And replace labeling engine dialog "Draw text as outlines" checkbox with a combobox
presenting the choice of always rendering as outlines OR text.
(This will allow us to easily add additional methods in future, e.g. potentially
a "Render as text wherever possible" setting, for defaulting to text objects
whenever it doesn't impact the rendering quality to do so)
The new QgsRenderContext::TextRenderFormat enum controls how text
should be handled during a render operation, e.g. whether to render
text as outlines (paths) or keep it as real text objects.
Deprecate previous arguments in QgsTextRenderer which handled
this same use case.
This allows us to make the setting vary per-render, instead of
having a single global flag controlling the setting. Ultimately
this will allow us to have different behaviour within the
canvas renders vs print layout exports.
Refs #3975
Also update a failing test case since when
we changed the priority in QgsVectorLayerUtils
CreateFeature.
Test was blacklisted on travis in nov 2017
but here they are anyway.
introduce MeshDataBlock
use mesh block in rendering
calculate magnitude
use new mesh API
fix bug for memory layer
fix SIP, sip is unable to work with qvector<qgsmeshvertex>
fix tests
implement new MDAL min/max api
improve mesh documentation
fix travis build
Iterates over the geometries in the collection, allowing this type
of code:
gc = QgsGeometryCollection()
gc.fromWkt('GeometryCollection( Point(1 2), Point(11 12), LineString(33 34, 44 45))')
for part in gc:
print(part.asWkt())
- Calling removeGeometry with an invalid index will now raise an IndexError
- Calling collection[0] will return the first geometry in the collection,
collection[1] the second, etc. And negative indices return from the end
of the collection, so collection[-1] returns the last geometry in the collection.
- Geometries can be deleted by calling `del collection[1]` (deletes the
second geometry from the collection). Also supports negative indices
to count from the end of the collection.
This algorithm takes an input (multi)line (or curve) layer, and splits
each feature into multiple parts such that no part is longer then
the specified maximum length.
Supports data-defined maximum length property, and edit in place operation.
Credit to @NathanW2 for the inspiration!
This new class QgsImageCache is the equivalent of QgsSvgCache
but for raster images.
QgsImageCache stores pre-rendered resampled versions of raster
image files, allowing efficient reuse without incurring the
cost of resampling on every render.
Additionally, it offers the other benefits QgsSvgCache has,
such as thread safety, ability to transparently download remote
images, and support for base64 encoded strings.
From the dox for QgsProcessing.TypeVector: "When used for a sink this indicates
the sink has no geometry.". If an algorithm is using this value to indicate
"maybe has geometry" for an output, then that algorithm is incorrect and
should be using TypeVectorAnyGeometry instead.
to show in numeric/distance widgets
E.g. to only show 2 decimal places:
# only show two decimal places in parameter's widgets, not 6:
param.setMetadata( {'widget_wrapper':
{ 'decimals': 2 }
})
