This fixes a deadlock
The iterator as a member as previously implemented kept connections
occupied for as long as the gatherer existed. This could lead to a
situation, where all connections have been reserved from finished
gatherers which were waiting to be deleted (through deleteLater),
but deleteLater would never happen because the main thread was
waiting for a connection to become available.
Fixes https://github.com/qgis/QGIS/issues/37496
Offers the same functionality as the Random points inside polygons
algorithm, and in addition it includes (possible to opt-out) the attributes
from the containing polygon and the ability to specify a seed for the random
number generator.
Could replace the Python algorithm Random points inside polygons, but
this C++ implementation only supports "Points count" ("number of points
for each feature") directly, while "Points density" is supported by allowing the
point count to be data defined (using the polygon area to calculate the
number of points).
[FEATURE] Introduces memory dataset groups for mesh layer. These dataset groups are temporary and are not kept when the project is closed.
Memory dataset groups can be created from the mesh calculator with a new option.
Allows the possibility to remove or save these memory dataset groups to a file with specified driver.
This algorithm allows models to create new file directories. Useful for
models which have to output results to certain dynamically created paths
(e.g. a folder containing the current date)
This allows for models which export print layouts from the current project. One
use case for this is allowing users to create an in-project model which exports
a particular set of layouts from the project to certain folders, so that they
can easily re-export the current project in a single operation instead of
having to manually open multiple layouts and export one-by-one.
Additionally, with the new capabilities to have expression based output files
inside models, you can automatically export the layouts to a folder with
the current date tag and include this in the exported file names!
and still be able to store generic labels on the item (like the field name and the field alias)
This fixes that there has been no example values in the expression builder on the fields when an alias has been used
