Umeshma/install test (#155)

In this PR
* Added a sample app to allow testing of nuget packages
* Added a  CodeOwners file
* Refactored IVocabCollection and IVocab into their own files
* Updated readmes for examples

.github/CODEOWNERS

@@ -0,0 +1,2 @@
+# @microsoft/TypeChat.Net-PR owns any files in this repo
+@microsoft/TypeChat.Net-PR

examples/CoffeeShop/ReadMe.md

@@ -1,18 +1,19 @@
 # CoffeeShop
-Demonstrates **JsonTranslator** with a [complex schema](CoffeeShopSchema.cs) for a coffee shop. It shows how GPT 3.5 and 4.0 can effectively be used to translate a broad range of user intent into strongly typed objects that can then be processed with conventional software.
+CoffeeShop demonstrates how to use **JsonTranslator** to create a natural language interface to take orders for a coffee shop. This requires a [complex schema](CoffeeShopSchema.cs). CoffeeShop shows how GPT 3.5 and 4.0 can effectively be used to translate a broad range of user intent into strongly typed objects that can then be processed with conventional software.
 
 CoffeeShop emulates natural language ordering at a Coffee Shop serving:
-- Coffee
-- Espresso
-- Lattes
-- Baked Good
+* Coffee
+* Espresso
+* Lattes
+* Baked Good
 
 A range of syrups and other familiar options are also offered. 
 
-Also demonstrates how to use the ***[JsonVocab]*** attribute to:
+CoffeeShop also demonstrates how to use the ***JsonVocab*** attribute to:
 * Specify string vocabularies like:
   ```[JsonVocab(whole milk | two percent milk | nonfat milk | soy milk | almond milk | oat milk)]```
-* Automatic validation and repair of strings using these vocabularies
+  * Vocabularies differ from enums because they can also be loaded on demand and/or be customized for each request (based on the request's context)
+* Automatic validation and repair of strings using these vocabularies.
 
 ## Target models
 Works with gpt-35-turbo and gpt-4.

examples/CoffeeShop2/Readme.md

@@ -1,10 +1,9 @@
 # CoffeeShop2
 
-CoffeeShop2 implements the same features as [CoffeeShop](../CoffeeShop/Readme.md), but with dynamic vocabularies - how to use IVocabCollection and IVocab. 
-- Vocab Strings are loaded from a file and bound dynamically. 
-- Your vocabularies could be loaded from a database or any other source. This lets you alter vocabularies WITHOUT recompiling your code.  
-- Each request could use a different translator and associated vocabulary (with associated schema generated on the fly). 
--- e.g. Premium Coffee Shop users could be offered more complex Syrups or Toppings. Vegan users would only see non-dairy milks. Etc.
+CoffeeShop2 implements the same features as [CoffeeShop](../CoffeeShop/Readme.md). But all vocabularies are dynamically loaded. CoffeeShop2 demonstrates how to use [IVocab](../../src/typechat/Schema/IVocab.cs) and [IVocabCollection](../../src/typechat/Schema/IVocabCollection.cs). 
+* Vocab Strings are loaded from a file and bound dynamically. The same vocabularies could also be loaded from a database or any other source. This lets you alter vocabularies without recompiling your code.
+* Each request could use a different translator and associated vocabulary. 
+  * E.g. Premium Coffee Shop users could be offered more complex Syrups or Toppings. Vegan users would only see non-dairy milks. Etc.
 
 ## Target models
 Works with gpt-35-turbo and gpt-4.

examples/HealthData/Readme.md

@@ -1,9 +1,9 @@
 # Health Data Agent
 This example requires GPT-4.
 
-Demonstrates a ***strongly typed*** Chat. Here, you work with an Agent to interactively enter medications or conditions. 
+Demonstrates a ***strongly typed*** Chat: a natural language interface for entering health information. You work with a Health Data Agent to interactively enter your medications or conditions. 
 
-Shows how to use **Agents (Chat) with history** to ***interactively*** acquire information needed for one or more Types from the user: **form filling**. 
+The Health Data Agent shows how strongly typed **Agents with history** could interact with a user to collect information needed for one or more data types ("form filling"). 
 
 ## Target models
 For best and consistent results, use **gpt-4**. 

examples/Math/Readme.md

@@ -1,6 +1,6 @@
 # Math
 
-The Math example shows how to use **TypeChat.Program** for program generation based on an API schema. This example translates calculations into simple programs given an [`API`](MathAPI.cs) type that can perform the four basic mathematical operations.
+The Math example shows how to use **TypeChat.Program** for program generation based on an API schema. This example translates calculations into simple programs given an [`API`](MathAPI.cs) type that can perform the basic mathematical operations.
 
 ## Target models
 Works with gpt-35-turbo and gpt-4.
@@ -9,7 +9,7 @@ Works with gpt-35-turbo and gpt-4.
 
 Example prompts can be found in [`input.txt`](input.txt).
 
-For example, we could use natural language to describe mathematical operations, and TypeChat.Program will generate a json program for the math API defined in the schema. It will then execute the program using a simple interpreter.
+For example, we could use natural language to describe mathematical operations, and TypeChat.Program will generate a json program for the math API defined in the schema. It will then execute the program.
 
 **Input**:
 

examples/Plugins/Readme.md

@@ -1,6 +1,8 @@
 # Plugins
 
-Demonstrates how to to generate programs (plans) that target Semantic Kernel and Open API plugins
+Demonstrates how to generate programs (plans) that target Semantic Kernel and Open API plugins. The Plugins example also shows how to:
+* Use a Program Interpreter to invoke plugins
+* Produce Schema from metadata that originates from outside of the .NET type system. Sources could include plugin manifests, Open API specs and other sources.
 
 For best and consistent results, use **GPT-4**. 
 

src/typechat/Schema/IVocab.cs

@@ -0,0 +1,21 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+namespace Microsoft.TypeChat.Schema;
+
+/// <summary>
+/// A Vocabulary is a string table whose values need not be hardcoded in code
+/// Example: lists of product names
+/// This means that each time a schema is generated, it can be different.
+///
+/// These string tables can be emitted in Typescript schema, e.g. like this:
+/// type FourSeasons: 'Winter' | 'Spring' | 'Summer' | 'Fall'
+///
+/// The string tables can also be automatically validated to ensure the model sends back know strings.
+/// Any error can be sent to the model for correction
+/// 
+/// </summary>
+public interface IVocab : IEnumerable<VocabEntry>
+{
+    bool Contains(VocabEntry entry);
+    bool Contains(VocabEntry entry, StringComparison comparison);
+}

src/typechat/Schema/IVocabCollection.cs

@@ -0,0 +1,13 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+namespace Microsoft.TypeChat.Schema;
+
+/// <summary>
+/// A collection of named vocabularies.
+/// Vocabulary collections can be dynamic and load vocabularies at runtime from data bases, files and so on
+/// </summary>
+public interface IVocabCollection : IEnumerable<NamedVocab>
+{
+    NamedVocab? Get(string name);
+}
+

src/typechat/Schema/Vocab.cs

@@ -4,24 +4,6 @@
 
 namespace Microsoft.TypeChat.Schema;
 
-/// <summary>
-/// A Vocabulary is a string table whose values need not be hardcoded in code
-/// Example: lists of product names
-/// This means that each time a schema is generated, it can be different.
-///
-/// These string tables can be emitted in Typescript schema, e.g. like this:
-/// type FourSeasons: 'Winter' | 'Spring' | 'Summer' | 'Fall'
-///
-/// The string tables can also be automatically validated to ensure the model sends back know strings.
-/// Any error can be sent to the model for correction
-/// 
-/// </summary>
-public interface IVocab : IEnumerable<VocabEntry>
-{
-    bool Contains(VocabEntry entry);
-    bool Contains(VocabEntry entry, StringComparison comparison);
-}
-
 /// <summary>
 /// An in-memory vocabulary.
 /// </summary>

src/typechat/Schema/VocabCollection.cs

@@ -4,15 +4,6 @@
 
 namespace Microsoft.TypeChat.Schema;
 
-/// <summary>
-/// A collection of named vocabularies.
-/// Vocabulary collections can be dynamic and load vocabularies at runtime from data bases, files and so on
-/// </summary>
-public interface IVocabCollection : IEnumerable<NamedVocab>
-{
-    NamedVocab? Get(string name);
-}
-
 /// <summary>
 /// An in-memory vocabulary collection
 /// </summary>

tests/EmojiApp/Emoji.csproj

@@ -0,0 +1,17 @@
+<Project Sdk="Microsoft.NET.Sdk">
+
+  <PropertyGroup>
+    <OutputType>Exe</OutputType>
+    <TargetFramework>net7.0</TargetFramework>
+    <ImplicitUsings>enable</ImplicitUsings>
+    <Nullable>enable</Nullable>
+  </PropertyGroup>
+
+  <ItemGroup>
+    <PackageReference Include="Microsoft.TypeChat" Version="0.1.230922.3-preview" />
+    <PackageReference Include="Microsoft.TypeChat.Dialog" Version="0.1.230922.3-preview" />
+    <PackageReference Include="Microsoft.TypeChat.Program" Version="0.1.230922.3-preview" />
+    <PackageReference Include="Microsoft.TypeChat.SemanticKernel" Version="0.1.230922.3-preview" />
+  </ItemGroup>
+
+</Project>

tests/EmojiApp/Emoji.sln

@@ -0,0 +1,25 @@
+
+Microsoft Visual Studio Solution File, Format Version 12.00
+# Visual Studio Version 17
+VisualStudioVersion = 17.7.34031.279
+MinimumVisualStudioVersion = 10.0.40219.1
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "Emoji", "Emoji.csproj", "{E05F478D-9736-4252-BEE1-8060E027C5A6}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Any CPU = Debug|Any CPU
+		Release|Any CPU = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{E05F478D-9736-4252-BEE1-8060E027C5A6}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{E05F478D-9736-4252-BEE1-8060E027C5A6}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{E05F478D-9736-4252-BEE1-8060E027C5A6}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{E05F478D-9736-4252-BEE1-8060E027C5A6}.Release|Any CPU.Build.0 = Release|Any CPU
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+	GlobalSection(ExtensibilityGlobals) = postSolution
+		SolutionGuid = {1709003B-ED8F-466F-9B59-E7D4FE42247A}
+	EndGlobalSection
+EndGlobal

tests/EmojiApp/EmojiProgram.cs

@@ -0,0 +1,95 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+//------------------------------------------------------
+//
+// A simple app that translates user requests into emojis
+// using all 3 options in TypeChat.
+// This application is used to test Package installs
+//
+//------------------------------------------------------
+
+using Microsoft.TypeChat;
+using Microsoft.TypeChat.Dialog;
+
+namespace Emoji;
+
+public class EmojiResponse
+{
+    public string EmojiTranslation { get; set; } = string.Empty;
+}
+
+public interface IEmojiService
+{
+    Task<string> ToEmoji(string userText);
+}
+
+internal class EmojiProgram : IEmojiService
+{
+    LanguageModel _model;
+    JsonTranslator<EmojiResponse> _translator;
+    public EmojiProgram()
+    {
+        _model = new LanguageModel(OpenAIConfig.FromEnvironment());
+        _translator = new JsonTranslator<EmojiResponse>(_model);
+    }
+
+    public async Task<string> ToEmoji(string line)
+    {
+        EmojiResponse response = await _translator.TranslateAsync(line);
+        return response.EmojiTranslation;
+    }
+
+    public async Task TestTranslate()
+    {
+        string? line;
+        Console.Write(">");
+
+        while ((line = Console.ReadLine()) != null)
+        {
+            string result = await ToEmoji(line);
+            Console.WriteLine(result);
+            Console.Write(">");
+        }
+    }
+
+    public async Task TestProgram()
+    {
+        string? line;
+        Console.Write(">");
+        Console.OutputEncoding = System.Text.Encoding.UTF8;
+
+        var api = new Api<IEmojiService>(this);
+        ProgramTranslator programGen = new ProgramTranslator<IEmojiService>(_model, api);
+        while ((line = Console.ReadLine()) != null)
+        {
+            using Program program = await programGen.TranslateAsync(line);
+            string result = await program.RunAsync(api);
+            Console.WriteLine(result);
+            Console.Write(">");
+        }
+    }
+
+    public async Task TestAgent()
+    {
+        Agent<EmojiResponse> agent = new AgentWithHistory<EmojiResponse>(_model);
+        agent.Instructions.Append("Respond to me in emojis");
+        string? line;
+        Console.Write(">");
+        while ((line = Console.ReadLine()) != null)
+        {
+            var response = await agent.GetResponseAsync(line);
+            Console.WriteLine(response.EmojiTranslation);
+            Console.Write(">");
+        }
+    }
+
+    public static async Task<int> Main(string[] args)
+    {
+        Console.OutputEncoding = System.Text.Encoding.UTF8;
+        EmojiProgram p = new EmojiProgram();
+        await p.TestTranslate();
+        //await p.TestProgram();
+        //await p.TestAgent();
+        return 0;
+    }
+}

tests/EmojiApp/addPackages.bat

@@ -0,0 +1,4 @@
+dotnet add package Microsoft.TypeChat --prerelease
+dotnet add package Microsoft.TypeChat.SemanticKernel --prerelease
+dotnet add package Microsoft.TypeChat.Program --prerelease
+dotnet add package Microsoft.TypeChat.Dialog --prerelease