Skip to content

Commit 6eb7b50

Browse files
alicejgibbonsalicejgibbons
authored
Merge pull request #1276 from siri-varma/users/svegiraju/java-http-ex
HTTP Java Conversation API quickstart
2 parents 54c5970 + fd1b8df commit 6eb7b50

File tree

5 files changed

+482
-0
lines changed

5 files changed

+482
-0
lines changed

‎conversation/java/http/README.md‎

Lines changed: 197 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,197 @@
1+
# Dapr Conversation API (Java HTTP)
2+
3+
In this quickstart, you'll send an input to a mock Large Language Model (LLM) using Dapr's Conversation API. This API is responsible for providing one consistent API entry point to talk to underlying LLM providers.
4+
5+
Visit [this](https://docs.dapr.io/developing-applications/building-blocks/conversation/conversation-overview/) link for more information about Dapr and the Conversation API.
6+
7+
> **Note:** This example leverages native HTTP client requests only. If you are looking for the example using the Dapr Client SDK (recommended) [click here](../sdk/).
8+
9+
This quickstart includes one app:
10+
11+
- `ConversationApplication.java`, responsible for sending an input to the underlying LLM and retrieving an output.
12+
13+
## Features Demonstrated
14+
15+
This quickstart demonstrates:
16+
17+
1. **Basic Conversation** - Send a simple message to an LLM and receive a response using the alpha2 API
18+
2. **Tool Calling** - Define tools/functions that the LLM can invoke, following OpenAI's function calling format
19+
20+
### Tool Calling
21+
22+
The conversation API supports advanced tool calling capabilities that allow LLMs to interact with external functions and APIs. This enables you to build sophisticated AI applications that can:
23+
24+
- Execute custom functions based on user requests
25+
- Integrate with external services and databases
26+
- Provide dynamic, context-aware responses
27+
28+
Tool calling follows [OpenAI's function calling format](https://developers.openai.com/api/docs/guides/function-calling), making it easy to integrate with existing AI development workflows.
29+
30+
## Pre-requisites
31+
32+
* [Dapr and Dapr Cli](https://docs.dapr.io/getting-started/install-dapr-cli/).
33+
* Java JDK 17 (or greater):
34+
* [Microsoft JDK 17](https://learn.microsoft.com/en-us/java/openjdk/download#openjdk-17)
35+
* [Oracle JDK 17](https://www.oracle.com/java/technologies/downloads/?er=221886#java17)
36+
* [OpenJDK 17](https://jdk.java.net/17/)
37+
* [Apache Maven](https://maven.apache.org/install.html) version 3.x.
38+
39+
## Run the app with the template file
40+
41+
Run the application using the [multi-app run template files](https://docs.dapr.io/developing-applications/local-development/multi-app-dapr-run/multi-app-overview/) and the Dapr CLI with dapr run -f ..
42+
43+
This example uses the default LLM Component provided by Dapr which simply echoes the input provided, for testing purposes. Here are other [supported Conversation components](https://docs.dapr.io/reference/components-reference/supported-conversation/).
44+
45+
### Build and run the Java application
46+
47+
1. Navigate to the `conversation` directory and build the Java application:
48+
49+
<!-- STEP
50+
name: Build Java file
51+
-->
52+
53+
```bash
54+
cd ./conversation
55+
mvn clean install
56+
cd ..
57+
```
58+
59+
<!-- END_STEP -->
60+
61+
2. Run the Java service app with Dapr using the multi app run template:
62+
63+
<!-- STEP
64+
name: Run multi app run template
65+
expected_stdout_lines:
66+
- '== APP - conversation == === Basic Conversation Example ==='
67+
- '== APP - conversation == Input sent: What is dapr?'
68+
- '== APP - conversation == Output response: What is dapr?'
69+
- '== APP - conversation =='
70+
- '== APP - conversation == === Tool Calling Example ==='
71+
- '== APP - conversation == Input sent: What is the weather like in San Francisco?'
72+
- '== APP - conversation == Tools defined: get_weather (location, unit)'
73+
- '== APP - conversation == LLM requested tool calls:'
74+
- '== APP - conversation == Tool ID: 0'
75+
- '== APP - conversation == Function: get_weather'
76+
- '== APP - conversation == Arguments: location,unit'
77+
- '== APP - conversation == Tool Result: {"temperature": 65, "unit": "fahrenheit", "description": "Sunny"}'
78+
- '== APP - conversation == '
79+
- '== APP - conversation == Note: The echo component echoes input for testing purposes.'
80+
- '== APP - conversation == For actual tool calling, configure a real LLM component like OpenAI.'
81+
expected_stderr_lines:
82+
output_match_mode: substring
83+
match_order: none
84+
background: true
85+
sleep: 15
86+
timeout_seconds: 30
87+
-->
88+
89+
```bash
90+
dapr run -f .
91+
```
92+
93+
The terminal console output should look similar to this:
94+
95+
```text
96+
== APP - conversation == === Basic Conversation Example ===
97+
== APP - conversation == Input sent: What is dapr?
98+
== APP - conversation == Output response: What is dapr?
99+
== APP - conversation ==
100+
== APP - conversation == === Tool Calling Example ===
101+
== APP - conversation == Input sent: What is the weather like in San Francisco?
102+
== APP - conversation == Tools defined: get_weather (location, unit)
103+
== APP - conversation == LLM requested tool calls:
104+
== APP - conversation == Tool ID: 0
105+
== APP - conversation == Function: get_weather
106+
== APP - conversation == Arguments: location,unit
107+
== APP - conversation == Tool Result: {"temperature": 65, "unit": "fahrenheit", "description": "Sunny"}
108+
== APP - conversation ==
109+
== APP - conversation == Note: The echo component echoes input for testing purposes.
110+
== APP - conversation == For actual tool calling, configure a real LLM component like OpenAI.
111+
```
112+
113+
<!-- END_STEP -->
114+
115+
3. Stop and clean up application processes.
116+
117+
<!-- STEP
118+
name: Stop multi-app run
119+
sleep: 5
120+
-->
121+
122+
```bash
123+
dapr stop -f .
124+
```
125+
126+
<!-- END_STEP -->
127+
128+
## Run the app with the Dapr CLI
129+
130+
1. Navigate to the `conversation` directory and build the Java application:
131+
132+
```bash
133+
cd ./conversation
134+
mvn clean install
135+
```
136+
137+
2. Run the application with Dapr:
138+
139+
```bash
140+
dapr run --app-id conversation --resources-path ../../../components -- java -jar target/ConversationService-0.0.1-SNAPSHOT.jar
141+
```
142+
143+
You should see the output:
144+
145+
```bash
146+
== APP - conversation == === Basic Conversation Example ===
147+
== APP - conversation == Input sent: What is dapr?
148+
== APP - conversation == Output response: What is dapr?
149+
== APP - conversation ==
150+
== APP - conversation == === Tool Calling Example ===
151+
== APP - conversation == Input sent: What is the weather like in San Francisco?
152+
== APP - conversation == Tools defined: get_weather (location, unit)
153+
== APP - conversation == LLM requested tool calls:
154+
== APP - conversation == Tool ID: 0
155+
== APP - conversation == Function: get_weather
156+
== APP - conversation == Arguments: location,unit
157+
== APP - conversation == Tool Result: {"temperature": 65, "unit": "fahrenheit", "description": "Sunny"}
158+
== APP - conversation ==
159+
== APP - conversation == Note: The echo component echoes input for testing purposes.
160+
== APP - conversation == For actual tool calling, configure a real LLM component like OpenAI.
161+
```
162+
163+
3. Stop the application:
164+
165+
```bash
166+
dapr stop --app-id conversation
167+
```
168+
169+
## Tool Calling with a Real LLM
170+
171+
To use actual tool calling functionality, configure a real LLM component (e.g., OpenAI, Anthropic). Here's an example OpenAI component configuration:
172+
173+
```yaml
174+
apiVersion: dapr.io/v1alpha1
175+
kind: Component
176+
metadata:
177+
name: openai
178+
spec:
179+
type: conversation.openai
180+
version: v1
181+
metadata:
182+
- name: key
183+
value: "<your-openai-api-key>"
184+
- name: model
185+
value: "gpt-4"
186+
```
187+
188+
Then update `CONVERSATION_COMPONENT_NAME` in the application to use your configured component.
189+
190+
When using a real LLM with tool calling:
191+
192+
1. The LLM analyzes the user request and available tools
193+
2. If a tool is needed, the response includes `finishReason: "tool_calls"` with tool call details
194+
3. Your application executes the tool and gets results
195+
4. Send results back to the LLM using an `ofTool` message to continue the conversation
196+
197+
For more details, see the [Conversation API reference](https://docs.dapr.io/reference/api/conversation_api/).

‎conversation/java/http/conversation/pom.xml‎

Lines changed: 44 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,44 @@
1+
<?xml version="1.0" encoding="UTF-8"?>
2+
<project xmlns="http://maven.apache.org/POM/4.0.0"
3+
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
4+
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
5+
<modelVersion>4.0.0</modelVersion>
6+
7+
<groupId>com.service</groupId>
8+
<artifactId>ConversationService</artifactId>
9+
<version>0.0.1-SNAPSHOT</version>
10+
<name>ConversationService</name>
11+
<description>Demo for Dapr Conversation API</description>
12+
<properties>
13+
<maven.compiler.source>17</maven.compiler.source>
14+
<maven.compiler.target>17</maven.compiler.target>
15+
</properties>
16+
<dependencies>
17+
<dependency>
18+
<groupId>com.fasterxml.jackson.core</groupId>
19+
<artifactId>jackson-databind</artifactId>
20+
<version>2.17.1</version>
21+
</dependency>
22+
</dependencies>
23+
<build>
24+
<plugins>
25+
<plugin>
26+
<groupId>org.springframework.boot</groupId>
27+
<artifactId>spring-boot-maven-plugin</artifactId>
28+
<version>3.3.1</version>
29+
<executions>
30+
<execution>
31+
<goals>
32+
<goal>repackage</goal>
33+
</goals>
34+
<configuration>
35+
<mainClass>
36+
com.service.ConversationApplication
37+
</mainClass>
38+
</configuration>
39+
</execution>
40+
</executions>
41+
</plugin>
42+
</plugins>
43+
</build>
44+
</project>

0 commit comments

Comments
 (0)