These changes are necessary for programs whose set of active attributes is not fixed at compile time by a template parameter pack, but rather varies based on the generated shader text at runtime. In such cases, the attribute location of a given named attribute may vary between instances of the same Program.
Previously, attribute bindings were implicitly associated with a location based on template parameter order, and -1 was used to indicate an inactive attribute. This left us unable to disable the appropriate attribute when it went from active to inactive. Now, the state tracker for bindings explicitly associates locations and state, and an empty optional is used to indicate an inactive attribute.
In addition, a gl::VertexArray class is now exposed, allowing more flexibility in the relationship between Programs, Segments, and attribute bindings. In this commit, that relationship does not change, but the subsequent commit adjusts it to match gl-js, reduce rebinds, and work around buggy VAO implementations.
VertexArray uses a pimpl idiom in order to support implementations that lack the VAO extension. In that case, all VertexArrays share global binding state, reflecting the platform reality in the absence of VAOs, while still providing a uniform API.
Make sure it is faster than std::to_value(). And it is, by ~30% for
the normal use case, and much faster when close to the double limits.
util::dtoa also offers a much better precision.
port https://github.com/mapbox/mapbox-gl-js/pull/4781
This improves legibility of labels that follow lines in pitched views.
The previous approach used the limited information in the shader to
calculate put the glyph in approximatelyright place. The new approach
does this more accurately by doing it on the cpu where we have access to
the entire line geometry.
* Add simple unit tests for SymbolSizeBinder
* Fix bug in symbol size uniform value calculation
For camera functions we were setting the zoom levels in "covering ranges" to
`[(zoom stop <= tile zoom), (zoom stop >= 1 + tile zoom)]`, but then evaluating
the function at `[tile_zoom, tile_zoom + 1]`.
* Check for it != end() before accessing it->first
Introduce a second level of immutability, over each of the collections held by a style: sources, images, and layers. Tracking immutability at this level allows us to short-circuit significant portions of the RenderStyle update logic via a simple equality check, greatly improving performance.
* Eliminate updateBatch in favor of diffing layers and detecting changes to properties upon which layout depends.
* Replace RenderSource::{update,remove,invalidate,reload}Tiles with a single update method
* Replace TilePyramid::{update,remove,invalidate,reload}Tiles with a single update method
* Remove Style& dependency TODO from GeometryTile and TileParameters
This adds a `convertJSON` template, to be used like:
Error error
optional<Foo> foo = convertJSON<Foo>(string, error);
Internally, it parses the string with RapidJSON and then calls `convert<Foo>(parsed, error)`.
While here, rationalize GeoJSON converters and fix error handling for Tileset conversion in OfflineDownload.
Combine Style::cascade, recalculate, relayout, and updateTiles into a single method. This allows multiple loops over sources and layers to be consolidated and prepares for additional Style-Map decoupling: rather than tracking pending updates via a set of Update flags held by the Map and passed to the Style, the Style can use its own data to determine what to update.
