feat(sources): OpenAlex, SemanticScholar, arXiv API clients

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

tests/sources/test_arxiv.py

@@ -0,0 +1,140 @@
+"""Tests for codex.sources.arxiv."""
+
+from __future__ import annotations
+
+import io
+import tarfile
+
+import httpx
+import pytest
+
+from codex.sources import arxiv
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+
+def _make_tar_gz(files: dict[str, bytes]) -> bytes:
+    """Build an in-memory .tar.gz archive from a dict of {name: content}."""
+    buf = io.BytesIO()
+    with tarfile.open(fileobj=buf, mode="w:gz") as tf:
+        for name, content in files.items():
+            info = tarfile.TarInfo(name=name)
+            info.size = len(content)
+            tf.addfile(info, io.BytesIO(content))
+    return buf.getvalue()
+
+
+# ---------------------------------------------------------------------------
+# fetch_source — success: in-memory tar.gz with one .tex
+# ---------------------------------------------------------------------------
+
+
+def test_fetch_source_returns_latex(monkeypatch: pytest.MonkeyPatch) -> None:
+    """fetch_source extracts and returns the LaTeX content from a .tar.gz."""
+    latex_content = b"\\documentclass{article}\n\\begin{document}\nHello world\n\\end{document}"
+    tar_bytes = _make_tar_gz({"main.tex": latex_content})
+
+    def mock_get(
+        url: str,
+        *,
+        timeout: int = 60,
+        follow_redirects: bool = True,
+    ) -> httpx.Response:
+        return httpx.Response(200, content=tar_bytes)
+
+    monkeypatch.setattr(httpx, "get", mock_get)
+
+    result = arxiv.fetch_source("2301.07041")
+
+    assert result is not None
+    assert "\\documentclass" in result
+    assert "Hello world" in result
+
+
+# ---------------------------------------------------------------------------
+# fetch_source — 404 → None
+# ---------------------------------------------------------------------------
+
+
+def test_fetch_source_404_returns_none(monkeypatch: pytest.MonkeyPatch) -> None:
+    """A 404 response should return None."""
+
+    def mock_get(
+        url: str,
+        *,
+        timeout: int = 60,
+        follow_redirects: bool = True,
+    ) -> httpx.Response:
+        return httpx.Response(404)
+
+    monkeypatch.setattr(httpx, "get", mock_get)
+
+    result = arxiv.fetch_source("nonexistent-id")
+    assert result is None
+
+
+# ---------------------------------------------------------------------------
+# fetch_source — no .tex in archive → None
+# ---------------------------------------------------------------------------
+
+
+def test_fetch_source_no_tex_returns_none(monkeypatch: pytest.MonkeyPatch) -> None:
+    """If the archive contains no .tex files, return None."""
+    tar_bytes = _make_tar_gz({"README.md": b"# Paper\nNo LaTeX here."})
+
+    def mock_get(
+        url: str,
+        *,
+        timeout: int = 60,
+        follow_redirects: bool = True,
+    ) -> httpx.Response:
+        return httpx.Response(200, content=tar_bytes)
+
+    monkeypatch.setattr(httpx, "get", mock_get)
+
+    result = arxiv.fetch_source("2301.07041")
+    assert result is None
+
+
+# ---------------------------------------------------------------------------
+# fetch_source — fallback to largest .tex when no \documentclass
+# ---------------------------------------------------------------------------
+
+
+def test_fetch_source_fallback_to_largest_tex(monkeypatch: pytest.MonkeyPatch) -> None:
+    """When no file has \\documentclass, the largest .tex is returned."""
+    small_tex = b"% small file\n\\section{Intro}"
+    large_tex = b"% large file\n" + b"x" * 500
+
+    tar_bytes = _make_tar_gz({"small.tex": small_tex, "large.tex": large_tex})
+
+    def mock_get(
+        url: str,
+        *,
+        timeout: int = 60,
+        follow_redirects: bool = True,
+    ) -> httpx.Response:
+        return httpx.Response(200, content=tar_bytes)
+
+    monkeypatch.setattr(httpx, "get", mock_get)
+
+    result = arxiv.fetch_source("2301.07041")
+
+    assert result is not None
+    assert "large file" in result
+
+
+# ---------------------------------------------------------------------------
+# fetch_pdf_url — pure computation, no mock needed
+# ---------------------------------------------------------------------------
+
+
+def test_fetch_pdf_url_pure_computation() -> None:
+    """fetch_pdf_url returns the correct URL without making any HTTP request."""
+    url = arxiv.fetch_pdf_url("2301.07041")
+    assert url == "https://arxiv.org/pdf/2301.07041.pdf"
+
+    url2 = arxiv.fetch_pdf_url("1234.56789")
+    assert url2 == "https://arxiv.org/pdf/1234.56789.pdf"