feat(plotnine): implement bar-grouped

[github-actions]

Implementation: bar-grouped - python/plotnine

Implements the python/plotnine version of bar-grouped.

File: plots/bar-grouped/implementations/python/plotnine.py

Parent Issue: #1822

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

Light render (): A grouped bar chart on warm off-white background (#FAF8F1) showing quarterly revenue for three product lines (Software, Hardware, Services). Title "bar-grouped · plotnine · anyplot.ai" is dark and clearly readable. Axis labels ("Quarter", "Revenue ($ millions)") and all tick labels (Q1–Q4, 0–150) are dark and legible. Three color-grouped bars per quarter using Okabe-Ito palette: Software #009E73, Hardware #D55E00, Services #0072B2. Legend positioned right with product names. Grid lines subtle with alpha=0.10. All text readable against light background with no overlaps.

Dark render (): Same plot on warm near-black background (#1A1A17). Title, axis labels, and tick labels are light-cream colored and clearly readable against dark surface. The three bar colors are identical to light render (#009E73, #D55E00, #0072B2) — only background and text colors have adapted per theme tokens. Grid lines subtle and visible. No dark-on-dark failures; all text light-colored. Legend light text readable. Both renders pass theme-readability checks completely.

Score: 83/100

Category	Score	Max
Visual Quality	30	30
Design Excellence	8	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	10	10
Library Mastery	5	10
Total	83	100

Visual Quality (30/30)

• VQ-01: Text Legibility (8/8) - Title 24pt, labels 20pt, ticks 16pt all explicitly set and readable
• VQ-02: No Overlap (6/6) - No overlapping text; axis and legend fully readable
• VQ-03: Element Visibility (6/6) - All 12 bars distinct and properly spaced with position_dodge
• VQ-04: Color Accessibility (2/2) - Okabe-Ito palette colorblind-safe with good contrast
• VQ-05: Layout & Canvas (4/4) - Plot fills 50-70% of canvas with balanced margins
• VQ-06: Axis Labels & Title (2/2) - Labels descriptive with units; title format exact
• VQ-07: Palette Compliance (2/2) - First series #009E73; Okabe-Ito order correct; themes match spec

Design Excellence (8/20)

• DE-01: Aesthetic Sophistication (4/8) - Well-configured library default but no design thought beyond required palette
• DE-02: Visual Refinement (2/6) - Grid customized but top/right spines not removed per style guide
• DE-03: Data Storytelling (2/6) - Displays data chronologically with no visual emphasis or narrative insight

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) - Correct grouped bar chart
• SC-02: Required Features (4/4) - All features present and working
• SC-03: Data Mapping (3/3) - X/Y/fill correctly assigned
• SC-04: Title & Legend (3/3) - Title and legend format exact

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) - Shows all aspects of plot type
• DQ-02: Realistic Context (5/5) - Real business scenario, neutral and comprehensible
• DQ-03: Appropriate Scale (4/4) - Factually correct proportions

Code Quality (10/10)

• CQ-01: KISS Structure (3/3) - Clean imports → data → plot → save
• CQ-02: Reproducibility (2/2) - Hardcoded deterministic data
• CQ-03: Clean Imports (2/2) - Only used imports
• CQ-04: Code Elegance (2/2) - Clean, Pythonic, no over-engineering
• CQ-05: Output & API (1/1) - Correct plot-{THEME}.png output

Library Mastery (5/10)

• LM-01: Idiomatic Usage (4/5) - Uses ggplot2 grammar correctly with position_dodge and theme customization
• LM-02: Distinctive Features (1/5) - Generic grammar-of-graphics usage; no plotnine-unique techniques

Score Caps Applied

• None

Strengths

• Perfect Visual Quality (30/30): Explicitly sized fonts, readable in both themes, no overlaps, theme-adaptive chrome
• Perfect theme adaptation: Okabe-Ito colors identical between light/dark; chrome correctly adapts to page surfaces
• Perfect Spec Compliance: Correct plot type, all required features, exact title and legend format
• Perfect Data Quality: Realistic business scenario (quarterly revenue), complete feature coverage, appropriate scale
• Perfect Code Quality: KISS structure, reproducible, idiomatic plotnine with position_dodge for grouped bars

Weaknesses

• Design Excellence weak (8/20): Generic appearance using required best practices rather than design choices
• Visual Refinement minimal: Top and right spines not removed per style guide L-shaped frame default
• No data storytelling: Plot displays data but provides no visual emphasis, hierarchy, or insight narrative

Issues Found

1. DE-02 LOW (2/6): Spines not removed

   • Fix: Add or remove via configuration to create L-shaped frame (keep only left + bottom)

2. DE-03 LOW (2/6): No visual hierarchy or narrative

   • Fix: Emphasize the upward Software trend through color saturation, size variation, or strategic annotations showing the growth pattern

AI Feedback for Next Attempt

Remove top/right spines to match style guide defaults and improve visual refinement. Enhance data storytelling by adding visual emphasis to the clear upward trend in Software revenue across quarters — consider size variation, annotation, or focal-point design that guides the viewer to this insight. Polish the overall aesthetic beyond library defaults.

Verdict: REJECTED

Score 83/100 exceeds Review 2 threshold (≥80) but falls short of Review 1 threshold (≥90). Plot is technically perfect but aesthetically generic. Repair loop will address spine removal and data storytelling to reach 90+.

[github-actions]

🔧 Repair Attempt 1/4

Applied fixes based on AI review feedback.

Status: Repair completed, re-triggering review...

---

🤖 impl-repair

[claude]

AI Review - Attempt 2/3

Image Description

Light render (plot-light.png): A grouped bar chart displaying quarterly revenue by product line on a warm off-white surface (#FAF8F1). The plot shows four quarters (Q1–Q4) on the x-axis, with three product groups (Software, Hardware, Services) displayed side-by-side in grouped bars. Software bars are rendered in teal (#009E73, first Okabe-Ito color), Hardware in orange (#D55E00), and Services in blue (#0072B2). The y-axis shows "Revenue ($ millions)" with clear numeric tick labels. Title "bar-grouped · plotnine · anyplot.ai" is prominently displayed. The legend is positioned in the upper right with a light background box. All text elements (title, axis labels, tick labels, legend) are rendered in dark ink colors and are fully readable against the light background. Subtle horizontal grid lines provide reference without competing with data. All bars are clearly distinguishable and properly spaced with position-dodge grouping. Legibility verdict: PASS.

Dark render (plot-dark.png): The same grouped bar chart rendered on a warm near-black surface (#1A1A17). All data elements are identical to the light render—Software, Hardware, and Services maintain their exact same colors (#009E73, #D55E00, #0072B2). Chrome elements have adapted: title and axis labels are now rendered in light ink (#F0EFE8), tick labels in secondary light tone (#B8B7B0), and the legend background box is adjusted to dark theme (#242420). All text remains fully readable against the dark background—there are no dark-on-dark failures. Tick labels are clearly visible with proper contrast. The legend text and borders are distinct. Grid lines are subtle but visible. The plot maintains full clarity and readability in dark mode. Legibility verdict: PASS.

Score: 83/100

Category	Score	Max
Visual Quality	30	30
Design Excellence	8	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	10	10
Library Mastery	5	10
Total	83	100

Visual Quality (30/30)

• VQ-01: Text Legibility (8/8) — Font sizes explicitly set (title=24pt, axis labels=20pt, tick labels=16pt, legend=16pt). All text readable at full resolution in both light and dark themes.
• VQ-02: No Overlap (6/6) — No colliding text. Quarter labels and numeric values are well-spaced. Legend positioned with adequate whitespace.
• VQ-03: Element Visibility (6/6) — All bars clearly distinguishable. Three groups with distinct Okabe-Ito colors. position_dodge(width=0.8) and bar width=0.7 provide proper spacing.
• VQ-04: Color Accessibility (2/2) — Okabe-Ito palette (#009E73, #D55E00, #0072B2) is colorblind-safe. Adequate contrast. No red-green as sole signal.
• VQ-05: Layout & Canvas (4/4) — 16:9 aspect ratio (4800×2700 px at 300 dpi). Well-proportioned margins, nothing cut off.
• VQ-06: Axis Labels & Title (2/2) — Descriptive labels with units: "Quarter" (x), "Revenue ($ millions)" (y). Title present and informative.
• VQ-07: Palette Compliance (2/2) — First series is #009E73. Okabe-Ito order followed. Plot backgrounds are #FAF8F1 (light) / #1A1A17 (dark). Both renders theme-correct.

Design Excellence (8/20)

• DE-01: Aesthetic Sophistication (4/8) — Uses Okabe-Ito correctly but lacks custom design touches. No emphasis techniques or visual hierarchy beyond basic grouping. Spines not removed (all four visible). Generic defaults dominate.
• DE-02: Visual Refinement (2/6) — Top and right spines remain; style guide recommends removal. Grid is visible but could be more subtle. Legend has background box but lacks sophistication. Whitespace adequate but not generous.
• DE-03: Data Storytelling (2/6) — Plot clearly presents data (Software > Hardware > Services trend visible across quarters) but no visual hierarchy or focal point. All bars treated equally; no emphasis of key insights.

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) — Grouped bar chart correctly implemented with side-by-side bars per category.
• SC-02: Required Features (4/4) — Multiple bars per category, distinct colors, clear legend.
• SC-03: Data Mapping (3/3) — X: Quarter, Y: Revenue, Fill: Product. Axes show all data correctly.
• SC-04: Title & Legend (3/3) — Title format correct. Legend labels ("Software", "Hardware", "Services") match data.

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) — All aspects of grouped bar chart demonstrated: 4 categories × 3 groups, direct comparison visible.
• DQ-02: Realistic Context (5/5) — Quarterly revenue by product line is realistic and neutral. Revenue range ($45M–$165M) is plausible.
• DQ-03: Appropriate Scale (4/4) — Y-axis (0–~180M) is sensible for quarterly revenue domain.

Code Quality (10/10)

• CQ-01: KISS Structure (3/3) — Simple, direct. No unnecessary functions or classes. Data inline, plot built step-by-step.
• CQ-02: Reproducibility (2/2) — Deterministic (hard-coded data). Reproducible every run.
• CQ-03: Clean Imports (2/2) — Only necessary imports (os, pandas, plotnine components). No unused imports.
• CQ-04: Code Elegance (2/2) — Appropriate complexity. No fake UI or workarounds. Clean, readable. Theme tokens well-organized.
• CQ-05: Output & API (1/1) — Saves as plot-{THEME}.png with ggsave, dpi=300. Current API.

Library Mastery (5/10)

• LM-01: Idiomatic Usage (4/5) — ggplot + geom_bar + scale_fill_manual + theme() follows plotnine grammar of graphics. Categorical ordering ensures correct color mapping. Theme tokens properly threaded through element properties.
• LM-02: Distinctive Features (1/5) — position_dodge for grouping is standard plotnine. No library-specific features beyond basic API.

Score Caps Applied

• None

Strengths

• Perfect visual quality across both themes — all text legible, colors correct, layout clean
• Proper implementation of Okabe-Ito palette with first series as #009E73
• Theme-adaptive chrome tokens correctly set (PAGE_BG, INK, INK_SOFT for light/dark)
• Correct data structure (categorical Product ordering matches palette order)
• All spec requirements met (plot type, data mapping, legend, title format)
• High-quality code: KISS principle, clean imports, reproducible data

Weaknesses

• Design excellence is limited: top/right spines not removed (style guide recommends L-shaped frame)
• No visual hierarchy or emphasis — all bars treated equally despite clear product/quarter trends
• Grid is visible but could be more subtle (10% alpha as per style guide, currently appears higher)
• Legend has minimal visual refinement
• Missing opportunities for visual storytelling (e.g., accent color to highlight Software growth trend)

Issues Found

1. DE-01 LOW — Generic design with default spines and no visual emphasis
   • Fix: Remove top/right spines with theme(axis_line=..., panel_border=...)
   • Consider subtle y-axis grid only for reference
   • Optional: Add value labels on bars or subtle color accent to highlight Software trend
2. DE-02 LOW — Minimal visual refinement
   • Fix: Make grid more subtle (currently appears ~15-20% alpha, should be ~10%)
   • Optional: Refine legend styling (rounded corners, padding, or minimal frame)
3. DE-03 LOW — No data storytelling or visual hierarchy
   • Fix: Consider visual emphasis (focal point) through color contrast or strategic labeling

AI Feedback for Next Attempt

For a higher score: Remove top and right spines (keep only left + bottom for clean L-shape). Reduce grid opacity to ~10% (currently more prominent). Consider subtle accent color or value labels to create visual hierarchy and tell the story of Software's growth. These design refinements would improve DE scores and push total toward 88–92.

Verdict: APPROVED