feat(ggplot2): implement scatter-connected-temporal

[github-actions]

Implementation: scatter-connected-temporal - r/ggplot2

Implements the r/ggplot2 version of scatter-connected-temporal.

File: plots/scatter-connected-temporal/implementations/r/ggplot2.R

Parent Issue: #4675

---

🤖 impl-generate workflow

[claude]

AI Review - Attempt 1/3

Image Description

Light render (plot-light.png): Warm off-white #FAF8F1 background (correct). Bold title "Phillips Curve Dynamics · scatter-connected-temporal · r · ggplot2 · anyplot.ai" in dark ink spans roughly 75% of canvas width — expected for the long mandated title format. Axis labels "Unemployment Rate" (x) and "Inflation Rate" (y) in dark ink at 10pt, tick labels showing 4%/6%/8%/10% and 0%/5%/10%/15% in muted dark tone. Connected scatter path from 1980 (brand green #009E73) to 2019 (blue #4467A3) via the imprint_seq gradient; 40 circular markers visible; bold year annotations at 1980, 1990, 2000, 2010, 2019; horizontal colorbar "Year" legend at bottom. All text is clearly readable against the off-white background — no light-on-light failures.

Dark render (plot-dark.png): Warm near-black #1A1A17 background (correct). Title and axis labels rendered in light cream/near-white tone, clearly readable against the dark surface. Tick labels and legend text in muted light tone (#B8B7B0), all legible. Year annotation labels switch to light color — no dark-on-dark failures. Data gradient colors are identical to the light render (brand green #009E73 to blue #4467A3) — only chrome flipped. Brand green #009E73 remains clearly visible on the dark surface. Both renders pass the theme-readability check.

Score: 87/100

Category	Score	Max
Visual Quality	28	30
Design Excellence	13	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	9	10
Library Mastery	7	10
Total	87	100

Visual Quality (28/30)

• VQ-01: Text Legibility (7/8) — All sizes explicitly set; title scaled via formula (10pt for 81-char string); axis labels 10pt, ticks 8pt, year annotations 3.2mm bold — all readable in both themes; title slightly compact at 10pt vs ideal 12pt default
• VQ-02: No Overlap (5/6) — No hard overlaps; minor crowding between "2019" and "2000" labels in the lower-left cluster (both near 3.5–4% unemployment)
• VQ-03: Element Visibility (6/6) — 40 points at size=2.8 with alpha=0.95 well-adapted to data density; path at linewidth=1.0 clearly visible
• VQ-04: Color Accessibility (2/2) — imprint_seq gradient (green→blue) is perceptually safe under CVD; luminance varies across the range
• VQ-05: Layout & Canvas (4/4) — 3200×1800 canvas confirmed; good margins; colorbar at bottom well-placed; no overflow or clipping
• VQ-06: Axis Labels & Title (2/2) — "Unemployment Rate" and "Inflation Rate" with percentage units via tick labels
• VQ-07: Palette Compliance (2/2) — Continuous data correctly uses imprint_seq (#009E73 to #4467A3); backgrounds #FAF8F1 / #1A1A17; chrome tokens flip correctly

Design Excellence (13/20)

• DE-01: Aesthetic Sophistication (5/8) — Above configured-default level: intentional gradient encoding for temporal direction, descriptive title prefix, clean chrome setup; not yet publication-ready — no start-path arrow, no additional visual emphasis beyond color gradient
• DE-02: Visual Refinement (4/6) — Good: L-shaped spine frame (panel.border removed + axis.line), minor grids removed, explicit generous margins, legend on ELEVATED_BG; grid could be subtler (linewidth 0.15 or rgba opacity string)
• DE-03: Data Storytelling (4/6) — Year annotations at economic milestones create narrative waypoints; color gradient guides eye through time; 1980 outlier naturally prominent; title signals the Phillips curve context

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) — Correct connected scatter: geom_path + geom_point
• SC-02: Required Features (4/4) — All features present: chronological path, visible markers, key-year annotations, temporal direction via color gradient
• SC-03: Data Mapping (3/3) — X=unemployment, Y=inflation, time=color; all 40 points (1980–2019) visible
• SC-04: Title & Legend (3/3) — "Phillips Curve Dynamics · scatter-connected-temporal · r · ggplot2 · anyplot.ai" ✓; Year colorbar appropriate

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) — Full 40-year path with multiple regime changes; inflation spans -0.4% to 13.5%; path crosses multiple regions
• DQ-02: Realistic Context (5/5) — Phillips curve is a canonical economic concept; neutral real-world scenario
• DQ-03: Appropriate Scale (4/4) — Values consistent with historical US economic data

Code Quality (9/10)

• CQ-01: KISS Structure (3/3) — Linear: imports → tokens → data → plot → save
• CQ-02: Reproducibility (2/2) — set.seed(42); hardcoded data is fully deterministic
• CQ-03: Clean Imports (1/2) — library(dplyr) imported but never used; all data manipulation uses base R
• CQ-04: Code Elegance (2/2) — Clean and readable; title font-size formula is well-applied; guide_colorbar customization appropriate
• CQ-05: Output & API (1/1) — sprintf("plot-%s.png", THEME) → correct output filenames; ragg::agg_png used correctly

Library Mastery (7/10)

• LM-01: Idiomatic Usage (4/5) — Correct use of geom_path() for ordered connected scatter (distinguishable from geom_line), proper aes(color=year) continuous mapping, scale_color_gradient with guide_colorbar, expansion()
• LM-02: Distinctive Features (3/5) — geom_path() for data-order connection, label_number(suffix="%") from scales, expansion(mult=...) for asymmetric axis padding — above generic default

Score Caps Applied

• None

Strengths

• Correct connected scatter using geom_path + geom_point with imprint_seq gradient (green-to-blue) for temporal direction encoding
• Full spec compliance: chronological path, visible markers, key-year text annotations, and colorbar legend all present
• Both themes implemented correctly with proper chrome token adaptation — no dark-on-dark issues
• Descriptive title prefix "Phillips Curve Dynamics" adds economic context
• Title font-size scaling formula correctly applied for the long mandated string

Weaknesses

• library(dplyr) is imported but never used — all data manipulation uses base R; remove this unused import
• Design sophistication at configured-default level: add an explicit start-of-path arrow (geom_segment with arrow=arrow()) pointing from the 1980 origin to reinforce temporal direction beyond color alone; increase geom_point size slightly (3.5) for more prominent year markers
• Minor label crowding in the lower-left cluster: "2019" and "2000" labels are close together — adjust nudge_x=-0.3 for 2019 or increase nudge_y separation to prevent visual crowding
• Grid lines use INK_MUTED at linewidth=0.2 which is functional but slightly prominent — reduce to linewidth=0.15 for subtler grid consistent with style-guide

Issues Found

1. CQ-03 MINOR: Unused import library(dplyr) — all data operations are base R
   • Fix: Remove library(dplyr) from imports
2. DE-01 BELOW TARGET: Design is above generic defaults but needs stronger visual emphasis to reach publication quality
   • Fix: Add geom_segment with arrow from the 1980 start point indicating the direction of the temporal path; slightly increase point size to 3.5
3. VQ-02 MINOR: "2019" and "2000" labels cluster in the bottom-left area
   • Fix: For year 2019 label, use nudge_x = -0.35, nudge_y = -0.3 to offset it below-left of the 2000 label rather than above-right

AI Feedback for Next Attempt

Remove unused library(dplyr). Add a start-path directional arrow using geom_segment(data=df[1,], aes(x=unemployment-0.3, y=inflation, xend=unemployment-0.05, yend=inflation), arrow=arrow(length=unit(0.15,'cm')), color=IMPRINT_PALETTE[1], linewidth=0.8) to reinforce temporal direction beyond the color gradient. Increase geom_point size to 3.5. Adjust the 2019 year label nudge to nudge_x=-0.35, nudge_y=-0.3 to separate it from the 2000 label. Reduce panel.grid.major linewidth to 0.15 for subtler grid lines.

Verdict: REJECTED

[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): Warm off-white background (#FAF8F1). Title "Phillips Curve Dynamics · scatter-connected-temporal · r · ggplot2 · anyplot.ai" renders at ~10pt (correctly scaled from default 12pt for the 80-char string), bold and dark — readable. X-axis "Unemployment Rate" and Y-axis "Inflation Rate" at 10pt in dark ink; tick labels with % suffixes at 8pt. The connected scatter path and points graduate from brand green #009E73 (1980, upper-right cluster of high inflation/varying unemployment) through teal to blue #4467A3 (recent years, lower-left cluster). A small arrow at the 1980→1981 segment reinforces the temporal direction. Year labels "1980", "1990", "2010" are well-spaced; "2000" and "2019" are nearly co-located and visually stack, nearly overlapping. Horizontal colorbar legend at bottom shows green-to-blue "Year" gradient. All text is readable against the light background.

Dark render (plot-dark.png): Warm near-black background (#1A1A17). Title, axis labels, tick labels, and legend text all render in light ink (#F0EFE8 / #B8B7B0) — clearly readable against the dark surface. No dark-on-dark failures observed. Data colors are identical to the light render — the green-to-blue gradient and the arrow overlay are pixel-perfect matches between themes; only the chrome (background, text, legend box) flipped. The "2000"/"2019" label proximity issue is equally present in this render. All text is readable against the dark background.

Both paragraphs are required. A review that only describes one render is invalid.

Score: 89/100

Category	Score	Max
Visual Quality	26	30
Design Excellence	14	20
Spec Compliance	15	15
Data Quality	15	15
Code Quality	10	10
Library Mastery	9	10
Total	89	100

Visual Quality (26/30)

• VQ-01: Text Legibility (7/8) — All sizes explicitly set (axis.title=10, axis.text=8, title=10 via scaling formula, legend.text=8, legend.title=9). Readable in both themes. Minor deduction for the "2000"/"2019" crowding creating a readability concern.
• VQ-02: No Overlap (3/6) — "2000" and "2019" labels are nearly co-located: lbl_y = 3.1 vs 3.0 on a 17%-range inflation axis (≈8 source-px gap vs ≈50px text height at size=3.2). Labels overlap in both renders.
• VQ-03: Element Visibility (6/6) — size=3.5 markers at alpha=0.95 with linewidth=1.0 path — well-adapted for 40 data points.
• VQ-04: Color Accessibility (2/2) — Green-to-blue imprint_seq gradient is CVD-safe; no red-green reliance.
• VQ-05: Layout & Canvas (4/4) — 3200×1800 landscape (8×4.5in, dpi=400). Good canvas utilization, bottom legend, generous margins.
• VQ-06: Axis Labels & Title (2/2) — Descriptive "Unemployment Rate" / "Inflation Rate"; % shown via label_number(suffix="%") on tick labels.
• VQ-07: Palette Compliance (2/2) — imprint_seq gradient (low="#009E73", high="#4467A3") correct for continuous temporal variable. Backgrounds #FAF8F1 / #1A1A17 set. All chrome theme-adaptive.

Design Excellence (14/20)

• DE-01: Aesthetic Sophistication (6/8) — Strong design: the imprint_seq gradient along path and points is visually striking and informative, clearly above library defaults. Clean minimal chrome, thoughtful arrow cue.
• DE-02: Visual Refinement (4/6) — Subtle grid (linewidth=0.15 with INK_MUTED), minor grid disabled, panel border removed, axis lines added for L-shape frame. Good refinement visible.
• DE-03: Data Storytelling (4/6) — Gradient + arrow + year annotations create a clear temporal narrative. The viewer is guided from the high-inflation 1980s through to the stable 2010s. Good visual hierarchy through color contrast.

Spec Compliance (15/15)

• SC-01: Plot Type (5/5) — Correct connected scatter with temporal path: geom_path for connections, geom_point for markers.
• SC-02: Required Features (4/4) — All required features present: chronological path (geom_path), key time-point annotations (1980/1990/2000/2010/2019), direction arrow (geom_segment with arrow), temporal color gradient, visible markers.
• SC-03: Data Mapping (3/3) — X=unemployment rate, Y=inflation rate, time encoded via gradient. Axes show full data range.
• SC-04: Title & Legend (3/3) — "Phillips Curve Dynamics · scatter-connected-temporal · r · ggplot2 · anyplot.ai" matches required format with optional descriptive prefix. Year colorbar legend appropriate.

Data Quality (15/15)

• DQ-01: Feature Coverage (6/6) — Shows full connected-scatter complexity: looping paths, direction reversals, clustering, 1980 high-inflation outlier, 2008-09 recession excursion.
• DQ-02: Realistic Context (5/5) — Classic Phillips curve (US unemployment vs inflation, 1980–2019). Real, comprehensible, neutral economic data.
• DQ-03: Appropriate Scale (4/4) — Unemployment 3.5–9.7%, inflation −0.4% to 13.5% — accurate historical US values.

Code Quality (10/10)

• CQ-01: KISS Structure (3/3) — Clean linear flow: theme tokens → data → plot layers → save. No functions or classes.
• CQ-02: Reproducibility (2/2) — set.seed(42) present; data is hardcoded historical values (fully deterministic).
• CQ-03: Clean Imports (2/2) — ggplot2, scales (label_number), ragg (agg_png) — all three used.
• CQ-04: Code Elegance (2/2) — Per-label nudge offsets are thoughtful. Title-size scaling formula is appropriate. No over-engineering.
• CQ-05: Output & API (1/1) — sprintf("plot-%s.png", THEME) → correct dual-theme output. ragg::agg_png device used.

Library Mastery (9/10)

• LM-01: Idiomatic Usage (5/5) — Expert ggplot2: aes(color=year) with scale_color_gradient for continuous mapping on path AND points simultaneously, guide_colorbar customization, geom_path (not geom_line) for temporal ordering, correct theme_minimal inheritance.
• LM-02: Distinctive Features (4/5) — geom_path temporal ordering, continuous aesthetic mapped across multiple geom layers, guide_colorbar with custom dimensions, geom_segment arrow — all distinctively ggplot2 grammar-of-graphics patterns. Minor deduction: no faceting or advanced grammar feature.

Score Caps Applied

• None

Strengths

• Correct geom_path() (not geom_line) preserves temporal order — idiomatic ggplot2 for connected scatter
• imprint_seq gradient (green #009E73 → blue #4467A3) elegantly encodes temporal progression along both path and points simultaneously
• Arrow overlay (geom_segment with arrow parameter) reinforces time direction without clutter
• Full theme-adaptive chrome: PAGE_BG, INK, INK_SOFT, INK_MUTED all correctly toggled between themes
• Title font scaling formula correctly handles the 80-char mandated title (scaled to 10pt)
• Real historical Phillips curve data (1980–2019) with accurate economic values provides compelling context
• guide_colorbar with barwidth/barheight customization makes the continuous legend compact

Weaknesses

• Label overlap: "2000" and "2019" are nearly co-located (lbl_y = 3.1 vs 3.0 — only 0.1% separation on a 17%-range axis ≈ 8 source-px gap vs ≈50px text height). Fix: increase y-offset for "2019" to at least inflation + 1.5, or nudge one label in x.
• Colorbar legend truncated: ticks show 1980, 1990, 2000, 2010 but not 2019. Minor cosmetic: widen barwidth or set explicit breaks on scale_color_gradient to include 2019.

Issues Found

1. VQ-02 MODERATE: "2000" and "2019" year labels overlap
   • df_labels$lbl_x for both = ~3.7; df_labels$lbl_y = 3.1 vs 3.0 — effectively same position at the rendered scale
   • Fix: change 2019 offset to c(0.2, 0.2, -0.3, 0.2, -0.3) for x and c(0.6, 0.5, 0.5, 0.5, -0.5) for y (push 2019 below its data point instead of above), or simply use c(0.5, 0.3, 2.5, 0.2, 1.0) for y-offsets to spread the labels

AI Feedback for Next Attempt

The implementation is high quality and approved at this attempt. If a repair is triggered anyway, focus on: (1) spreading the "2000" and "2019" year labels — nudge "2019" downward (negative y-offset) rather than upward, since both 2000 and 2019 points are in the same low-unemployment low-inflation cluster; (2) add breaks = c(1980, 1990, 2000, 2010, 2019) to scale_color_gradient to show the full year range in the colorbar.

Verdict: APPROVED