asmhatre/spring-ai-2-migration
1
0
Fork 0
Code Activity

Spring AI 1.x to 2.0 migration: breaking-change reference and offline verification suite

Browse Source 
Companion code for the ankurm.com guide. Verified on Spring AI 2.0.0, Spring Boot 4.1.0,
JDK 25.0.3. run.txt is unedited mvn test output: 5 tests, 0 failures.

Runs with NO API key, no network and no provider account. A canned ChatModel returns fixed
responses; ChatClient, advisors and chat memory all sit above the model, so everything that
changed is still exercised faithfully. TestChatModel records each Prompt it receives, which
lets the tests assert what Spring AI SENT rather than what a model replied.

docs/01  complete breaking-change reference: platform (Boot 4 mandatory, Jackson 2 to 3),
         artifact renames (spring-ai-advisors-vector-store -> spring-ai-vector-store-advisor,
         spring-ai-core split, OCI and Minimax removals, MCP transports moved into Spring AI),
         options builders replacing setters, the dropped .options property prefix, tool
         calling (internalToolExecutionEnabled and toolNames removed outright, not renamed),
         chat memory, structured output schema changes, and a migration order that works.

Tested here
  - Chat memory becomes stricter: 2.0 removes the remaining default-ID and builder-based
    configuration paths (ChatMemory.DEFAULT_CONVERSATION_ID and .conversationId() are both
    gone). Explicit per-request conversation ids - introduced during the 1.x line and already
    the recommended approach - become the only migration target.
  - CONVERSATION_ID survives as the metadata KEY. It and the removed DEFAULT_CONVERSATION_ID
    look interchangeable and are opposites; swapping them to clear a compile error restores
    exactly the shared-conversation behaviour that removing it was meant to take away.
  - PromptChatMemoryAdvisor is gone; MessageChatMemoryAdvisor replaces it, and the difference
    is not cosmetic (system-prompt text vs real Message objects, so token counts change).
  - A test asserts two conversation ids cannot see each other.
This commit is contained in:
Ankur
2026-08-01 10:44:31 +05:30
commit cb777068a7
8 changed files with 606 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

70
README.md Normal file
View File

@@ -0,0 +1,70 @@
# Spring AI 1.x → 2.0 migration
Companion repository for **[Spring AI 1.x to 2.0: The Migration Guide](https://ankurm.com/spring-ai-1-to-2-migration-guide/)**
on [ankurm.com](https://ankurm.com).
Verified against **Spring AI 2.0.0**, **Spring Boot 4.1.0**, **JDK 25.0.3**. Full output in
[`run.txt`](run.txt).
---
## Quick start
**No API key, no network, no provider account.** A canned `ChatModel` returns fixed responses, which
is enough to exercise every API that changed — `ChatClient`, advisors and chat memory all sit
*above* the model.
```console
git clone https://ankurm.com/git.app/asmhatre/spring-ai-2-migration.git
cd spring-ai-2-migration
mvn test
```
5 tests, under a second.
---
## What is here
| | |
|---|---|
| [Breaking-change reference](docs/01-breaking-changes.md) | The complete list, organised by area, with the migration order that works |
| [`ChatMemoryMigrationTest`](src/test/java/com/ankurm/ai/ChatMemoryMigrationTest.java) | The chat-memory changes, asserted — including a test proving two conversations cannot see each other |
| [`TestChatModel`](src/main/java/com/ankurm/ai/TestChatModel.java) | The offline model. Records every `Prompt` it receives, so tests assert **what Spring AI sent**, not what a model replied |
---
## The five things most likely to bite
1. **Boot 4 is mandatory.** Spring AI 2.0 is built on the Boot 4 dependency model and *cannot* load
in a Boot 3 context. Migrate Boot first, as a separate step.
2. **Jackson 2 → 3** (`com.fasterxml.jackson` → `tools.jackson`), which changes default date format
and property order. This is the change least likely to be caught by a compiler and most likely to
break a downstream consumer.
3. **`internalToolExecutionEnabled` is removed** — not renamed. Per-model internal tool execution no
longer exists in any `ChatModel`, so if you relied on the model driving the tool loop, that
orchestration moves up into the advisor layer.
4. **Chat memory gets stricter.** 2.0 removes the remaining default-ID and builder-based
configuration paths — `ChatMemory.DEFAULT_CONVERSATION_ID` and `.conversationId()` are both
gone. Explicit per-request conversation IDs, introduced during the 1.x line, become the only
safe migration target. Leaning on the old default put every caller in one shared conversation.
5. **Structured-output schema changes are silent.** Optional Kotlin properties and
`@JsonProperty` without an explicit `required` are no longer in the JSON Schema `required` array,
which changes what the model returns without producing any error.
---
## A trap worth its own line
`ChatMemory.CONVERSATION_ID` still exists. `ChatMemory.DEFAULT_CONVERSATION_ID` does not.
They look like the same thing and are opposites: the survivor is the **metadata key** you use to
pass an id, the removed one was the **value** `"default"` used when you passed nothing. Swapping one
for the other to clear a compile error restores exactly the shared-conversation behaviour that
removing it was meant to take away.
---
## Licence
MIT. See [LICENSE](LICENSE).