Module: Jamf::Utility

Includes:
Constants
Included in:
Jamf
Defined in:
lib/jamf/utility.rb

Overview

A collection of useful utility methods. Mostly for converting values between formats, parsing data, and user interaction. This module should be extended into the Jamf Module so all methods become module methods

Constant Summary collapse

OS_TEN_MAXS =

Hash of 'minor' => 'maint' The maximum maint release for macOS 10.minor.maint e.g the highest release of 10.6 was 10.6.8, the highest release of 10.15 was 10.15.7

12 is the default for the current OS and higher (and hoping apple doesn't release 10.16.13)

{
  2 => 8,
  3 => 9,
  4 => 11,
  5 => 8,
  6 => 8,
  7 => 5,
  8 => 5,
  9 => 5,
  10 => 5,
  11 => 6,
  12 => 6,
  13 => 6,
  14 => 6,
  15 => 7
}
MAC_OS_MAXS =

Hash of 'major' => 'minor' The maximum minor release for macOS major.minor e.g. the highest release of 11 is 11.12

12 is the default for the current OS and higher (and hoping apple doesn't release, e.g., 11.13)

{
  11 => 12,
  12 => 12,
  13 => 12,
  14 => 12,
  15 => 12,
  16 => 12,
  17 => 12,
  18 => 12,
  19 => 12,
  20 => 12
}

Instance Method Summary collapse

Instance Method Details

#api_object_class(name) ⇒ Class

Given a name, singular or plural, of a Jamf::APIObject subclass as a String or Symbol (e.g. :computer/'computers'), return the class itself (e.g. Jamf::Computer) The available names are the RSRC_LIST_KEY and RSRC_OBJECT_KEY values for each APIObject subclass.

Parameters:

  • name (String, Symbol) —

    The name of a Jamf::APIObject subclass, singluar or plural

Returns:

  • (Class) —

    The class

Raises:



402
403
404
405
406
407
 
# File 'lib/jamf/utility.rb', line 402

def api_object_class(name)
  klass = api_object_names[name.downcase.to_sym]
  raise Jamf::InvalidDataError, "Unknown API Object Class: #{name}" unless klass

  klass
end

#api_object_names ⇒ Hash

APIObject subclasses have singular names, and are, of course capitalized, e.g. 'Computer' But we often want to refer to them in the plural, or lowercase, e.g. 'computers' This method returns a Hash of the RSRC_LIST_KEY (a plural symbol) and the RSRC_OBJECT_KEY (a singular symbol) of each APIObject subclass, keyed to the class itself, such that both :computer and :computers are keys for Jamf::Computer and both :policy and :policies are keys for Jamf::Policy, and so on.

Returns:

  • (Hash) —

    APIObject subclass names to Classes



421
422
423
424
425
426
427
428
429
430
431
432
433
434
 
# File 'lib/jamf/utility.rb', line 421

def api_object_names
  return @api_object_names if @api_object_names

  @api_object_names ||= {}
  JSS.constants.each do |const|
    klass = JSS.const_get const
    next unless klass.is_a? Class
    next unless klass.ancestors.include? Jamf::APIObject

    @api_object_names[klass.const_get(:RSRC_LIST_KEY).to_sym] = klass if klass.constants.include? :RSRC_LIST_KEY
    @api_object_names[klass.const_get(:RSRC_OBJECT_KEY).to_sym] = klass if klass.constants.include? :RSRC_OBJECT_KEY
  end
  @api_object_names
end

#array_to_rexml_array(element, list) ⇒ Array<REXML::Element>

Given an element name and an array of content, generate an Array of REXML::Element objects with that name, and matching content. Given element name 'foo' and the array ['bar','morefoo'] The array of REXML elements would render thus:

<foo>bar</foo>
<foo>morefoo</foo>

Parameters:

  • element (#to_s) —

    an element_name like :foo

  • list (Array<#to_s>) —

    an Array of element content such as [“bar”, :morefoo]

Returns:

  • (Array<REXML::Element>)

Raises:



464
465
466
467
468
469
470
471
472
473
 
# File 'lib/jamf/utility.rb', line 464

def array_to_rexml_array(element, list)
  raise Jamf::InvalidDataError, 'Arg. must be an Array.' unless list.is_a? Array

  element = element.to_s
  list.map do |v|
    e = REXML::Element.new(element)
    e.text = v
    e
  end
end

#devmode(setting) ⇒ Boolean

un/set devmode mode. Useful when coding - methods can call JSS.devmode? and then e.g. spit out something instead of performing some action.

Parameters:

  • Set (Symbol) —

    devmode :on or :off

Returns:

  • (Boolean) —

    The new state of devmode



633
634
635
 
# File 'lib/jamf/utility.rb', line 633

def devmode(setting)
  @devmode = setting == :on
end

#devmode? ⇒ Boolean

is devmode currently on?

Returns:

  • (Boolean) 


641
642
643
 
# File 'lib/jamf/utility.rb', line 641

def devmode?
  @devmode
end

#epoch_to_time(epoch) ⇒ Time?

Converts JSS epoch (unix epoch + milliseconds) to a Ruby Time object

Parameters:

  • epoch (String, Integer, nil)

Returns:

  • (Time, nil) —

    nil is returned if epoch is nil, 0 or an empty String.



383
384
385
386
387
 
# File 'lib/jamf/utility.rb', line 383

def epoch_to_time(epoch)
  return nil if NIL_DATES.include? epoch

  Time.at(epoch.to_i / 1000.0)
end

#escape_xml(string) ⇒ String

Given a string of xml element text, escape any characters that would make XML unhappy.

* & => &amp;
* " => &quot;
* < => &lt;
* > => &gt;
* ' => &apos;

Parameters:

  • string (String) —

    the string to make xml-compliant.

Returns:

  • (String) —

    the xml-compliant string



447
448
449
 
# File 'lib/jamf/utility.rb', line 447

def escape_xml(string)
  string.gsub(/&/, '&amp;').gsub(/"/, '&quot;').gsub(/>/, '&gt;').gsub(/</, '&lt;').gsub(/'/, '&apos;')
end

#expand_min_os(min_os) ⇒ Array

Converts an OS Version into an Array of equal or higher OS versions, up to some non-existant max, hopefully far in the future, currently 20.12.10

This array can then be joined with commas and used as the value of the os_requirements for Packages and Scripts.

It's unlikely that this method, as written, will still be in use by the release of macOS 20.12.10, but currently thats the upper limit.

Hopefully well before then JAMF will implement a “minimum OS” in Jamf Pro itself, then we could avoid the inherant limitations in using a method like this.

When the highest maint. release of an OS version is not known, because its the currently released OS version or higher, then this method assumes '12' e.g. '10.16.12', '11.12', '12.12', etc.

Apple has never released more than 11 updates to a version of macOS (that being 10.4), so hopefully 12 is enough

Since Big Sur might report itself as either '10.16' or '11.x.x', this method will allow for both possibilities, and the array will contain whatever iterations needed for both version numbers

Examples:

JSS.expand_min_os ">=10.9.4" # => returns this array
 # ["10.9.4",
 #  "10.9.5",
 #  "10.10.x"
 #  ...
 #  "10.16.x",
 #  "11.x",
 #  "12.x",
 #  ...
 #  "20.x"]

Parameters:

  • min_os (String) —

    the mimimum OS version to expand, e.g. “>=10.9.4” or “11.1”

Returns:

  • (Array) —

    Nearly all potential OS versions from the minimum to 20.12.10



122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
 
# File 'lib/jamf/utility.rb', line 122

def expand_min_os(min_os)
  min_os = min_os.delete '>='

  # split the version into major, minor and maintenance release numbers
  major, minor, maint = min_os.split('.')
  minor = 'x' if minor.nil? || minor == '0'
  maint = 'x' if maint.nil? || maint == '0'

  ok_oses = []

  # Deal with 10.x.x up to 10.16
  if major == '10'

    # In big sur with SYSTEM_VERSION_COMPAT
    # set, it will only ever report as `10.16`
    # So if major is 10 and minor is 16, ignore maint
    # and start explicitly at '10.16'
    if minor == '16'
      ok_oses << '10.16'

    # But for Catalina and below, we need to
    # expand things out
    else
      # e.g. 10.14.x
      # doesn't expand to anything
      if maint == 'x'
        ok_oses << "10.#{minor}.x"

      # e.g. 10.15.5
      # expand to 10.15.5, 10.15.6, 10.15.7
      else
        max_maint_for_minor = OS_TEN_MAXS[minor.to_i]

        (maint.to_i..max_maint_for_minor).each do |m|
          ok_oses << "#{major}.#{minor}.#{m}"
        end # each m
      end # if maint == x

      # now if we started below catalina, account for everything
      # up to 10.15.x
      ((minor.to_i + 1)..15).each { |v| ok_oses << "10.#{v}.x" } if minor.to_i < 15

      # and add big sur with SYSTEM_VERSION_COMPAT
      ok_oses << '10.16'
    end # if minor == 16

    # now reset these so we can go higher
    major = '11'
    minor = 'x'
    maint = 'x'
  end # if major == 10

  # if the min os is 11.0.0 or equiv, and we aven't added 10.16
  # for SYSTEM_VERSION_COMPAT, add it now
  ok_oses << '10.16' if ['11', '11.x', '11.x.x', '11.0', '11.0.0'].include?(min_os) && !ok_oses.include?('10.16')

  # e.g. 11.x, or 11.x.x
  # expand to 11.x, 12.x, 13.x, ... 20.x
  if minor == 'x'
    ((major.to_i)..20).each { |v| ok_oses << "#{v}.x" }

  # e.g. 11.2.x
  # expand to 11.2.x, 11.3.x, ... 11.12.x,
  #   12.x, 13.x,  ... 20.x
  elsif maint == 'x'
    # first expand the minors out to their max
    # e.g. 11.2.x, 11.3.x, ... 11.12.x
    max_minor_for_major = MAC_OS_MAXS[major.to_i]
    ((minor.to_i)..max_minor_for_major).each do |m|
      ok_oses << "#{major}.#{m}.x"
    end # each m

    # then add the majors out to 20
    ((major.to_i + 1)..20).each { |v| ok_oses << "#{v}.x" }

  # e.g. 11.2.3
  # expand to 11.2.3, 11.2.4, ... 11.2.10,
  #   11.3.x, 11.4.x, ... 11.12.x,
  #   12.x, 13.x, ... 20.x
  else
    # first expand the maints out to 10
    # e.g. 11.2.3, 11.2.4, ... 11.2.10
    ((maint.to_i)..10).each { |mnt| ok_oses << "#{major}.#{minor}.#{mnt}" }

    # then expand the minors out to their max
    # e.g. 11.3.x, ... 11.12.x
    max_minor_for_major = MAC_OS_MAXS[major.to_i]
    ((minor.to_i + 1)..max_minor_for_major).each { |min| ok_oses << "#{major}.#{min}.x" }

    # then add the majors out to 20
    ((major.to_i + 1)..20).each { |v| ok_oses << "#{v}.x" }
  end

  ok_oses
end

#hash_to_rexml_array(hash) ⇒ Array<REXML::Element>

Given a simple Hash, convert it to an array of REXML Elements such that each key becomes an element, and its value becomes the text content of that element

Examples:

my_hash = {:foo => "bar", :baz => :morefoo}
xml = JSS.hash_to_rexml_array(my_hash)
xml.each{|x| puts x }

<foo>bar</foo>
<baz>morefoo</baz>

Parameters:

  • hash (Hash{#to_s => #to_s}) —

    the Hash to convert

Returns:

  • (Array<REXML::Element>) —

    the Array of REXML elements.

Raises:



491
492
493
494
495
496
497
498
499
500
501
 
# File 'lib/jamf/utility.rb', line 491

def hash_to_rexml_array(hash)
  raise InvalidDataError, 'Arg. must be a Hash.' unless hash.is_a? Hash

  ary = []
  hash.each_pair do |k, v|
    el = REXML::Element.new k.to_s
    el.text = v
    ary << el
  end
  ary
end

#humanize_secs(secs) ⇒ String

Very handy! lifted from stackoverflow.com/questions/4136248/how-to-generate-a-human-readable-time-range-using-ruby-on-rails

Turns the integer 834756398 into the string “26 years 23 weeks 1 day 12 hours 46 minutes 38 seconds”

Parameters:

  • secs (Integer) —

    a number of seconds

Returns:

  • (String) —

    a human-readable (English) version of that number of seconds.



655
656
657
658
659
660
661
662
663
 
# File 'lib/jamf/utility.rb', line 655

def humanize_secs(secs)
  [[60, :second], [60, :minute], [24, :hour], [7, :day], [52.179, :week], [1_000_000_000, :year]].map do |count, name|
    next unless secs > 0

    secs, n = secs.divmod(count)
    n = n.to_i
    "#{n} #{n == 1 ? name : (name.to_s + 's')}"
  end.compact.reverse.join(' ')
end

#item_list_to_rexml_list(list_element, item_element, item_list, content = :name) ⇒ REXML::Element

Given an Array of Hashes with :id and/or :name keys, return a single REXML element with a sub-element for each item, each of which contains a :name or :id element.

e.g. :computers

e.g. :computer

Examples:

comps = [{:id=>2,:name=>'kimchi'},{:id=>5,:name=>'mantis'}]
xml = JSS.item_list_to_rexml_list(:computers, :computer , comps, :name)
puts xml
# output manually formatted for clarity. No newlines in the real xml string
<computers>
  <computer>
    <name>kimchi</name>
  </computer>
  <computer>
    <name>mantis</name>
  </computer>
</computers>

# if content is :id, then, eg. <name>kimchi</name> would be <id>2</id>

Parameters:

  • list_element (#to_s) —

    the name of the XML element that contains the list.

  • item_element (#to_s) —

    the name of each XML element in the list,

  • item_list (Array<Hash>) —

    an Array of Hashes each with a :name or :id key.

  • content (Symbol) (defaults to: :name) —

    which hash key should be used as the content of if list item? Defaults to :name

Returns:

  • (REXML::Element) —

    the item list as REXML



535
536
537
538
539
540
541
 
# File 'lib/jamf/utility.rb', line 535

def item_list_to_rexml_list(list_element, item_element, item_list, content = :name)
  xml_list = REXML::Element.new  list_element.to_s
  item_list.each do |i|
    xml_list.add_element(item_element.to_s).add_element(content.to_s).text = i[content]
  end
  xml_list
end

#os_ok?(requirement, os_to_check = nil) ⇒ Boolean

Scripts and packages can have OS limitations. This method tests a given OS, against a requirement list to see if the requirement is met.

Parameters:

  • requirement (String, Array) —

    The os requirement list, a comma-seprated string or array of strings of allows OSes. e.g. 10.7, 10.8.5 or 10.9.x

  • processor (String) —

    the os to check, defaults to the os of the current machine.

Returns:

  • (Boolean) —

    can this pkg be installed with the processor given?



252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
 
# File 'lib/jamf/utility.rb', line 252

def os_ok?(requirement, os_to_check = nil)
  return true if requirement.to_s =~ /none/i
  return true if requirement.to_s == 'n'

  requirement = JSS.to_s_and_a(requirement)[:arrayform]
  return true if requirement.empty?

  os_to_check ||= `/usr/bin/sw_vers -productVersion`.chomp

  # convert the requirement array into an array of regexps.
  # examples:
  #   "10.8.5" becomes  /^10\.8\.5$/
  #   "10.8" becomes /^10.8(.0)?$/
  #   "10.8.x" /^10\.8\.?\d*$/
  req_regexps = requirement.map do |r|
    if r.end_with?('.x')
      /^#{r.chomp('.x').gsub('.', '\.')}(\.?\d*)*$/

    elsif r =~ /^\d+\.\d+$/
      /^#{r.gsub('.', '\.')}(.0)?$/

    else
      /^#{r.gsub('.', '\.')}$/
    end
  end

  req_regexps.each { |re| return true if os_to_check =~ re }
  false
end

#parse_jss_version(version) ⇒ Hash{Symbol => String, Gem::Version}

Parse a JSS Version number into something comparable.

This method returns a Hash with these keys:

  • :major => the major version, Integer

  • :minor => the minor version, Integor

  • :maint => the revision, Integer (also available as :patch and :revision)

  • :build => the revision, String

  • :version => a Gem::Version object built from :major, :minor, :revision which can be easily compared with other Gem::Version objects.

NOTE: the :version value ignores build numbers, so comparisons only compare major.minor.maint

Parameters:

  • version (String) —

    a JSS version number from the API

Returns:

  • (Hash{Symbol => String, Gem::Version}) —

    the parsed version data.

Raises:



560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
 
# File 'lib/jamf/utility.rb', line 560

def parse_jss_version(version)
  major, second_part, *_rest = version.split('.')
  raise Jamf::InvalidDataError, 'JSS Versions must start with "x.x" where x is one or more digits' unless major =~ /\d$/ && second_part =~ /^\d/

  release, build = version.split(/-/)

  major, minor, revision = release.split '.'
  minor ||= 0
  revision ||= 0

  {
    major: major.to_i,
    minor: minor.to_i,
    revision: revision.to_i,
    maint: revision.to_i,
    patch: revision.to_i,
    build: build,
    version: Gem::Version.new("#{major}.#{minor}.#{revision}")
  }
end

#parse_plist(plist, symbol_keys: false) ⇒ Object

Parse a plist into a Ruby data structure. The plist parameter may be a String containing an XML plist, or a path to a plist file, or it may be a Pathname object pointing to a plist file. The plist files may be XML or binary.

Parameters:

  • plist (Pathname, String) —

    the plist XML, or the path to a plist file

  • symbol_keys (Boolean) (defaults to: false) —

    should any Hash keys in the result be converted into Symbols rather than remain as Strings?

Returns:

  • (Object) —

    the parsed plist as a ruby hash,array, etc.



335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
 
# File 'lib/jamf/utility.rb', line 335

def parse_plist(plist, symbol_keys: false)
  require 'cfpropertylist'

  # did we get a string of xml, or a string pathname?
  case plist
  when String
    return CFPropertyList.native_types(CFPropertyList::List.new(data: plist).value, symbol_keys) if plist.include? '</plist>'

    plist = Pathname.new plist
  when Pathname
    true
  else
    raise ArgumentError, 'Argument must be a path (as a Pathname or String) or a String of XML'
  end # case plist

  # if we're here, its a Pathname
  raise Jamf::MissingDataError, "No such file: #{plist}" unless plist.file?

  CFPropertyList.native_types(CFPropertyList::List.new(file: plist).value, symbol_keys)
end

#parse_time(a_datetime) ⇒ Object

a wrapper around Time.parse that returns nil for nil, zero, and empty values.



317
318
319
320
321
 
# File 'lib/jamf/utility.rb', line 317

def parse_time(a_datetime)
  return nil if NIL_DATES.include? a_datetime

  Time.parse a_datetime.to_s
end

#processor_ok?(requirement, processor = nil) ⇒ Boolean

Scripts and packages can have processor limitations. This method tests a given processor, against a requirement to see if the requirement is met.

Parameters:

  • requirement (String) —

    The processor requirement. either 'ppc', 'x86', or some variation on “none”, nil, or empty

  • processor (String) (defaults to: nil) —

    the processor to check, defaults to the processor of the current machine. Any flavor of intel

    is (i486, i386, x86-64, etc) is treated as "x86"

Returns:

  • (Boolean) —

    can this pkg be installed with the processor given?



232
233
234
235
236
237
 
# File 'lib/jamf/utility.rb', line 232

def processor_ok?(requirement, processor = nil)
  return true if requirement.to_s.empty? || requirement =~ /none/i

  processor ||= `/usr/bin/uname -p`
  requirement == (processor.to_s.include?('86') ? 'x86' : 'ppc')
end

#prompt_for_password(message) ⇒ String

Prompt for a password in a terminal.

Parameters:

  • message (String) —

    the prompt message to display

Returns:

  • (String) —

    the text typed by the user



612
613
614
615
616
617
618
619
620
621
622
623
 
# File 'lib/jamf/utility.rb', line 612

def prompt_for_password(message)
  begin
    $stdin.reopen '/dev/tty' unless $stdin.tty?
    $stderr.print "#{message} "
    system '/bin/stty -echo'
    pw = $stdin.gets.chomp("\n")
    puts
  ensure
    system '/bin/stty echo'
  end # begin
  pw
end

#stdin(line = 0) ⇒ String?

Retrive one or all lines from whatever was piped to standard input.

Standard input is read completely the first time this method is called and the lines are stored as an Array in the module var @stdin_lines

Parameters:

  • line (Integer) (defaults to: 0) —

    which line of stdin is being retrieved. The default is zero (0) which returns all of stdin as a single string.

Returns:

  • (String, nil) —

    the requested ling of stdin, or nil if it doesn't exist.



597
598
599
600
601
602
603
604
 
# File 'lib/jamf/utility.rb', line 597

def stdin(line = 0)
  @stdin_lines ||= ($stdin.tty? ? [] : $stdin.read.lines.map { |l| l.chomp("\n") })

  return @stdin_lines.join("\n") if line <= 0

  idx = line - 1
  @stdin_lines[idx]
end

#superuser? ⇒ Boolean

Returns is this code running as root?.

Returns:

  • (Boolean) —

    is this code running as root?



583
584
585
 
# File 'lib/jamf/utility.rb', line 583

def superuser?
  Process.euid.zero?
end

#to_s_and_a(somedata) ⇒ Hash{:stringform => String, :arrayform => Array}

Given a list of data as a comma-separated string, or an Array of strings, return a Hash with both versions.

Some parts of the JSS require lists as comma-separated strings, while often those data are easier work with as arrays. This method is a handy way to get either form when given either form.

Examples:

JSS.to_s_and_a "foo, bar, baz" # Hash => {:stringform => "foo, bar, baz", :arrayform => ["foo", "bar", "baz"]}

JSS.to_s_and_a ["foo", "bar", "baz"] # Hash => {:stringform => "foo, bar, baz", :arrayform => ["foo", "bar", "baz"]}

Parameters:

  • somedata (String, Array) —

    the data to parse, of either class,

Returns:

  • (Hash{:stringform => String, :arrayform => Array}) —

    the data as both comma-separated String and Array



298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
 
# File 'lib/jamf/utility.rb', line 298

def to_s_and_a(somedata)
  case somedata
  when nil
    valstr = ''
    valarr = []
  when String
    valstr = somedata
    valarr = somedata.split(/,\s*/)
  when Array
    valstr = somedata.join ', '
    valarr = somedata
  else
    raise Jamf::InvalidDataError, 'Input must be a comma-separated String or an Array of Strings'
  end # case
  { stringform: valstr, arrayform: valarr }
end

#xml_plist_from(data) ⇒ String

Convert any ruby data to an XML plist.

NOTE: Binary data is tricky. Easiest way is to pass in a Pathname or IO object (anything that responds to `read` and returns a bytestring) and then the CFPropertyList.guess method will read it and convert it to a Plist <data> element with base64 encoded data. For more info, see CFPropertyList.guess

Parameters:

  • data (Object) —

    the data to be converted, usually a Hash

Returns:

  • (String) —

    the object converted into an XML plist



370
371
372
373
374
375
 
# File 'lib/jamf/utility.rb', line 370

def xml_plist_from(data)
  require 'cfpropertylist'
  plist = CFPropertyList::List.new
  plist.value = CFPropertyList.guess(data, convert_unknown_to_string: true)
  plist.to_str(CFPropertyList::List::FORMAT_XML)
end