script help

jwojda
Valued Contributor II

I have a script that I'm using for computer renaming...but one part of it where it looks at whether the model name is MacBook (and then assigning ML) or not (assigning MD) - the results are saying everything is not a MacBook (assigning MD).

#!/bin/bash

ModelType=$(system_profiler SPHardwareDataType | grep 'Model Name:' | awk '{ print $3 }')
if [ "$ModelType" = "MacBook" ]; then
    TT=ML
else
    TT=MD
fi

As a side note, further down in the script, it pops up a prompt for user input via osascript - which is all working fine, but if the user isn't fast enough typing in it goes to the next prompt .... can someone advise how to have it wait for the input before skipping to the next portion?

BB=$(/usr/bin/osascript -e 'tell application "System Events" to set BRAND to text returned of (display dialog "Please input the brand - AF , AI, AU" default answer "" with icon 2)')
7 REPLIES 7

sdagley
Esteemed Contributor II

@jwojda How about this for determining Mac type:

model=$(ioreg -l | grep "product-name" | cut -d "=" -f 2 | sed -e s/[^[:alnum:]]//g | sed s/[0-9]//g)
# Default to a Desktop
TT="MD"
isMacbook=$(echo "$model" | grep "Book")
if [[ $isMacbook != "" ]]; then
    TT="ML"
fi

jwojda
Valued Contributor II

@sdagley same result. I did have to put $isMacBook into "$isMacbook" to make it not error out, but after doing that it still went to MD

Sichas
Contributor

I gave the code from your original post a run and it's properly returning ML for me. Might I suggest echoing out $ModelType before the IF statement and see what value it's entering that statement with? Maybe awk is cutting things off that we're not expecting?

Also, just a side note for defining variables, TT would be a string, so I'd recommend redoing those assignments to:

TT="MD"

Not the most essential thing in the world, and certainly isn't the issue here, but it's a decent practice to make sure your variable types are always correctly defined :)

WellsJtech
New Contributor III

This is what I use as a template when I want to set different things depending if it's a macbook or not...

# Detects if this Mac is a laptop or not by checking the model ID for the word "Book" in the name.
IS_LAPTOP=$(/usr/sbin/system_profiler SPHardwareDataType | grep "Model Identifier" | grep "Book")
modelName=$(/usr/sbin/system_profiler SPHardwareDataType | awk -F': ' ' /Model Name/{print $NF}')

if [ "$IS_LAPTOP" != "" ]; then
    /usr/bin/pmset -b sleep 15 disksleep 10 displaysleep 5 halfdim 1
    /usr/bin/pmset -c sleep 0 disksleep 0 displaysleep 30 halfdim 1
else    
    /usr/bin/pmset sleep 0 disksleep 0 displaysleep 30 halfdim 1
fi

tlarkin
Honored Contributor

I like reading the system_profiler data into Python's plistlib because then you can just parse the dictionary outputs pretty easily, an example I just whipped up:

#!/usr/bin/python

import subprocess
import plistlib

def get_device_model():
    cmd = ['/usr/sbin/system_profiler', 'SPHardwareDataType', '-xml']
    proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
    out, err = proc.communicate()
    data = plistlib.readPlistFromString(out)
    model = data[0]['_items'][0]['machine_model']
    return model


model = get_device_model()

if 'MacBook' in model:
    print('MacBook Found')

When I run the code on my MacBook Pro:

/usr/bin/python2.7 /Users/tlarkin/IdeaProjects/test/get_model.py
MacBook Found

All of these items could be parsed and tested against in code using a simple dictionary as well. This is truncated output in plist format from the shell, it puts them into a dictionary with the key as _items and then key value for the rest of the data

<key>_items</key>
        <array>
            <dict>
                <key>_name</key>
                <string>hardware_overview</string>
                <key>boot_rom_version</key>
                <string>220.270.99.0.0 (iBridge: 16.16.6568.0.0,0)</string>
                <key>cpu_type</key>
                <string>Intel Core i9</string>
                <key>current_processor_speed</key>
                <string>2.9 GHz</string>
                <key>l2_cache_core</key>
                <string>256 KB</string>
                <key>l3_cache</key>
                <string>12 MB</string>
                <key>machine_model</key>
                <string>MacBookPro15,1</string>
                <key>machine_name</key>
                <string>MacBook Pro</string>
                <key>number_processors</key>
                <integer>6</integer>
                <key>packages</key>
                <integer>1</integer>
                <key>physical_memory</key>
                <string>32 GB</string>
                <key>platform_UUID</key>
                <string>53685B89-8388-55AF-A129-3B8EC1ED00C1</string>
                <key>platform_cpu_htt</key>
                <string>htt_enabled</string>
                <key>serial_number</key>
                <string>C02XG4K6JGH5</string>
            </dict>
        </array>

Hope this helps some of y'all

tlarkin
Honored Contributor

For waiting for a prompt you can build a loop around that and break the loop once you have desired input, and for bonus points you can count how many times the loop runs, or give it a time out before calling break in code and moving forward.

Eigger
Contributor III

This is what we use for renaming.

#!/bin/bash

##############################################################################################################
# renameMacCSV.sh
# Modified by Reggie Santos 
# Last modified 05/31/2018
# add auto-prefix function using system_profiler SPHardwareDataType | grep "Model Identifier" if not using Custom Prefix in parameter $4

#Written by: Frederick Abeloos | Field Technician | Jamf 
#Created On: March 8th, 2018 / script is provided "AS IS" /

#Purpose: Jamf Pro @enrollment script to rename the mac prior to binding to AD (optional)

#Put this script in a policy with option AFTER (important when using .pkg)

#Upload a pkg in JamfPro which puts an assettag.csv file in /tmp/assettags.csv
#First column serialnumber and second column the desired assettag

#OR use CURL to download any .csv from a public link

# Optional: Create a policy with the directory binding payload and custom trigger 'bindToAD'

##############################################################################################################

### RENAMING SETTINGS ########################################################################################

# Don't change any settings here. Use $4-$9 parameters in Jamf Pro script payload options.

# First choose a naming convention for your Mac.
# Custom Prefix can be set via the $4 parameter in the Jamf Pro policy.
# Leave the parameter $4 in Jamf Pro empty if you like to auto-prefix computername base on "Model Identifier", e.g. type "MBP-" (without the quotes). 
prefix="$4"
# Suffix will be set via the $5 parameter in the Jamf Pro policy (script payload).
# Leave the parameter empty if no suffix is required. e.g. type "Mac_" (without the quotes).
suffix="$5"
# Choose serialnumber or assettag via the $6 parameter in the Jamf Pro policy (script payload).
# Set it to "serialnumber" or "assettag" (without the quotes).
namingType="$6"
# Define if you are adding the .pkg with the /tmp/assettags.csv file or using curl to download any .csv file.
# When using curl, the script below will download and rename the file to /tmp/assettags.csv
# Type will be set via $7 paremeter in the Jamf Pro Policy. Type "curl" or "package" (without the quotes).
csvType="$7"
# Define the public link to the .csv file. Leave blank when using .pkg
# Link will be set via th $8 parameter in the Jamf Pro policy. Add the full "https://URL" (without the quotes).
csvURL="$8"
# Path to local csv file, via package in same policy: pkg putting assettags.csv in /tmp/
csvPath="/tmp/assettags.csv"
# Do you want to bind or not? Set to "bind" or leave blank
bindSetting="$9"

### RENAMING SCRIPT ########################################################################################
MBA=`/usr/sbin/system_profiler SPHardwareDataType | grep "Model Identifier" | grep "BookAir"`

MBP=`/usr/sbin/system_profiler SPHardwareDataType | grep "Model Identifier" | grep "BookPro"`

if [ "$4" == "" ] && [ "$MBP" == "" ] && [ "$MBA" != "" ]; then
    prefix="MBA"
elif [ "$4" == "" ] && [ "$MBA" == "" ] && [ "$MBP" != "" ]; then
    prefix="MBP"
elif [ "$4" == "" ] && [ "$MBA" == "" ] && [ "$MBP" == "" ]; then
    prefix="MAC"

fi

if [ "$csvType" = "curl" ]; then

    curl -L -o $csvPath $csvURL

fi



if [ "$namingType" = "serialnumber" ]; then

    echo "using serialnumber"
    jamf setComputerName -useSerialNumber -prefix $prefix -suffix $suffix

elif  [ "$namingType" = "assettag" ]; then

    echo "using assettag" 
    jamf setComputerName -fromFile "$csvPath" -prefix $prefix -suffix $suffix

fi

### AD BINDING ###########################################################################################

if [ "$bindSetting" = "bind" ]; then

    echo "Binding to AD"

    jamf policy -event bindToAD

else echo "Not binding"

fi


### CLEAN UP #############################################################################################

#delete the assettags.csv

rm "$csvPath"

#update Jamf Pro inventory
jamf recon