Release version 0.9.2

CHANGELOG.md

@@ -0,0 +1,32 @@
+# Changelog
+All notable changes to this project since its first released version (0.9.0) are documented in this file.
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/).
+
+## [1.0.0] – Not yet released
+
+### Added
+- Form API for HTML5: set up forms from CRUD action metadata
+
+## [0.9.2] – 2023-12-27
+
+### Fixed
+- Connection to URLs ending with slash
+- RxJS integration
+
+## [0.9.1] – 2023-07-28
+
+### Added
+- Installation instructions
+
+### Changed
+- Clearer error messages
+
+### Fixed
+- Authentication loop
+- Readme file contents
+
+### Removed
+- The Angular UI is no longer part of Portofino. It is now a separate project.
+    - Given the lack of built-in UI, several new helper projects have been started to help build Portofino-based
+      applications, including a command-line tool and libraries to integrate Portofino into web applications. Please
+      refer to [the website](https://manydesigns.github.io/Portofino/).

README.md

@@ -1,20 +1,21 @@
 [![npm version](https://badge.fury.io/js/portofino-commander.svg)](https://badge.fury.io/js/portofino-commander)
 
-# portofino-commander
+# Portofino-commander
 
-A JavaScript library to query, inspect and modify a [Portofino](https://github.com/ManyDesigns/Portofino) application.
+A JavaScript library to query, inspect and modify a [Portofino](https://github.com/ManyDesigns/Portofino) service. We can use this to build an application
+on top of a Portofino backend.
 
-It has few dependencies:
-- The standard `fetch` API
-- RxJS
-- jwt-decode
-- i18next
+Portofino-commander has few dependencies:
+- The standard `fetch` API, for HTTP requests;
+- RxJS, for reactive APIs;
+- jwt-decode, to handle authentication;
+- i18next, for internationalization.
 
-These all work both on the browser and in Node, therefore Portofino-commander runs in both environments.
+These all work on the browser as well as in Node, therefore Portofino-commander runs in both environments.
 
 ## Installation
 
-portofino-commander is available from the NPM Registry so we can use tools like NPM or Yarn to install it:
+Portofino-commander is available from the NPM Registry, so we can use tools like NPM or Yarn to install it:
 
 ```
 npm install --save portofino-commander
@@ -30,7 +31,7 @@ For convenience, we also provide a minified bundle of Portofino-commander and it
 so you can include it like so:
 
 ```html
-<script src="./static/bundle/portofino-commander-0.9.1-bundle.js" charset="UTF-8" defer></script>
+<script src="./static/bundle/portofino-commander-0.9.2-bundle.js" charset="UTF-8" defer></script>
 ```
 
 You can find the bundle in the [Releases page](https://github.com/alessiostalla/portofino-commander/releases).
@@ -54,7 +55,12 @@ const portofino = Portofino.connect(
 ```
 
 The above applies to Portofino 5 and 6 services (based on Spring Boot).
-Of course, a service could have a non-standard configuration, and its address could be different.
+
+**Note** that, in a production or staging application, the service is going to be exposed with a public or internal
+URL, most probably behind a reverse proxy. So, the actual URL will vary accordingly to the server's configuration.
+
+#### Non-standard Service URL
+Even on a development machine, a service could have a non-standard configuration, and its address could be different.
 
 If unsure, check the log messages emitted while the service starts up and look for lines like the following:
 
@@ -82,16 +88,19 @@ if the application is deployed at the root (e.g., on Tomcat, if the .war file is
 
 `/api` is the REST API root, because if we access `http://localhost:8080/demo-tt` we'll receive the HTML of the 
 application's home page. This path can be customized in the application's `web.xml` file, but `/api` is the
-default and it rarely gets changed.
+default, and it rarely gets changed.
+
+Again, this only applies to local usage in a development environment. A deployed .war application will have a different
+URL, that depends on the configuration of the server. 
 
 ### Making Requests
 
 Once we've got a connection, we can make requests to the server. We do this conceptually in two steps:
-- First, we obtain a _resource_ (e.g. a CRUD `ResourceAction`) – this is a _class_ of the application;
+- First, we obtain a _resource_ (e.g. a CRUD `ResourceAction`) – this is a _class_ in the application;
 - Then, we invoke an _operation_ on the resource (e.g. "save" on an CRUD) – this is a _method_ of the resource.
 
-Portofino-commander implements access to a resource and invocation of an operation with an RxJS Observable. 
-These are proxied so that we don't have to use the RxJS APIs (such as `pipe` or `subscribe`) to chain them.
+Portofino-commander implements both access to a resource and invocation of an operation with an RxJS Observable. 
+Observables are proxied so that we don't have to use the RxJS APIs (such as `pipe` or `subscribe`) to chain them.
 Some examples follow.
 
 Let's first define an _observer_ that will just print the result of an operation to the console:
@@ -149,7 +158,7 @@ portofino.upstairs.get("database/tables")
     .subscribe(observer);
 ```
 
-**Note:** on Portofino 5, we have to replace `.load()` with `.op_get()` (see below).
+**Note:** when connecting to a Portofino 5 service or application, we have to replace `.load()` with `.op_get()` (see below).
 
 When we're done, we should terminate the session:
 
@@ -158,7 +167,7 @@ portofino.logout().subscribe();
 ```
 
 Note how calls to `subscribe` are needed to actually perform the HTTP requests to the backend. This is because RxJS is
-"lazy" and doesn't run any code until someone looks at the results.
+"lazy" and doesn't run any code until someone looks at the results (i.e. subscribes to the observable).
 
 Portofino-commander automatically discovers operations such as _load()_ above for CRUD, or _getTablesInSchema(db, schema)_,
 by querying the Portofino service. That's why `resource.get(subresource)` returns an _Observable_. The resource is only
@@ -223,7 +232,7 @@ invoking operations on the server (such as _load_ in the previous example) is an
 
 ## Compatibility
 
-portofino-commander is developed and tested against Portofino 6.
+Portofino-commander is developed and tested against Portofino 6.
 
 While portofino-commander's general approach works perfectly well with Portofino 5, some REST APIs in P5 weren't 
 designed with such a client in mind, and require some extra handling to invoke them.
@@ -261,7 +270,7 @@ Portofino-commander is licensed under the GNU AGPL. In layman terms, if you buil
 it, you need to release its source code.
 If you'd like to use portofino-commander as a component in a tool or service and require a more business-friendly license,
 please open an issue or contact me directly. I'm open to licensing this to specific organizations so that they can use it,
-but I prefer to have a say in that.
+even free of charge, but I prefer to have a say in that.
 
 ## Donations
 

package.json

@@ -1,7 +1,7 @@
 {
   "name": "portofino-commander",
   "author": "Alessio Stalla",
-  "version": "1.0.0",
+  "version": "0.9.2",
   "license": "AGPL-3.0-or-later",
   "repository": {
     "url": "https://github.com/alessiostalla/portofino-commander/"

src/portofino.ts

@@ -98,7 +98,7 @@ export class ResourceAction {
 
     get<T extends ResourceAction>(
         segment: string,
-        type: new (...args) => T = ResourceAction as any,
+        type: new (...args: any[]) => T = ResourceAction as any,
         setup: (o: Observable<T>) => Observable<T> = o => o) {
         if (segment.startsWith("/")) {
             return this.root.get(segment);