[QGIS-Developer] Updating qgis:rastercalculator to work better in models

[Luigi Pirelli]

Hi @Rudi von Staden <rudivs at gmail.com> , @Nyall Dawson
<nyall.dawson at gmail.com> and @Matthias Kuhn <matthias at opengis.ch>

in the meantime I had to resolve compilation problems and OS update I'll
show you what I'm doing to resurrect rastercalculator in modeler trying to
evaluate if I'm doing well.

the principal problem (but not the only one) to run raster calculator in a
modeler is to map input value to the formula. This is a generic problem
that can be applied to any algorithm that need to map input in something
internal to the algorithm.

The main problems is that the algorithm context scope hadn't information to
allow this mapping. I added:

context.setExpressionContext(expContext);
here
https://github.com/qgis/QGIS/blob/master/src/core/processing/models/qgsprocessingmodelalgorithm.cpp#L276
and naming the createExpressionContextScopeForChildAlgorithm scope
"algorithm_inputs" (actually the added scope does not have name, because
was not necessary)

and add this mapping in ::variablesForChildAlgorithm
https://github.com/qgis/QGIS/blob/master/src/core/processing/models/qgsprocessingmodelalgorithm.cpp#L431
with value:
<var name>: <maplayer instance>
In this moment only <var name>_[extent] variable are added in this context
scope

the idea, in the processAlgorithm python part, is to get the expression
context from ProcessingContext and substitude <var name> with <maplayer
source> in the formula that will be executed by QgsRasterCalculator

conceptually it works, but many other problems in context and parameter
settings are present that seems they are more generic modeler problems I'm
still trying to solve and reproduce them)

What I'm not sure is to use the expressionContext to store this information
instead of a different or new contextScope. the advantages is that the new
input parameters are available in the expression editor because available
in the context.

as soon as I can I'll give you a PR where to talk about real code... in
this moment I doing tests if my approach is (almost) correct and finding
all limitation using these pathces in simple test models

Luigi Pirelli

**************************************************************************************************
* LinkedIn: https://www.linkedin.com/in/luigipirelli
* Stackexchange: http://gis.stackexchange.com/users/19667/luigi-pirelli
* GitHub: https://github.com/luipir
* Mastering QGIS 2nd Edition:
*
https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition
* Hire me: http://goo.gl/BYRQKg
**************************************************************************************************


On Wed, 4 Jul 2018 at 00:24, Luigi Pirelli <luipir at gmail.com> wrote:

> I was supposing is was subject to the GSOC about qgis/grass... but I can
> be wrong.
> Luigi Pirelli
>
>
> **************************************************************************************************
> * LinkedIn: https://www.linkedin.com/in/luigipirelli
> * Stackexchange: http://gis.stackexchange.com/users/19667/luigi-pirelli
> * GitHub: https://github.com/luipir
> * Mastering QGIS 2nd Edition:
> *
> https://www.packtpub.com/big-data-and-business-intelligence/mastering-qgis-second-edition
> * Hire me: http://goo.gl/BYRQKg
>
> **************************************************************************************************
>
>
> On Tue, 3 Jul 2018 at 18:31, Giovanni Manghi <giovanni.manghi at gmail.com>
> wrote:
>
>> Hi Luigi,
>>
>> > I know that some power users are stuk to 2.18 for the lack r.calc in
>> grass7
>> > integration in qgis (that seems will be solved soon)
>>
>> what is greatly missing (in GRASS7) is r.mapcalculator.py, a script
>> made on top of r.mapcalc that among the other things allowed us to
>> have in Processing a nice way to use r.mapcalc (because it worked via
>> normal description files while r.mpalcalc does not, and any attempt to
>> make it work in Processing was not very succesful).
>>
>> Are you saying that this script is being ported in GRASS7?
>>
>> cheers
>>
>> -- G --
>> _______________________________________________
>> QGIS-Developer mailing list
>> QGIS-Developer at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-developer
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-developer/attachments/20180709/f6b99c66/attachment-0001.html>