QtLocation map updates

[Dirk Hohndel]

> On Aug 1, 2017, at 3:31 PM, Linus Torvalds <torvalds at linux-foundation.org> wrote:
> 
> On Tue, Aug 1, 2017 at 2:55 PM, Lubomir I. Ivanov <neolit123 at gmail.com> wrote:
>> 
>> someone already did this:
>> https://github.com/vladest/googlemaps
>> 
>> not sure how i've missed that in the google search results????
>> 
>> and it works; at least after testing it for 10 minutes - more zoom
>> levels, no missing tiles, offline caching, terrain / hybrid maps:
>> https://gifyu.com/images/google_maps_provider.gif
> 
> Looking good.

Indeed.

>> the source code is MIT licensed. the last update is from the 10th of
>> June and is to make it work with Qt 5.9.
>> i was able to build it without a problem on Windows using Qt 5.9.0.
> 
> The other good part about it is just the size. The project size right
> now is just 2636 lines total, and that's with gitignore files and
> license etc.

Nice. I would love to be able to see if this could be made to work with 
Qt 5.7 (or even better 5.6) - otherwise we are leaving a lot of distributions 
behind. Requiring 5.9 is no problem for Windows and Mac, but for the
various Linux distros (about 20% or so of our users) it's a no go. Most
are not even on 5.8 at this point.

> The individual source files are a couple of hundred of lines of code.
> The license apparently *used* to be LGPL 3, but got switched to MIT
> immediately.
> 
> I didn't really look at the source code very much, but what I looked
> at didn't look horrible. I'm a *bit* worried about things like this:
> 
>    // Google version strings
>    _versionGoogleMap            = "m at 338000000";
>    _versionGoogleSatellite      = "198";
>    _versionGoogleLabels         = "h at 336";
>    _versionGoogleTerrain        = "t at 132,r at 338000000";
>    _secGoogleWord               = "Galileo";
> 
> which to me implies there is some hardcoded special magic going on,
> but it looks like it updates those strings based on the replies, and
> then never actually *uses* them, so maybe it's just debugging.
> 
> So this does seem to be one very possible solution to all the map issues.

Yes, I'd love this solution :-)

Lubomir, are you going to send a PR for this? What are the next steps?

/D