Add getting-started module: ChatClient fluent API, system/user prompts, streaming, and provider switching via spring.ai.model.chat

getting-started/scripts/run-all.sh

@@ -0,0 +1,13 @@
+#!/usr/bin/env bash
+# Regenerates every reproducible file under output/. Run from the module root
+# (spring-ai/getting-started). Needs Maven with the dependencies already resolved once online.
+#
+# output/05-*.txt is NOT regenerated here -- it needs a live network path to api.openai.com and a
+# running instance of the app started separately. See its own header for the exact commands.
+set -eu
+cd "$(dirname "$0")/.."
+
+mvn -q -o test
+echo
+echo "Regenerated output/01-plain-call.txt, 02-system-template.txt, 03-streaming.txt, 04-provider-switch.txt"
+echo "output/05-real-network-round-trip.txt was captured manually -- see its header."

getting-started/scripts/run.sh

@@ -0,0 +1,19 @@
+#!/usr/bin/env bash
+# Starts the application, killing any previous instance first.
+#
+# Usage:
+#   OPENAI_API_KEY=sk-... ./scripts/run.sh
+#   CHAT_PROVIDER=ollama OLLAMA_BASE_URL=http://localhost:11434 ./scripts/run.sh
+#
+# Then, in another shell:
+#   curl 'http://localhost:8080/api/chat?message=What+is+a+ChatClient?'
+#   curl 'http://localhost:8080/api/chat/as?voice=pirate&message=Where+is+my+jar+cached'
+#   curl -N 'http://localhost:8080/api/chat/stream?message=Stream+this'
+set -eu
+cd "$(dirname "$0")/.."
+
+for p in $(ps -eo pid,cmd | grep '[G]ettingStartedApplication' | awk '{print $1}'); do
+  kill -9 "$p"
+done
+
+mvn -q -o org.springframework.boot:spring-boot-maven-plugin:run