v2.16.0 (#798)

Co-authored-by: Hdandria <henri.dandria@mistral.ai>
Co-authored-by: Liam Lyons <65613603+lyons-liam@users.noreply.github.com>
Co-authored-by: Mathias Gesbert <mathias.gesbert@mistral.ai>
Co-authored-by: Pierre Rossinès <pierre.rossines@mistral.ai>
Co-authored-by: Quentin <quentin.torroba@mistral.ai>
Co-authored-by: allansimon-mistral <allan.simon@ext.mistral.ai>
Co-authored-by: angelapopopo <angele.lenglemetz@mistral.ai>
Co-authored-by: Mistral Vibe <vibe@mistral.ai>

tests/core/test_sse.py

@@ -0,0 +1,75 @@
+from __future__ import annotations
+
+from collections.abc import AsyncIterator
+
+import httpx
+import pytest
+
+from vibe.core.utils.sse import iter_sse_lines
+
+
+class _ChunkedStream(httpx.AsyncByteStream):
+    def __init__(self, chunks: list[bytes]) -> None:
+        self._chunks = chunks
+
+    async def __aiter__(self) -> AsyncIterator[bytes]:
+        for chunk in self._chunks:
+            yield chunk
+
+
+async def _collect(chunks: list[bytes]) -> list[str]:
+    response = httpx.Response(
+        status_code=200,
+        stream=_ChunkedStream(chunks),
+        request=httpx.Request("POST", "https://example.com"),
+    )
+    return [line async for line in iter_sse_lines(response)]
+
+
+class TestIterSseLines:
+    @pytest.mark.asyncio
+    async def test_splits_on_lf_and_crlf(self) -> None:
+        lines = await _collect([b"data: a\r\ndata: b\n\ndata: c\n"])
+        assert lines == ["data: a", "data: b", "", "data: c"]
+
+    @pytest.mark.asyncio
+    @pytest.mark.parametrize("separator", ["\u2028", "\u2029", "\x85", "\x0b", "\x0c"])
+    async def test_keeps_unicode_line_breaks_within_line(self, separator: str) -> None:
+        payload = f'data: {{"arguments": "{separator}value"}}\n'
+        lines = await _collect([payload.encode()])
+        assert lines == [payload.removesuffix("\n")]
+
+    @pytest.mark.asyncio
+    async def test_buffers_partial_lines_across_chunks(self) -> None:
+        lines = await _collect([b"data: one", b"two\ndata: three", b"four\n"])
+        assert lines == ["data: onetwo", "data: threefour"]
+
+    @pytest.mark.asyncio
+    async def test_multibyte_char_split_across_chunks(self) -> None:
+        encoded = "data: café\n".encode()
+        lines = await _collect([encoded[:9], encoded[9:]])
+        assert lines == ["data: café"]
+
+    @pytest.mark.asyncio
+    async def test_yields_trailing_line_without_newline(self) -> None:
+        lines = await _collect([b"data: a\ndata: b"])
+        assert lines == ["data: a", "data: b"]
+
+    @pytest.mark.asyncio
+    async def test_crlf_split_across_chunks(self) -> None:
+        lines = await _collect([b"data: a\r", b"\ndata: b\n"])
+        assert lines == ["data: a", "data: b"]
+
+    @pytest.mark.asyncio
+    async def test_splits_on_lone_cr(self) -> None:
+        lines = await _collect([b"data: a\rdata: b\r"])
+        assert lines == ["data: a", "data: b"]
+
+    @pytest.mark.asyncio
+    async def test_empty_stream(self) -> None:
+        assert await _collect([]) == []
+
+    @pytest.mark.asyncio
+    async def test_replaces_undecodable_bytes(self) -> None:
+        lines = await _collect([b"data: a\xff b\n"])
+        assert lines == ["data: a<> b"]