참고소스 수정본

참고/instructor-main/tests/llm/test_bedrock/conftest.py

@@ -0,0 +1,29 @@
+from __future__ import annotations
+import base64
+import pytest
+
+
+@pytest.fixture(scope="session")
+def tiny_png_bytes() -> bytes:
+    return base64.b64decode(
+        b"iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR4nGNgYAAAAAMA"
+        b"ASsJTYQAAAAASUVORK5CYII="
+    )
+
+
+@pytest.fixture(scope="session")
+def tiny_png_data_url(tiny_png_bytes: bytes) -> str:
+    return "data:image/png;base64," + base64.b64encode(tiny_png_bytes).decode("utf-8")
+
+
+@pytest.fixture(scope="session")
+def image_url() -> str:
+    # Public test asset used across the suite
+    return "https://raw.githubusercontent.com/instructor-ai/instructor/main/tests/assets/image.jpg"
+
+
+@pytest.fixture(scope="session")
+def tiny_pdf_bytes() -> bytes:
+    return base64.b64decode(
+        b"JVBERi0xLjQKJSVPRgoAAAAQAAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA"
+    )

참고/instructor-main/tests/llm/test_bedrock/test_bedrock_native_passthrough.py

@@ -0,0 +1,20 @@
+from __future__ import annotations
+from instructor.providers.bedrock.utils import _to_bedrock_content_items
+
+
+def test_bedrock_native_text_passthrough():
+    content = [{"text": "Bedrock-native text"}]
+    items = _to_bedrock_content_items(content)
+    assert items == [{"text": "Bedrock-native text"}]
+
+
+def test_bedrock_native_image_passthrough(tiny_png_bytes: bytes):
+    native = {"image": {"format": "png", "source": {"bytes": tiny_png_bytes}}}
+    items = _to_bedrock_content_items([native])
+    assert items[0] == native
+
+
+def test_bedrock_native_document_passthrough(tiny_pdf_bytes: bytes):
+    native = {"document": {"format": "pdf", "source": {"bytes": tiny_pdf_bytes}}}
+    items = _to_bedrock_content_items([native])
+    assert items[0] == native

참고/instructor-main/tests/llm/test_bedrock/test_normalize.py

@@ -0,0 +1,26 @@
+from __future__ import annotations
+import pytest
+from instructor.providers.bedrock.utils import _normalize_bedrock_image_format
+
+
+@pytest.mark.parametrize(
+    "inp,expected",
+    [
+        ("image/jpeg", "jpeg"),
+        ("image/jpg", "jpeg"),
+        ("jpg", "jpeg"),
+        ("jpeg", "jpeg"),
+        ("image/pjpeg", "jpeg"),
+        ("image/png", "png"),
+        ("png", "png"),
+        ("image/gif", "gif"),
+        ("gif", "gif"),
+        ("image/webp", "webp"),
+        ("webp", "webp"),
+        ("", "jpeg"),
+        (None, "jpeg"),
+        ("image/whatever", "jpeg"),
+    ],
+)
+def test_normalize_bedrock_image_format(inp, expected):
+    assert _normalize_bedrock_image_format(inp) == expected

참고/instructor-main/tests/llm/test_bedrock/test_openai_image_conversion.py

@@ -0,0 +1,52 @@
+from __future__ import annotations
+import base64
+import pytest
+from instructor.providers.bedrock.utils import (
+    _openai_image_part_to_bedrock,
+    _to_bedrock_content_items,
+)
+
+
+def test_openai_image_part_to_bedrock_data_url(tiny_png_data_url: str):
+    part = {"type": "image_url", "image_url": {"url": tiny_png_data_url}}
+    out = _openai_image_part_to_bedrock(part)
+    assert "image" in out
+    assert out["image"]["format"] in {"png", "jpeg", "gif", "webp"}  # png expected
+    assert out["image"]["source"]["bytes"] == base64.b64decode(
+        tiny_png_data_url.split(",", 1)[1]
+    )
+
+
+def test_openai_image_part_to_bedrock_https(image_url: str):
+    part = {"type": "image_url", "image_url": {"url": image_url}}
+    out = _openai_image_part_to_bedrock(part)
+    assert "image" in out
+    # GitHub raw returns jpeg for the sample. Normalize is handled in utils.
+    assert out["image"]["format"] in {"jpeg", "png", "gif", "webp"}
+    assert isinstance(out["image"]["source"]["bytes"], (bytes, bytearray))
+    assert len(out["image"]["source"]["bytes"]) > 0
+
+
+@pytest.mark.parametrize(
+    "text_part",
+    [
+        {"type": "text", "text": "What is in this image?"},
+        {"type": "input_text", "text": "Describe the image."},
+    ],
+)
+@pytest.mark.parametrize("image_kind", ["data", "https"])
+def test_to_bedrock_content_items_openai_combo(
+    text_part, image_kind, tiny_png_data_url: str, image_url: str
+):
+    if image_kind == "data":
+        image_part = {"type": "image_url", "image_url": {"url": tiny_png_data_url}}
+    else:
+        image_part = {"type": "image_url", "image_url": {"url": image_url}}
+
+    content = [text_part, image_part]
+    items = _to_bedrock_content_items(content)
+
+    assert items[0] == {"text": text_part["text"]}
+    assert "image" in items[1]
+    assert isinstance(items[1]["image"]["source"]["bytes"], (bytes, bytearray))
+    assert len(items[1]["image"]["source"]["bytes"]) > 0

참고/instructor-main/tests/llm/test_bedrock/test_prepare_kwargs.py

@@ -0,0 +1,59 @@
+from __future__ import annotations
+import pytest
+from instructor.providers.bedrock.utils import _prepare_bedrock_converse_kwargs_internal
+
+
+def test_prepare_bedrock_kwargs_openai_text_plus_image(tiny_png_data_url: str):
+    call_kwargs = {
+        "model": "anthropic.claude-3-5-sonnet",
+        "temperature": 0.3,
+        "max_tokens": 256,
+        "top_p": 0.9,
+        "stop": ["<END>"],
+        "system": [{"text": "You are helpful."}],
+        "messages": [
+            {
+                "role": "user",
+                "content": [
+                    {"type": "text", "text": "hi"},
+                    {"type": "image_url", "image_url": {"url": tiny_png_data_url}},
+                ],
+            },
+        ],
+    }
+
+    out = _prepare_bedrock_converse_kwargs_internal(call_kwargs)
+
+    assert out["modelId"] == "anthropic.claude-3-5-sonnet"
+    inf = out["inferenceConfig"]
+    assert inf["temperature"] == 0.3
+    assert inf["maxTokens"] == 256
+    assert inf["topP"] == 0.9
+    assert inf["stopSequences"] == ["<END>"]
+    assert out["system"][0]["text"] == "You are helpful."
+
+    parts = out["messages"][0]["content"]
+    assert parts[0] == {"text": "hi"}
+    assert parts[1]["image"]["format"] in {"jpeg", "png", "gif", "webp"}
+    assert isinstance(parts[1]["image"]["source"]["bytes"], (bytes, bytearray))
+    assert len(parts[1]["image"]["source"]["bytes"]) > 0
+
+
+def test_prepare_bedrock_kwargs_openai_image_url_rejects_http():
+    """Remote HTTP(S) URLs are not fetched (SSRF prevention) — must use data: URLs."""
+    call_kwargs = {
+        "model": "anthropic.claude-3-5-sonnet",
+        "messages": [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "image_url",
+                        "image_url": {"url": "https://example.com/image.jpg"},
+                    }
+                ],
+            }
+        ],
+    }
+    with pytest.raises(ValueError, match="Unsupported image_url scheme for Bedrock"):
+        _prepare_bedrock_converse_kwargs_internal(call_kwargs)