Fwd: Re: OVMS App missing most charge points, detail info wrong
FYI -------- Weitergeleitete Nachricht -------- Betreff: Re: OVMS App missing most charge points, detail info wrong Datum: Mon, 12 Jan 2015 18:30:38 +0800 Von: Christopher Cook (Open Charge Map) <christopher.cook@openchargemap.net> An: Michael Balzer <dexter@expeedo.de> Hi Michael, I'm also currently trying to get my android dev setup working again (for the OCM app). Not easy especially as they have now retired eclipse and moved to the new gradle build system in Android Studio, so there's lots of trial and error required to get things building again. The API prefix is also wrong in both cases, it should be: http://api.openchargemap.io/v2/referencedata/ (for pulling the reference data for dropdown lists etc) And the POI query should really be of the form: http://api.openchargemap.io/v2/poi/?output=json&verbose=false&<filter params> Max results filter can be much higher (2500 etc) but will potentially return more data (and take longer), my own preference would be to default the search to a smaller distance. API docs are here: http://openchargemap.org/site/develop/api It's also wrong on iOS, I did submit a pull request a while back but I don't think anything happened to it. Chris On 12/01/2015 18:13, Michael Balzer wrote:
Hi Christopher,
thank you for your help!
From home this morning ( 51.263201, 7.150157 ) I had 494/495 stations on the map. From office now ( 51.321685, 7.343394 ) I have 507 stations. The same stations are missing.
The search radius according to the App should already be switchable to the vehicle range, but that switch has no effect on the number of stations.
So this definitely looks like the 500 results limit.
Search: https://github.com/openvehicles/Open-Vehicle-Android/search?utf8=%E2%9C%93&q...
Seems to me it's this API call: https://github.com/openvehicles/Open-Vehicle-Android/blob/81971beb45defd87e4...
That also explains why the range switch has no effect, the distance is fixed in the call.
There's another call in ui/FragMap, but that's fixed (some init/default maybe).
I'll try to change the maxresults parameter, but I first need to get a working Android development environment -- haven't developed Apps up to now.
Maybe the maxresults parameter should be user configurable.
Regards, Michael
Am 12.01.2015 um 01:57 schrieb Christopher Cook (Open Charge Map):
Hi Michael,
I'll download the code for the app from github and have a look, I haven't developed any of the OCM integration myself and I think (certainly in the case of iOS, not sure about Android) that 3rd party developers were paid to deliver the integration. Presumably then the hope is that the community will refine their efforts via github - I'm assuming the path is very much clear for enthusiastic developers to submit patches for these apps directly. I can help more on integration questions like calling the OCM APIs etc (such as your filtering question). I've held off contributing directly to the apps because I'm limited for time with OCM and other projects.
If I find a simple fix for the API request to OCM I'll submit a pull request to github.
For the example you've given, where is the center of your search (lat/long if possible)? By default the app is search from your current lat/long with a radius of 285 KM, max 500 results. Is it possible there are more than 500 charging locations in your search area?
Best regards, Christopher Cook Open Charge Map
On 12/01/2015 00:36, Michael Balzer wrote:
Hi Christopher,
I first posted this on the OVMS developer list (I'm the Twizy developer). J-C Saint-Po told me to forward this to you:
I'm missing about 80% of all OCM charge stations on the OVMS map display (Android).
I cannot find any criteria in the station data for this effect. Example: visible: http://www.openchargemap.org/site/poi/details/32598 missing: http://www.openchargemap.org/site/poi/details/32607
Both stations are from the same provider with identical capabilities.
The connection filter does not affect this, neither does the range filter or clustering.
Also the detail info on visible points is wrong. See attached screen shots for the map label and detail info on #32598 as shown in the App.
If I can help by more examples, tell me.
This is no new issue, has been this way from the beginning. I just decided to add some missing stations from my region and then discovered they are not really missing in the OCM database.
Regards, Michael
-- Christopher Cook Open Charge Map
Michael, Chris, Regarding Chris's iOS pull request, it seems I missed it. Github is supposed to notify, but in that case it didn't. I've found it, and merged it in. Just a small change: @@ -13,7 +13,7 @@ #import "EntityName.h" #define NULL_TO_NIL(obj) ({ __typeof__ (obj) __obj = (obj); __obj == [NSNull null] ? nil : obj; }) -#define BASE_URL @"http://www.openchargemap.org/api/?output=json" +#define BASE_URL @"http://api.openchargemap.io/v2/poi/?output=json&verbose=false" #define SWF(format, args...) [NSString stringWithFormat:format, args] //#define UPDATE_SEC 120.0 I'll make sure it gets included with the next build for iOS. For Android, I still have my eclipse based build environment working. If anyone wants/needs me to build a test apk let me know. Regards, Mark. On 12 Jan, 2015, at 7:14 pm, Michael Balzer <dexter@expeedo.de> wrote:
FYI
-------- Weitergeleitete Nachricht -------- Betreff: Re: OVMS App missing most charge points, detail info wrong Datum: Mon, 12 Jan 2015 18:30:38 +0800 Von: Christopher Cook (Open Charge Map) <christopher.cook@openchargemap.net> An: Michael Balzer <dexter@expeedo.de>
Hi Michael,
I'm also currently trying to get my android dev setup working again (for the OCM app). Not easy especially as they have now retired eclipse and moved to the new gradle build system in Android Studio, so there's lots of trial and error required to get things building again.
The API prefix is also wrong in both cases, it should be: http://api.openchargemap.io/v2/referencedata/ (for pulling the reference data for dropdown lists etc)
And the POI query should really be of the form: http://api.openchargemap.io/v2/poi/?output=json&verbose=false&<filter params>
Max results filter can be much higher (2500 etc) but will potentially return more data (and take longer), my own preference would be to default the search to a smaller distance. API docs are here: http://openchargemap.org/site/develop/api
It's also wrong on iOS, I did submit a pull request a while back but I don't think anything happened to it.
Chris On 12/01/2015 18:13, Michael Balzer wrote:
Hi Christopher,
thank you for your help!
From home this morning ( 51.263201, 7.150157 ) I had 494/495 stations on the map. From office now ( 51.321685, 7.343394 ) I have 507 stations. The same stations are missing.
The search radius according to the App should already be switchable to the vehicle range, but that switch has no effect on the number of stations.
So this definitely looks like the 500 results limit.
Search: https://github.com/openvehicles/Open-Vehicle-Android/search?utf8=%E2%9C%93&q...
Seems to me it's this API call: https://github.com/openvehicles/Open-Vehicle-Android/blob/81971beb45defd87e4...
That also explains why the range switch has no effect, the distance is fixed in the call.
There's another call in ui/FragMap, but that's fixed (some init/default maybe).
I'll try to change the maxresults parameter, but I first need to get a working Android development environment -- haven't developed Apps up to now.
Maybe the maxresults parameter should be user configurable.
Regards, Michael
Am 12.01.2015 um 01:57 schrieb Christopher Cook (Open Charge Map):
Hi Michael,
I'll download the code for the app from github and have a look, I haven't developed any of the OCM integration myself and I think (certainly in the case of iOS, not sure about Android) that 3rd party developers were paid to deliver the integration. Presumably then the hope is that the community will refine their efforts via github - I'm assuming the path is very much clear for enthusiastic developers to submit patches for these apps directly. I can help more on integration questions like calling the OCM APIs etc (such as your filtering question). I've held off contributing directly to the apps because I'm limited for time with OCM and other projects.
If I find a simple fix for the API request to OCM I'll submit a pull request to github.
For the example you've given, where is the center of your search (lat/long if possible)? By default the app is search from your current lat/long with a radius of 285 KM, max 500 results. Is it possible there are more than 500 charging locations in your search area?
Best regards, Christopher Cook Open Charge Map
On 12/01/2015 00:36, Michael Balzer wrote:
Hi Christopher,
I first posted this on the OVMS developer list (I'm the Twizy developer). J-C Saint-Po told me to forward this to you:
I'm missing about 80% of all OCM charge stations on the OVMS map display (Android).
I cannot find any criteria in the station data for this effect. Example: visible: http://www.openchargemap.org/site/poi/details/32598 missing: http://www.openchargemap.org/site/poi/details/32607
Both stations are from the same provider with identical capabilities.
The connection filter does not affect this, neither does the range filter or clustering.
Also the detail info on visible points is wrong. See attached screen shots for the map label and detail info on #32598 as shown in the App.
If I can help by more examples, tell me.
This is no new issue, has been this way from the beginning. I just decided to add some missing stations from my region and then discovered they are not really missing in the OCM database.
Regards, Michael
-- Christopher Cook Open Charge Map
_______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
Hi there, Has anyone tried implementing anything for VW eUP, VW eGolf and Renault Zoe?I tried looking at the forks on github, but don't seem to find anything related. Also - is anyone using the Leaf code? Is it reliable? Any problems? Thanks in advance,Nikolay
Nikolay, I haven't heard of any work on those vehicles. The leaf code is very very basic at the moment. SOC, location, but that is pretty much all. Regards, Mark. On 15 Jan, 2015, at 7:46 am, Nikolay Shishkov <nshishkov@yahoo.com> wrote:
Hi there,
Has anyone tried implementing anything for VW eUP, VW eGolf and Renault Zoe? I tried looking at the forks on github, but don't seem to find anything related.
Also - is anyone using the Leaf code? Is it reliable? Any problems?
Thanks in advance, Nikolay
_______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
Thanks Mark, Do you know what the production status is on the OVMS modules? They seem to be sold out on fasttech's website. Nikolay On Thursday, January 15, 2015 2:16 AM, Mark Webb-Johnson <mark@webb-johnson.net> wrote: Nikolay, I haven't heard of any work on those vehicles. The leaf code is very very basic at the moment. SOC, location, but that is pretty much all. Regards, Mark. On 15 Jan, 2015, at 7:46 am, Nikolay Shishkov <nshishkov@yahoo.com> wrote: Hi there, Has anyone tried implementing anything for VW eUP, VW eGolf and Renault Zoe?I tried looking at the forks on github, but don't seem to find anything related. Also - is anyone using the Leaf code? Is it reliable? Any problems? Thanks in advance,Nikolay _______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk http://lists.teslaclub.hk/mailman/listinfo/ovmsdev _______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
Nikolay, Latest batch is stuck in manufacturing, trying to squeeze production in before Chinese New Year. I am told that they should be ready by the end of this month (January). Regards, Mark.
On 15 Jan, 2015, at 9:33 pm, Nikolay Shishkov <nshishkov@yahoo.com> wrote:
Thanks Mark, Do you know what the production status is on the OVMS modules? They seem to be sold out on fasttech's website.
Nikolay
On Thursday, January 15, 2015 2:16 AM, Mark Webb-Johnson <mark@webb-johnson.net> wrote:
Nikolay,
I haven't heard of any work on those vehicles.
The leaf code is very very basic at the moment. SOC, location, but that is pretty much all.
Regards, Mark.
On 15 Jan, 2015, at 7:46 am, Nikolay Shishkov <nshishkov@yahoo.com <mailto:nshishkov@yahoo.com>> wrote:
Hi there,
Has anyone tried implementing anything for VW eUP, VW eGolf and Renault Zoe? I tried looking at the forks on github, but don't seem to find anything related.
Also - is anyone using the Leaf code? Is it reliable? Any problems?
Thanks in advance, Nikolay
_______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk <mailto:OvmsDev@lists.teslaclub.hk> http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
_______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk <mailto:OvmsDev@lists.teslaclub.hk> http://lists.teslaclub.hk/mailman/listinfo/ovmsdev <http://lists.teslaclub.hk/mailman/listinfo/ovmsdev>
_______________________________________________ OvmsDev mailing list OvmsDev@lists.teslaclub.hk http://lists.teslaclub.hk/mailman/listinfo/ovmsdev
participants (3)
-
Mark Webb-Johnson -
Michael Balzer -
Nikolay Shishkov