[Qgis-user] Can we make QGIS stop lying?

[Carlos Cerdán]

Hi Tobias:

I understand your frustation (I've lived it), but QGIS doesn't lie, it is
different from the way that some topics are done in another GIS.

You can get right metric values even if your layer and project are in
Lat-long units. In your particular case, if your project and layer CRS is
EPSG 4326, please try with this expression in field calculator:

length(transform($geometry, 'EPSG:4326', 'EPSG:25832') )

If your project CRS is 25832 (on the the fly CRS enabled) but your layer is
still EPSG 4326 or 25832, you can use:

 length($geometry)

If your project CRS is 25832 (on th fly CRS disabled) and your layer is
EPSG 25832, you can use:

$length


Yes, it's something complicated, but it's the price for more liberty. With
areas it's similar. Perhaps it's better don't use "$length" or "$area". I
recomend "length" and "area" fuctions.

For difference beteween "$_function" and "function" see help for fuctions
in field calculator

Best whishes from Perú.

Carlos Cerdán







2017-05-25 5:47 GMT-05:00 Tobias Wendorff <tobias.wendorff at tu-dortmund.de>:

> Hi there!
>
> Yesterday, I gave an introductory workshop at University on
> how to download OSM-data off its website and work on it in QGIS.
> The students are in master's degree now and learned ArcGIS in
> their study runs only. So we have to make QGIS tasty to them.
>
> I've explained the students, how to switch to a metric projection
> (EPSG:25832) instead of the geographical one (EPSG:4326). Then
> we've filtered out shops
>
> Then the students should filter out all the shops in their hometown
> and calculate the airlining distance to their flat. I've used hub
> lines for this. The dialog informs about distance in meters.
>
> It's a lie. The resulting distance is in decimal degrees. So 600
> meters ere about 0.1 [whatever]. Of course, it's easy to calculate
> the real distance using "$length" in field calculator, but why?!
>
> QGIS functions should detect that coordinates are in geographic
> projections and should inform that the result isn't metric OR
> switch to a processing system, which gives metric results. There
> are many libraries, which can calculate VERY accurate and quick
> on ellipsoids. With less than 0.01 mm of errors.
>
> Most of us might be aware that's normal behavior in QGIS, but
> most of new users will not be aware that meters in GUI aren't
> meters and dataset.
>
> Best regards,
> Tobias
>
> _______________________________________________
> Qgis-user mailing list
> Qgis-user at lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20170525/b58c6651/attachment.html>