Re: modRana on Android
 

Update, re the requests:

The exit seems to be clean when killing modRana's window directly, thus the Exit button is not such a dire need.

However the non-working rotation is a problem...

EDIT: woa, you're faster, just seen the above post, thanks for the Quit button, will test...

Re: modRana on Android
 

Of course, I'll write a detailed guide, with details on setting up the chroot in debian also etc. in a couple of weeks, as I'll be on vacation, but these details are basically all one needs to know to get it running.
Thanks, I'll test.
Yeah, xrandr will be the way to go, I'll post info on this...
I'll see what I can find on Android sound, I'd think there is a way to do it, let's see...
I remember I promised :O)

I'll do it, just have to also set up icewm or fluxbox in the chroot, so I'll get you some nice photos, as StumpWM doesn't give you any eye-candy... I just have to install these, since I don't use them at all, but I'll do it so I can take the screenshots. In a couple of days, it's gonna be done, when I head for vacation... :O)

directions_filter.csv
 

BTW, Martin please correct the following in the directions_filter.csv file:

should be

Пер.;переулок
пер.;переулок

not "prospekt" as you have there currently.

Also, further additions:

Пл.;Площадь
пл.;площадь
Пр.;Проспект
пр.;проспект
Просп.;Проспект
просп.;проспект
Pkwy ;Parkway
Pl ;Place
Plz ;Plaza
Rd ;Road
Sq ;Square
St.;Street
St ;Street
Tce ;Terrace
Trk ;Track
ул.;улица
1-й; первый
1-я; первая
2-й; второй
2-я; вторая
3-й; третий
3-я; третья
4-й; четвертый
4-я; четвертая
5-й; пятый
5-я; пятая
шо.; шоссе

Screenshots of running modRana in chroot on Android
 

As promised... :O)

I installed IceWM in chroot for this purpose, as it's the lightest on dependencies. Hope you like these...

http://sumoudou.org/pics/1a.png

http://sumoudou.org/pics/2a.png

http://sumoudou.org/pics/3a.png

Re: modRana on Android
 

OK, there is a problem with gpsd in chroot on Android.

Basically, doing what I outlined above, i.e. sharing the GPS data via BlueNMEA, modRana's position that it gets from gpsd is stuck on the original position.

Meaning, when you start modRana, it correctly shows you your location, however then as you drive, the GPS data doesn't get updated.

This could be a problem with:

- BlueNMEA not updating the GPS data via TCP
- with TCP transfer
- with gpsd in debian chroot having issues accessing the GPS data etc.

However, I confirmed, while driving that telneting to BlueNMEA 3542 port from inside the chroot worked, even while modRana's position from gpsd was stuck.

Sample output from the telnet:
Debugging gpsd, tcp, BlueNMEA etc. could be quite lengthy.

As we know that BlueNMEA correctly updates the GPS coordinates on its 4352 port at localhost, could we bypass gpsd, when running modRana in Android chroot, Martin?

I.e. access the telnet data directly, since then the problem with the GPS position being stuck would not have to be lengthily debugged, and modRana could be used...

Re: [Announce] modRana: a flexible GPS navigation system
 

I have a couple of questions:

1) Is it possible to predownload based on gps coordination? If I look correctly, it is either (1) around the route or (2) around the current postition on the map or (3) around the area on the map.
This way, it would be very difficult to download, say Vietnam (very big diference in latituted and small difference in longtitude).

2) How is downloading around view working? That should download what I see on the scree, right, or do I misunderstand something? But why does offer to download so few maps when I select a large area (around 100 000 km2) butnnly 1km? I thought that was supposed to be the margin to download around the are plus the are itself.

3) Does downloading from google even work? It does not seem to work for me.

Re: modRana on Android
 

OK, I ran more experiments.

It seems to me the bottleneck is modRana's interface to gpsd or gpsd itself.

Why?

I eliminated gpsd access to NMEA data via tcp, instead created a regular file with socat at /tmp/gps and then started gpsd reading from this file.

The GPS coordinates get updated, so I was able ascertain that the problem doesn't lay in BlueNMEA.

Anyhow, this way, I did get an initial fix, and pursuant updates of the GPS position on the map in modRana...

However, it seems that there is heavy processing going on, either on the modRana side, the GTK side, or, probably, the gpsd side, as I'd have the GPS point in modRana disappear for let's say 10 seconds at a time, then the position would update, the circle appear, then disappear again.

When it'd disappear, modRana would not react to screen taps, leading me to believe that there is some heavy processor load going on.

In fact a couple of times, modrana's window would go completely gray without displaying anything, and I had to kill modRana.

This seemed to disapper after I disabled automatic rotation in the direction of the car movement, leading me to believe the processing load is too heavy for VNC access to X on Android.

Then I got it to update the position at 10-20 second intervals, not very good.

This would improve dramatically, when the moving car would come to a stop at one point, whereby the point would appear right away, and modRana's interface would appear...

Either it's the GTK window redrawing in TightVNC which is causing the delays, or, what I'm hoping for, it's the gpsd interaction somehow...

Therefore, my proposal stands, to access the NMEA data directly, without the gpsd middleman, probably most comfortably via the regular files at /tmp/gps which I can create via socat...

That way we could determine if there would be a speedup in modRana's processing and whether it'd be usable on Android. Martin?

If not, then messing with TightVNC remains the option, lowering the number of colors of TightVNC display, this is an option, as it can interface at 256 colors, there would probably have to be some filter in modRana to only use 256 colors when started in a specific way, that could lead to processing speed increase...

Anyhow, I've already determined that android-vnc-viewer is compiled without xrandr support, so windows cannot be rotated in X chroot, so am having to look into an alternative VNC viewer anyway...

However, cutting out gpsd might still be worth it, though the main hog is probably X VNC access...

Re: [Announce] modRana: a flexible GPS navigation system
 

...who is using VNC anyway today? It's very resource demanding both for network I/O as well as memory I/O, caused by it's implementation of uncompressed full framebuffer transmits. I know that under Android it seems the only way to access chroots and the like, caused by the crippled architecture / integration of Android which doesn't allow simple, efficient X redirects. That's one of the points why quite some people avoid Android if you don't rely on games'n'stuff.

Re: directions_filter.csv
 

Thanks ! It's already committed to the SVN version an will be released in the next update. :)

Nice ! :) Can I include the screenshots on the project website and Maemo wiki article (once written :) ). ?

Can you check if the position is updated if you restart modRana ?

Also, try to enable "Show gpsd debug output" in Options->Debug->GPS, modRana will start printing the information it gets from gpsd to stdout.

I have found a few other things to check in the gpsd troubleshooting guide:

• stopping gpsd & running it manually with debug output:
  
  Code:

  ---

  gpsd -N -D3 -F /var/run/gpsd.sock

  ---

• using cgps or xgps utilities to check the output from gpsd (they are in the package gpsd-clients, at least on Ubuntu)

This certainly also possible - modRana already has a worker thread (in the location module that resides in modules/mod_location.py that reads position data from GPSD. The worker thread could be modified to connect to BlueNMEAs socket and parsing the NMEA sentences it provides. There is just a single stable source (BlueNMEA) without the usual quirks of standard GPS devices, the parsing should not be that difficult, at least for basic location data (latitude, longitude, elevation, speed, direction(?)).

Good point !
I see two possible solutions for this uscase:

• entering two lat:lon coordinate sets, representing the upper left and lower right corners of the area to download
• interactively placing a rectangle on the map over the area to download

Imagine a crosshair in the middle of the screen - thats where the download is centered when using Download around view.

That's because the maps consist of square tiles organized in zoom-levels. On zoom level 0, one tile covers the whole world, on zl 1 4 tiles cover the world, on zl 2 its 16, ... , zl 17 has 17.179.869.184 tiles, etc. :)
So the number of tiles modRana downloads mainly depends on the zoom level range selected.

Google does not really like external use of their tiles and temporarily stops serving them if it detects excessive usage from an IP address.

Re: [Announce] modRana: a flexible GPS navigation system
 

I agree with you man, but I switched to Droid 3, as I'm a heavy Emacs user, and Emacs is faster on it in the chroot, much much faster, than on the N900... And it has a MUCH better keyboard than the N900.

So not only those who rely on games :O)

But yeah, if I could get my hands on an N950, I'd probably be much happier with it, but it can't be had.

So I'm stuck with Android, and that's how I got to experimenting with modRana in it... And VNC is the only way unfortunately...