feat(letsplot): implement bar-3d

[github-actions]

Implementation: bar-3d - letsplot

Implements the letsplot version of bar-3d.

File: plots/bar-3d/implementations/letsplot.py

Parent Issue: #2857

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

The plot displays a 3D bar chart showing quarterly revenue data ($M) across five product categories: Electronics, Clothing, Home, Sports, and Books. The bars are rendered as 3D boxes using an isometric projection, with colors mapped to revenue values using a viridis color scale (purple/dark blue for lower values ~20, through teal/green, to yellow for higher values ~65). The chart includes a floor grid, axis indicator lines in the bottom-left corner, product category labels on the left side, quarter labels (Q1-Q4) at the bottom front, and "Revenue ($M)" label on the z-axis. The title "bar-3d · letsplot · pyplots.ai" appears in the top-left. A legend on the right shows the revenue color scale.

Quality Score: 82/100

Criteria Checklist

Visual Quality (32/40 pts)

• VQ-01: Text Legibility (7/10) - Title and legend text are readable; category and quarter labels are visible but somewhat small relative to the canvas
• VQ-02: No Overlap (6/8) - Product and quarter labels overlap/crowd together in the isometric projection area; "Sports" and "Books" labels overlap with the bars
• VQ-03: Element Visibility (7/8) - Bars are well-sized and clearly visible with good alpha for depth
• VQ-04: Color Accessibility (5/5) - Viridis colormap is colorblind-safe with excellent contrast
• VQ-05: Layout Balance (4/5) - Good use of canvas space; slight imbalance with empty area on far right
• VQ-06: Axis Labels (2/2) - "Revenue ($M)" label present with units
• VQ-07: Grid & Legend (1/2) - Floor grid is subtle (good), but legend is positioned with some isolation from main plot area

Spec Compliance (23/25 pts)

• SC-01: Plot Type (8/8) - Correct 3D bar chart implemented with isometric projection
• SC-02: Data Mapping (5/5) - X (quarters), Y (products), Z (revenue) correctly mapped
• SC-03: Required Features (4/5) - Has depth, color gradient reinforcement, but lacks interactive rotation (spec notes it as beneficial)
• SC-04: Data Range (3/3) - All bars and data fully visible
• SC-05: Legend Accuracy (2/2) - Legend correctly shows "Revenue ($M)" scale
• SC-06: Title Format (2/2) - Uses correct format "bar-3d · letsplot · pyplots.ai"

Data Quality (17/20 pts)

• DQ-01: Feature Coverage (6/8) - Shows variation across products and quarters; demonstrates comparison capability but all categories show similar upward Q4 trend
• DQ-02: Realistic Context (7/7) - Quarterly sales by product category is a real, neutral business scenario
• DQ-03: Appropriate Scale (4/5) - Revenue values ($15M-$65M) are plausible for business context

Code Quality (6/10 pts)

• CQ-01: KISS Structure (1/3) - Uses a helper function project_3d_to_2d() which violates KISS principle (no functions/classes rule)
• CQ-02: Reproducibility (3/3) - Uses np.random.seed(42) and deterministic data
• CQ-03: Clean Imports (2/2) - All imports are used
• CQ-04: No Deprecated API (0/1) - Uses deprecated ggsave import path (should be from lets_plot.export)
• CQ-05: Output Correct (0/1) - Saves correctly but code actually does use correct import path

Library Features (4/5 pts)

• LF-01: Uses distinctive library features (4/5) - Creative use of geom_polygon for 3D effect, layer_tooltips for interactivity, scale_fill_viridis, theme customization

Strengths

• Creative implementation of 3D bars using polygon-based isometric projection in a 2D library
• Effective use of viridis color scale to reinforce z-values for depth perception
• Interactive tooltips provide hover information (Product, Quarter, Revenue)
• Clean floor grid and axis indicators aid depth perception
• Proper depth sorting ensures correct rendering order (back-to-front)

Weaknesses

• Uses helper function project_3d_to_2d() which violates KISS code structure requirement
• Product and quarter labels overlap with bar positions in the crowded center area
• Some labels ("Sports", "Books") are difficult to read against the bars
• Fixed perspective angle - no interactive rotation despite spec noting it as beneficial

Verdict: REJECTED

[github-actions]

🔧 Repair Attempt 1/3

Applied fixes based on AI review feedback.

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

---

🤖 impl-repair

[claude]

AI Review - Attempt 2/3

Image Description

The plot displays a 3D bar chart with an isometric perspective. The title "bar-3d · letsplot · pyplots.ai" appears in bold at the top left. Twenty 3D bars are arranged in a 5x4 grid representing quarterly sales (Q1-Q4) across five product categories (Electronics, Clothing, Home, Sports, Books). Each bar uses the viridis color scale (purple→teal→green→yellow) to encode revenue values ranging from ~15M to 65M. The bars have three visible faces (front, top, right) with dark outlines creating clear 3D definition. A subtle gray floor grid provides depth reference. Product labels are stacked vertically on the left, quarter labels (Q1-Q4) appear along the front edge, and "Revenue ($M)" labels the z-axis. A color legend on the right maps the gradient to revenue values.

Quality Score: 91/100

Criteria Checklist

Visual Quality (36/40 pts)

• VQ-01: Text Legibility (10/10) - Title at 32pt, labels at 11-12pt are all clearly readable at full resolution
• VQ-02: No Overlap (8/8) - All labels well separated, no text overlap
• VQ-03: Element Visibility (7/8) - Bars have good size and 0.85 alpha, depth sorting works well; minor deduction for some back bars partially occluded
• VQ-04: Color Accessibility (5/5) - Viridis is colorblind-safe with excellent perceptual uniformity
• VQ-05: Layout Balance (4/5) - Good canvas utilization, slight asymmetry with labels on left
• VQ-06: Axis Labels (2/2) - "Revenue ($M)" has units, descriptive category labels
• VQ-07: Grid & Legend (0/2) - Floor grid is subtle at alpha 0.5, legend well placed; however axis lines are somewhat hidden

Spec Compliance (25/25 pts)

• SC-01: Plot Type (8/8) - Correct 3D bar chart with bars rising from a 2D plane
• SC-02: Data Mapping (5/5) - X=quarters, Y=products, Z=revenue correctly assigned
• SC-03: Required Features (5/5) - Two categorical dimensions, numeric height, color gradient for depth perception
• SC-04: Data Range (3/3) - All 20 bars fully visible with no clipping
• SC-05: Legend Accuracy (2/2) - Color legend accurately shows Revenue ($M) range
• SC-06: Title Format (2/2) - "bar-3d · letsplot · pyplots.ai" matches required format

Data Quality (18/20 pts)

• DQ-01: Feature Coverage (7/8) - Shows variation across quarters and products, Electronics Q4 peak at 65M, Books consistently lower; could show more extreme variation
• DQ-02: Realistic Context (7/7) - Quarterly sales by product category is a realistic business analytics scenario
• DQ-03: Appropriate Scale (4/5) - Revenue values 15-65M are plausible; good variation but range could be wider

Code Quality (9/10 pts)

• CQ-01: KISS Structure (3/3) - Linear flow: imports → data → projection math → plot → save
• CQ-02: Reproducibility (3/3) - Uses np.random.seed(42)
• CQ-03: Clean Imports (2/2) - All imports are used (pandas, numpy, lets_plot components)
• CQ-04: No Deprecated API (1/1) - Current lets-plot API
• CQ-05: Output Correct (0/1) - Saves both plot.png and plot.html (HTML is a bonus, not a defect)

Library Features (3/5 pts)

• LF-01: Uses distinctive library features (3/5) - Good use of geom_polygon, layer_tooltips for interactivity, scale_fill_viridis, theme customization; could better utilize lets-plot's native grammar of graphics features

Strengths

• Excellent 3D isometric projection with clear depth perception using geom_polygon for bar faces
• Good use of viridis color scale reinforcing z-values for colorblind accessibility
• Clean separation of product categories and quarters with readable labels
• Interactive tooltips provide rich data exploration when viewing HTML export
• Well-structured code with proper depth sorting for correct face rendering

Weaknesses

• Product labels on left side appear stacked vertically rather than aligned with their corresponding bar rows in the 3D space

Verdict: APPROVED