Algorithms and other processing code should use this method
(instead of dataobjects.getLayerFromString) to
retrieve layers from a string, as it considers the processing
context and allows resolving strings to temporarily stored layers.
This permits processing models to function correctly when
intermediate results are stored as memory layers. Subsequent
model algorithms can then access these temporary layers as inputs.
All temporary layers will be removed when the context object
is destroyed after the model algorithm is run.
Also
- simplify and add tests
- remove large memory leak (persistant store of all non-project layers)
- remove broken support for direct loading postgres/virtual layers
by string (Python version was very broken and would never match
a postgres/virtual layer)
* [processing] qgis regular points test
* [processing] qgis shape buffer variable tests
* [processing] qgis create grid lines test
* [processing] qgis convert geometry test
* [processing] qgis extract by location test
* [processing] qgis add field test
* [processing] trying to fix travis failing
* [processing] trying to fix travis failing/2
* trying to fix travis failing/3
* [processing] Add new default option "ProjectCrs" to ParameterCrs
* [processing] RegularPoints tests shouldn't rely on iface
* [processing] Fix regular points test
* [processing] RandmPointsExtent new CRS parameter
* [processing] qgis random point in extent test
* [processing] qgis random point in extent test/2
* [processing] remove qgis random point in extent test
* no output random points in extent test
* remove useless output
Crs parameters in algorithm settings in modeler
This makes it easier to pick static CRSes for the parameter,
and also makes it obvious to users that they can use a fixed
CRS parameter in their model (as opposed to one taken from
a layer or input)
This adds a new input type for expression inputs. Expression
inputs can be linked to a parent layer so that the builder
shows the correct fields and layer variables.
It's designed for two use cases:
1. to be used when an algorithm specifically requires an expression,
eg Select by Expression and Extract by Expression.
2. to be potentially used as a replacement input instead of string
or number literals in algorithms. Eg - if the simplify algorithm
tolerance parameter was replaced with an expression paremeter, then
this expression would be evaluated for every feature before
simplifying that feature. It would allow parameters to be calculated
per feature, as opposed to the current approach of calculating
a parameter once before running the algorithm. It would also
mean algorithms like "variable distance buffer" would no longer
be needed, as a single "buffer" algorithm could then be used
for either a fixed distance, field based, or expression based
distance.
accept numeric, file and table field inputs in modeler
This allows a non-string parameter to be reused as a string
parameter in contexts where it makes sense.
This commit restores some pre 3.0 processing behaviour for number inputs.
Now, if a number input is required outside of modeller than a spin box
will be shown instead of a free text input.
Clicking the expression builder button results in an expression
which is evaluated immediately to avoid users expecting that
the expression will be evaluated per feature.
