I recently exported some very low quality vectordrawables from figma. The result was a number of bugs being revealed in the current implementation, which this diff addresses:
FillType should be supported by clip-paths
Despite the spec saying that they aren't supported, certain attributes are accepted for clip-paths, which we'll silently ignore
Certain sequences of elements could break parsing the XML. This fixes that by switching to a more reliable method of determining if we are in a self-closing tag.
I recently exported some very low quality vectordrawables from figma. The result was a number of bugs being revealed in the current implementation, which this diff addresses: