Since the underlying issues with the Python bindings are now fixed,
in most cases we can safely default to allowing an algorithm to
run in a background thread!!
So now we make this the default, and require individual algorithms
which are NOT thread safe to declare this. This includes algorithms
which directly manipulate the current project or layers (such as
setting layer styles), alter the selections in layers, or which
rely on 3rd party libraries (for now, SAGA and GRASS algorithms
are marked as not thread safe... TODO - someone more familiar with
these libraries can investigate and remove the flag if appropriate).
Also models are marked as non-thread safe. TODO: only flag an
individual model as thread-unsafe if any of its child algorithms
report this flag.
initAlgorithm() method
This allows 2 benefits:
- algorithms can be subclassed and have subclasses add additional
parameters/outputs to the algorithm. With the previous approach
of declaring parameters/outputs in the constructor, it's not
possible to call virtual methods to add additional parameters/
outputs (since you can't call virtual methods from a constructor).
- initAlgorithm takes a variant map argument, allowing the algorithm
to dynamically adjust its declared parameters and outputs according
to this configuration map. This potentially allows model algorithms which
can be configured to have variable numbers of parameters and
outputs at run time. E.g. a "router" algorithm which directs
features to one of any number of output sinks depending on some
user configured criteria.
have specific icons, instead of generic qgis icon
We consider these 'top level' algorithms, and using the
standard algorithm icon should help reflect this and
differentiate them from 3rd party algorithms.
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)
- don't use setFilterFid() within loops to fetch features one
at time (as it's extremely slow), instead use setFilterFids()
outside the loop
- don't fetch unused attributes/geometry when it can be
avoided
Makes SelectByAttribute orders of magnitude faster for providers which
support compiled expressions (eg Postgres), also allows algorithm
to take advantage of database indexes created on attribute.
For a 4 million point PostGIS layer (with an index on attribute)
BEFORE: cancelled after 20 mins
AFTER: ~2 seconds
