These are chat archives for nightscout/intend-to-bolus

10th
Jul 2016
clevens
@clevens
Jul 10 2016 00:16
hmm.. i didn't mention a converter
@SirBinky any recommendation on a converter
Paul Martin
@Spazholio
Jul 10 2016 00:17
@scottleibrand If I have auto-sens enabled (and I do) should it run during every iteration of my loop? I just realized that I haven’t seen the row o’ symbols that I normally do, and my settings/auto-sens.json file doesn’t seem to have updated in the last 7 hours - right around the time I think I updated my oref0#merge-ama install. However, I’m not seeing errors in my loop.log file. Does merge-ama have any known issues with auto-sens?
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:19
you still need to invoke the auto-sens report
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:21
I'm testing merge-ama, been working good, I was worried about the change to a longer average delta, but so far it's looking ok
Paul Martin
@Spazholio
Jul 10 2016 00:21
I wouldn’t sweat it. Buying an SD card separately is likely going to be cheaper than the $25 difference.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:21
the sd card in the kit aren't that good anyway
clevens
@clevens
Jul 10 2016 00:21
@Spazholio great ty! i just bought a separate SD and SD converter to regular size one too
Paul Davis
@SirBinky
Jul 10 2016 00:22
@clevens i just meant a thing you plug your sd card into a usb device so you just plug that into the comp
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:22
a fast 16g or bigger card is good
clevens
@clevens
Jul 10 2016 00:22
@jasoncalabrese good to know! i've never heard of a single one of these parts before lol. makes it so confusing
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:23
the sd cards get a lot of wear if you make frequent backups
clevens
@clevens
Jul 10 2016 00:23
@SirBinky oh! is that what my microSD to SD memory card adapter can do?
Paul Davis
@SirBinky
Jul 10 2016 00:23
@jasoncalabrese you only need 8gb and normal speed, bigger and faster sd's are for photographers that want to take rapid shots. a standard 8gb micro sd is fine
Paul Martin
@Spazholio
Jul 10 2016 00:24
@clevens Yes, provided your computer/laptop has a slot for an SD card...
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:24
maybe it's imagined, but it seemed like the loop ran a little faster with a faster card
Paul Martin
@Spazholio
Jul 10 2016 00:24
I can’t see how it wouldn’t. I mean, it certainly couldn’t HURT...
Paul Davis
@SirBinky
Jul 10 2016 00:24
@clevens plug the micro sd into a device that lets you insert it into a standard usb port to initially install the raspberry linux distro
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:24
could help when creating and restoring backups too
Paul Davis
@SirBinky
Jul 10 2016 00:26
@jasoncalabrese it prob did run a little quicker it wouldn't hurt just doesn't make a difference. only if you need instant storing for quick photo saving. for this set up a standard is fine, quicker will be slightly quicker but that isn't a requirement. always back up frequently to a slow or faster card is no difference
Paul Martin
@Spazholio
Jul 10 2016 00:36
@scottleibrand I think I see why I’m having issues with auto-sens.json. It’s got nothing to do with the merge-ama bit, but possibly with using your retry-loop. Here’s the alias I’m using: retry-loop = ! bash -c "openaps wait-loop || openaps loop”. When I look through the aliases, both wait-loop and loop look very similar. Can you give me a quick breakdown of the differences or why you have it set up this way? I’ve been using it for quite some time and it works really well, I’m just unsure of the ins and outs of it.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 00:41
the 1 I'm testing now is https://github.com/jasoncalabrese/indy-e3b/blob/master/openaps.ini, but that part looks the same
I have the 24 hour reports only run in the wait-loop, if it fails it does the non-waiting loop
I don't really like how that is setup, I want to move to something like what @scottleibrand is doing, but hopefully find a way to keep more of the logic outside the ini
Paul Martin
@Spazholio
Jul 10 2016 00:46
So in your context, when is “wait-loop” activated? And should my loop alias JUST be for short stuff, since it’s apparently only invoked whenever something in the wait-loop alias fails?
Paul Davis
@SirBinky
Jul 10 2016 00:47
@Spazholio is it just if the loop fails it draws on longer term data so the recommendation it more accurate ignoring short term loss of data?
or not fails just lacks data
Paul Martin
@Spazholio
Jul 10 2016 00:48
Couldn’t say. That’s sorta why I’m asking, because I’ve committed a cardinal DIYPS sin here - I’ve implemented without fully understanding.
Paul Davis
@SirBinky
Jul 10 2016 00:49
my entire experience so far has been enacting without fully understanding but realising they kept the enact stuff hidden so you have to ask to learn it. then i very very carefully ran some open loops for suggestions
clevens
@clevens
Jul 10 2016 00:53
i went to download openAPS from pypi and it told me there was no application to open it. any advice?
Paul Davis
@SirBinky
Jul 10 2016 00:54
@Spazholio the other thing is what's the best way to do it. there's so many options and ways, and openaps adds combination options etc. A couple of times i've felt like i'm trying to run in 3 different directions having none of them to work then trying to concentrate on one or two ways
@clevens your pypi?
clevens
@clevens
Jul 10 2016 00:56
thanks i'll look at that @SirBinky
Paul Davis
@SirBinky
Jul 10 2016 00:57
@clevens you just starting out?
clevens
@clevens
Jul 10 2016 00:57
@SirBinky yes!
Paul Davis
@SirBinky
Jul 10 2016 00:57
@clevens what hardware do you have?
clevens
@clevens
Jul 10 2016 00:58
@SirBinky Medtronic 723, raspberry pi model 3, Dexcom CG4M with Share, a 8 GB USB, a mini SD to regular converter.
@SirBinky i went to that link, but it's all code i'm not familiar with. any advice? thanks!
Paul Davis
@SirBinky
Jul 10 2016 01:00
you have the rapsberry pi linux distro linked to your home network and you've ssh'd into the pi from a computer?
clevens
@clevens
Jul 10 2016 01:01
@SirBinky not yet! i was just in the guide to understanding trying to get prepared ( https://openaps.readthedocs.io/en/latest/docs/openaps-guide/installing.html )
it's also still on it's way haha. (arrives monday) and also the TI stick, forgot to mention
oh and i bought a cc debugger
Paul Davis
@SirBinky
Jul 10 2016 01:02
right so first get the sd card put it in adapater to your computer
Paul Martin
@Spazholio
Jul 10 2016 01:02
@clevens Do you have ANY hardware yet? You don’t, do you? Still waiting for it to arrive?
clevens
@clevens
Jul 10 2016 01:03
@Spazholio just the pump so far! i guess i'm a little ahead of myself until it arrives monday. i was just trying to get all prepared
@SirBinky thanks for all of your help! i think it looks like i need to wait till it's all here. i thought there might be some things i need to download first
Paul Martin
@Spazholio
Jul 10 2016 01:03
No, that’s a great thing, and it’ll likely help once you get started. =)
Paul Davis
@SirBinky
Jul 10 2016 01:04
download the raspberry pi linux distro. once that's on the card, but it into the pi, attach an hdmi cable to a monitor and an usb keyboard and mouse. let it load to the homescreen. sign into your home network. then you can unplug your keyboard mouse and monitor, get putty ssh install on your main comp link into it from it's assigned ip address
clevens
@clevens
Jul 10 2016 01:04
@Spazholio but it won't let you download it until you have the stuff right?
Paul Davis
@SirBinky
Jul 10 2016 01:04
@v
Paul Martin
@Spazholio
Jul 10 2016 01:04
No, you can. You just won’t really be able to DO anything with it. It’s best to wait with that aspect of things.
Paul Davis
@SirBinky
Jul 10 2016 01:05
@clevens you can download the distro file you just need to put it on to the sd card then plug it into the pi
Paul Martin
@Spazholio
Jul 10 2016 01:05
@SirBinky She has literally no hardware yet.
Paul Davis
@SirBinky
Jul 10 2016 01:05
@Spazholio yup
clevens
@clevens
Jul 10 2016 01:06
i'll come back once everything has arrived! thanks SO much to both of you!! i can't tell you how much i appreciate it!
Paul Davis
@SirBinky
Jul 10 2016 01:07
@clevens once it comes, come back here, before that, maybe just read through some of the docs. even if you don't get it, just read it a few times, try to learnt the patterns it'll help when you start
clevens
@clevens
Jul 10 2016 01:07
@SirBinky yes, that's exactly what i plan to do. i have A LOT to learn
Paul Davis
@SirBinky
Jul 10 2016 01:09
@clevens also you can wipe everything out start again don't worry about mistakes just take your time learn. best plan is get a second or even 3rd memory cards, learn on one, when you understand something, write it all down in an diary then swap the cards and implement on a second place. then swap out go back to the first keep testing stuff
Jason Calabrese
@jasoncalabrese
Jul 10 2016 01:23
@Spazholio you probably need to run them 1 at a time, I used the new flags for determine-basal and get-profile
might be that it doesn't work the same with out the flags
Paul Martin
@Spazholio
Jul 10 2016 01:23
I don’t think I follow? Can you give me an example or paste what you have so I can compare?
Jason Calabrese
@jasoncalabrese
Jul 10 2016 01:24
for example, --meal
Paul Martin
@Spazholio
Jul 10 2016 01:32
What do those do? Specifically the —model flag?
Chris Oattes
@cjo20
Jul 10 2016 01:32
The model flag in get-profile is used in determine-basal to work out what basal rates the pump is capable of setting
Paul Martin
@Spazholio
Jul 10 2016 01:33
…huh. That’s interesting.
Chris Oattes
@cjo20
Jul 10 2016 01:33
Specifically, whether the smallest dose it can provide for basal rates < 1u/hr is 0.05u/hr or 0.025u/hr
Ben West
@bewest
Jul 10 2016 01:33
anyone looking at gym.openapi.com ?
Paul Davis
@SirBinky
Jul 10 2016 01:34
all you americans you use mmol?
Chris Oattes
@cjo20
Jul 10 2016 01:35
Americans (usually) use mg/dL
Paul Davis
@SirBinky
Jul 10 2016 01:35
oh ok
Chris Oattes
@cjo20
Jul 10 2016 01:36
@bewest that looks cool
Paul Martin
@Spazholio
Jul 10 2016 01:36
3>/dev/null <— I’m unfamilar with the “3” in this context.
Paul Davis
@SirBinky
Jul 10 2016 01:36
we're mmol in the uk but it's been so well supported with all the differnet aps etc
cjo20 @cjo20 is in the UK too
Chris Oattes
@cjo20
Jul 10 2016 01:36
@Spazholio Do you have more context than that?
Paul Martin
@Spazholio
Jul 10 2016 01:37
upload = ! bash -c "echo Uploading... && ( echo -e \nNS STATUS && openaps upload-ns-status; openaps report invoke enact/suggested.json 2>/dev/null; openaps pebble; openaps upload-pumphistory-entries; openaps upload-recent-treatments ) 3>/dev/null >/dev/null”
Paul Davis
@SirBinky
Jul 10 2016 01:37
wasn't sure if it was a standard thing or just us, it was so well supported i thought you guys did it too but the machines didn't or something!
Paul Martin
@Spazholio
Jul 10 2016 01:37
Tell me what everything is doing after the closing )
(please?)
Ben West
@bewest
Jul 10 2016 01:37
3> looks like a typo, it will redirect the third file descriptor for the process to /dev/null
Chris Oattes
@cjo20
Jul 10 2016 01:38
I suspect it's meant to be a 2
Paul Martin
@Spazholio
Jul 10 2016 01:38
Oh, good. I could only find explanations for 0, 1, and 2. 3 was right out.
Ben West
@bewest
Jul 10 2016 01:38
by default third file descriptor isn't bound to anything
you can bind it to stuff using socat and other tricks
Paul Martin
@Spazholio
Jul 10 2016 01:39
And those are: STDIN, STDOUT, and STDERR ?
Ben West
@bewest
Jul 10 2016 01:39
but it's meaningless here ;-)
Paul Martin
@Spazholio
Jul 10 2016 01:39
In that order?
Ben West
@bewest
Jul 10 2016 01:39
yeah, 0, 1, 2
it shows up often in oref0 scripts as /dev/fd/1
or /dev/stdout
Paul Martin
@Spazholio
Jul 10 2016 01:40
And to combine all output, I could to 1&2>/dev/null - right?
Ben West
@bewest
Jul 10 2016 01:40
1, /dev/fd1, and /dev/stdout are all symlinks basically
Chris Oattes
@cjo20
Jul 10 2016 01:40
> /dev/null 2>&1
Ben West
@bewest
Jul 10 2016 01:40
no, 2>&1 is what you want
Paul Martin
@Spazholio
Jul 10 2016 01:41
Can you explain what that does? I don’t follow the syntax.
I mean, I get what it does just not quite how.
Chris Oattes
@cjo20
Jul 10 2016 01:41
2>1 would be redirecting to a file called 1
Paul Martin
@Spazholio
Jul 10 2016 01:41
…it’s been a while since I’ve gotten such a well-deserved RTFM. Kudos, @bewest. And thanks. =)
Paul Davis
@SirBinky
Jul 10 2016 01:42
the two nightscout guide sections are so different and complex yet visualisation is important to this process has anyone tried to make a guide to just upload info to nightscout from a rapsberry rig?
Paul Martin
@Spazholio
Jul 10 2016 01:43
Oh, wait…so I’d chain that to/dev/null if I wanted to eliminate all output? 2>&1>/dev/null?
Paul Davis
@SirBinky
Jul 10 2016 01:45
rtf oh wait the m is totally new now nevermind?
Ben West
@bewest
Jul 10 2016 01:45
right, the 2>&1 ties the first stream to the second
rather the second to the first, really
so makes everytthing go out stdout
then > /dev/null redirects everything to/dev/null
Chris Oattes
@cjo20
Jul 10 2016 01:46
I'm not sure that works entirely how Paul thinks it does
Paul Martin
@Spazholio
Jul 10 2016 01:46
I like to have echo statements peppered throughout my loop so I can see what it’s doing where. Unfortuantely, if I enable any output from that alias I pasted, it’s stupidly noisy. =)
@cjo20 Oh?
Chris Oattes
@cjo20
Jul 10 2016 01:46
The lack of spaces isn't great
Paul Davis
@SirBinky
Jul 10 2016 01:47
@cjo20 read the fucking oh wait the manual changed?
oh sorry you mean martin
Paul Martin
@Spazholio
Jul 10 2016 01:47
I mean…conceivably, I could just do > /dev/null right? With no file descriptors? And then it would all go to /dev/null?
Chris Oattes
@cjo20
Jul 10 2016 01:47
No, if you have >/dev/null it implicitly does 1>/dev/null
Paul Martin
@Spazholio
Jul 10 2016 01:47
@SirBinky Sorry dude, I’m calling dibs. Now you’re gonna have to change your name. =)
Chris Oattes
@cjo20
Jul 10 2016 01:48
So with just >/dev/null you'd only have stdout redirected to /dev/null
I was just a bit unsure by the way you wrote 2>&1>/dev/null
Paul Davis
@SirBinky
Jul 10 2016 01:49
@Spazholio i had this once before, we agreed in furturama style. he was universe A guy and i was universe 1 so Paul 1 and Paul A ;)
Paul Martin
@Spazholio
Jul 10 2016 01:49
@cjo20 Because I don’t know quite what I’m doing and sometimes what I’m even asking. =) Just wrapping my head around it.
For instance, here: http://www.cyberciti.biz/faq/how-to-redirect-output-and-errors-to-devnull/ shows that you can do command > /dev/null 2>&1
So you can put the descriptors after the destination?
Paul Davis
@SirBinky
Jul 10 2016 01:50
@Spazholio the error messages are really quite useful for basic stuff i've found. what you're asking deeper coding
Paul Martin
@Spazholio
Jul 10 2016 01:50
Oh, I know. That’s why I’m modifying some of the aliases I have.
Paul Davis
@SirBinky
Jul 10 2016 01:52
ive ended up with two sets of alias's. the originals i was told to put in, then to try to upload nightscout, i've got bewest's alias and select reports. so i worked back from the alias invokes to make reports that fit it.
Chris Oattes
@cjo20
Jul 10 2016 01:52
I'd say it's just a good idea to keep spaces between different parts of the redirection. I'm not sure that 1>&2>/dev/nullwould work in the same way, but I could be wrong
Paul Martin
@Spazholio
Jul 10 2016 01:52
Nah, I like spaces. Makes it more readable.
Paul Davis
@SirBinky
Jul 10 2016 01:53
as i learnt bewest's report names for the commands and what they need i've sorted my openaps.ini to work with my original alias's and the new ones
Ben West
@bewest
Jul 10 2016 01:58
eventually i started a new one from scratch because I had so many old reports with different names as the system has matured
the new one I always use the oref0 template tool and it's much easier
Paul Davis
@SirBinky
Jul 10 2016 02:02
oref0 template?
Paul Martin
@Spazholio
Jul 10 2016 02:03
Damn you, I’m gonna have to give those a shot one of these days. I really need to get this dev rig up and running soon. =)
Tweaking all my aliases and reports now just feels like polishing brass on the Titanic.
Paul Davis
@SirBinky
Jul 10 2016 02:04
@Spazholio why? you got the basic original commands you're sorting
Paul Martin
@Spazholio
Jul 10 2016 02:05
As you’ve noticed, things change pretty quickly around here. And I like to be current and using the best/latest techniques and code.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 02:06
I just updated our main production rig to merge-ama, looking good so far
Paul Davis
@SirBinky
Jul 10 2016 02:06
i've noticed that, the system i learnt that's been out moded, was actually fine and worked. the commands for the pump and cgm are the same
the only difference is the report names ppl gave them
Ben West
@bewest
Jul 10 2016 02:07
it's not outmoded, it works exactly the same way
Paul Davis
@SirBinky
Jul 10 2016 02:08
you need the same reports and json files but names have changed
Ben West
@bewest
Jul 10 2016 02:08
the new stuff allows you to kind of skip some steps and reduce the amount of typing you have to do
my suggestion if you have an existing instance is to runopenaps init another-one and try the template stuff out to see what it does
Paul Davis
@SirBinky
Jul 10 2016 02:09
@bewest honestly you could make a program that works off the bat, only thing you'd need to do is specify the cgm device and add the pump with it's serial name
Ben West
@bewest
Jul 10 2016 02:09
I've done that too
complete with web app to go with it
Paul Davis
@SirBinky
Jul 10 2016 02:10
@bewest er ok...
Ben West
@bewest
Jul 10 2016 02:11
the templates are easier to use though
Paul Davis
@SirBinky
Jul 10 2016 02:12
@bewest i dunno when i first started on this, i saw the options the differnet routes. i ignored nightscout. i thought constant web access was a flaw. i wanted an offline closed system its' the only one that made sense to me.
@bewest now isee why nighscout basal statements make sense and back ups etc
Ben West
@bewest
Jul 10 2016 02:13
yeah, to get data into NS, you need to collect cgm values and pump-history
Paul Davis
@SirBinky
Jul 10 2016 02:13
na with the dexcom g5, the iphone reciever shares data and the nightscout site gets the bg readings
Ben West
@bewest
Jul 10 2016 02:13
if you've got that, you can use my ns reports
ok in that case you just need medtronic pump-history to upload to NS to get started with unified visualization
I'm using g4 + share ble
so it's different for everyone, see?
Paul Davis
@SirBinky
Jul 10 2016 02:14
from the help files i need 4 things
Ben West
@bewest
Jul 10 2016 02:15
you need model and pump history
openaps use ns shell format-recent-history-treatments monitor/pump-history.json model.json
need a zoned pump-history and model
Paul Davis
@SirBinky
Jul 10 2016 02:16
model is easy
Ben West
@bewest
Jul 10 2016 02:17
oref0 template mint reports medtronic-pump | openaps import sets all of them up
Paul Davis
@SirBinky
Jul 10 2016 02:17
pumphistory
ok i ran that

pi@raspberrypi:~/sabina $ oref0 template mint reports medtronic-pump | openaps import
pump://JSON/read_bg_targets/raw-pump/bg-targets-raw.json
units://JSON/bg_targets/settings/bg-targets.json
pump://JSON/read_insulin_sensitivities/raw-pump/insulin-sensitivities-raw.json
units://JSON/insulin_sensitivities/settings/insulin-sensitivities.json
pump://JSON/read_clock/raw-pump/clock-raw.json
tz://JSON/clock/monitor/clock.json
pump://JSON/read_temp_basal/monitor/temp-basal-status.json
determine-basal://text/shell/oref0-predict/oref0.json
pump://JSON/iter_pump_hours/raw-pump/pump-history-raw.json
tz://JSON/rezone/monitor/pump-history.json
pump://JSON/model/model.json
pump://JSON/reservoir/monitor/reservoir.json
pump://JSON/read_status/monitor/status.json
pump://JSON/read_battery_status/monitor/battery.json
pump://JSON/set_temp_basal/oref0-enacted/enacted-temp-basal.json
oref0://JSON/shell/settings/settings.json
pump://JSON/read_settings/raw-pump/settings.json
oref0://JSON/shell/settings/selected-basal-profile.json
pump://JSON/read_selected_basal_profile/raw-pump/selected-basal-profile.json
pump://JSON/read_temp_basal/raw-pump/temp-basal-status.json
pump://JSON/reservoir/raw-pump/reservoir.json
pump://JSON/read_status/raw-pump/status.json
pump://JSON/read_battery_status/raw-pump/battery.json
Ben West
@bewest
Jul 10 2016 02:18
oref0 template mint alias common | openaps import should create bunch of alias
openaps gather-pump-data should get the zoned history
Paul Davis
@SirBinky
Jul 10 2016 02:19
pi@raspberrypi:~/sabina $ openaps alias show
gather-profile report invoke settings/settings.json settings/bg_targets_raw.json settings/bg_targets.json settings/insulin_sensitivities_raw.json settings/insulin_sensitivities.json settings/basal_profile.json settings/profile.json
monitor-pump report invoke monitor/clock.json monitor/temp_basal.json monitor/pumphistory.json monitor/iob.json
monitor-cgm report invoke monitor/glucose.json
paul ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps monitor-pump && openaps report invoke monitor/iob.json && openaps report invoke enact/suggested.json"
paul1 ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps monitor-pump && openaps report invoke monitor/iob.json"
gather ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps monitor-pump && openaps report invoke monitor/iob.json"
preflight ! bash -c "(echo PREFLIGHT ) && openaps warmup 2>&1 >/dev/null && grep -q T monitor/clock.json && echo PREFLIGHT OK || openaps fail-warmup"
rm-warmup ! bash -c "rm -f model.json monitor/clock.json > /dev/null"
warmup report invoke model.json monitor/clock.json monitor/clock-zoned.json
fail-warmup ! bash -c "echo PREFLIGHT FAIL; exit 1"
monitor-pump-history report invoke monitor/pumphistory.json
get-basal-status report invoke monitor/temp-basal-status.json
get-pump-details report invoke monitor/reservoir.json monitor/status.json monitor/battery.json
get-settings report invoke settings/bg_targets_raw.json settings/bg_targets.json settings/insulin_sensitivities_raw.json settings/insulin_sensitivities.json settings/basal_profile.json settings/profile.json settings/settings.json
gather-pump-data ! bash -c "openaps get-basal-status; openaps get-pump-details; openaps monitor-pump-history;  openaps get-settings"
gather-clean-data ! bash -c "openaps monitor-cgm && openaps gather-pump-data"
do-oref0 report invoke settings/profile.json monitor/iob.json enact/suggested.json
enact-oref0 report invoke enact/enacted.json
do-everything ! bash -c "(openaps preflight && openaps gather-clean-data && openaps do-oref0 && openaps enact-oref0);  openaps report-nightscout"
ping ! bash -c "echo PONG! $*" --
pong ! echo PONG
hello ! echo hello
okbar ! bash -c "echo $1" --
report-nightscout report invoke nightscout/preflight.json nightscout/recent-treatments.json nightscout/uploaded.json nightscout/recent-missing-entries.json nightscout/uploaded-entries.json
enact report invoke enact/enacted.json
do-mostthing ! bash -c "(openaps preflight && openaps gather-clean-data && openaps do-oref0);  openaps report-nightscout"
that's openaps alias show so far do i still need to run your oref0 line?
Ben West
@bewest
Jul 10 2016 02:20
openaps use ns shell format-recent-history-treatments monitor/pump-history.json model.json should print json you can upload to NS
Paul Davis
@SirBinky
Jul 10 2016 02:20
pi@raspberrypi:~/sabina $ openaps use ns shell format-recent-history-treatments monitor/pump-history.json model.json
Could not parse pumphistory:  { [Error: Cannot find module '/home/pi/sabina/monitor/pump-history.json'] code: 'MODULE_NOT_FOUND' }
No JSON object could be decoded

Traceback (most recent call last):
  File "/usr/local/bin/openaps-use", line 4, in <module>
    __import__('pkg_resources').run_script('openaps==0.1.5', 'openaps-use')
  File "/usr/local/lib/python2.7/dist-packages/setuptools-22.0.0-py2.7.egg/pkg_resources/__init__.py", line 719, in run_script
  File "/usr/local/lib/python2.7/dist-packages/setuptools-22.0.0-py2.7.egg/pkg_resources/__init__.py", line 1504, in run_script
  File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-use", line 63, in <module>
    app( )
  File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/cli/__init__.py", line 51, in __call__
    self.run(self.args)
  File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-use", line 58, in run
    reporter(output)
  File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/reports/reporters/__init__.py", line 33, in __call__
    self.blob = self.serialize(data)
  File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/reports/reporters/__init__.py", line 31, in serialize
    return self.method.serialize(render(data), self)
  File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/vendors/process.py", line 56, in prerender_json
    return json.loads(data)
  File "/usr/lib/python2.7/json/__init__.py", line 338, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
pi@raspberrypi:~/sabina $
Ben West
@bewest
Jul 10 2016 02:21
right, you need to actually get the history
oref0 template mint alias common | openaps import should create bunch of alias
openaps gather-pump-data should get the zoned history
Martin Haeberli
@mhaeberli
Jul 10 2016 02:22
@bewest - dumb question #6 - can I somehow get pump history “synthetically” - RileyLink -> (mongo) -> NightScout -> openaps?
Ben West
@bewest
Jul 10 2016 02:22
I think we were talking about this earlier, the NS treatments data is available you were fetching it
openaps use ns shell get treatments.json count=10 works
Martin Haeberli
@mhaeberli
Jul 10 2016 02:23
@bewest - yes
Ben West
@bewest
Jul 10 2016 02:23
the issue is that oref0 expects data in a different format
Martin Haeberli
@mhaeberli
Jul 10 2016 02:23
but how do I turn that into the “zoned history” that oref0 wants ?
(I could write a python script?)
Scott Leibrand
@scottleibrand
Jul 10 2016 02:23
@Spazholio you might prefer the simplicity of https://github.com/scottleibrand/openaps-sh/blob/refresh-loops/setup.sh
Ben West
@bewest
Jul 10 2016 02:23
issue isn't the zoning, it's that the organization of the records is actually different
Paul Davis
@SirBinky
Jul 10 2016 02:24
@bewest pump data didn't work cause pump isn't near me
Paul Martin
@Spazholio
Jul 10 2016 02:24
@scottleibrand Is that the “refresh” one? I need to look into what those do, specifically.
Martin Haeberli
@mhaeberli
Jul 10 2016 02:24
@SirBinky my problem exactly !
Ben West
@bewest
Jul 10 2016 02:25
https://github.com/openaps/oref0/blob/dev/bin/mm-format-ns-treatments.sh this script is used to reformat event stream from pump for Nightscout
you would need to do the opposite
you mean the pi or edison isn't near the pump?
Jason Calabrese
@jasoncalabrese
Jul 10 2016 02:26
it's for setting up a test?
Martin Haeberli
@mhaeberli
Jul 10 2016 02:26
@jasoncalabrese @bewest indeed - the pump and RileyLink are in RedwoodCity; the linux box is in Palo Alto
Ben West
@bewest
Jul 10 2016 02:26
@SirBinky I'm confused, what happens with openaps use pump model?
Paul Davis
@SirBinky
Jul 10 2016 02:28
@bewest i mean the pump is away. Sorry ben I don't mean to be difficult. i'm trying to do this all for my partner. Mostly i can try to set up situations and test invokes etc and then get her to be close to test the invokes and uploading etc
Ben West
@bewest
Jul 10 2016 02:28
https://github.com/openaps/oref0/blob/dev/bin/oref0-normalize-temps.js this script is also used in tandem... most of the major re-organization is happening here
Paul Davis
@SirBinky
Jul 10 2016 02:30
@bewest you're a leading programmer. @scottleibrand and Dana are too
are you 3 all in the same drive forward?
Jason Calabrese
@jasoncalabrese
Jul 10 2016 02:32
@scottleibrand I want to change the autosens limit, think that should happen in merge-ama, or advanced-meal-assist and then merged to merge-ama
Paul Davis
@SirBinky
Jul 10 2016 02:32
i've tried from my limited knowledge to sort out the early stages of the docs, but the new walk thorough stuff is big steps.
can someone look at that and explain some of the additional info
the blue top bar and the secon lower graph?
Jeremy M Lucas (GetRileyLink.org)
@jlucasvt
Jul 10 2016 02:44
It's very well laid out in the Nightscout docs let me see if I can find the page
Jarred Yaw
@jyaw
Jul 10 2016 02:44
@scottleibrand hey was looking at your refresh-loops approach... is the intent to start the cgm, ns and pump loops in parallel (like every 5 min or so)? Have you observed the case where the pump loop gets the suggest before cgm/ns refreshes BG values?
Scott Leibrand
@scottleibrand
Jul 10 2016 02:48
@jasoncalabrese either way: whatever is most convenient for you.
@jyaw they all refresh as fast as they can. I'll probably add some more logic to make the pump refreshes trigger right after the BG comes in, but it's a lot faster and more reliable than it used to be regardless.
Paul Davis
@SirBinky
Jul 10 2016 02:50
@scottleibrand how much of the walkthrough docs are being changed? they're so different to what i knew. orignally, i adpated bewests nightscout aliases to work and added new report
Jason Calabrese
@jasoncalabrese
Jul 10 2016 02:50
I just set them in my fork, but I'm thinking about making it possible to set the limits in prefs.json
Paul Davis
@SirBinky
Jul 10 2016 02:51
@scottleibrand i wanted to rewrite some of it to make it clearer and i tried. but the new stuff can be diving in at the deep en
d
Jarred Yaw
@jyaw
Jul 10 2016 02:52
@scottleibrand ah ok. thanks
Dana Lewis
@danamlewis
Jul 10 2016 02:54
@SirBinky it's always going to be somewhat complicated. That's the DIY nature. To @bewest earlier point, everyone wants to do it slightly different, and because of his foundational work, it can be. But because everyone's sharing what they're doing, updates get made pretty quickly. Which is awesome, but does add to the complexity of sorting through which route you want to go.
Paul Davis
@SirBinky
Jul 10 2016 02:55
@scottleibrand i have no experience of programming etc and the difference between python json stuff then bash commands etc
Dana Lewis
@danamlewis
Jul 10 2016 02:57
@mhaeberli I think you flagged me yesterday evening, but I can't find it. Did you get an answer to that particular Q?
Jeremy M Lucas (GetRileyLink.org)
@jlucasvt
Jul 10 2016 02:58
@SirBinky Blue line is Basal and changes to basal made by OpenAPS, bottom graph is the full 24 hr data you can slide the box on the lower right to look at the previous data, or choose a view window scale 3hr, 6hr, 12hr, 24hr
Paul Davis
@SirBinky
Jul 10 2016 02:58
@danamlewis it is going to be complicated, and i think it should be also to ppl understand what they're doing, how complicated it is i have no idea about. i'm not diabetic i'm doing it for my partner, so learning the ins and outs of this system but how bg levels are important, and ranges and bolus and basal levels, what is a basal level
Martin Haeberli
@mhaeberli
Jul 10 2016 02:58
@danamlewis thx - no, not yet, although the trend looks right - there seem to be some suggestions from @bewest since ...
@danamlewis what I had hoped for was a clean process for running real-time data, clean, predict, recommend remotely from the pump / cgm via Nightscout. sounds like i’ll need to do some experimentation as is understandable with the “homebrew” nature of things
Paul Martin
@Spazholio
Jul 10 2016 03:00
The problem is (I think) that this whole thing is a great big beta test, but the beta testers are using this thing in production for something fairly essential, AND our beta-testing pool is somewhat small compared to most beta tests. So we’re updating everything according to newfound best practices, which makes it difficult for people just starting out to keep up sometimes.
Martin Haeberli
@mhaeberli
Jul 10 2016 03:00
@danamlewis - my intent being to get started before we do the looping part
Paul Davis
@SirBinky
Jul 10 2016 03:00
@jlucasvt think i have it wrong blue line is at 13 bg oh ok i get the bottom i move that slider it shows the bg range
Dana Lewis
@danamlewis
Jul 10 2016 03:00
@Spazholio swap in alpha though for beta
And this is where the community comes in.
Paul Martin
@Spazholio
Jul 10 2016 03:01
“alpha” can have negative connotations, and I was looking to avoid that, but…yes. =)
Jason Calabrese
@jasoncalabrese
Jul 10 2016 03:01
exactly
Dana Lewis
@danamlewis
Jul 10 2016 03:01
We need people to step up and work through the docs over and over and fixing what's confusing or broken.
There's 90+ loopers and many in progress, so lots of people fully capable of contributing, too :)
Paul Martin
@Spazholio
Jul 10 2016 03:02
The issue with fixing the docs seems to be that it’s a moving target.
Like…a rapidly moving target.
Paul Davis
@SirBinky
Jul 10 2016 03:02
@danamlewis i'm coming at this from a non diabetic none progammer view
Jason Calabrese
@jasoncalabrese
Jul 10 2016 03:02
my comment was to Dana's, but alpha/beta are more product terms, openaps isn't a product
Paul Martin
@Spazholio
Jul 10 2016 03:03
@jasoncalabrese You don’t think so? I mean, it’s never going to be a finished product, but I tend to look at my loop as a product. Also a process, but still.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 03:03
everyone needs to be a contributor, setup their own experiments, make adjustments, improve their setup over time, etc
Dana Lewis
@danamlewis
Jul 10 2016 03:03
you may have a product, but we are not responsible for a product
Paul Martin
@Spazholio
Jul 10 2016 03:03
I can see how that’s a fairly important semantic distinction, given what we’re doing here.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 03:04
you can't ever finish
Paul Davis
@SirBinky
Jul 10 2016 03:04
@danamlewis i know a bit of computers and linux very very basic linux, hence why i'm trying to help teh docs but they've changed alot since i've started out
Dana Lewis
@danamlewis
Jul 10 2016 03:04
If anyone wants to contribute, that's great. That's the only way we make this better for anyone else. But there's no one person who's in charge, other than you for YOUR loop. And if no one contributes back, it stagnates.
So help where you can :) and we'll all move forward more quickly.
Paul Martin
@Spazholio
Jul 10 2016 03:04
@jasoncalabrese Agreed. To date, I have never “finished” any of my code.
Dana Lewis
@danamlewis
Jul 10 2016 03:06
@SirBinky yep. It all moves quickly and we are all constantly learning as we go
Ditto what @jasoncalabrese said re: everyone needs to
@mhaeberli can you elaborate what you mean re wanting a clean process?
Paul Davis
@SirBinky
Jul 10 2016 03:11
@danamlewis just you got some docs written by ppl using _ between words others using - then totally different json files for the results of the device commands. eg inter-glucose-hours 4 = pumphistory etc
Dana Lewis
@danamlewis
Jul 10 2016 03:12
@SirBinky yea. If you want to clean it up, that's great - please submit Pr. And if not, this is why using critical thinking skills is a requirement of this project. There's no press a button and get an AP for a reason.
Martin Haeberli
@mhaeberli
Jul 10 2016 03:12
@danamlewis - at this stage I feel like I’m taking more than I’m contributing - so I’ll probably wander in the desert some more - but what I meant was a setup process that walks me through configuring stuff for the standalone non-local scenario - that is, i’m just getting data into oref0 through Nightscout… (and not trying to suggest anything is ‘unclean’ - I just remain confused ...
Dana Lewis
@danamlewis
Jul 10 2016 03:12
@mhaeberli ah, I see.
Wandering is a part of the learning process, so don't be discouraged :)
Paul Davis
@SirBinky
Jul 10 2016 03:15
@danamlewis i do want to clean it up. i've tried with the couple of pr's ive done. i've also from my nativity advised the less than ideal paths. I tried to start to update the docs from a beginners view and did to an extent missing the importance of some aspects
@mhaeberli keep asking that's not a problem. keep hanging here though. if you see someone asking a question you know that, you might think it simple or something answer it advise and help them. you might give more back to one person with the right few words than staying and trying to deliberately trying to help a 100 ppl.
Martin Haeberli
@mhaeberli
Jul 10 2016 03:19
@danamlewis @SirBinky thx - :thumbsup:
Paul Davis
@SirBinky
Jul 10 2016 03:21
Also there is no product here, just some ppl chatting
the last 18months i've known my t1d gf, i've seen at least 10 cures plash stories, oh but it's only in this small test on mice or something. This openaps thing is the first thing i saw that said, yes we can make a change this can work. So after careful research I joined and want to help and advance it and learn the thing you guys have managed
Dana Lewis
@danamlewis
Jul 10 2016 03:26
:+1:
Paul Davis
@SirBinky
Jul 10 2016 03:29
@danamlewis i dropped out the last week cause i was worried about the implications. Just found out sabs is pregnant her insulin resistance etc will fluctuate and every normal figure is in flux to a lesser or greater degree. One thing is she's getting low bg's in the night alot. if i can set this up to monitor her bg at home and reduce her basal at night if need be and it stops her having to wake up a couple of times a night it'll be worth it
Dana Lewis
@danamlewis
Jul 10 2016 03:30
:+1:
Paul Davis
@SirBinky
Jul 10 2016 03:31
@danamlewis its exhasting for her to deal with it. i find it tiring to wake roll over go to sleep, she's testing herself getting sugar etc
jaylagorio
@jaylagorio
Jul 10 2016 03:46
I might be having overheating issues on my Pi 2. It has a case and a heat sink but the bag I've put together is made of leather. After a while the Pi goes dark but for the power LED and the heat sink is super hot. Is anyone underclocking to save power? I think that might be one way to reduce the temperature but don't want to lower the clock rate so much it stops working.
I'm hoping that I didn't just put a lot of work into this carry bag that now I can't use because it overheats the device.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 03:56
I think @audiefile is underclocking her Pi
jaylagorio
@jaylagorio
Jul 10 2016 03:57
Thanks! @audiefile, What speed are you underclocking to?
Jason Calabrese
@jasoncalabrese
Jul 10 2016 03:57
Edison runs a lot cooler...
Paul Davis
@SirBinky
Jul 10 2016 03:57
@jaylagorio can you not hole punch the leather pouch ?
jaylagorio
@jaylagorio
Jul 10 2016 03:59
This is my first rig, I'd like to get it working as best it can before trying to reduce the size and do something completely new.
Paul Davis
@SirBinky
Jul 10 2016 04:00
hole punch the leather then
jaylagorio
@jaylagorio
Jul 10 2016 04:01
@SirBinky It's not so much a pouch as it is this: https://www.amazon.com/gp/product/B005HOMZTI/ref=oh_aui_search_detailpage?ie=UTF8&psc=1. I've sewn in all kinds of wiring conduits and elastic straps to hold the Pi, a 26000mAH battery, a cell modem and booster antenna, the CareLink, and the Dexcom Receiver.
Elaborate, yes. But I can throw it in my work bag, leave the house, and not worry about much of anything. Power-wise it can run for a couple days without charging. I also sewed USB pass-throughs so I can plug into connectors that are exposed on the side of the bag for charging rather than taking everything out.
As long as I can take care of the cooling issue I was going to document it and the parts on GitHub.
Jason Calabrese
@jasoncalabrese
Jul 10 2016 04:05
Sounds like a nice rig
Pictures?
jaylagorio
@jaylagorio
Jul 10 2016 04:11
Yeah, I can do that. It's getting a bit on the late side here so I'll take a note to do that tomorrow, throw it up on Imgur, and send a link your way. A writeup would include product links but if you had questions on that I would be happy to direct you to more Amazon links.
To be honest, it's probably a lot bigger than what most people would want to carry or have put together but I have a backpack with me almost always and don't mind the size since I can just throw it in there.
Paul Davis
@SirBinky
Jul 10 2016 04:20
never through heat dissipation would be an issue!
jaylagorio
@jaylagorio
Jul 10 2016 04:22
Yup. Neither did I so it never occurred to me that this wouldn't work. It would be more than a little disappointing.
Paul Davis
@SirBinky
Jul 10 2016 04:24
is it for you?
i'm building it for my gf but the bumbag idea seemed to be right yet now you say it'll overheat
jaylagorio
@jaylagorio
Jul 10 2016 04:27
It might be a problem for me if I can't solve the heat problem. Or I'll have to go back to the drawing board on the bag. It wouldn't be the biggest deal in the world but I won't be particularly happy to have to do it.
All that said I think this is a problem unique to me and not something most people will encounter.
Paul Davis
@SirBinky
Jul 10 2016 04:29
i thought the default was rig in a bumbag
pi2 is slower than 3 so the older bumbag rigs were ok
even if it was low air flow it would only heat up a bit not damage the capacity of the pi
jaylagorio
@jaylagorio
Jul 10 2016 04:32
I think everyone carries differently. @CrushingT1D had an Edison and TI stick elasticed together in his pocket and I'm sure he didn't have heating issues. Other people may use more breathable material than what I picked.
Paul Davis
@SirBinky
Jul 10 2016 04:33
you need leather it's the best
jaylagorio
@jaylagorio
Jul 10 2016 04:34
Like I said, it's getting late for me - I'm going to bed down for the night, I'll try to take those pictures tomorrow after errands.
Paul Davis
@SirBinky
Jul 10 2016 04:35
cool, i honestly don't think upgrading the pi3 would help
so it's making the rig work for you
Ben West
@bewest
Jul 10 2016 04:36
@mhaeberli you'll need a script to reverse the steps the script I showed earlier performs
ns-reassemble-mm-history
it's worth doing, something similar will be needed to support more pumps
oref0 tools could probably be updated to accept different styles of history perhaps
in last week, I've been looking more at openaps-predict and mmhistorytools because it fits nicely with the machine learning stuff
Jason Calabrese
@jasoncalabrese
Jul 10 2016 04:39
We already pull in carbs from NS, pulling boluses isn't very different
Martin Haeberli
@mhaeberli
Jul 10 2016 04:40
@bewest thx
Jason Calabrese
@jasoncalabrese
Jul 10 2016 04:40
I like how it could be used to setup tests too
Martin Haeberli
@mhaeberli
Jul 10 2016 04:40
@bewest I’ll keep you posted … but probably a few days before I get traction ...
Ben West
@bewest
Jul 10 2016 04:42
specifically, it's calculate-iob, it should be taught to use yargs to handle arguments and possibly to accept NS treatments as input
Jim is CrushingT1D
@CrushingT1D
Jul 10 2016 04:42
@jaylagorio no heat issues until the moment i might have had a fire in my pocket. I never had an issue but am fairly sure what i was doing was a terrible idea and i was lucky
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:43
Hey guys, I'm stuck - any suggestions on how to fix:
pi@raspberrypi:~/giarepo $ openaps report show
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 4, in <module>
import('pkg_resources').run_script('openaps==0.1.5', 'openaps-report')
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 719, in run_script
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 1504, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-report", line 82, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/cli/init.py", line 47, in call
self.configure_parser(self.parser)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-report", line 50, in configure_parser
available = devices.get_device_map(self.config)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/init.py", line 21, in get_device_map
for device in Device.FromConfig(vendors, conf):
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/device.py", line 79, in FromConfig
device.read(config=config)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/device.py", line 36, in read
self.extra.fields.update(dict(extra.items(self.section_name( ))))
File "/usr/lib/python2.7/ConfigParser.py", line 642, in items
raise NoSectionError(section)
ConfigParser.NoSectionError: No section: u'device "giapump"'
Paul Davis
@SirBinky
Jul 10 2016 04:45
what do you run to get that?
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:45
openaps report show
Ben West
@bewest
Jul 10 2016 04:45
openaps device add giapump medtronic
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:46
it returned:
pi@raspberrypi:~/giarepo $ openaps device add giapump medtronic
Traceback (most recent call last):
File "/usr/local/bin/openaps-device", line 4, in <module>
import('pkg_resources').run_script('openaps==0.1.5', 'openaps-device')
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 719, in run_script
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 1504, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-device", line 35, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/cli/init.py", line 47, in call
self.configure_parser(self.parser)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-device", line 21, in configure_parser
self.commands = devices.configure_commands(parser, parent=self)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/init.py", line 36, in configure_commands
parent.devices = get_device_map(parent.config)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/init.py", line 21, in get_device_map
for device in Device.FromConfig(vendors, conf):
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/device.py", line 79, in FromConfig
device.read(config=config)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/device.py", line 36, in read
self.extra.fields.update(dict(extra.items(self.section_name( ))))
File "/usr/lib/python2.7/ConfigParser.py", line 642, in items
raise NoSectionError(section)
ConfigParser.NoSectionError: No section: u'device "giapump"'
Paul Davis
@SirBinky
Jul 10 2016 04:47
giapump doesn't exist yet?
Ben West
@bewest
Jul 10 2016 04:47
that last command should create it
Paul Davis
@SirBinky
Jul 10 2016 04:48
openaps device add giapump medtronic
that command
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:48
that is what i used
Screen Shot 2016-07-09 at 9.49.57 PM.png
snapshot from bitbucket
Ben West
@bewest
Jul 10 2016 04:52
use openaps-import:
curl -s https://gist.githubusercontent.com/bewest/d3db9ca1c144b845382c885138a8f66e/raw/e1e69c3fe4dc06e7d3490b4dce9f6380daf8704c/pump.json | openaps import
oops
haha
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:53
did i jump the gun?
pi@raspberrypi:~/giarepo $ curl -s https://gist.githubusercontent.com/bewest/d3db9ca1c144b845382c885138a8f66e/raw/e1e69c3fe4dc06e7d3490b4dce9f6380daf8704c/pump.json | openaps import
medtronic://pump
pi@raspberrypi:~/giarepo $ openaps report show
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 4, in <module>
import('pkg_resources').run_script('openaps==0.1.5', 'openaps-report')
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 719, in run_script
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 1504, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-report", line 82, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/cli/init.py", line 47, in call
self.configure_parser(self.parser)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-report", line 50, in configure_parser
available = devices.get_device_map(self.config)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/init.py", line 21, in get_device_map
for device in Device.FromConfig(vendors, conf):
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/device.py", line 79, in FromConfig
device.read(config=config)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/devices/device.py", line 36, in read
self.extra.fields.update(dict(extra.items(self.section_name( ))))
File "/usr/lib/python2.7/ConfigParser.py", line 642, in items
raise NoSectionError(section)
ConfigParser.NoSectionError: No section: u'device "giapump"'
hmm
there's some report in openaps.ini that is referencing giapump
curl -s https://gist.githubusercontent.com/bewest/d3db9ca1c144b845382c885138a8f66e/raw/57816c09cb83904b772db4fe8411ab282a0a21a7/custom_pump.json | openaps import
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:54
pi@raspberrypi:~/giarepo $ curl -s https://gist.githubusercontent.com/bewest/d3db9ca1c144b845382c885138a8f66e/raw/57816c09cb83904b772db4fe8411ab282a0a21a7/custom_pump.json | openaps import
Traceback (most recent call last):
File "/usr/local/bin/openaps-import", line 4, in <module>
import('pkg_resources').run_script('openaps==0.1.5', 'openaps-import')
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 719, in run_script
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 1504, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-import", line 89, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-import", line 81, in run
incoming = self.Make(candidate)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-import", line 63, in Make
required.append(configurable.get(field, core.get(field)))
AttributeError: 'NoneType' object has no attribute 'get'
Ben West
@bewest
Jul 10 2016 04:55
oh
ok
curl -s https://gist.githubusercontent.com/bewest/d3db9ca1c144b845382c885138a8f66e/raw/ca5c14f879e3294c879ef61480a7e7c50900a233/custom_pump.json | openaps import
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 04:56
pi@raspberrypi:~/giarepo $ curl -s https://gist.githubusercontent.com/bewest/d3db9ca1c144b845382c885138a8f66e/raw/ca5c14f879e3294c879ef61480a7e7c50900a233/custom_pump.json | openaps import
Traceback (most recent call last):
File "/usr/local/bin/openaps-import", line 4, in <module>
import('pkg_resources').run_script('openaps==0.1.5', 'openaps-import')
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 719, in run_script
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 1504, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-import", line 89, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-import", line 81, in run
incoming = self.Make(candidate)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.5-py2.7.egg/EGG-INFO/scripts/openaps-import", line 63, in Make
required.append(configurable.get(field, core.get(field)))
AttributeError: 'NoneType' object has no attribute 'get'
Terri Lyman
@tatigger
Jul 10 2016 04:59
This is where I am starting this process. So I have to partition the disk on my Macbook Pro and Install Ubuntu to run Linux? I might need the remedial chat group
Paul Davis
@SirBinky
Jul 10 2016 04:59
oh no
you download the linux file
then use installer to install ont he microusb card
Ben West
@bewest
Jul 10 2016 05:00
I actually recommend starting on your mac osx
install openaps and learn how to use it there
then get all the hardware and run linux on the raspberrypi
and copy the configuration over
Paul Davis
@SirBinky
Jul 10 2016 05:01
so get the 8gb micro card, put it into an adapter to your computer, download raspberry linux distro (full version) install it onto the card,then put the card into the pi and boot with an hdmi cable, usb keyboard and usb mouse attached to the pi
log into your wifi network then disconnect everything put the powersupply and use ssh client on your computer to access it
Terri Lyman
@tatigger
Jul 10 2016 05:03
I see, but @bewest I just save it into a file in documents? or does Ubuntu have to be on my computer>
Ben West
@bewest
Jul 10 2016 05:03
@GreggHaroldson2 not sure how you got that error, it might be possible to use git to restore to a previous version ,but not sure what commands led up to getting to that error
openaps will run on mac osx and linux both
you don't need linux
Paul Davis
@SirBinky
Jul 10 2016 05:03
u don't need ubuntu get the rapsberry pi linux version
Terri Lyman
@tatigger
Jul 10 2016 05:04
Awesome, thanks both of you : )
Ben West
@bewest
Jul 10 2016 05:04
to set up raspberrypi, you install linux on the rpi using mac osx
Terri Lyman
@tatigger
Jul 10 2016 05:04
I will be able to write "OpenAPS for Dummies" :)
Paul Davis
@SirBinky
Jul 10 2016 05:08
i thought that, doesn't work that way. its horrible and hard to understand you struggle, but actually you need to know it inside and out to write a guide
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 05:10

@bewest - I've had a series of unfortunate events & had to git clone from my repo - not sure if this is the culprit:
pi@raspberrypi:~/giaopenaps $ openaps monitor-pump
giapump://JSON/read_clock/monitor/clock.json
openapreporting monitor/clock.json
giapump://JSON/read_temp_basal/monitor/temp_basal.json
reporting monitor/temp_basal.json
giapump://JSON/iter_pump_hours/monitor/pumphistory.json
reporting monitor/pumphistory.json
calculate-iob://JSON/shell/monitor/iob.json

module.js:340
throw err;
^
Error: Cannot find module '/home/pi/giaopenaps/settings/profile.json'
at Function.Module._resolveFilename (module.js:338:15)
at Function.Module._load (module.js:280:25)
at Module.require (module.js:364:17)
at require (module.js:380:17)
at Object.<anonymous> (/usr/local/lib/node_modules/oref0/bin/oref0-calculate-iob.js:43:22)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Function.Module.runMain (module.js:497:10)
No JSON object could be decoded

GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 05:15
that was captured a few days ago (i only have a few hours here and there to work on this)
Paul Davis
@SirBinky
Jul 10 2016 05:16
why did you need to git clone?
you should work on it then copy the directory as often as possible
it's only like 15mb
GreggHaroldson2
@GreggHaroldson2
Jul 10 2016 05:18
pi crapped out - 8gb sd card (cana kit) said it was out of memory at 6gb - bought a couple 16gb card and had to image them, yada yada
Paul Davis
@SirBinky
Jul 10 2016 05:19
i'd start fresh know what you're meant to do a bit more requently back up in ew sites the files
Mitchell Slep
@moomoobloo
Jul 10 2016 05:19
openaps report invoke nightscout/recent-treatments.json is empty ([]) even though monitor/pump-history.json has treatments, any idea why?
Paul Davis
@SirBinky
Jul 10 2016 05:21
what's the report?
Ben West
@bewest
Jul 10 2016 05:22
it might be empty
it's detecting stuff not in NS yet
Mitchell Slep
@moomoobloo
Jul 10 2016 05:23
you're saying it diffs against what's already in nightscout?
Ben West
@bewest
Jul 10 2016 05:23
if NS is already up to date (there are no new treatments to upload in last five minutes) then it can be empty
Mitchell Slep
@moomoobloo
Jul 10 2016 05:25
well i don't have any treatments in NS yet, but the pump history is more than 5 minutes old so I'll try with a more recent treatment
Ben West
@bewest
Jul 10 2016 05:25
yes, it calls out to NS and does a gap sync to determine how much new data is available
Terri Lyman
@tatigger
Jul 10 2016 05:25
I found the READ MEs and got them open!!! Only took a half hour. Now that everything is in the download folder, what do I do with them?
Paul Davis
@SirBinky
Jul 10 2016 05:26
ignore them...
Mitchell Slep
@moomoobloo
Jul 10 2016 05:28
openaps report invoke nightscout/recent-missing-entries.jsongives me a call stack that ends with AttributeError: 'NoneType' object has no attribute 'read'
I saw some people had this problem earlier but not sure if there was a resolution
(I'm doing the nightscout walkthrough and running into a bunch of issues)
Paul Davis
@SirBinky
Jul 10 2016 05:31
which walk through?
Mitchell Slep
@moomoobloo
Jul 10 2016 05:31
some mashup of the old one and the new simpler one, but neither of them are working :)
Paul Davis
@SirBinky
Jul 10 2016 05:32
yeah think there are 3 versions really
old one stumped me, new one was even worse
relies on alias of alot of different report files from the earlier walk through reports
Mitchell Slep
@moomoobloo
Jul 10 2016 05:33
yeah i've got everything set up and looping except for the nightscout part
this part is definitely the most confusing for me so far
Mitchell Slep
@moomoobloo
Jul 10 2016 05:46

@bfaloona @Spazholio I'm hitting the same issue you posted here, did you ever figure it out?

https://gist.github.com/bfaloona/6e2e9ea3ee734614631d8f99f0f7d528

Brandon Faloona
@bfaloona
Jul 10 2016 05:48
I did not. I realized that I don't need that report.
Mitchell Slep
@moomoobloo
Jul 10 2016 05:50
what is it supposed to do?
@bfaloona when you deleted it, did you also delete nightscout/uploaded-entries.json? uploaded-entries seems to have some dependency on recent-missing-entries (at least that
Mitchell Slep
@moomoobloo
Jul 10 2016 05:56
's what I think the remainder = entries.json nightscout/recent-missing-entries.json line means)
Brandon Faloona
@bfaloona
Jul 10 2016 06:53
@moomoobloo I can't check my config now as I'm traveling. A fairly recent version should be up at https://github.com/bfaloona/openaps-lp
Ben West
@bewest
Jul 10 2016 06:58
neat, looks like templates worked for you
hmm your ns api secret is out in public, which is not a big deal, but you should be aware of that :-)
nice, I like having the raw stuff separated out and everything organized
Mitchell Slep
@moomoobloo
Jul 10 2016 07:30

@bewest I'm seeing that recent-treatments.json is empty even with a new treatment. Here's how I configured the report (based on the walkthrough):

[report "nightscout/recent-treatments.json"]
oper = format-recent-history-treatments
use = shell
reporter = JSON
json_default = True
device = ns
remainder = monitor/pumphistory.json settings/model.json

Since it uploads treatments from pumphistory.json, does that mean I either need to fetch the pump history again after I enact, or else the treatment will only get uploaded on the next iteration?

now i'm realizing you said it only uploads treatments from the past 5 minutes. so if my loop is running every 5 minutes, that means i need to fetch pumphistory again, otherwise treatments won't get uploaded?
Jarred Yaw
@jyaw
Jul 10 2016 13:51
Wild morning, had my first git issue after a battery swap. Oref0-reset-git couldn't repair an empty commit file... so I eventually did git reset --hard
Any reason why oref0-reset-git wouldn't detect an issue?
Jason Calabrese
@jasoncalabrese
Jul 10 2016 15:59
There was a recent change to reset-git, are you current with dev?
@jyaw ^
Jarred Yaw
@jyaw
Jul 10 2016 16:31
@jasoncalabrese thanks for letting me know. I'm on the ama branch... would I still have that change if I do a pull?
Terri Lyman
@tatigger
Jul 10 2016 16:59
starting from absolute zero. downloaded the openaps source, and now I have no idea what to do to get an instance setup on my MacBook
Paul Davis
@SirBinky
Jul 10 2016 17:00
i thought it needed a linux base?
@tatigger normally you set up the system as a raspberry pi based mini system you carry with you
Terri Lyman
@tatigger
Jul 10 2016 17:04
Yes, but I am trying to learn from the beginning, so installing on my Mac to run there
Jason Calabrese
@jasoncalabrese
Jul 10 2016 17:07
@jyaw you can try merge-ama I'm running it in production now, but still needs some testing, it's merging in everything from dev
Scott Leibrand
@scottleibrand
Jul 10 2016 17:11
@tatigger I disagree with Ben on that: I would only advise setting up openaps on a Mac if you're a command-line guru who doesn't have a Pi yet. Much easier IMO to follow the walkthrough and do it on the Pi other wise.
Paul Davis
@SirBinky
Jul 10 2016 17:13
otherwise you need to be within a few feet of the mac constantly for the carelink to work if you're using that
Terri Lyman
@tatigger
Jul 10 2016 17:16
Ok. Will try that :)
Eric
@ecc1
Jul 10 2016 17:20
Anyone know of an Android Wear watchface for openaps visualization, or is everyone using Pebbles?
Paul Davis
@SirBinky
Jul 10 2016 17:25
openaps or nightscout?
nightscout is a site, you set it up if you can view a website you can see it
Jason Calabrese
@jasoncalabrese
Jul 10 2016 17:27
I don't know of anything other than Urchin for pebble
Dana Lewis
@danamlewis
Jul 10 2016 17:36
@ecc1 no one has set anything up yet, it's either pebble (99%) or I know of one person who has an Apple Watch vis
Eric
@ecc1
Jul 10 2016 17:54
Thanks, that's what it was looking like from my googling.
Jarred Yaw
@jyaw
Jul 10 2016 18:10
@jasoncalabrese will give that a try then. Thanks!
ssaint1
@ssaint1
Jul 10 2016 20:27
Anyone know why the "configuring and learning to use Openaps tools" page is not working?
Dana Lewis
@danamlewis
Jul 10 2016 20:28
Because someone needs to update the readme to reflect the new page org :)
Use the RTD link in the mean time for the docs
ssaint1
@ssaint1
Jul 10 2016 20:29
Hi Dana. Not sure what that means. I tried to navigate there through Git and couldnt find it either. Thanks Paul will try that.
Paul Davis
@SirBinky
Jul 10 2016 20:29
worked fine for me?
ssaint1
@ssaint1
Jul 10 2016 20:29
@SirBinky Thanks, worked perfect.
Dana Lewis
@danamlewis
Jul 10 2016 20:29
@ssaint1 there was some content moved arohnd
The readme with its outline needs to be updated to match
Paul Davis
@SirBinky
Jul 10 2016 20:30
@danamlewis so much moved around atm
Dana Lewis
@danamlewis
Jul 10 2016 20:30
Good opportunity to submit a Pr, @SirBinky
ssaint1
@ssaint1
Jul 10 2016 20:30
OK, thanks. My rig is up and running and doing everything EXCEPT reporting "treatments" to nightscout (status is reporting fine). I'm struggling a bit with this one.
Dana Lewis
@danamlewis
Jul 10 2016 20:31
"if you see something, say something" and in OpenAPS world, the next step is "do something" :) if you can help fix
Paul Davis
@SirBinky
Jul 10 2016 20:32
@danamlewis i've tried i thought i was tweaking stuff but i had the wrong idea and attitude for some stuff. now i'm going back over the doc's they've changed so much. everyone names the reports differently. i think it would really help if the json files were just the commands ie pump history is iter_glucose_hours 4 etc
also i think @scottleibrand uses -'s to name files and @bewest uses _'s so you need to be careful to unify what you're doing
danamlewis @danamlewis endlessly suggests filing PRs
Ben West
@bewest
Jul 10 2016 20:36
fwiw, there are permanent archives of the older docs
what dates are missing? I just restored four or five checkpoints
Dana Lewis
@danamlewis
Jul 10 2016 20:37
Dates, @bewest?
Paul Davis
@SirBinky
Jul 10 2016 20:37
i can and will do i'm learning new stuff constantly etc. also we've got a g5 and medtronic pump i didn't get the internet nighscout stuff thought it was a weakness wanted to have a totally offline setup
Dana Lewis
@danamlewis
Jul 10 2016 20:37
@SirBinky you can have a setup that works offline but that also takes advantage of some powerful vis tools for when you do have connectivity.
It's not an absolute pick between the two
Paul Davis
@SirBinky
Jul 10 2016 20:38
@danamlewis yeah i wanted a totally offine set up ignore the advantages of having an occasionally online visualisation set up, now i've realised my mistake and i'm trying to do that
Ben West
@bewest
Jul 10 2016 20:39
blob
here's a link to an older version of the docs
Dana Lewis
@danamlewis
Jul 10 2016 20:39
Oh, excellent @bewest
Ben West
@bewest
Jul 10 2016 20:39
that's one of the reasons we're doing things this way with these tools ;)
danamlewis @danamlewis cheers
Ben West
@bewest
Jul 10 2016 20:40
any time you want to create a permanently archived version, hit the release button on github
since we're not doing dev merges anymore, I suggest just doing it every few weeks
Dana Lewis
@danamlewis
Jul 10 2016 20:40
:+1: so best practice is doing that before a merge dev to master?
K
Ben West
@bewest
Jul 10 2016 20:40
after merge to master
it's done when master changes in theory
Paul Davis
@SirBinky
Jul 10 2016 20:40
@bewest @danamlewis $ rm -f settings/ monitor/ enact/*
$ openaps gather-profile
$ openaps monitor-pump
$ openaps monitor-cgm
$ openaps report invoke monitor/iob.json
$ openaps report invoke enact/suggested.json
Ben West
@bewest
Jul 10 2016 20:41
did you try using the templates?
ssaint1
@ssaint1
Jul 10 2016 20:41
OK, I'm throwing in the towel on trying to do this one on my own. As I mentioned earlier, having a hard time getting treatments uploaded to NS. I skipped that stage as I set everything up. Everything else is working fine. Fully closed loop! I have a ns-status device but when I try to run the nightscout autoconfigure-device-crud I get the following:added process://ns/nightscout/ns NIGHTSCOUT_HOST API_SECRET
usage: openaps-device show [-h] [--ini] [--json]
[{calculate-iob,cgm,curl,determine-basal,get-profile,ns-status,oref0,pump,tz,}]
openaps-device show: error: argument name: invalid choice: 'ns' (choose from u'calculate-iob', u'cgm', u'curl', u'determine-basal', u'get-profile', u'ns-status', u'oref0', u'pump', u'tz', '
')
Traceback (most recent call last):
File "/usr/local/bin/openaps-import", line 4, in <module>
import('pkg_resources').run_script('openaps==0.1.0', 'openaps-import')
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 719, in run_script
File "build/bdist.linux-armv7l/egg/pkgresources/_init.py", line 1504, in run_script
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.0-py2.7.egg/EGG-INFO/scripts/openaps-import", line 89, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.0-py2.7.egg/openaps/cli/init.py", line 51, in call
self.run(self.args)
File "/usr/local/lib/python2.7/dist-packages/openaps-0.1.0-py2.7.egg/EGG-INFO/scripts/openaps-import", line 77, in run
candidates = json.load(args.input)
File "/usr/lib/python2.7/json/init.py", line 290, in load
**kw)
File "/usr/lib/python2.7/json/init.py", line 338, in loads
return _default_decoder.decode(s)
File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
obj, end = self.raw_decode(s, idx=_w(s, 0).end())
File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
Paul Davis
@SirBinky
Jul 10 2016 20:42
the reports and alias's are one thing, then the nightscout intergration docs make you import totally new reports only but you have to import your complete aliases which are totally named differently from the old docs report setups
Ben West
@bewest
Jul 10 2016 20:42
the templates create all the matching reports and aliases
@ssaint1 sudo pip install -U openaps openaps-contrib
Paul Davis
@SirBinky
Jul 10 2016 20:45
the please test this easier setup templates create the ns reports and then copy all your alias's, but the alais's don't match up to the old doc report names at all. i've spent alot of time figuring out what the inital command is for the alias names you make. oref0 stuff wasn't at all in the old doc's. it's split settings, monitor, enact
Ben West
@bewest
Jul 10 2016 20:45
I'm confused, the templates offer a way to create devices, reports and aliases that all match
Paul Davis
@SirBinky
Jul 10 2016 20:46
@bewest the try this new nightscout set up guide page
Ben West
@bewest
Jul 10 2016 20:46
if you can point out which one doesn't match which doc, perhaps we can fix it
ssaint1
@ssaint1
Jul 10 2016 20:46
@bewest Happy to do that, can you give me a small idea of what that will do before I try it? I am new to linux and not sure of when I am doing something unrecoverable. Thanks.
Paul Davis
@SirBinky
Jul 10 2016 20:46
gets you to download the report file then merge it
then the alias file and merge it
Ben West
@bewest
Jul 10 2016 20:47
@ssaint1 it will update openaps
looks like you're using an older version
ssaint1
@ssaint1
Jul 10 2016 20:47
@bewest Fair enough, thanks very much.
Paul Davis
@SirBinky
Jul 10 2016 20:47
but the json report names are set up totally different to how the report names are referenced in the alias's
Tim Mecklem
@tmecklem
Jul 10 2016 20:47
I've tried getting the https://github.com/openaps/openxshareble BLE->dexcom G4 with share working a few times and thrown my hands up in frustration. I'm trying again, will soon have a few questions for the parts where I get stuck.
Ben West
@bewest
Jul 10 2016 20:48
@SirBinky I'm not sure I understand: AFAICT the docs suggest using the templates, the templates create reports and aliases which are all consistent
in the case where they aren't, I need help with a specific link and reference to specific template
Paul Davis
@SirBinky
Jul 10 2016 20:49
ok
paul ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps
monitor-pump && openaps report invoke monitor/iob.json && openaps report invoke enact/suggested.json"
Ben West
@bewest
Jul 10 2016 20:49
a few months ago there were no aliases, and before that there were no reports either
the templates don't create that alias
ssaint1
@ssaint1
Jul 10 2016 20:49

@bewest Now I got this:

added process://ns/nightscout/ns NIGHTSCOUT_HOST API_SECRET
Traceback (most recent call last):
File "/usr/local/bin/openaps-device", line 35, in <module>
app( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 52, in call
self.epilog( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 75, in epilog
self.create_git_commit( )
File "/usr/local/lib/python2.7/dist-packages/openaps/cli/init.py", line 89, in create_git_commit
self.repo.git.add([diff.b_path], write_extension_data=False)
File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 463, in <lambda>
return lambda args, **kwargs: self._call_process(name, args, kwargs)
File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 929, in _call_process
return self.execute(make_call(),
_kwargs)
File "/usr/local/lib/python2.7/dist-packages/git/cmd.py", line 713, in execute
raise GitCommandError(command, status, stderr_value)
git.exc.GitCommandError: 'git add openaps.ini' returned with exit code 128
stderr: 'fatal: Unable to create '/home/pi/seanaps/.git/index.lock': File exists.

If no other git process is currently running, this probably means a
git process crashed in this repository earlier. Make sure no other git
process is running and remove the file manually to continue.'
process://ns/nightscout/ns https://XXX.azurewebsites.net XXX

Ben West
@bewest
Jul 10 2016 20:49
the docs encourage using the templates
Paul Davis
@SirBinky
Jul 10 2016 20:49
yes
but the old walkthrough told me to make a gather alias
Ben West
@bewest
Jul 10 2016 20:50
ok, well the new one says use templates
did you try the template?
ssaint1
@ssaint1
Jul 10 2016 20:50
@bewest Just noticed, there is a NS device now in openaps.ini though. That's new.
Ben West
@bewest
Jul 10 2016 20:50
the old docs are still available, you don't have to switch
Paul Davis
@SirBinky
Jul 10 2016 20:50
but your docs added a get settings
Ben West
@bewest
Jul 10 2016 20:51
@ssaint1 yeah, seems like new problem
Paul Davis
@SirBinky
Jul 10 2016 20:51
@bewest ok after doing the walk through then importing your alias's for the nightscout i've ended up with

gather-profile report invoke settings/settings.json settings/bg_targets_raw.json settings/bg_targets.json settings/insulin_sensitivities_raw.json settings/insulin_sensitivities.json settings/basal_profile.json settings/profile.json

monitor-pump report invoke monitor/clock.json monitor/temp_basal.json monitor/pumphistory.json monitor/iob.json

monitor-cgm report invoke raw-cgm/glucose-raw.json monitor/glucose.json

paul ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps
monitor-pump && openaps report invoke monitor/iob.json && openaps report invoke enact/suggested.json"

paul1 ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps monitor-pump && openaps report invoke monitor/iob.json"

gather ! bash -c "rm -f monitor/*; openaps gather-profile && openaps monitor-cgm && openaps monitor-pump && openaps report invoke monitor/iob.json"

preflight ! bash -c "(echo PREFLIGHT ) && openaps warmup 2>&1 >/dev/null && grep -q T monitor/clock.json && echo PREFLIGHT OK || openaps fail-warmup"

rm-warmup ! bash -c "rm -f model.json monitor/clock.json > /dev/null"

warmup report invoke model.json monitor/clock.json monitor/clock-zoned.json

fail-warmup ! bash -c "echo PREFLIGHT FAIL; exit 1"

monitor-pump-history report invoke raw-pump/pump-history-raw.json monitor/pump-history.json
get-basal-status report invoke monitor/temp-basal-status.json

get-pump-details report invoke monitor/reservoir.json monitor/status.json monitor/battery.json

get-settings report invoke settings/bg_targets_raw.json settings/bg_targets.json settings/insulin_sensitivities_raw.json settings/insulin_sensitivities.json settings/basal_profile.json settings/profile.json settings/settings.json

gather-pump-data ! bash -c "openaps get-basal-status; openaps get-pump-details; openaps monitor-pump-history; openaps get-settings"

gather-clean-data ! bash -c "openaps monitor-cgm && openaps gather-pump-data"

do-oref0 report invoke oref0-monitor/profile.json oref0-monitor/iob.json oref0-predict/oref0.json

enact-oref0 report invoke oref0-enacted/enacted-temp-basal.json

do-everything ! bash -c "(openaps preflight && openaps gather-clean-data && openaps do-oref0 && openaps enact-oref0); openaps report-nightscout"

ping ! bash -c "echo PONG! $*" --

pong ! echo PONG

hello ! echo hello

okbar ! bash -c "echo $1" --

report-nightscout report invoke nightscout/preflight.json nightscout/recent-treatments.json nightscout/uploaded.json nightscout/recent-missing-entries.json nightscout/uploaded-entries.json

Ben West
@bewest
Jul 10 2016 20:52
@ssaint1 try oref-fix-git-corruption perhaps, hmmm looks like the git dir got squirrely
jaylagorio
@jaylagorio
Jul 10 2016 20:53
@jasoncalabrese You wanted to see the bag I carry everything in: http://imgur.com/a/Npsz1
I can throw it in a backpack and head out the door along with the stuff I usually throw in there.
Ben West
@bewest
Jul 10 2016 20:54
so now you need a raw-cgm/glucose-raw.json monitor/glucose.json reports
Paul Davis
@SirBinky
Jul 10 2016 20:54
@bewest anyway doesn't matter hugely it's made me understand how things work etc, maybe all the new docs are more consistent
Ben West
@bewest
Jul 10 2016 20:54
how are you getting glucose data?
what problem are you actually experiencing/trying to solve right now? now that you have imported the alias templates, if you've also imported the medtronic and oref0 reports, you only need to define how to get cgm data
Paul Davis
@SirBinky
Jul 10 2016 20:55
@bewest i get my glucose from a g5 reciever so only have monitor/glucose.json not the raw data
Ben West
@bewest
Jul 10 2016 20:55
specifically, you should define reports to put cgm data in raw-cgm/glucose-raw.json monitor/glucose.json and it should work just fine
ok, so you should do openaps use cgm oref0_glucose --no-raw --hours 2 or similar
and save that as raw-cgm/glucose-raw.json
or if you're using iter_glucose or similar, either way it should be called raw-cgm/glucose-raw.json
this is to separate raw-data from devices from the more polished, reformatted versions
then you should format it using openaps use tz rezone --astimezone ... and save that as monitor/glucose.json
then everything will work
Paul Davis
@SirBinky
Jul 10 2016 20:58
yeah i didn't get why you had it as raw data until i tried to do the nightscout stuff and how you need to format it
Ben West
@bewest
Jul 10 2016 20:58
we simply don't offer a copy/paste walkthrough for people
there's also issues with adjusting the timestamps to all be zoned the same way
and other steps to throw out bad data
Paul Davis
@SirBinky
Jul 10 2016 20:59
@bewest i know you don't i'm not complaining you don't. the way you've set up the docs makes perfect sense, you don't want an idiot copying it all wrong and pumping alot of insulin in to themselves or a loved one
Ben West
@bewest
Jul 10 2016 21:00
well, there are links to both the new docs and the old docs
Paul Davis
@SirBinky
Jul 10 2016 21:00
also a few times i've had errors it spits out this command is wrong your options are these
Ben West
@bewest
Jul 10 2016 21:00
there's also a guide for exploring how to use openaps and to help learn what the errors mean
and I do encourage running that on mac osx just to figure out how openaps works without the complexity of installing OS or working with an embedded device
Paul Davis
@SirBinky
Jul 10 2016 21:01
mostly the error's i've had is missing json files
@bewest ugh no you lost now :P no mac
ssaint1
@ssaint1
Jul 10 2016 21:02
@bewest I am clearly impaired. How do I use that command? I tried every combination of openaps, use, oref0 and that command I could think of. Thanks
Ben West
@bewest
Jul 10 2016 21:02
it just runs as is
ssaint1
@ssaint1
Jul 10 2016 21:02
I tried that too. No luck
Paul Davis
@SirBinky
Jul 10 2016 21:02
openaps use
Ben West
@bewest
Jul 10 2016 21:02
ah ok
sudo npm install -g git://github.com/openaps/oref0.git'#dev'
@SirBinky running the reports requires access to the devices to fetch the data
if openaps use pump model doesn't work, it won't run, and it it does run it will complain about missing data
ssaint1
@ssaint1
Jul 10 2016 21:04
@bewest Just checking since you are chatting with two of us, did you want me to run the npm install? What is that?
Paul Davis
@SirBinky
Jul 10 2016 21:04
@bewest when openaps use pump model didn't work it just showed me "" instead of "458"
Ben West
@bewest
Jul 10 2016 21:05
@ssaint1 it updates oref0, it sounds like you don't have that command (because your version is older?)
Paul Davis
@SirBinky
Jul 10 2016 21:05
@ssaint1 that's all aimed at you
Ben West
@bewest
Jul 10 2016 21:05
@SirBinky right, that means it won't work because you won't have any pump history or glucose data (or any other data)
what should it do if it doesn't get any data from the pump?
Paul Martin
@Spazholio
Jul 10 2016 21:06
Mental note - don’t forget to renew your letsencrypt SSL certs ontime. Various parts of your loop will go ka-flooey if you forget. Grrr.
Paul Davis
@SirBinky
Jul 10 2016 21:07
@bewest i know, i was just saying that it didn't spit alot of data at me it just did "" in that circumstance i got my gf to take her pump out and place it closer to the usb carelink so it'll connect
Ben West
@bewest
Jul 10 2016 21:07
highly recommend the TI sticks, or even the rileylinks to greatly extend range
ssaint1
@ssaint1
Jul 10 2016 21:07
Thanks guys, will give it a shot. just to be clear, everything else works. Pump is happily looping away as we type.
Paul Davis
@SirBinky
Jul 10 2016 21:07
ti?
can cover multiple rooms, larger spaces
Paul Davis
@SirBinky
Jul 10 2016 21:09
this is the other thing when you start to dive into all of this stuff. there are so so many options
it's great but a bit overwhelming
Paul Martin
@Spazholio
Jul 10 2016 21:10
If some of my pump data isn’t making it into NS, that will affect visualization ONLY, correct? The loop doesn’t use any pump data pulled from NS, just directly from the pump right?
(my SSL certs expired and things went all sideways for an hour before I realized)
ssaint1
@ssaint1
Jul 10 2016 21:11

I finally noticed that the command (which I cut and pasted) said oref not oref0, so I tried that and got this:
pi@raspberrypi:~/seanaps $ oref0-fix-git-corruption
Git repo does not appear to be corrupt.

I guess that is good. Now that I have the NS device, some of the other stuff seems to be working. I added the NS treatment reports. Now just figuring what to do with them. Progress.

Ben West
@bewest
Jul 10 2016 21:12
@ssaint1 interesting, try: git commit -avm 'clear'
you might be good to go: try openaps use ns shell preflight
Paul Davis
@SirBinky
Jul 10 2016 21:12
@Spazholio where do you get your bg info from? if it's a reciever connected to the raspberry it'll get just that
Jason Calabrese
@jasoncalabrese
Jul 10 2016 21:17
Nice @jaylagorio, any luck with the underclocking?
Paul Davis
@SirBinky
Jul 10 2016 21:18
@bewest you can make an xdrip upload to an android need internet, use a dexcom receiver too or an iphone and upload to nightscout, you can pull bg readings from the dexcom receiver direct or from the nightscout site. you can use a pi or edison or other device or combination of nightscout and receiver etc it can be quite overwhelming to start with
Ben West
@bewest
Jul 10 2016 21:19
I'm sorry if that's a problem
Paul Davis
@SirBinky
Jul 10 2016 21:20
@bewest lol no not at all it's great helps everyone you've not restricted it
ssaint1
@ssaint1
Jul 10 2016 21:22

@bewest I am no Git expert, I ran the command and got:

pi@raspberrypi:~/seanaps $ git commit -avm 'clear'
On branch master
Your branch is ahead of 'origin/master' by 19482 commits.
(use "git push" to publish your local commits)
Untracked files:
.gitignore
calculate-iob.ini
cgm.ini
determine-basal.ini
get-profile.ini
ns-status.ini
ns.ini
oref0.ini
preferences.json
pump-session.json
seansdex.ini

nothing added to commit but untracked files present

Which makes me think I am doing something a little wrong. I am saving my stuff to my own Git account.

Also, things seem to be working now, although I am still sort of figuring out what needs to be done.

Paul Davis
@SirBinky
Jul 10 2016 21:22
@bewest i can try to update the docs to help others, but you guys are delving into the code and making this stuff work. openaps is a magnificent work of art
jaylagorio
@jaylagorio
Jul 10 2016 21:22
@jasoncalabrese I haven't tried it yet. So far keeping that vent on the top unblocked seems to be helping, but if by tomorrow morning it freezes again I'll attempt the underclock.
I'm optimistic that that's all I need to do but will try something else if proven otherwise.
Paul Davis
@SirBinky
Jul 10 2016 21:23
seanaps? not openaps?
oh that's your dir
ssaint1
@ssaint1
Jul 10 2016 21:24
@SirBinky Yeah, when I first set the thing up it wasnt clear that I should name it that, so I was sort of nmaming things whatever. Mistake in hindsight. Have of me wants to start completely over and fresh.
Paul Davis
@SirBinky
Jul 10 2016 21:24
new sd card is only a few quid
or dollars
ssaint1
@ssaint1
Jul 10 2016 21:24
Its not the cost, its the time.
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:25

So if my monitor/glucose.json contains data but I get this error.

pi@artificialpancreas:~/awesomepancreas
openaps report invoke enact/suggested.json
determine-basal://text/shell/enact/suggested.json
Could not parse input data:  [SyntaxError: /home/pi/awesomepancreas/monitor/glucose.json: Unexpected end of input]
reporting enact/suggested.json

it must not contain the right data?

Tim Mecklem
@tmecklem
Jul 10 2016 21:25
Ok. I'm to a place where I need help with openxshareble. Should I ask here or somewhere else less noisy or more specific to the task?
Paul Martin
@Spazholio
Jul 10 2016 21:25
@SirBinky I don’t use a Pi, and it’s not connected to my receiver. And…that’s not at all what I asked. =)
Paul Davis
@SirBinky
Jul 10 2016 21:26
@ssaint1 it's making it work right. i'm learning alot each time i try stuff, i've added alot of junk and excess, once i've got it all working right i'm going to start over fresh and yeah it'll take some time but if i can start over fresh make it work right first time it means i know what i'm doing
@Spazholio i was asking where you get your bg so i could answer how it would affect your situation
ssaint1
@ssaint1
Jul 10 2016 21:28
@SirBinky That sounds about right not me. I am feeling very comfortable with aspects of this right now (like the openaps.ini file) but very uncomfortable with other aspects. Like Git, updating openaps, oref0, etc.
Paul Martin
@Spazholio
Jul 10 2016 21:28
But my BG isn’t my concern at all. It’s data from my pump.
Tim Mecklem
@tmecklem
Jul 10 2016 21:29
openaps use share list_dexcom looks like it returns the dexcom I have in pairing mode, and openaps use share configure shows serial and mac entries match the dexcom. When I run openaps use share iter_glucose_hours 3, it doesn't see the device and errors with
AttributeError: 'NoneType' object has no attribute 'iter_records'
Sulka Haro
@sulkaharo
Jul 10 2016 21:29
@shgrady any time the loop complains about a file, I suggest just checking the file contents. In my experience the issues is often immediately obvious when seeing the contents; it's likely to be empty or an error message
Tim Mecklem
@tmecklem
Jul 10 2016 21:30
Is there an actual pairing that I need to do outside the setup process? I'm not really up on BLE and how that pairing works. The dexcom is still flashing the Bluetooth symbol.
Paul Davis
@SirBinky
Jul 10 2016 21:30
@Spazholio oh right sorry i thought the pump just had more steady data. your set up should just pull the data direct from the pump. an extra step if upload data and format for nightscout etc
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:30
@sulkaharo yeah, I am looking at the contents, but other than showing I am having a rough day with unexplained hyperglycemia, not telling me whats wrong with it....
Sulka Haro
@sulkaharo
Jul 10 2016 21:31
Right. :) Where are you pulling the file from?
Paul Davis
@SirBinky
Jul 10 2016 21:31
@Spazholio so it could pull it from the pump ok but then something messes up the nightscout data
Tim Mecklem
@tmecklem
Jul 10 2016 21:32
I'd like to mention again that I have access to a nearby Microcenter and they've been pretty stocked with Pi Zeros. If anyone would like me to buy one and ship it to you at cost for the openaps project, I'm willing.
Hard to beat a $5 tiny computer.
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:33

last object of monitor/glucose.json

{
    "filtered": 244000,
    "direction": "NOT COMPUTABLE",
    "noise": 1,
    "type": "sgv",
    "_id": "5782ac3f622181a4e69e7cb1",
    "sysTime": "2016-07-10T16:11:17.911-0400",
    "dateString": "2016-07-10T20:11:17.911000+00:00",
    "sgv": 249,
    "device": "xDrip-DexcomShare",
    "xDrip_filtered": 244,
    "unfiltered": 239000,
    "xDrip_raw": 239,
    "date": 1468181477911,
    "rssi": 100,
    "xDrip_calculated_current_slope": 1.2281475271863614e-05,
    "xDrip_age_adjusted_raw_value": 239,
    "xDrip_calculated_value": 249,
    "glucose": 249
  }

@sulkaharo

Brett
@burgertime14
Jul 10 2016 21:34
@tmecklem I was just looking for one on amazon/adafruit...
Sulka Haro
@sulkaharo
Jul 10 2016 21:34
@shgrady hmm that looks fine
Tim Mecklem
@tmecklem
Jul 10 2016 21:34
They're still in short supply online unless you want to be price gouged
(Or buy a bundle that includes stuff you don't need and costs 4-5 times as much as a bare zero)
Brett
@burgertime14
Jul 10 2016 21:35
@tmeIcklem I may take you up on your offer, I am working in tandem with a buddy and I will see if he wants to go that way, did you solder a wifi board on to it?
Tim Mecklem
@tmecklem
Jul 10 2016 21:36
No, I have a Slice of Radio and an edimax adapter.
Paul Davis
@SirBinky
Jul 10 2016 21:36
@tmecklem lol it'a not got wifi or bt it's slower than the 3 less ram...
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:36
so, from my earlier error @sulkaharo is the problem somewhere else than I thought, I was thinking monitor/glucose.json the problem but maybe not reading it right?
Tim Mecklem
@tmecklem
Jul 10 2016 21:36
Trying to lose the edimax wifi and add a bluetooth adapter to get dexcom and share iphone internet
goal is tiny pi with tiny battery. Don't need wifi or quad core processing with the energy cost it adds.
Paul Davis
@SirBinky
Jul 10 2016 21:38
@burgertime14 cool
Tim Mecklem
@tmecklem
Jul 10 2016 21:38
That looks interesting. I'd like to punt and not have wifi at all
Brett
@burgertime14
Jul 10 2016 21:38
I still don't have a receiver for my dex so I have to have access to the internets
Sulka Haro
@sulkaharo
Jul 10 2016 21:39
@shgrady are you sure you're passing all arguments in the right order?
Tim Mecklem
@tmecklem
Jul 10 2016 21:39
anyone in here successfully use the openxshareble device and willing to be a rubber duck?
Sulka Haro
@sulkaharo
Jul 10 2016 21:39

Also, you can try running

json --validate -f /home/pi/awesomepancreas/monitor/glucose.json

to see if the file is valid

Tim Mecklem
@tmecklem
Jul 10 2016 21:40
@burgertime14 let me know via DM if you end up wanting to take me up on the pi zero offer.
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:40

@sulkaharo I was just thinking the same thing, because after invoking the monitor/glucose.json, I got a different error this time

pi@artificialpancreas:~/awesome
pancreas $ openaps report invok
e enact/suggested.json
determine-basal://text/shell/enact/suggested.json
BG data is too old, or clock set incorrectly Sun Jul 10 2016 21:16:16 GMT+0000 (UTC) vs Sun Jul 10 2016 21:38:03 GMT+0000 (UTC)
reporting enact/suggested.json

running suggested command now

Brett
@burgertime14
Jul 10 2016 21:41
@tmecklem awesome, thanks, wish we had a micro center here :(
ssaint1
@ssaint1
Jul 10 2016 21:41
@bewest OK, making real progress here. I was able to define the reports finally (nightscout/recent-treatments.json and nightscout/uploaded.json). I ran the uses and the first made the file it was supposed to and the second (upoad-non-empty-treatments) seemed to work, it echoed the uploaded file to the screen with no errors. So, based on the description of what this does in the documentation I would have expected treatments to now be in nightscout, which they don't seem to be. Any suggestions?
One thing I was wondering about is it seems that most times I deal with nightscout I have to give it my host and api, I dont do that here. Where is it grabbing them? When I type "env" the host and api are in there, although the api is hashed. Not sure if it needs to be or not. In my crontab file they are also both there (non hased api). Not sure where I should be storing this stuff, or for that matter how it got in the "env".
Brett
@burgertime14
Jul 10 2016 21:41
or a frye's
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:42
@sulkaharo validate command gave no response, is that a negative?
Sulka Haro
@sulkaharo
Jul 10 2016 21:43
That's a positive; file is valid JSON
Or however way negative / positive goes :D
The loop refuses to act on old BG data, the latter error is about the data being too old
Is it possible the earlier error was caused by the file being updated concurrently with the loop trying to act on the data?
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:47
@sulkaharo NS appears to be showing current data, and I just ran that report thinking the clock is actually off
Sulka Haro
@sulkaharo
Jul 10 2016 21:49
(right, hitting bed!)
Sarah Hodson Grady
@shgrady
Jul 10 2016 21:53
Thanks @sulkaharo
sometimes I just need someone to talk me through the obvious
Sulka Haro
@sulkaharo
Jul 10 2016 21:59
We're all here to help! ❤️
Jason Calabrese
@jasoncalabrese
Jul 10 2016 22:00
@shgrady it could also be that you're not zoning the data or not passing the zoned data to determine-basal
Tim Mecklem
@tmecklem
Jul 10 2016 22:57
I'll take a look. I think that contains the missing pieces for me. Thanks!
Chris Oattes
@cjo20
Jul 10 2016 23:16
I've got a few options for what I do about my Edison.
  1. Forget about it, use my (very bulky) pi
  2. Just buy a new edison, go back to using the sparkfun board I've got, use a big external battery and OTG cable
  3. Buy a new edison + intel breakout board so I get to use my lipo and UART cc1111
Which do I go for?
Paul Davis
@SirBinky
Jul 10 2016 23:19
@cjo20 you're def leaning to the edison route
Chris Oattes
@cjo20
Jul 10 2016 23:20
Yup. I already had an edison, and I fried it.
Paul Davis
@SirBinky
Jul 10 2016 23:21
@cjo20 i know that's why you're wanting new device. I went for pi and try out, but you know edison and how it is. even if you fried it your sd card should be ok?so just straight swap it?
Chris Oattes
@cjo20
Jul 10 2016 23:22
The memory is built in to the edison
Paul Davis
@SirBinky
Jul 10 2016 23:23
oh
Chris Oattes
@cjo20
Jul 10 2016 23:24
I haven't lost anything valuable on it though. It'll take me less than an hour to get a new one up and running properly
Paul Davis
@SirBinky
Jul 10 2016 23:24
@cjo20 mate it's up to you, whatever you can make work and you're happiest with...
Dana Lewis
@danamlewis
Jul 10 2016 23:28
@cjo20 you probably already know your answer from the three options ;)
Chris Oattes
@cjo20
Jul 10 2016 23:29
I know what I'd like to do :P I also know how much it costs
mariano42
@mariano42
Jul 10 2016 23:30
hi all.. hoping someone could help me with an issue I am having with adding some oref0. I think I missed a step somewhere along the line that creates the glucose.json (and possibly some other files in /monitor). when I try "openaps report add monitor/glucose.json JSON cgm iter_glucose 5" it seems to complete, but nothing gets to the glucose.json file i manually created in /monitor. When I try to run"openaps monitor-cgm" i get an error "cgm://JSON/iter_glucose/monitor/glucose.json
monitor/glucose.json raised run_mainloop_with() got an unexpected keyword argument 'quit_with_loop'". I haven't actually hooked up my dexcom at all, my cgm is setup with the device openxshareble. I am not sure that should matter. So far I only see that the scripts created 4 files in the /monitor directory.. I noticed in bewest's example on github there were 7 json files. I am worried I missed something, but backtracked and haven't identified any place like that. Anyone have any ideas?
Paul Davis
@SirBinky
Jul 10 2016 23:32
@cjo20 yeah my gf has a pump from the nhs but anything else we have to pay for, she got new g5 sensors and the transmitter will run out in 2 months now. £240
Chris Oattes
@cjo20
Jul 10 2016 23:32
Which pump does she have?
Paul Davis
@SirBinky
Jul 10 2016 23:33
@mariano42 you're getting the bg from xshareable?
@cjo20 erm dunno one that's compatable to the usual files i think maybe 428?
mariano42
@mariano42
Jul 10 2016 23:34
yes.. i can see data in reports @SirBinky
Paul Davis
@SirBinky
Jul 10 2016 23:36
@mariano42 you see the data in what reports? use that report file instead of monitor_glucose
or make a new report that gets the bg data from your cgm and create monitor/glusoce.json
mariano42
@mariano42
Jul 10 2016 23:39
@SirBinky sorry, i misspoke.. i see data if do a get entries.json, I am not sure what report to check
Paul Davis
@SirBinky
Jul 10 2016 23:41
you see suggested basal from entries? what do you want to check?
Ben West
@bewest
Jul 10 2016 23:42
creating monitor/glucose.json is usually a two step task
step 1 is to figure out which device is generating the data, and save it in a "raw" file
step 2 is to use one of the tools to ensure it's the right format and everything is good with it
Paul Davis
@SirBinky
Jul 10 2016 23:44
@bewest i worked that out from your alias's but i thought with a g5 you don't need to format the data?
Ben West
@bewest
Jul 10 2016 23:44
if data is being fetched from nightscout, consider openaps use ns shell oref0_glucose tz count=20 or similar
you generally always need to format the data
Paul Davis
@SirBinky
Jul 10 2016 23:44
it's from a g5 receiver usb'd to the pi
Ben West
@bewest
Jul 10 2016 23:44
step 2 is generally doing openaps use tz rezone --astimezone
especially since data from receiver doesn't have time zone
Paul Davis
@SirBinky
Jul 10 2016 23:46
so i get data from the device then openaps use tz rezone "glucose file"
mariano42
@mariano42
Jul 10 2016 23:46
@bewest thanks.. i will check that out
Ben West
@bewest
Jul 10 2016 23:47
use tz rezone --astimezone yup
Paul Davis
@SirBinky
Jul 10 2016 23:49
the openaps instance you guys make is so simple and useful. the "programming" i need to do is mainly setting up reports and alias's