Add topostats file helper class

notebooks/topostats_file_helper_example.ipynb

@@ -0,0 +1,141 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Import needed libraries\n",
+    "import numpy as np\n",
+    "from topostats.io import TopoFileHelper\n",
+    "from IPython.display import clear_output\n",
+    "import matplotlib.pyplot as plt"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Load topostats file"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Load topostats file\n",
+    "file = \"../tests/resources/file.topostats\"\n",
+    "helper = TopoFileHelper(file)\n",
+    "# Clear logging output\n",
+    "clear_output(wait=False)\n",
+    "print(\"File loaded\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Print the structure of the file"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Print the structure of the file\n",
+    "helper.pretty_print_structure()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Find data within the file"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Find the name of the data we want, we know it contains \"ordered_trace_heights\" and we want grain 2, but don't know\n",
+    "# what keys precede it\n",
+    "helper.find_data([\"ordered_trace_heights\", \"2\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Retrieve data from the file"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Get some data from the file\n",
+    "cropped_image = helper.get_data(\"grain_trace_data/above/cropped_images/2\")\n",
+    "ordered_trace_heights = helper.get_data(\"grain_trace_data/above/ordered_trace_heights/2\")\n",
+    "cumulative_distances = helper.get_data(\"grain_trace_data/above/ordered_trace_cumulative_distances/2\")\n",
+    "ordered_traces = helper.get_data(\"grain_trace_data/above/ordered_traces/2\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Use the retrieved data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Plot the image\n",
+    "plt.imshow(cropped_image)\n",
+    "# Create a basic colour scale for the moleucle trace\n",
+    "c = np.arange(0, len(ordered_traces))\n",
+    "# Plot the molecule trace\n",
+    "plt.scatter(ordered_traces[:, 1], ordered_traces[:, 0], c=c, s=10)\n",
+    "plt.show()\n",
+    "# Plot the height of the molecule trace against the cumulative distance in nanometres\n",
+    "plt.plot(cumulative_distances, ordered_trace_heights)\n",
+    "plt.xlabel(\"Cumulative distance (nm)\")\n",
+    "plt.ylabel(\"Height (nm)\")\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "topo-unet",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.9"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

tests/test_io.py

@@ -1,5 +1,7 @@
 """Tests of IO."""
 
+from __future__ import annotations
+
 import argparse
 import json
 import logging
@@ -15,6 +17,7 @@
 
 from topostats.io import (
     LoadScans,
+    TopoFileHelper,
     convert_basename_to_relative_paths,
     dict_to_hdf5,
     dict_to_json,
@@ -1379,3 +1382,41 @@ def test_dict_to_json(dictionary: dict, target: dict, tmp_path: Path) -> None:
 
     with outfile.open("r", encoding="utf-8") as f:
         assert target == json.load(f)
+
+
+class TestTopoFileHelper:
+    """Test the TopoFileHelper class."""
+
+    @pytest.mark.parametrize(
+        ("file_path_or_string"),
+        [
+            pytest.param(
+                "tests/resources/file.topostats",
+                id="String file path",
+            ),
+            pytest.param(
+                Path("tests/resources/file.topostats"),
+                id="Path object path",
+            ),
+        ],
+    )
+    def test_init(self, file_path_or_string: Path | str) -> None:
+        """Test the __init__ method of the TopoFileHelper class."""
+        topo_file_helper = TopoFileHelper(file_path_or_string)
+        assert isinstance(topo_file_helper, TopoFileHelper)
+        assert isinstance(topo_file_helper.data, dict)
+
+    def test_get_data(self) -> None:
+        """Test the get_data method of the TopoFileHelper class."""
+        topo_file_helper = TopoFileHelper("tests/resources/file.topostats")
+        cropped_image = topo_file_helper.get_data("grain_trace_data/above/cropped_images/2")
+        assert isinstance(cropped_image, np.ndarray)
+
+
+# This test only works when not part of the TestTopoFileHelper class
+def test_pretty_print_structure(caplog) -> None:
+    """Test the pretty_print_structure method of the TopoFileHelper class."""
+    topo_file_helper = TopoFileHelper("tests/resources/file.topostats")
+    topo_file_helper.pretty_print_structure()
+    assert "filename" in caplog.text
+    assert "keys with numpy arrays as values" in caplog.text