/bar

Goal

Generate a publication-ready bar chart with optional error bars, grayscale styling, and Chinese font support.

Preparation

Collect the following parameters from the user:

Parameter	Required	Default	Description
data	Yes	-	Data source: CSV file path
x	Yes	-	Column name for categories (X-axis)
y	Yes	-	Column name for values
error	No	-	Column name for error values (for error bars)
group	No	-	Column name for grouping (creates grouped bars)
title	No	-	Chart title
xlabel	No	-	X-axis label
ylabel	No	-	Y-axis label
orientation	No	vertical	Bar orientation: vertical or horizontal
show_values	No	true	Show value labels on bars
output	No	paper/figures/bar_chart	Output file path
width	No	6	Figure width in inches
height	No	4	Figure height in inches

Execution Steps

1. Load and prepare data

import pandas as pd
import matplotlib.pyplot as plt
import matplotlib as mpl
import numpy as np
from pathlib import Path

data = pd.read_csv('{{data}}')

2. Configure academic style

plt.rcParams['font.sans-serif'] = ['STFangsong', 'SimSun', 'PingFang SC', 'SimHei']
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['axes.unicode_minus'] = False

COLORS = ['#000000', '#4D4D4D', '#7F7F7F', '#B2B2B2', '#D9D9D9']
HATCHES = ['', '///', '...', 'xxx', '\\\\\\']

mpl.rcParams.update({
    'font.size': 11,
    'axes.linewidth': 0.8,
    'axes.spines.top': False,
    'axes.spines.right': False,
    'axes.grid': True,
    'grid.alpha': 0.3,
    'legend.frameon': False,
    'savefig.dpi': 300,
    'savefig.bbox': 'tight'
})

3. Generate bar chart

fig, ax = plt.subplots(figsize=({{width}}, {{height}}))

x_col = '{{x}}'
y_col = '{{y}}'
error_col = '{{error}}' if '{{error}}' else None
group_col = '{{group}}' if '{{group}}' else None
orientation = '{{orientation}}' or 'vertical'
show_values = '{{show_values}}' != 'false'

categories = data[x_col].unique()
x_pos = np.arange(len(categories))

if group_col:
    # Grouped bar chart
    groups = data[group_col].unique()
    n_groups = len(groups)
    width = 0.8 / n_groups

    for i, group in enumerate(groups):
        subset = data[data[group_col] == group]
        values = [subset[subset[x_col] == cat][y_col].values[0] if len(subset[subset[x_col] == cat]) > 0 else 0
                  for cat in categories]
        errors = None
        if error_col:
            errors = [subset[subset[x_col] == cat][error_col].values[0] if len(subset[subset[x_col] == cat]) > 0 else 0
                      for cat in categories]

        offset = (i - n_groups/2 + 0.5) * width
        bars = ax.bar(x_pos + offset, values, width,
                      label=group,
                      color=COLORS[i % len(COLORS)],
                      edgecolor='black', linewidth=0.8,
                      hatch=HATCHES[i % len(HATCHES)] if i > 0 else '',
                      yerr=errors, capsize=4,
                      error_kw={'linewidth': 1, 'color': 'black'})
else:
    # Simple bar chart
    values = data.groupby(x_col)[y_col].mean().reindex(categories).values
    errors = None
    if error_col:
        errors = data.groupby(x_col)[error_col].mean().reindex(categories).values

    bars = ax.bar(x_pos, values,
                  color=COLORS[2],
                  edgecolor='black', linewidth=0.8,
                  yerr=errors, capsize=4,
                  error_kw={'linewidth': 1, 'color': 'black'})

    # Add value labels above bars (above error bars if present)
    if show_values:
        for bar, val, err in zip(bars, values, errors if errors is not None else [0]*len(values)):
            height = val + (err if err else 0) + (max(values) * 0.03)
            ax.text(bar.get_x() + bar.get_width()/2, height,
                    f'{val:.0f}%' if val < 100 else f'{val:.0f}',
                    ha='center', va='bottom', fontsize=9)

# Configure axes
ax.set_xticks(x_pos)
ax.set_xticklabels(categories, rotation=15 if len(str(categories[0])) > 4 else 0, ha='right' if len(str(categories[0])) > 4 else 'center')
ax.set_xlabel('{{xlabel}}' or '')
ax.set_ylabel('{{ylabel}}' or y_col)

if '{{title}}':
    ax.set_title('{{title}}')

if group_col:
    ax.legend()

# Adjust y-axis to fit labels
if show_values:
    ymax = ax.get_ylim()[1]
    ax.set_ylim(0, ymax * 1.1)

plt.tight_layout()

4. Save output

output_path = Path('{{output}}' or 'paper/figures/bar_chart')
output_path.parent.mkdir(parents=True, exist_ok=True)

for fmt in ['pdf', 'png']:
    fig.savefig(f"{output_path}.{fmt}")
    print(f"Saved: {output_path}.{fmt}")

plt.close()

Example Usage

/academic-figures:bar survey.csv --x category --y support_rate --error std_error --title "政策支持率" --ylabel "支持率 (%)"

Output

• Grayscale bar chart with optional error bars
• Value labels positioned above error bars (not overlapping)
• PDF and PNG formats