ruby-jss 2.0.0a11 → 2.0.0b5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 89a063540145b18ab23771f87ad6eb8db80c154e911138b75a7a233117fb6dc6
-  data.tar.gz: 139bc4655da8a6a1e4bed21add476f6bfaa844dbda4fb0e477fdcee45079cc9e
+  metadata.gz: b224082c515d1c12d50736092c859f7f21f1acbb8d6d3d98b861221362a64f7d
+  data.tar.gz: 7c10c43a7842be9fee1819bcc9a92457e1dce0d338e97f6a00eeb27d1f75416b
 SHA512:
-  metadata.gz: 33d23b52e0f0bb2f42a4a606ae34288b4b7ceb35e3d051ccb1281456709cdee7d4bcaea26681f30126239e30bde8de55bc550fe021d2e59a9cd0f05fdeb40d65
-  data.tar.gz: b8b8c46c351833c47493ad2512349b952522d2916bcac4b0143a0c49aa140187635be582ad5971da0a63e599fbd263900eecb01bce24b88da6f88a8457ea82fc
+  metadata.gz: 32378337e65ccc32c836217434cdfa7fe060d400268848174a675fcbb163157e0d0965e4141fbbe35eef60bd042cca58777189db18d641e1dd1d6ab7ed4d2f37
+  data.tar.gz: d24ca66815d7e4f0b5bc4a212463e985e490fe499c0b299b6df41f12f6943f156f2681ebf5696198a66853e593cbfe9fdd248948e02e4d2c67ccd910cccc900f

data/CHANGES.md

@@ -26,6 +26,7 @@ Here are the high-level changes and there are many many others. For more details
   - Base classes for JamfPro API objects are automatically generated from the OAPI3 Schema available at https://your.jamf.server/api/schema. This make it much simpler for ruby-jss to implement and update new and changed objects and endpoints in the Jamf Pro API.
 - Code is auto-loaded from disk
   - Because the auto-generated classes add hundreds of files to the gem, ruby-jss now uses [Zeitwerk](https://github.com/fxn/zeitwerk) to auto-load only the files it needs when they are needed.
+- Support for Ruby 3.x
 
 
 ## \[1.6.7] - 2022-02-22

data/README-2.0.0.md

@@ -2,13 +2,13 @@
 
 Version 2.0.0 is a major refactoring of ruby-jss. While attempting to provide as much backward compatibility as possible, there are some significant changes under the hood. **_PLEASE TEST YOUR CODE EXTENSIVELY_**
 
-This document discusses the high-level changes, listing the detailed changes that have already happened, as well as planned changes and deprecations. It also provides some discussion and background around all of this. It is a work-in-progress at the moment, and hopefully will prompt discussion and decision-making in the #ruby-jss channel of MacAdmins Slack (Please join us!)
+This document discusses the high-level changes, listing the changes that have already happened, as well as planned changes and deprecations. It also provides some discussion and background around all of this. It is a work-in-progress at the moment, and will probably remain so well after initial release. Hopefully this document will prompt discussion and decision-making in the #ruby-jss channel of MacAdmins Slack (Please join us!)
 
-These changes have been in mind for some time, but the ability (soon to be requirement) for the Classic API to authenticate with Bearer Tokens from the Jamf Pro API means that the time has come, so here we are!
+These changes have been in mind for some time, but the requirement in late 2022 for the Classic API to authenticate with Bearer Tokens from the Jamf Pro API means that the time has come, so here we are!
 
 **CONTENTS**
 
-<!-- TOC depthFrom:2 depthTo:6 withLinks:1 updateOnSave:1 orderedList:0 -->
+<!-- TOC -->
 
 - [Requirements](#requirements)
 - [High level changes](#high-level-changes)
@@ -17,24 +17,25 @@ These changes have been in mind for some time, but the ability (soon to be requi
     - [A single Connection class](#a-single-connection-class)
       - [Connecting to the API](#connecting-to-the-api)
         - [The default connection](#the-default-connection)
-    - [A single namespace `Jamf`](#a-single-namespace-jamf)
+    - [A single namespace Jamf](#a-single-namespace-jamf)
       - [Inherant differences between the APIs](#inherant-differences-between-the-apis)
       - [Which API does an object come from?](#which-api-does-an-object-come-from)
   - [Automatic code generation](#automatic-code-generation)
   - [Autoloading with Zeitwerk](#autoloading-with-zeitwerk)
 - [Notable changes from ruby-jss 1.x](#notable-changes-from-ruby-jss-1x)
   - [Paged queries to the Jamf Pro API](#paged-queries-to-the-jamf-pro-api)
-  - [API data are no longer cached (?)](#api-data-are-no-longer-cached-)
+  - [API data are no longer cached ?](#api-data-are-no-longer-cached-)
   - [No Attribute aliases for Jamf Pro API objects](#no-attribute-aliases-for-jamf-pro-api-objects)
   - [Class/Mixin hierarchy for Jamf Pro API objects](#classmixin-hierarchy-for-jamf-pro-api-objects)
+  - [Support for 'Sticky Sessions' in Jamf Cloud](#support-for-sticky-sessions-in-jamf-cloud)
 - [Planned deprecations](#planned-deprecations)
   - [Use of the term 'api' in method names, parameter names, and attributes](#use-of-the-term-api-in-method-names-parameter-names-and-attributes)
-  - [`.map_all_ids_to` method for Classic API collection classes](#map_all_ids_to-method-for-classic-api-collection-classes)
-  - [Using `.make`, `#create`, and `#update` for Classic API objects](#using-make-create-and-update-for-classic-api-objects)
+  - [.map_all_ids_to method for Classic API collection classes](#map_all_ids_to-method-for-classic-api-collection-classes)
+  - [Using .make, #create, and #update for Classic API objects](#using-make-create-and-update-for-classic-api-objects)
   - [JSS::CONFIG](#jssconfig)
-  - [Jamf::Connection instance methods `#next_refresh`, `#secs_to_refresh`, &  `#time_to_refresh`](#jamfconnection-instance-methods-next_refresh-secs_to_refresh---time_to_refresh)
+  - [Jamf::Connection instance methods #next_refresh, #secs_to_refresh, &  #time_to_refresh](#jamfconnection-instance-methods-next_refresh-secs_to_refresh---time_to_refresh)
   - [Cross-object validation in setters](#cross-object-validation-in-setters)
-  - [.fetch :random](#fetch-random)
+  - [fetch :random](#fetch-random)
 - [Documentation](#documentation)
 - [How to install for testing](#how-to-install-for-testing)
 - [Contact](#contact)
@@ -51,19 +52,19 @@ This means it will work with the OS-supplied /usr/bin/ruby in macOS 10.15 Catali
 
 ### Ruby 3.x support
 
-The plan is for ruby-jss 2.0+ to be compatible with ruby 2.7 and higher, including ruby 3.x
+The plan is for ruby-jss 2.0+ to be compatible with ruby 2.6.3 and higher, including ruby 3.x
 
 As of this writing, basic access to the API seems to be working in ruby 3, but much much more testing is needed.
 
-It looks like the biggest changes have been dealing with keyword arguments as Hashs.  Methods defined with `def methodname([...] foo = {}` need to be changed to `def methodname([...] **foo` and calls to those methods, even in your own code, need to be changed to `methodname([...] **foo)` when `foo` is a hash of keyword args.
+It looks like the biggest changes have been dealing with keyword arguments as Hashs.  Methods defined with `def methodname([...] foo = {})` need to be changed to `def methodname([...] **foo)` and calls to those methods, even in your own code, need to be changed to `methodname([...] **foo)` when `foo` is a hash of keyword args.
 
-**IMPORTANT**: do not pass raw hashes as 'keyword' args.  Instead use the double-splat: `methodname(**hash)` which should be compatible with ruby 3.x and 2.7.x
+**IMPORTANT**: do not pass raw hashes as 'keyword' args.  Instead use the double-splat: `methodname(**hash)` which should be compatible with ruby 3.x and 2.6.x
 
 for more info see [Separation of positional and keyword arguments in Ruby 3.0](https://www.ruby-lang.org/en/news/2019/12/12/separation-of-positional-and-keyword-arguments-in-ruby-3-0/)
 
 ### Combined API access
 
-ruby-jss has always used the `JSS` module to encapsulate all access to the Classic API. When the Jamf Pro API became a thing, the `Jamf` module was created as the way to interact with that API as it grew and developed.
+Previous versions of ruby-jss used the `JSS` module to encapsulate all access to the Classic API. When the Jamf Pro API became a thing, it was a vary different beast, so the `Jamf` module was created as the way to interact with that API as it grew and developed.
 
 Even though the latest Jamf Pro release notes say the Jamf Pro API is still officially "open for user testing", it has stablized enough that it is used by many folks for production work.
 
@@ -75,23 +76,24 @@ There is now one `Jamf::Connection` class, instances of which are connections to
 
 ##### Connecting to the API
 
-Most of the previous methods and parameters for making API connections to either API should still work, including using a URL rather than individual connection parameters. So both of these are valid and identical:
+Most of the previous methods and parameters for making connections to either API should still work, including using a URL rather than individual connection parameters. So both of these are valid and identical:
 
 ```ruby
 a_connection = Jamf::Connection.new 'https://apiuser@my.jamf.server:8443/', pw: :prompt
 
 another_connection = Jamf::Connection.new host: 'my.jamf.server', port: 8443, user: 'apiuser', pw: :prompt
 ```
-
 Other connection parameters can be passed in as normal.
 
 ###### The default connection
 
-The top-level module methods for accessing the 'default' connection are still available and are now synonyms: `Jamf.cnx` and `JSS.api` both return the current default Jamf::Connection instance. There is also a top-level methods`Jamf.connect` which is the same as `Jamf.cnx.connect`. The top-level methods for changing the default connection are still there. The use of `JSS::API` has been deprecated for years now, and still is (see below).
+The top-level module methods for accessing the 'default' connection are still available and are now synonyms: `Jamf.cnx` and `JSS.api` both return the current default Jamf::Connection instance. There is also a top-level methods`Jamf.connect` which is the same as `Jamf.cnx.connect`. The top-level methods for changing the default connection are still there. 
+
+NOTE: The use of `JSS::API` has been deprecated for years now, and still is (see below).
 
 #### A single namespace `Jamf`
 
-Version 2.0.0 combines the `JSS` module and the `Jamf` module into a single `Jamf` module, with `JSS` aliased to it. This means you can use them interchangably to refer to the Jamf module, and existing code that used either should still work. The module name no longer indicates which API you're working with.
+Version 2.0.0 combines the `JSS` module and the `Jamf` module into a single `Jamf` module, with `JSS` aliased to it. This means you can use the names interchangably to refer to the Jamf module, and existing code that used either should still work. The module name no longer indicates which API you're working with.
 
 For example, the `JSS::Computer` class, from the Classic API, is still a thing, but now just points to the `Jamf::Computer` class, still from the Classic API.  The `Jamf::InventoryPreloadRecord` class, from the Jamf Pro API remains as is, but can also be referred to as `JSS::InventoryPreloadRecord`
 
@@ -101,9 +103,9 @@ In theory, you shouldn't need to worry about which classes and objects come from
 
 However, in reality the two APIs have different functionality, some of which must be reflected in the ruby classes that represent objects in those APIs.
 
-Take, for example, the classes for 'Collection Resources' - API endpoints that let you deal with collections of objects like Computers, or Inventory Preload Records.  These classes implement a `.all` class method, which retrieves a list of some data about all members of the collection.
+Take, for example, the classes for 'Collection Resources' - API endpoints that let you deal with collections of objects like Computers, or Inventory Preload Records.  These classes implement a `.all` class method, which retrieves an Array of some data about all members of the collection.
 
-Not only is the data returned in such lists very different between the APIs, but in the Jamf Pro API, you can ask the server to return the list already sorted, possibly filtered, or 'paged' in groups of some number of items. None of that is possible in the Classic API.
+Not only is the data returned in such Arrays very different between the APIs, but in the Jamf Pro API, you can ask the server to return the list already sorted, possibly filtered, or 'paged' in groups of some number of items. None of that is possible in the Classic API.
 
 The `.all` method, and its relatives like `.all_ids`, `.all_names`, etc. exist for Collection Resources in both APIs, but the methods take different parameters, e.g. to deal with sorting and filtering. Jamf Pro API classes have a `.pager` method which returns an object from which you can retrieve the 'pages' of your query.
 
@@ -119,13 +121,15 @@ The API documentation you see at your own Jamf Pro server at https://your.jamf.s
 
 In ruby-jss 2.0 and up, the OAPI spec is used to automatically generate hundreds of 'base' classes in ruby-jss, each with automatically generated attribute getters, setters, validators, and other useful methods. These base classes can then be used as the superclasses of the Jamt Pro API objects we implement for direct use in ruby-jss - and the majority of the coding is already done! The subclasses implementing objects in ruby-jss can then be expanded and customized beyond the simple, auto-generated superclass.
 
-Not only does this make it fast and simple to implement new objects in ruby-jss, but allows fast and simple updates to existing objects, when new functionality is introduced to the API.
+Not only does this make it fast and simple to implement new objects in ruby-jss, but allows fast and simple updates to existing objects, when new functionality is introduced to the API. 
+
+Hopefully it will also allow a single version of ruby-jss to work with a wide range of Jamf Pro versions, as API endpoint versions are added and deprecated (details of how that'll work are TBD)
 
-If you develop ruby-jss, please see (documentation link TBA) for more info about how to use the auto-generated classes.
+If you develop ruby-jss, please see (documentation link will go here) for more info about how to use the auto-generated classes, or reach out to us. (See [Contact](#contact), below)
 
 ### Autoloading with Zeitwerk
 
-Because the classes generated from the OAPI spec number in the hundreds, it's a waste of memory and time to load all of them in every time you `require ruby-jss`, since most of them will never be used for any given application.
+Because the classes generated from the OAPI spec number in the hundreds, it's a waste of memory and time to load all of them in every time you `require 'ruby-jss'`, since most of them will never be used for any given application.
 
 To deal with this, ruby-jss now uses the wonderfully cool [Zeitwerk gem](https://github.com/fxn/zeitwerk) to automatically load only the files needed for classes and modules as they are used.
 
@@ -148,26 +152,32 @@ The `.all` method will never deliver paged results, however if you give it a `fi
 
 ### API data are no longer cached (?)
 
-**NOTE:** As of this writing, caching has been removed for the objects from the Jamf Pro API, but caching remains in the Classic API. Its removal, or the re-instatement of caching for JP API objects, are pending discussion with users of ruby-jss.
+**NOTE:** As of this writing, caching has been removed for the objects from the Jamf Pro API, but caching remains in the Classic API. Its removal, or the re-instatement of caching for JP API objects, are pending discussion with other users of ruby-jss.
 
 Pre-2.0, methods that would fetch large datasets from the server would always cache that data in the Connection object, and by default use the cache in future calls unless a `refresh` parameter is given. These datasets included:
 
 - collection lists, used by `.all` and friends, like `.all_ids` and `.valid_id`
-- Extension Attribute, definitions, used for validating Extension Attribute values
+- Extension Attribute definitions, used for validating Extension Attribute values
 
 In 2.0+, that caching has been removed for objects from them Jamf Pro API.
 
- If you want to avoid repeated GET requests to the server when you aren't worried that the resulting data may have changed, you can store the results of `.all` in a variable, and either use it yourself, or pass it in to other methods via the `cached_list:` parameter. Passing in a cached_list wil prevent those methods from calling `.all` and reaching out to the server again.
+If you want to avoid repeated GET requests to the server when you aren't worried that the resulting data may have changed, you can store the results of `.all` in a variable, and either use it yourself, or pass it in to other methods via the `cached_list:` parameter. Passing in a cached_list wil prevent those methods from calling `.all` and reaching out to the server again.
 
-**WARNING**: Be careful that the list you pass in via `cached_list` contains the correct data structure for the class, and came from the desired Connection instance.
+**WARNING**: Caching a list yourself and using it with `cached_list` can cause all kinds of problems if you don't ensure these points:
+- the cached_list contains the correct data structure for the class
+  - i.e. pass in the results of `.all` for that class, not `.all_ids` or other sub-lists.
+- the cached_list came from the correct Connection instance
+- the cached_list is sufficiently up to date.
 
 ### No Attribute aliases for Jamf Pro API objects
 
 Objects from the Jamf Pro API will no longer define aliases for the attribute names that come from the API itself. This means, e.g., to get the name of a ComputerPrestage or MobileDevicePrestage, you have to ask for its `displayName` not its `name`, since the property comes from the API as `displayName`. To see a list  of all the names, you must use `.all_displayNames` not `.all_names`.  For objects with a 'name' property (most of them) then you can use `.name` and `.all_names`.
 
-The reason behind this is twofold: 1) to simplify ruby-jss's code and automate as much as possible; 2) to reflect what Jamf actually gives us in the APIs.
+The reason behind this is twofold: 
+1) to simplify ruby-jss's code and automate as much as possible
+2) to reflect what Jamf actually gives us in the APIs
 
-This is a breaking change from earlier ruby-jss versions, for which Jamf Pro API objects had the potential for aliases of their attribute names.
+**IMPORTANT** This is a breaking change from earlier ruby-jss versions, for which Jamf Pro API objects had the potential for aliases of their attribute names.
 
 ### Class/Mixin hierarchy for Jamf Pro API objects
 
@@ -175,7 +185,17 @@ If you contribute to ruby-jss, be aware that the structure of superclasses, subc
 
 There's a lot to document about these changes, and much of the current documentation is out of date, referring to how things were done when the Jamf module was separate and only talked to the Jamf Pro API.
 
-Give us time and we'll get everything updated. In the meantime, feel free to reach out in the #ruby-jss channel of Macadmins Slack, or open an issue on GitHub, or email ruby-jss@pixar.com.
+Give us time and we'll get everything updated. In the meantime, feel free to reach out for assistance or questions. (See [Contact](#contact), below)
+
+### Support for 'Sticky Sessions' in Jamf Cloud
+
+If you are connecting to a Jamf Cloud server, you can specifcy `sticky_sessions: true` when calling `Jamf::Connection.new` or `Jamf::Connection#connect`. If you already have a connected Connection object, you can enable or disable Sticky Sessions using `my_connection_object.sticky_session =` with a boolean value (for the default connection use `Jamf.cnx.sticky_session = <boolean>`). To see the actual cookie being sent to enable sticky sessions, use `Jamf::Connection#sticky_session_cookie`. 
+
+Attempting to enable a sticky session with a connection to an on-prem server (host not ending in 'jamfcloud.com') will raise an error.
+
+For details about Sticky Sessions see [Sticky Sessions for Jamf Cloud](https://developer.jamf.com/developer-guide/docs/sticky-sessions-for-jamf-cloud) at the Jamf Developer site. 
+
+**WARNING:** Jamf recommends NOT using sticky sessions unless they are needed. Using them inappropriately may negatively impact performance, especially for large automated processes.
 
 ## Planned deprecations
 
@@ -201,11 +221,15 @@ comp = JSS::Computer.fetch id: 12, cnx: my_connection
 comp.cnx # => my_connection
 ```
 
-In ruby-jss < 2.0, the term `api` is used with the Classic API in method names, method parameters, instance variables, attributes, and constants. It is used to pass, access, or hold instances of JSS::APIConnnection, e.g. so a method that talks to the server would use the passed connection rather than the module-wide default connection.  But, the thing being passed is a 'connection' not an 'API', and now that there are actually two APIs at play, that usage is even less appropriate.
+In ruby-jss < 2.0, the term `api` is used with the Classic API in method names, method parameters, instance variables, attributes, and constants. It is used to pass, access, refer to, or hold instances of JSS::APIConnnection, e.g. so a method that talks to the server would use the passed connection rather than the module-wide default connection.  
+
+But, the thing being passed is a 'connection' not an 'API', and now that there are actually two APIs at play, that usage is even less appropriate.
+
+The original Jamf module, which accessed only the Jamf Pro API, has always used the better-suited abbreviation `cnx` for this, and now that is standard everywhere. 
 
-The Original Jamf module, which accessed only the Jamf Pro API, has always used the better-suited abbreviation `cnx` for this, and now that is standard everywhere. For now `api` should continue to work, but it will be removed 'eventually', so please start changing your code now.
+For now `api` should continue to work, but it will be removed 'eventually', so please start changing your code now.
 
-Accordingly, `JSS::API` (which should never have been a constant to begin with) is also deprecated. To access the default connection, use `Jamf.cnx`
+Accordingly, `JSS::API` (which should never have been a constant to begin with) has been deprecated for years in favor of `JSS.api`, which is now also deprecated. To access the default connection, use `Jamf.cnx`
 
 ### `.map_all_ids_to` method for Classic API collection classes
 
@@ -217,7 +241,7 @@ For now `map_all_ids_to` still works, however it's just a wrapper for `map_all`.
 
 Use `.create` and `#save` instead, as with the Jamf Pro API objects
 
-In previous ruby-jss, both APIs avoided the use of the ruby-standard `.new` on Collection Resource classes, because the word 'new' in this context is ambiguous: are you creating a new instance of the class in ruby (which might already exist on the server), or are you creating a new object in Jamf Pro that doesn't yet exist on the server?
+All versions of ruby-jss have avoided the use of the ruby-standard `.new` on Collection Resource classes, because the word 'new' in this context is ambiguous: are you creating a new instance of the class in ruby (which might already exist on the server), or are you creating a new object in Jamf Pro that doesn't yet exist on the server?
 
 In v2.0.0 we are standardizing on the behavior of the previous Jamf module:
 
@@ -273,7 +297,7 @@ existing_policy.save
 
 ### JSS::CONFIG
 
-This also should never have been a constant.  Use Jamf.config.  JSS::CONFIG will go away eventually.
+This should never have been a constant.  Use Jamf.config.  JSS::CONFIG will go away eventually.
 
 ### Jamf::Connection instance methods `#next_refresh`, `#secs_to_refresh`, &  `#time_to_refresh`
 
@@ -307,7 +331,7 @@ You can still fetch random objects from a collection, but use `.fetch random: tr
 
 The YARD documentation for v2.0.0 is available before release in the [github.io site for ruby-jss](http://pixaranimationstudios.github.io/ruby-jss/docs/v2.x/index.html).
 
-All the documentation for ruby-jss is in need of updating, from the top-level README to the auto-generated YARD docs.  If you have questions that aren't answered there, please reach out (see Contact, below)
+All the documentation for ruby-jss is in serious need of updating, from the top-level README to the auto-generated YARD docs. Forgive us while we slowly get it up to snuff. If you have questions that aren't answered there, please reach out (see [Contact](#contact), below)
 
 ## How to install for testing