09:55

Saving pump history into LoopKit and delegating upload to Nightscout (#…

ba63f6b

…128) * Adding a current glucose HUD * DoseStore upload delegation * Updating interface; fixing watch * Reverting font weight change * Better display of history values

0.11.0

ccf05cb

Hide the navigation bar on the status screen to give more vertical sp…

5115dca

…ace (#137) * Hide the navigation bar on the status screen to give more vertical space * Updating the launch screen to match

Adding a current glucose HUD

1d030ac

Reverting font weight change

d352ff0

Getting insulin dosing data from pump event history

4753b48

Update Amplitude.framework

326952c

Update Amplitude.framework (#143)

7628aff

Updating LoopKit to 0.6.0.

47c8ba9

Preventing a loop from starting if reservoir data is bad and history fetch fails. Fixes #142 Fixes #134 Fixes #24

LoopKit 0.6.0: Event History preference and safety considerations (#144)

0f28464

* Getting insulin dosing data from pump event history * Updating LoopKit to 0.6.0. Preventing a loop from starting if reservoir data is bad and history fetch fails. Fixes #142 Fixes #134 Fixes #24

Version 0.12.0

d2cefde

Provide bolus amount suggestions for all watch boluses, and honor max…

7d536c0

… bolus. Progress toward #30

Provide bolus amount suggestions for all watch boluses, and honor max…

5e260d1

… bolus. (#145) Progress toward #30

loudnate mentioned this pull request Aug 31, 2016

Loop seems to do the wrong thing when reservoir reports 0 #142

Closed

Grammar

1cfc97d

loudnate changed the title ~~v0.12.0~~ v0.11.0 Aug 31, 2016

loudnate added 2 commits August 31, 2016 07:57

0.11.0... oops.

ee27fa8

Merge branch 'dev' of github.com:loudnate/Loop into dev

3573ed9

Fix order of pump events delivered to NightscoutUploadKit (#147)

35090a0

TempBasal and TempBasalDuration must appear in the same order as in the pump

loudnate merged commit 2ebe5d9 into master Sep 3, 2016

ps2 pushed a commit that referenced this pull request Aug 4, 2020

LOOP-1485: move temporary correction range editor to LoopKit (#146)

74431ff

v0.11.0 #146

v0.11.0 #146

Uh oh!

Conversation

loudnate commented Aug 31, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sarahspins commented Aug 31, 2016

Uh oh!

sarahspins commented Aug 31, 2016

Uh oh!

loudnate commented Aug 31, 2016

Uh oh!

sarahspins commented Aug 31, 2016

Uh oh!

loudnate commented Aug 31, 2016

Uh oh!

sarahspins commented Aug 31, 2016

Uh oh!

loudnate commented Aug 31, 2016

Uh oh!

sarahspins commented Aug 31, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

sarahspins commented Aug 31, 2016

Uh oh!

loudnate commented Sep 1, 2016

Uh oh!

sarahspins commented Sep 1, 2016

Uh oh!

sarahspins commented Sep 1, 2016

Uh oh!

sarahspins commented Sep 1, 2016

Uh oh!

ps2 commented Sep 2, 2016

Uh oh!

sarahspins commented Sep 2, 2016

Uh oh!

ps2 commented Sep 2, 2016 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

channemann commented Sep 2, 2016

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

loudnate commented Aug 31, 2016 •

edited

Loading

sarahspins commented Aug 31, 2016 •

edited

Loading

ps2 commented Sep 2, 2016 •

edited

Loading