Re: [gpsd-dev] The carnivorous config bug

[Hal Murray]

address@hidden said:
> Unfortunately, there are perfectly ligitimate reasons for that to happen.
> Absence of optional packages like bluez and libusb, for example. 

I was hoping that I could tell it something like bluez=False on the command 
line and that would suppress the message.  I tried a bit but either I didn't 
get it right or it didn't do what I wanted.

I'll hack a script to scan the log file.

--------

Either I'm confused or scons -c is broken.

Sorry for the clutter if this is still on your list.  I thought it or it's 
first cousin was recently fixed.  Maybe this is a handy test case.

A) scons -c goes through the configuration stuff.  Is there a reason for that?

B) a second run of scons -c gets a different configuration.

First try after rm .sco*:

Checking for bluez... no
Checking for C header file sys/timepps.h... yes
You have kernel PPS available.
Checking for C header file linux/can.h... yes
You have kernel CANbus available.

Altered configuration variables:
bluez = False (default True): BlueZ support for Bluetooth devices
scons: done reading SConscript files.

That's what I expect.

Second try:

Checking for bluez... (cached) no
Checking for C header file sys/timepps.h... (cached) no
You do not have kernel PPS available.
Checking for C header file linux/can.h... (cached) no
You do not have kernel CANbus available.

Altered configuration variables:
nmea2000 = False (default True): NMEA2000/CAN support
pps = False (default True): PPS time syncing support
bluez = False (default True): BlueZ support for Bluetooth devices
scons: done reading SConscript files.

Running "scons" a second time after "rm .sco*" doesn't get confused like that.


-- 
These are my opinions.  I hate spam.