Specifying Kernels in ALE

Ale has a few options for getting kernels:

1. Search for kernels locally in $ALESPICEROOT (default).
2. Search for kernels online (with -w or props={'web' : True}).
3. Specify kernels manually.

Search for Kernels Locally

Searching for a local kernel is the default behavior in ALE's isd_generate. Just input your cube's name, no need for additional flags.

shellpython

isd_generate B10_013341_1010_XN_79S172W.cub

# Output: B10_013341_1010_XN_79S172W.json

import ale
isd_dictionary = ale.load('B10_013341_1010_XN_79S172W.cub')

# Output: Python Dictionary (Use ale.loads for a string)

Kernel Download and Setup Required

For ALE to successfully find kernels locally:

1. The NAIF Kernels for your mission must be downloaded
2. The $ALESPICEROOT environmental variable must be set to point to the NAIF Data.
3. The metakernel for your cube in the NAIF Data must be setup.

See Setting Up NAIF Data for more info.

Search for Kernels Online

Use the -w flag to search for kernels online with the USGS SpiceQL web service. This is a good option if you don't want to download and setup kernels locally on your computer.

shellpython

isd_generate -w B10_013341_1010_XN_79S172W.cub

import ale
isd_dictionary = ale.load('B10_013341_1010_XN_79S172W.cub', props={'web' : True})

Specify Kernels Manually

List in Metakernel File

To give ALE a list of kernels, use a metakernel file. In the metakernel file, in the data section (which starts at \begindata):

1. Set PATH_VALUES to the directory containing the kernels
2. List the kernels under KERNELS_TO_LOAD.
   (Use the value of PATH_SYMBOLS to point to the directory you specified in PATH_VALUES)

You can also specify single kernel file or a cube to use as the kernels. But typically, you will need to use a metakernel file to list multiple kernels, for the sake of complete information in the ISD you are generating.

shellpython

isd_generate -k mro_B10_72W.tm B10_013341_1010_XN_79S172W.cub

import ale

ale.load('B10_013341_1010_XN_79S172W.cub', props={'kernels' : 'mro_B10_72W.tm'})

Metakernel Format

This metakernel has a path value of ., which tells it to look for other kernels in its own folder. You could change . to another location if the rest of your kernels were somewhere else.

...
metakernel information section
...

\begindata

   PATH_VALUES     = ( '.' )

   PATH_SYMBOLS    = ( 'KERNELS' )

   KERNELS_TO_LOAD = (
                       '$KERNELS/naif0012.tls'
                       '$KERNELS/pck00008.tpc'
                       '$KERNELS/mro_sclkscet_00082_65536.tsc'
                       '$KERNELS/mro_v16.tf'
                       '$KERNELS/mro_ctx_v11.ti'
                       '$KERNELS/B10_013341_1010_XN_79S172W_0.bsp'
                       '$KERNELS/B10_013341_1010_XN_79S172W_1.bsp'
                       '$KERNELS/mro_sc_psp_090526_090601_0_sliced_-74000.bc'
                       '$KERNELS/mro_sc_psp_090526_090601_1_sliced_-74000.bc'
                     )

\begintext

As props in Python

If you are using ale.load or ale.loads (but not isd_generate), you can list out all your kernels in the kernels prop.

python

import ale

props = {
    'kernels': [
        'B10_013341_1010_XN_79S172W_0.bsp',
        'B10_013341_1010_XN_79S172W_1.bsp',
        'mro_ctx_v11.ti',
        'mro_sc_psp_090526_090601_0_sliced_-74000.bc',
        'mro_sc_psp_090526_090601_1_sliced_-74000.bc',
        'mro_sclkscet_00082_65536.tsc',
        'mro_v16.tf',
        'naif0012.tls',
        'pck00008.tpc'
    ]
}

ale.load('B10_013341_1010_XN_79S172W.cub', props=props)

---