Some of the tools currently have confusing names. An idea would be to try to align the tool names to common names in python. For example, the apply tool applies a function to the values in a stream, which is what python's map does, so perhaps this tool should just be called map. Note that the actual tool class will be Map, so there's no conflict with builtins.
N.B. in this case one could argue that this should actually be map_values (and MapValues) since it's only applying the function to the values, and there could be corresponding map_timestamps and map_instances tools.
This is problematic for deployed hyperstream instances, since workflows that depended on those tools would now be broken. However this is probably the right time to do it before there are too many deployed instances.
Some of the tools currently have confusing names. An idea would be to try to align the tool names to common names in python. For example, the
apply
tool applies a function to the values in a stream, which is what python'smap
does, so perhaps this tool should just be calledmap
. Note that the actual tool class will beMap
, so there's no conflict with builtins.N.B. in this case one could argue that this should actually be
map_values
(andMapValues
) since it's only applying the function to the values, and there could be correspondingmap_timestamps
andmap_instances
tools.This is problematic for deployed hyperstream instances, since workflows that depended on those tools would now be broken. However this is probably the right time to do it before there are too many deployed instances.