GitLab

completely took out all of the constant correctness enahcements. Also
added a powerUp() method.

order to minimize noisy/false message reception. Removed begin call
from constructor as begin now has delay which creates timer race
initialization. The delay exists to allow for power on/reset radio
settling which can otherwise prevent CONFIG bits from holding. Added a
method to allow CRC hardware validation to be disabled. This is a
requirement for various promiscuous listening RF applications. Setting
CRC width now forces CRC hardware validation enable.

for specific pipelines per suggestion.

location. Changed some bit value interpretation.

Now auto-detects P-vaiants and is queryable. setDataRate now returns a
boolean indicating success or failure of the new data rate
request. The wide_band state indicator has been added and now tracks
the state, including on error fallback. Its now possible to query the
current data rate via getDataRate().

The AA retry duration has been changed to 1500us, which is the lowest
interval allowed when running at 250Kbs, with AA enabled, and 32-byte
packets.

Some additional constantness cleanup.

Added the concept of wide band. Currently not used but state is
properly tracked in accordance with data rate selection.
Default channel is changed to 64 in begin method. That needs to move
anyways.
Automatic ACKs can now be adjusted on a per pipeline bases.

New P variant feature support with corresponding defines. RF_DR is now broken into two bits to support P variant and is now obsolete. RF_DR is now RF_DR_HIGH and RF_DR_LOW; with RF_DR_HIGH corresponding to the original RF_DR bit. Data rate can now select 250kbs mode on supporting hardware.

Its now possible to both get and set the current PA level. LNA is obsolete in P variant.

CD is obsoleted in P variant. Can now check both CD and RDP levels. Will follow through in the future to determine if there is any real difference. Aside from dBm floor and slight changes in internal behavior, they appear identical in interface and function.

Fixes a bug in openReadingPipe which was excluding the 5th pipline. We have 6-pipelines, 0-5, so excluding the 5th (<5) would only allow for four active pipelines.

Environment-specific changes for me.  Testing out latest arduino from git, and added a 3rd upload option