feat(bokeh): implement venn-labeled-items

[github-actions]

Implementation: venn-labeled-items - python/bokeh

Implements the python/bokeh version of venn-labeled-items.

File: plots/venn-labeled-items/implementations/python/bokeh.py

Parent Issue: #5364

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

Light render (plot-light.png): The plot renders on a warm off-white #FAF8F1 background. Three overlapping circles use Okabe-Ito colors in canonical order: green (#009E73) for "Overhyped" (top-left), orange/vermillion (#D55E00) for "Actually Useful" (top-right), and blue (#0072B2) for "Secretly Loved" (bottom-center). Each circle has a 22% alpha semi-transparent fill with a matching solid outline, creating clear visual overlap zones. Category names are rendered in large italic serif font outside each circle in the circle's own color. Fifteen item labels are placed as dark-ink serif text inside their assigned zones across all 7 regions. The title "Tech Vibes 2026 · venn-labeled-items · bokeh · anyplot.ai" sits at top, and an italic serif subtitle anchors the bottom. No axes or gridlines. All text is clearly readable against the warm off-white background — no light-on-light failures.

Dark render (plot-dark.png): The same diagram renders on a warm near-black #1A1A17 background. The three Okabe-Ito circle colors are identical to the light render — only chrome flips as required. The title and subtitle are rendered in light cream (#F0EFE8 / #B8B7B0), clearly readable against the dark surface. Category names retain their vibrant Okabe-Ito colors and are easily legible. Item labels inside the circles use INK = #F0EFE8 (light cream), appearing as white/light text on dark colored circle fills — readable throughout, including in the triple-overlap ABC zone where the compounded semi-transparent fills create a very dark region. No dark-on-dark failures detected. Brand green #009E73 is clearly visible in the Overhyped circle and its label.

Score: 86/100

Category	Score	Max
Visual Quality	26	30
Design Excellence	15	20
Spec Compliance	14	15
Data Quality	15	15
Code Quality	10	10
Library Mastery	6	10
Total	86	100

Visual Quality (26/30)

• VQ-01: Text Legibility (7/8) — All font sizes explicitly set (title 54pt, category names 64pt, items 38pt, subtitle 30pt). All readable in both themes. Item labels slightly small for the 3600×3600 canvas.
• VQ-02: No Overlap (5/6) — No visible overlaps. Minor tightness in AC zone (Crocs / Pumpkin Spice) but no actual collision.
• VQ-03: Element Visibility (5/6) — Circles well-sized, semi-transparent fills clear. Items in the triple-overlap zone slightly harder to distinguish on dark render due to compounded fills.
• VQ-04: Color Accessibility (2/2) — Okabe-Ito is CVD-safe; no red-green sole signal.
• VQ-05: Layout & Canvas (3/4) — Square 3600×3600 format appropriate. Diagram fills ~70% of canvas with title/subtitle framing. Minor empty space at sides.
• VQ-06: Axis Labels & Title (2/2) — No axis labels needed for Venn diagram. Title present and descriptive.
• VQ-07: Palette Compliance (2/2) — First circle #009E73, canonical Okabe-Ito order throughout. Backgrounds #FAF8F1 / #1A1A17. Theme-adaptive chrome correct in both renders.

Design Excellence (15/20)

• DE-01: Aesthetic Sophistication (6/8) — Strong editorial aesthetic: italic serif category labels in their circle colors, semi-transparent circle fills, witty subtitle, clean surface colors. Clearly above library defaults, magazine-print feel is convincingly achieved.
• DE-02: Visual Refinement (5/6) — No grid, no axes, outline_line_color=None, background and border fill unified. Excellent chrome minimalism.
• DE-03: Data Storytelling (4/6) — The zone choices tell a coherent cultural story (NFTs/Metaverse as purely Overhyped, Google Maps as purely Useful, Dolly Parton bridging Useful+Loved). Typography hierarchy (large italic category names vs. small item labels) naturally guides the viewer's reading order.

Spec Compliance (14/15)

• SC-01: Plot Type (5/5) — Three-circle symmetric Venn with labeled items — exactly as specified.
• SC-02: Required Features (4/4) — Semi-transparent fills, italic serif category labels outside circles, text-only item placement, all 7 zones populated, gridless editorial layout.
• SC-03: Data Mapping (3/3) — All 15 items placed in geometrically correct zones.
• SC-04: Title & Legend (2/3) — Title is Tech Vibes 2026 · venn-labeled-items · bokeh · anyplot.ai. Required format is {spec-id} · {library} · anyplot.ai. The witty editorial prefix "Tech Vibes 2026 ·" belongs in a separate subtitle Label, not prepended to the main title.

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) — All 7 zones populated: A×3, B×2, C×2, AB×2, AC×2, BC×2, ABC×2. 15 items total, within spec's 10–25 range.
• DQ-02: Realistic Context (5/5) — Culturally recognizable pop-culture items. Neutral — no politics, religion, or sensitive comparisons. NFTs/Metaverse/Web3 as "Overhyped" is a widely shared, non-partisan view.
• DQ-03: Appropriate Scale (4/4) — Categorical membership plot; zone distribution is balanced and appropriate.

Code Quality (10/10)

• CQ-01: KISS Structure (3/3) — Linear flow: imports → tokens → geometry → data → figure → render → save. No functions or classes.
• CQ-02: Reproducibility (2/2) — All data hardcoded. Fully deterministic.
• CQ-03: Clean Imports (2/2) — All five imports used.
• CQ-04: Code Elegance (2/2) — Clean, Pythonic geometry and iteration.
• CQ-05: Output & API (1/1) — Saves plot-{THEME}.png + plot-{THEME}.html. Current Bokeh API.

Library Mastery (6/10)

• LM-01: Idiomatic Usage (3/5) — Correct Bokeh API: figure(), ellipse(), add_layout(Label()), export_png, output_file/save. However, item labels are placed as individual Label objects rather than using ColumnDataSource + p.text() — the idiomatic Bokeh pattern for collections of same-type glyphs.
• LM-02: Distinctive Features (3/5) — HTML export alongside PNG and match_aspect=True are Bokeh-distinctive. Per-glyph Label styling is more expressive than matplotlib annotations. However, no HoverTool — which would be the most compelling showcase of Bokeh's interactive advantage over static libraries for this plot type.

Score Caps Applied

• None

Strengths

• Beautiful editorial magazine aesthetic: italic serif category labels in each circle's own Okabe-Ito color, semi-transparent fills, witty cultural subtitle
• All 7 Venn zones populated with culturally recognizable, non-controversial items (15 total)
• Perfect palette compliance: #009E73 → #D55E00 → #0072B2 canonical order; correct #FAF8F1/#1A1A17 theme surfaces; full theme-adaptive chrome
• Completely clean canvas — axes hidden, grid hidden, no outline border — appropriate editorial minimalism
• Perfect code quality: linear KISS structure, deterministic hardcoded data, clean imports, correct dual PNG+HTML output

Weaknesses

• Title format must be exactly {spec-id} · {library} · anyplot.ai — move "Tech Vibes 2026" to a separate Label subtitle; keep main title as venn-labeled-items · bokeh · anyplot.ai
• Item label font size (38pt) is slightly small for a 3600×3600 canvas — increase to 44–48pt for bolder readability at full resolution
• No HoverTool — adding one showing zone membership (e.g. "Overhyped + Actually Useful") would demonstrate Bokeh's unique interactive value versus static libraries
• Items placed as individual Label objects instead of ColumnDataSource + p.text() — the idiomatic Bokeh pattern for collections of text glyphs

Issues Found

1. SC-04 PARTIAL: Title prefix breaks required format
   • Fix: Set p.title.text = "venn-labeled-items · bokeh · anyplot.ai" and add a new Label for the witty editorial title (e.g. placed at y=2.5, large italic serif) as a visual header above the circles
2. LM-02 PARTIAL: No Bokeh-distinctive interactivity
   • Fix: Add HoverTool with a ColumnDataSource that includes zone and description columns; show zone membership on hover — this is the one feature that makes this Bokeh version genuinely superior to a static matplotlib version
3. VQ-01 MINOR: Item labels slightly small
   • Fix: Increase item text_font_size from "38pt" to "44pt" or "48pt"

AI Feedback for Next Attempt

Fix the title: p.title.text = "venn-labeled-items · bokeh · anyplot.ai" and add the witty editorial header as a separate Label object. Increase item font sizes to 44–48pt. Most importantly, add a HoverTool using ColumnDataSource for items — include zone and a short description column — this is what separates the Bokeh version from every static alternative and should push Library Mastery from 6 to 9. The editorial aesthetic is already excellent; these are the three targeted fixes needed to cross 90.

Verdict: REJECTED