feat(F-09): rich parsing — formula + figure extraction
- codex/parsing/mathpix.py: pix2tex (local, CPU) primary + MathPix API optional; bbox heuristic h>15px, math-char-count>5; singleton model cache - codex/parsing/figures.py: pymupdf embedded-image extraction → PNG; caption detection via proximity + "Figure/Fig./Abbildung" prefix - codex/models.py: FormulaChunk + FigureChunk dataclasses (R-10/R-11) - codex/ingest.py: --rich flag wires formula+figure extraction post-ingest - codex/cli.py: search_app sub-typer (paper + formula subcommands), --rich flag on ingest; wiki_app from F-12 preserved intact - codex/config.py: mathpix_app_id/key, pix2tex_fallback, figures_dir - infra/schema.sql: formulas + figures tables with HNSW pgvector indexes - pyproject.toml: pymupdf>=1.24, pix2tex>=0.1.4 - tests/parsing/test_mathpix.py + test_figures.py: 31 tests (mock pix2tex + MathPix HTTP, real pymupdf on synthetic PDF) Gate: 158 passed, ruff clean, mypy clean (20 files) Requirements: R-10 R-11 R-12 R-13 R-14 → done Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Tarik Moussa
386
tests/parsing/test_mathpix.py
Normal file
386
tests/parsing/test_mathpix.py
Normal file
@@ -0,0 +1,386 @@
|
||||
"""Tests for codex.parsing.mathpix — formula extraction module.
|
||||
|
||||
All external dependencies (fitz, pix2tex, httpx, get_settings) are mocked
|
||||
so the suite runs offline without real API calls or model downloads.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
from typing import Any
|
||||
from unittest.mock import MagicMock, patch
|
||||
|
||||
import pytest
|
||||
|
||||
from codex.models import FormulaChunk
|
||||
|
||||
|
||||
def _mock_open_pdf() -> MagicMock:
|
||||
"""Return a mock that satisfies ``open(path, 'rb')`` as a context manager."""
|
||||
mock_file = MagicMock()
|
||||
mock_file.__enter__ = MagicMock(return_value=mock_file)
|
||||
mock_file.__exit__ = MagicMock(return_value=False)
|
||||
mock_open = MagicMock(return_value=mock_file)
|
||||
return mock_open
|
||||
|
||||
|
||||
class TestMathPixAPIPath:
|
||||
"""Tests for the MathPix cloud API backend."""
|
||||
|
||||
def test_mathpix_200_returns_formulas(self) -> None:
|
||||
"""extract_formulas routes to MathPix when creds are provided and parses result."""
|
||||
mmd_content = "\\[\n\\alpha + \\beta\n\\]\n\\[\nE = mc^2\n\\]"
|
||||
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch("builtins.open", _mock_open_pdf()),
|
||||
patch("httpx.post") as mock_post,
|
||||
patch("httpx.get") as mock_get,
|
||||
patch("time.sleep"),
|
||||
):
|
||||
mock_post.return_value = MagicMock(
|
||||
status_code=200,
|
||||
json=lambda: {"pdf_id": "abc123"},
|
||||
)
|
||||
mock_post.return_value.raise_for_status = MagicMock()
|
||||
|
||||
mock_get.return_value = MagicMock(
|
||||
status_code=200,
|
||||
text=mmd_content,
|
||||
)
|
||||
|
||||
from codex.parsing.mathpix import _extract_formulas_mathpix
|
||||
|
||||
results = _extract_formulas_mathpix("paper.pdf", "app_id", "app_key")
|
||||
|
||||
assert len(results) == 2
|
||||
assert all(isinstance(r, FormulaChunk) for r in results)
|
||||
assert results[0].raw_latex == "\\[\n\\alpha + \\beta\n\\]"
|
||||
|
||||
def test_mathpix_empty_pdf_id_returns_empty(self) -> None:
|
||||
"""When MathPix returns no pdf_id, result is empty list."""
|
||||
with (
|
||||
patch("builtins.open", _mock_open_pdf()),
|
||||
patch("httpx.post") as mock_post,
|
||||
patch("time.sleep"),
|
||||
):
|
||||
mock_post.return_value = MagicMock(
|
||||
status_code=200,
|
||||
json=lambda: {}, # No pdf_id
|
||||
)
|
||||
mock_post.return_value.raise_for_status = MagicMock()
|
||||
|
||||
from codex.parsing.mathpix import _extract_formulas_mathpix
|
||||
|
||||
results = _extract_formulas_mathpix("paper.pdf", "app_id", "app_key")
|
||||
|
||||
assert results == []
|
||||
|
||||
def test_mathpix_paper_id_from_stem(self) -> None:
|
||||
"""paper_id in FormulaChunk is derived from the PDF filename stem."""
|
||||
mmd_content = "\\[\nx^2\n\\]"
|
||||
|
||||
with (
|
||||
patch("builtins.open", _mock_open_pdf()),
|
||||
patch("httpx.post") as mock_post,
|
||||
patch("httpx.get") as mock_get,
|
||||
patch("time.sleep"),
|
||||
):
|
||||
mock_post.return_value = MagicMock(
|
||||
status_code=200,
|
||||
json=lambda: {"pdf_id": "xyz"},
|
||||
)
|
||||
mock_post.return_value.raise_for_status = MagicMock()
|
||||
mock_get.return_value = MagicMock(status_code=200, text=mmd_content)
|
||||
|
||||
from codex.parsing.mathpix import _extract_formulas_mathpix
|
||||
|
||||
results = _extract_formulas_mathpix("/tmp/2301.07041.pdf", "id", "key")
|
||||
|
||||
assert results[0].paper_id == "2301.07041"
|
||||
|
||||
def test_mathpix_page_counter_increments(self) -> None:
|
||||
"""Page counter increments on MathPix page markers in MMD."""
|
||||
mmd_content = "\\[\nA\n\\]\n<!-- Page 2 -->\n\\[\nB\n\\]"
|
||||
|
||||
with (
|
||||
patch("builtins.open", _mock_open_pdf()),
|
||||
patch("httpx.post") as mock_post,
|
||||
patch("httpx.get") as mock_get,
|
||||
patch("time.sleep"),
|
||||
):
|
||||
mock_post.return_value = MagicMock(
|
||||
status_code=200,
|
||||
json=lambda: {"pdf_id": "p1"},
|
||||
)
|
||||
mock_post.return_value.raise_for_status = MagicMock()
|
||||
mock_get.return_value = MagicMock(status_code=200, text=mmd_content)
|
||||
|
||||
from codex.parsing.mathpix import _extract_formulas_mathpix
|
||||
|
||||
results = _extract_formulas_mathpix("p.pdf", "id", "key")
|
||||
|
||||
assert results[0].page == 1
|
||||
assert results[1].page == 2
|
||||
|
||||
def test_mathpix_http_error_propagates(self) -> None:
|
||||
"""HTTP errors from MathPix are raised (tenacity reraises after retries)."""
|
||||
import httpx
|
||||
|
||||
with (
|
||||
patch("builtins.open", _mock_open_pdf()),
|
||||
patch("httpx.post") as mock_post,
|
||||
):
|
||||
mock_post.return_value = MagicMock(status_code=401)
|
||||
mock_post.return_value.raise_for_status.side_effect = httpx.HTTPStatusError(
|
||||
"401", request=MagicMock(), response=MagicMock()
|
||||
)
|
||||
|
||||
from codex.parsing.mathpix import _extract_formulas_mathpix
|
||||
|
||||
with pytest.raises(httpx.HTTPStatusError):
|
||||
_extract_formulas_mathpix("p.pdf", "id", "key")
|
||||
|
||||
|
||||
class TestPix2TexFallbackPath:
|
||||
"""Tests for the local pix2tex backend."""
|
||||
|
||||
def _make_fitz_page(
|
||||
self,
|
||||
blocks: list[Any],
|
||||
pixmap_samples: bytes = b"\xff" * (30 * 100 * 3),
|
||||
pixmap_w: int = 100,
|
||||
pixmap_h: int = 30,
|
||||
) -> MagicMock:
|
||||
"""Build a mock fitz page with given text blocks and a fake pixmap."""
|
||||
mock_pix = MagicMock()
|
||||
mock_pix.width = pixmap_w
|
||||
mock_pix.height = pixmap_h
|
||||
mock_pix.samples = pixmap_samples
|
||||
|
||||
mock_page = MagicMock()
|
||||
mock_page.get_text.return_value = blocks
|
||||
mock_page.get_pixmap.return_value = mock_pix
|
||||
return mock_page
|
||||
|
||||
def test_pix2tex_fallback_invoked_without_creds(self) -> None:
|
||||
"""When no MathPix creds, pix2tex fallback is called."""
|
||||
math_block = (0.0, 0.0, 200.0, 50.0, "∑ α∫β∇γ∂δ∞ε", 0, 0)
|
||||
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = True
|
||||
mock_settings.figures_dir = "/tmp/figs"
|
||||
|
||||
mock_model = MagicMock(return_value=r"\sum \alpha")
|
||||
|
||||
mock_page = self._make_fitz_page([math_block])
|
||||
mock_doc = MagicMock()
|
||||
mock_doc.__len__.return_value = 1
|
||||
mock_doc.__getitem__ = MagicMock(return_value=mock_page)
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch("codex.parsing.mathpix._get_pix2tex_model", return_value=mock_model),
|
||||
patch("fitz.open", return_value=mock_doc),
|
||||
patch("fitz.Matrix"),
|
||||
patch("fitz.Rect"),
|
||||
):
|
||||
from PIL import Image
|
||||
|
||||
with patch.object(Image, "frombytes", return_value=MagicMock()):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
results = extract_formulas("paper.pdf")
|
||||
|
||||
assert len(results) == 1
|
||||
assert results[0].raw_latex == r"\sum \alpha"
|
||||
|
||||
def test_pix2tex_bbox_size_filter(self) -> None:
|
||||
"""Blocks smaller than MIN_HEIGHT_PT or MIN_WIDTH_PT are skipped."""
|
||||
tiny_block = (0.0, 0.0, 10.0, 5.0, "∑∫∂∞∇ε", 0, 0) # 10pt wide, 5pt tall
|
||||
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
mock_model = MagicMock(return_value=r"\sum")
|
||||
mock_page = self._make_fitz_page([tiny_block])
|
||||
mock_doc = MagicMock()
|
||||
mock_doc.__len__.return_value = 1
|
||||
mock_doc.__getitem__ = MagicMock(return_value=mock_page)
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch("codex.parsing.mathpix._get_pix2tex_model", return_value=mock_model),
|
||||
patch("fitz.open", return_value=mock_doc),
|
||||
patch("fitz.Matrix"),
|
||||
patch("fitz.Rect"),
|
||||
):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
results = extract_formulas("paper.pdf")
|
||||
|
||||
assert results == []
|
||||
|
||||
def test_pix2tex_math_ratio_filter(self) -> None:
|
||||
"""Blocks with low math-character ratio are skipped."""
|
||||
plain_text_block = (0.0, 0.0, 200.0, 50.0, "This is just regular text with no math.", 0, 0)
|
||||
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
mock_model = MagicMock(return_value=r"\alpha")
|
||||
mock_page = self._make_fitz_page([plain_text_block])
|
||||
mock_doc = MagicMock()
|
||||
mock_doc.__len__.return_value = 1
|
||||
mock_doc.__getitem__ = MagicMock(return_value=mock_page)
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch("codex.parsing.mathpix._get_pix2tex_model", return_value=mock_model),
|
||||
patch("fitz.open", return_value=mock_doc),
|
||||
patch("fitz.Matrix"),
|
||||
patch("fitz.Rect"),
|
||||
):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
results = extract_formulas("paper.pdf")
|
||||
|
||||
assert results == []
|
||||
|
||||
def test_pix2tex_exception_swallowed(self) -> None:
|
||||
"""If pix2tex raises, the block is skipped (no exception propagates)."""
|
||||
math_block = (0.0, 0.0, 200.0, 50.0, "∑ α∫β∇γ∂δ∞ε", 0, 0)
|
||||
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
mock_model = MagicMock(side_effect=RuntimeError("GPU OOM"))
|
||||
mock_page = self._make_fitz_page([math_block])
|
||||
mock_doc = MagicMock()
|
||||
mock_doc.__len__.return_value = 1
|
||||
mock_doc.__getitem__ = MagicMock(return_value=mock_page)
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch("codex.parsing.mathpix._get_pix2tex_model", return_value=mock_model),
|
||||
patch("fitz.open", return_value=mock_doc),
|
||||
patch("fitz.Matrix"),
|
||||
patch("fitz.Rect"),
|
||||
):
|
||||
from PIL import Image
|
||||
|
||||
with patch.object(Image, "frombytes", return_value=MagicMock()):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
results = extract_formulas("paper.pdf") # Must not raise
|
||||
|
||||
assert results == []
|
||||
|
||||
|
||||
class TestRouteSelection:
|
||||
"""Tests for extract_formulas route selection logic."""
|
||||
|
||||
def test_mathpix_selected_when_creds_present(self) -> None:
|
||||
"""extract_formulas calls MathPix backend when both creds are set."""
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = "my_id"
|
||||
mock_settings.mathpix_app_key = "my_key"
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_mathpix", return_value=[]
|
||||
) as mock_mathpix,
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_pix2tex", return_value=[]
|
||||
) as mock_pix2tex,
|
||||
):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
extract_formulas("paper.pdf")
|
||||
|
||||
mock_mathpix.assert_called_once_with("paper.pdf", "my_id", "my_key")
|
||||
mock_pix2tex.assert_not_called()
|
||||
|
||||
def test_pix2tex_selected_without_creds(self) -> None:
|
||||
"""extract_formulas calls pix2tex when no MathPix creds are set."""
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_mathpix", return_value=[]
|
||||
) as mock_mathpix,
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_pix2tex", return_value=[]
|
||||
) as mock_pix2tex,
|
||||
):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
extract_formulas("paper.pdf")
|
||||
|
||||
mock_pix2tex.assert_called_once_with("paper.pdf")
|
||||
mock_mathpix.assert_not_called()
|
||||
|
||||
def test_empty_string_creds_treated_as_absent(self) -> None:
|
||||
"""Empty string credentials fall through to pix2tex (not MathPix)."""
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = ""
|
||||
mock_settings.mathpix_app_key = ""
|
||||
mock_settings.pix2tex_fallback = True
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_mathpix", return_value=[]
|
||||
) as mock_mathpix,
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_pix2tex", return_value=[]
|
||||
) as mock_pix2tex,
|
||||
):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
extract_formulas("paper.pdf")
|
||||
|
||||
mock_pix2tex.assert_called_once()
|
||||
mock_mathpix.assert_not_called()
|
||||
|
||||
def test_formula_extraction_disabled_when_fallback_false(self) -> None:
|
||||
"""No extraction when pix2tex_fallback=False and no creds."""
|
||||
mock_settings = MagicMock()
|
||||
mock_settings.mathpix_app_id = None
|
||||
mock_settings.mathpix_app_key = None
|
||||
mock_settings.pix2tex_fallback = False
|
||||
|
||||
with (
|
||||
patch("codex.parsing.mathpix.get_settings", return_value=mock_settings),
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_mathpix", return_value=[]
|
||||
) as mock_mathpix,
|
||||
patch(
|
||||
"codex.parsing.mathpix._extract_formulas_pix2tex", return_value=[]
|
||||
) as mock_pix2tex,
|
||||
):
|
||||
from codex.parsing.mathpix import extract_formulas
|
||||
|
||||
results = extract_formulas("paper.pdf")
|
||||
|
||||
assert results == []
|
||||
mock_mathpix.assert_not_called()
|
||||
mock_pix2tex.assert_not_called()
|
||||
Reference in New Issue
Block a user