theseusplot 0.1.0__tar.gz → 0.1.1__tar.gz

{theseusplot-0.1.0 → theseusplot-0.1.1}/.github/workflows/publish.yml

@@ -24,6 +24,9 @@ jobs:
       - name: Install build dependencies
         run: python -m pip install --upgrade build twine
 
+      - name: Prepare README for PyPI
+        run: python scripts/prepare_pypi_readme.py
+
       - name: Build source and wheel distributions
         run: python -m build
 

{theseusplot-0.1.0 → theseusplot-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: theseusplot
-Version: 0.1.0
+Version: 0.1.1
 Summary: Python port of TheseusPlot for decomposing differences in rate metrics.
 Author: TheseusPlot.py contributors
 License: MIT
@@ -23,8 +23,9 @@ Description-Content-Type: text/markdown
 
 <!-- badges: start -->
 
+[![PyPI version](https://img.shields.io/pypi/v/theseusplot.svg)](https://pypi.org/project/theseusplot/)
+[![Downloads](https://static.pepy.tech/badge/theseusplot)](https://pepy.tech/project/theseusplot)
 [![CI](https://github.com/hoxo-m/TheseusPlot_py/actions/workflows/ci.yml/badge.svg)](https://github.com/hoxo-m/TheseusPlot_py/actions/workflows/ci.yml)
-
 <!-- badges: end -->
 
 ## 1. Overview
@@ -58,7 +59,7 @@ Thus, the contribution of the female group is -0.2 percentage points.
 
 When visualized, the results appear as follows:
 
-<img src="README-figures/overview-1.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/overview-1.png" alt="" width="500" />
 
 From this plot, we can see that the decline in the metric is primarily
 driven by the male group. We call this visualization the “Theseus Plot.”
@@ -68,18 +69,25 @@ Theseus Plots for various attributes.
 
 ## 2. Installation
 
-You can install the development version from
-[GitHub](https://github.com/hoxo-m/TheseusPlot_py) with:
+You can install the **theseusplot** package from
+[PyPI](https://pypi.org/project/theseusplot/) with:
 
 ``` bash
-python -m pip install "git+https://github.com/hoxo-m/TheseusPlot_py.git"
+python -m pip install theseusplot
 ```
 
 You can install the optional dependencies for examples and documentation
 data with:
 
 ``` bash
-python -m pip install "theseusplot[examples] @ git+https://github.com/hoxo-m/TheseusPlot_py.git"
+python -m pip install "theseusplot[examples]"
+```
+
+You can install the development version from
+[GitHub](https://github.com/hoxo-m/TheseusPlot_py) with:
+
+``` bash
+python -m pip install "git+https://github.com/hoxo-m/TheseusPlot_py.git"
 ```
 
 ## 3. Details
@@ -168,7 +176,7 @@ fig, ax = ship.plot("origin")
 fig.show()
 ```
 
-<img src="README-figures/plot_origin-3.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_origin-3.png" alt="" width="500" />
 
 New York City has three major airports, and Newark Liberty International
 Airport (EWR) accounted for the largest share of the decline in the
@@ -209,7 +217,7 @@ fig, ax = ship.plot_flip("carrier")
 fig.show()
 ```
 
-<img src="README-figures/plot_carrier-5.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_carrier-5.png" alt="" width="500" />
 
 When the number of subgroups is large, those with small contributions
 are automatically grouped together. By default, this happens when there
@@ -221,7 +229,7 @@ fig, ax = ship.plot_flip("carrier", n=6)
 fig.show()
 ```
 
-<img src="README-figures/plot_carrier_n-7.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_carrier_n-7.png" alt="" width="500" />
 
 From this plot, JetBlue Airways and United Air Lines appear to have the
 largest contributions to the decline in on-time arrival rate.
@@ -237,7 +245,7 @@ fig, ax = ship.plot_flip("dep_delay")
 fig.show()
 ```
 
-<img src="README-figures/plot_dep_delay-9.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_dep_delay-9.png" alt="" width="500" />
 
 By default, continuous variables are discretized so that each subgroup
 has roughly equal sample sizes, with the number of bins set to 10. You
@@ -251,7 +259,7 @@ fig, ax = ship.plot_flip("dep_delay", continuous=continuous_config(n=3))
 fig.show()
 ```
 
-<img src="README-figures/plot_dep_delay_n-11.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_dep_delay_n-11.png" alt="" width="500" />
 
 This result shows that both a decrease in on-time departures and an
 increase in delayed departures contributed to the decline in on-time
@@ -305,7 +313,7 @@ fig, ax = ship.plot("segment")
 fig.show()
 ```
 
-<img src="README-figures/factor_column-13.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/factor_column-13.png" alt="" width="500" />
 
 Even if the contribution of `"High"` is larger than that of `"Low"` or
 `"Medium"`, the rows and bars are shown in the order

{theseusplot-0.1.0 → theseusplot-0.1.1}/README.Rmd

@@ -18,12 +18,20 @@ knitr::opts_chunk$set(
 )
 
 library(reticulate)
+py_install(envname = "theseusplot-env", packages = ".", pip = TRUE)
 use_virtualenv("theseusplot-env", required = TRUE)
+
+readme_base_url <- Sys.getenv("THESEUSPLOT_README_BASE_URL", unset = "")
+if (nzchar(readme_base_url)) {
+  knitr::opts_knit$set(base.url = readme_base_url)
+}
 ```
 
 # TheseusPlot: Visualizing Decomposition of Differences in Rate Metrics
 
 <!-- badges: start -->
+[![PyPI version](https://img.shields.io/pypi/v/theseusplot.svg)](https://pypi.org/project/theseusplot/)
+[![Downloads](https://static.pepy.tech/badge/theseusplot)](https://pepy.tech/project/theseusplot)
 [![CI](https://github.com/hoxo-m/TheseusPlot_py/actions/workflows/ci.yml/badge.svg)](https://github.com/hoxo-m/TheseusPlot_py/actions/workflows/ci.yml)
 <!-- badges: end -->
 
@@ -87,18 +95,25 @@ The **TheseusPlot** package is designed to make it easy to generate Theseus Plot
 
 ## 2. Installation
 
-You can install the development version from
-[GitHub](https://github.com/hoxo-m/TheseusPlot_py) with:
+You can install the **theseusplot** package from
+[PyPI](https://pypi.org/project/theseusplot/) with:
 
 ```bash
-python -m pip install "git+https://github.com/hoxo-m/TheseusPlot_py.git"
+python -m pip install theseusplot
 ```
 
 You can install the optional dependencies for examples and documentation data
 with:
 
 ```bash
-python -m pip install "theseusplot[examples] @ git+https://github.com/hoxo-m/TheseusPlot_py.git"
+python -m pip install "theseusplot[examples]"
+```
+
+You can install the development version from
+[GitHub](https://github.com/hoxo-m/TheseusPlot_py) with:
+
+```bash
+python -m pip install "git+https://github.com/hoxo-m/TheseusPlot_py.git"
 ```
 
 ## 3. Details

{theseusplot-0.1.0 → theseusplot-0.1.1}/README.md

@@ -5,8 +5,9 @@
 
 <!-- badges: start -->
 
+[![PyPI version](https://img.shields.io/pypi/v/theseusplot.svg)](https://pypi.org/project/theseusplot/)
+[![Downloads](https://static.pepy.tech/badge/theseusplot)](https://pepy.tech/project/theseusplot)
 [![CI](https://github.com/hoxo-m/TheseusPlot_py/actions/workflows/ci.yml/badge.svg)](https://github.com/hoxo-m/TheseusPlot_py/actions/workflows/ci.yml)
-
 <!-- badges: end -->
 
 ## 1. Overview
@@ -40,7 +41,7 @@ Thus, the contribution of the female group is -0.2 percentage points.
 
 When visualized, the results appear as follows:
 
-<img src="README-figures/overview-1.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/overview-1.png" alt="" width="500" />
 
 From this plot, we can see that the decline in the metric is primarily
 driven by the male group. We call this visualization the “Theseus Plot.”
@@ -50,18 +51,25 @@ Theseus Plots for various attributes.
 
 ## 2. Installation
 
-You can install the development version from
-[GitHub](https://github.com/hoxo-m/TheseusPlot_py) with:
+You can install the **theseusplot** package from
+[PyPI](https://pypi.org/project/theseusplot/) with:
 
 ``` bash
-python -m pip install "git+https://github.com/hoxo-m/TheseusPlot_py.git"
+python -m pip install theseusplot
 ```
 
 You can install the optional dependencies for examples and documentation
 data with:
 
 ``` bash
-python -m pip install "theseusplot[examples] @ git+https://github.com/hoxo-m/TheseusPlot_py.git"
+python -m pip install "theseusplot[examples]"
+```
+
+You can install the development version from
+[GitHub](https://github.com/hoxo-m/TheseusPlot_py) with:
+
+``` bash
+python -m pip install "git+https://github.com/hoxo-m/TheseusPlot_py.git"
 ```
 
 ## 3. Details
@@ -150,7 +158,7 @@ fig, ax = ship.plot("origin")
 fig.show()
 ```
 
-<img src="README-figures/plot_origin-3.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_origin-3.png" alt="" width="500" />
 
 New York City has three major airports, and Newark Liberty International
 Airport (EWR) accounted for the largest share of the decline in the
@@ -191,7 +199,7 @@ fig, ax = ship.plot_flip("carrier")
 fig.show()
 ```
 
-<img src="README-figures/plot_carrier-5.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_carrier-5.png" alt="" width="500" />
 
 When the number of subgroups is large, those with small contributions
 are automatically grouped together. By default, this happens when there
@@ -203,7 +211,7 @@ fig, ax = ship.plot_flip("carrier", n=6)
 fig.show()
 ```
 
-<img src="README-figures/plot_carrier_n-7.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_carrier_n-7.png" alt="" width="500" />
 
 From this plot, JetBlue Airways and United Air Lines appear to have the
 largest contributions to the decline in on-time arrival rate.
@@ -219,7 +227,7 @@ fig, ax = ship.plot_flip("dep_delay")
 fig.show()
 ```
 
-<img src="README-figures/plot_dep_delay-9.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_dep_delay-9.png" alt="" width="500" />
 
 By default, continuous variables are discretized so that each subgroup
 has roughly equal sample sizes, with the number of bins set to 10. You
@@ -233,7 +241,7 @@ fig, ax = ship.plot_flip("dep_delay", continuous=continuous_config(n=3))
 fig.show()
 ```
 
-<img src="README-figures/plot_dep_delay_n-11.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/plot_dep_delay_n-11.png" alt="" width="500" />
 
 This result shows that both a decrease in on-time departures and an
 increase in delayed departures contributed to the decline in on-time
@@ -287,7 +295,7 @@ fig, ax = ship.plot("segment")
 fig.show()
 ```
 
-<img src="README-figures/factor_column-13.png" alt="" width="500" />
+<img src="https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/README-figures/factor_column-13.png" alt="" width="500" />
 
 Even if the contribution of `"High"` is larger than that of `"Low"` or
 `"Medium"`, the rows and bars are shown in the order

{theseusplot-0.1.0 → theseusplot-0.1.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "theseusplot"
-version = "0.1.0"
+version = "0.1.1"
 description = "Python port of TheseusPlot for decomposing differences in rate metrics."
 readme = "README.md"
 requires-python = ">=3.10"

theseusplot-0.1.1/scripts/prepare_pypi_readme.py

@@ -0,0 +1,21 @@
+"""Prepare README image URLs for PyPI rendering."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+RAW_BASE_URL = "https://raw.githubusercontent.com/hoxo-m/TheseusPlot_py/main/"
+README_PATH = Path("README.md")
+
+
+def main() -> None:
+    readme = README_PATH.read_text(encoding="utf-8")
+    readme = readme.replace(
+        'src="README-figures/',
+        f'src="{RAW_BASE_URL}README-figures/',
+    )
+    README_PATH.write_text(readme, encoding="utf-8")
+
+
+if __name__ == "__main__":
+    main()

{theseusplot-0.1.0 → theseusplot-0.1.1}/src/theseusplot/_ship.py

@@ -20,8 +20,8 @@ _MIN_BREAK_COUNT = 2
 _REFITTED_COLOR = "#00BFC4"
 _ORIGINAL_SIZE_COLOR = "#7CAE00"
 _REFITTED_SIZE_COLOR = "#C77CFF"
-_POSITIVE_COLOR = "#F8766D"
-_NEGATIVE_COLOR = "#00BFC4"
+_POSITIVE_COLOR = "#00BFC4"
+_NEGATIVE_COLOR = "#F8766D"
 
 
 class ShipOfTheseus:
@@ -695,7 +695,6 @@ class ShipOfTheseus:
         ax.axvline(0, color="#333333", linewidth=0.8)
         ax.set_yticks(positions)
         ax.set_yticklabels(waterfall["items"].astype(str))
-        ax.invert_yaxis()
         ax.set_xlabel(self.y_label or "")
         ax.set_ylabel("")
         ax.set_title(column)
@@ -733,7 +732,6 @@ class ShipOfTheseus:
                 row["scaled_n"],
                 width=width,
                 color=colors[group],
-                alpha=0.35,
                 linewidth=0,
                 zorder=1,
             )
@@ -760,7 +758,6 @@ class ShipOfTheseus:
                 row["scaled_n"],
                 height=height,
                 color=colors[group],
-                alpha=0.35,
                 linewidth=0,
                 zorder=1,
             )
@@ -774,7 +771,7 @@ class ShipOfTheseus:
         colors = [
             _REFITTED_COLOR
             if row["kind"] == "total"
-            else (_POSITIVE_COLOR if row["amount"] >= 0 else _NEGATIVE_COLOR)
+            else self._contribution_color(float(row["amount"]))
             for _, row in waterfall.iterrows()
         ]
         ax.bar(
@@ -813,7 +810,7 @@ class ShipOfTheseus:
         colors = [
             _REFITTED_COLOR
             if row["kind"] == "total"
-            else (_POSITIVE_COLOR if row["amount"] >= 0 else _NEGATIVE_COLOR)
+            else self._contribution_color(-float(row["amount"]))
             for _, row in waterfall.iterrows()
         ]
         ax.barh(
@@ -827,10 +824,12 @@ class ShipOfTheseus:
             zorder=3,
         )
         for position, (_, row) in zip(positions, waterfall.iterrows(), strict=True):
-            value = self._format_plot_value(float(row["amount"]))
+            amount = float(row["amount"])
+            display_amount = -amount if row["kind"] == "contribution" else amount
+            value = self._format_plot_value(display_amount)
             x = float(row["bottom"]) + float(row["height"])
             ha = "left"
-            if float(row["amount"]) < 0:
+            if amount < 0:
                 x = float(row["bottom"])
                 ha = "right"
             ax.text(
@@ -843,6 +842,10 @@ class ShipOfTheseus:
                 zorder=4,
             )
 
+    @staticmethod
+    def _contribution_color(amount: float) -> str:
+        return _NEGATIVE_COLOR if amount < 0 else _POSITIVE_COLOR
+
     @staticmethod
     def _draw_connectors(
         ax: Any,

theseusplot-0.1.1/tests/test_plot.py

@@ -0,0 +1,255 @@
+import pandas as pd
+import pytest
+
+from theseusplot import create_ship
+
+
+def _require_pyplot():
+    matplotlib = pytest.importorskip("matplotlib")
+    matplotlib.use("Agg", force=True)
+    return pytest.importorskip("matplotlib.pyplot")
+
+
+def _waterfall_colors(ax) -> list[str]:
+    colors = pytest.importorskip("matplotlib.colors")
+    return [
+        colors.to_hex(patch.get_facecolor())
+        for patch in ax.patches
+        if patch.get_zorder() == 3
+    ]
+
+
+def _size_bar_styles(ax) -> list[tuple[str, float]]:
+    colors = pytest.importorskip("matplotlib.colors")
+    return [
+        (colors.to_hex(patch.get_facecolor()), patch.get_facecolor()[3])
+        for patch in ax.patches
+        if patch.get_zorder() == 1
+    ]
+
+
+def _visible_yticklabels(ax) -> list[str]:
+    labels = []
+    for tick in ax.get_yticklabels():
+        _, display_y = ax.transData.transform((0, tick.get_position()[1]))
+        labels.append((display_y, tick.get_text()))
+    return [label for _, label in sorted(labels, reverse=True)]
+
+
+def _plot_texts(ax) -> list[str]:
+    return [text.get_text() for text in ax.texts]
+
+
+def test_plot_returns_matplotlib_figure_and_axes() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [1, 1, 1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [0, 0, 1, 1]})
+    ship = create_ship(data1, data2, labels=("Before", "After"), y_label="Rate")
+
+    fig, ax = ship.plot("group")
+
+    try:
+        assert ax.get_title() == "group"
+        assert ax.get_ylabel() == "Rate"
+        assert [tick.get_text() for tick in ax.get_xticklabels()] == [
+            "Before",
+            "A",
+            "B",
+            "After",
+        ]
+        assert fig is ax.figure
+    finally:
+        plt.close(fig)
+
+
+def test_plot_respects_explicit_levels() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot("group", levels=["B", "A"])
+
+    try:
+        assert [tick.get_text() for tick in ax.get_xticklabels()] == [
+            "Original",
+            "B",
+            "A",
+            "Refitted",
+        ]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_validates_main_item() -> None:
+    _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
+    ship = create_ship(data1, data2)
+
+    with pytest.raises(ValueError, match="main_item"):
+        ship.plot("group", main_item="C")
+
+
+def test_plot_colors_negative_contributions_red_and_positive_blue() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame(
+        {"group": ["A"] * 10 + ["B"] * 10, "y": [0] * 10 + [1] * 10},
+    )
+    data2 = pd.DataFrame(
+        {"group": ["A"] * 10 + ["B"] * 30, "y": [1] * 10 + [0] * 30},
+    )
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot("group")
+
+    try:
+        assert _waterfall_colors(ax) == [
+            "#00bfc4",
+            "#f8766d",
+            "#00bfc4",
+            "#00bfc4",
+        ]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_size_bars_use_r_colors_without_alpha() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot("group")
+
+    try:
+        assert _size_bar_styles(ax) == [
+            ("#7cae00", 1.0),
+            ("#7cae00", 1.0),
+            ("#c77cff", 1.0),
+            ("#c77cff", 1.0),
+        ]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_flip_returns_horizontal_matplotlib_plot() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [1, 1, 1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [0, 0, 1, 1]})
+    ship = create_ship(data1, data2, labels=("Before", "After"), y_label="Rate")
+
+    fig, ax = ship.plot_flip("group")
+
+    try:
+        assert ax.get_title() == "group"
+        assert ax.get_xlabel() == "Rate"
+        assert _visible_yticklabels(ax) == [
+            "Before",
+            "A",
+            "B",
+            "After",
+        ]
+        assert fig is ax.figure
+    finally:
+        plt.close(fig)
+
+
+def test_plot_flip_colors_original_negative_contributions_red() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame(
+        {"group": ["A"] * 10 + ["B"] * 10, "y": [0] * 10 + [1] * 10},
+    )
+    data2 = pd.DataFrame(
+        {"group": ["A"] * 10 + ["B"] * 30, "y": [1] * 10 + [0] * 30},
+    )
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot_flip("group")
+
+    try:
+        assert _waterfall_colors(ax) == [
+            "#00bfc4",
+            "#00bfc4",
+            "#f8766d",
+            "#00bfc4",
+        ]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_flip_labels_use_original_contribution_signs() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame(
+        {"group": ["A"] * 10 + ["B"] * 10, "y": [0] * 10 + [1] * 10},
+    )
+    data2 = pd.DataFrame(
+        {"group": ["A"] * 10 + ["B"] * 30, "y": [1] * 10 + [0] * 30},
+    )
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot_flip("group")
+
+    try:
+        assert _plot_texts(ax) == ["25", "37.5", "-62.5", "50"]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_flip_size_bars_use_r_colors_without_alpha() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot_flip("group")
+
+    try:
+        assert _size_bar_styles(ax) == [
+            ("#7cae00", 1.0),
+            ("#7cae00", 1.0),
+            ("#c77cff", 1.0),
+            ("#c77cff", 1.0),
+        ]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_flip_respects_reversed_explicit_levels() -> None:
+    plt = _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
+    ship = create_ship(data1, data2)
+
+    fig, ax = ship.plot_flip("group", levels=["B", "A"])
+
+    try:
+        assert _visible_yticklabels(ax) == [
+            "Original",
+            "B",
+            "A",
+            "Refitted",
+        ]
+    finally:
+        plt.close(fig)
+
+
+def test_plot_flip_validates_main_item() -> None:
+    _require_pyplot()
+
+    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
+    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
+    ship = create_ship(data1, data2)
+
+    with pytest.raises(ValueError, match="main_item"):
+        ship.plot_flip("group", main_item="C")

theseusplot-0.1.0/tests/test_plot.py

@@ -1,118 +0,0 @@
-import pandas as pd
-import pytest
-
-from theseusplot import create_ship
-
-
-def _require_pyplot():
-    matplotlib = pytest.importorskip("matplotlib")
-    matplotlib.use("Agg", force=True)
-    return pytest.importorskip("matplotlib.pyplot")
-
-
-def test_plot_returns_matplotlib_figure_and_axes() -> None:
-    plt = _require_pyplot()
-
-    data1 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [1, 1, 1, 1]})
-    data2 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [0, 0, 1, 1]})
-    ship = create_ship(data1, data2, labels=("Before", "After"), y_label="Rate")
-
-    fig, ax = ship.plot("group")
-
-    try:
-        assert ax.get_title() == "group"
-        assert ax.get_ylabel() == "Rate"
-        assert [tick.get_text() for tick in ax.get_xticklabels()] == [
-            "Before",
-            "A",
-            "B",
-            "After",
-        ]
-        assert fig is ax.figure
-    finally:
-        plt.close(fig)
-
-
-def test_plot_respects_explicit_levels() -> None:
-    plt = _require_pyplot()
-
-    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
-    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
-    ship = create_ship(data1, data2)
-
-    fig, ax = ship.plot("group", levels=["B", "A"])
-
-    try:
-        assert [tick.get_text() for tick in ax.get_xticklabels()] == [
-            "Original",
-            "B",
-            "A",
-            "Refitted",
-        ]
-    finally:
-        plt.close(fig)
-
-
-def test_plot_validates_main_item() -> None:
-    _require_pyplot()
-
-    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
-    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
-    ship = create_ship(data1, data2)
-
-    with pytest.raises(ValueError, match="main_item"):
-        ship.plot("group", main_item="C")
-
-
-def test_plot_flip_returns_horizontal_matplotlib_plot() -> None:
-    plt = _require_pyplot()
-
-    data1 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [1, 1, 1, 1]})
-    data2 = pd.DataFrame({"group": ["A", "A", "B", "B"], "y": [0, 0, 1, 1]})
-    ship = create_ship(data1, data2, labels=("Before", "After"), y_label="Rate")
-
-    fig, ax = ship.plot_flip("group")
-
-    try:
-        assert ax.get_title() == "group"
-        assert ax.get_xlabel() == "Rate"
-        assert [tick.get_text() for tick in ax.get_yticklabels()] == [
-            "After",
-            "B",
-            "A",
-            "Before",
-        ]
-        assert fig is ax.figure
-    finally:
-        plt.close(fig)
-
-
-def test_plot_flip_respects_reversed_explicit_levels() -> None:
-    plt = _require_pyplot()
-
-    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
-    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
-    ship = create_ship(data1, data2)
-
-    fig, ax = ship.plot_flip("group", levels=["B", "A"])
-
-    try:
-        assert [tick.get_text() for tick in ax.get_yticklabels()] == [
-            "Refitted",
-            "A",
-            "B",
-            "Original",
-        ]
-    finally:
-        plt.close(fig)
-
-
-def test_plot_flip_validates_main_item() -> None:
-    _require_pyplot()
-
-    data1 = pd.DataFrame({"group": ["A", "B"], "y": [1, 1]})
-    data2 = pd.DataFrame({"group": ["A", "B"], "y": [0, 1]})
-    ship = create_ship(data1, data2)
-
-    with pytest.raises(ValueError, match="main_item"):
-        ship.plot_flip("group", main_item="C")