Replace copyrighted slide figures with freely-licensed equivalents

[d-morrison]

Summary

This PR consolidates several changes that accumulated alongside the introductory lecture slides work in PR #527. The primary purpose is replacing unlicensed figures, but additional bug fixes and CI improvements are included.

1. Replace copyrighted slide figures (main purpose)

Three figures in the methodology lecture deck either carried explicit copyright notices or came from unclear sources.

Deleted (no reuse rights):

File	Source
41579_2024_1126_Fig1_HTML.png	© Springer Nature — Nature Reviews Microbiology (2024)
340149_1_En_16_Fig6_HTML.webp	© Springer Nature — book chapter
burden-typhoid-vaccines.png	mapchart.net, unclear license

Replacements & slide changes:

• Shigella slide — Springer NRM choropleth removed; Shigella_stool.jpg (CDC/PHIL, public domain) kept with explicit attribution caption.
• Scrub-typhus slide — Springer range map removed; Trombicula-larva-stylostome.jpg (CC BY-SA 3.0, Alan R Walker) kept with proper attribution.
• Burden-data-gap slide — burden-typhoid-vaccines.png replaced by an R code chunk generating a TCV-adoption world map using map_data("world") (ggplot2/maps) and the CDC MMWR 72(7):182–188 (2023) country list (US federal government work, public domain).

Caption fixes for existing Wikimedia Commons images:

• Fievre_typhoide.png was miscaptioned as "historic typhoid fever illustration" — it is a typhoid-distribution world map (CC BY-SA 3.0). Caption corrected and Commons/license URLs added.
• SalmonellaTyphiFlagellarStain.jpg and Shigella_stool.jpg — attribution updated to "CDC/PHIL (public domain)" with Commons links.
• Trombicula-larva-stylostome.jpg — author (Alan R Walker) and CC BY-SA 3.0 link added.

Springer removed from inst/WORDLIST. Spell check passes cleanly.

2. Cross-platform simulation reproducibility fix

sim_pop_data() and sim_pop_data_multi() previously produced different results across macOS, Windows, and Linux because log() returns platform-specific floating-point values that shift the L'Ecuyer-CMRG RNG stream out of sync. Inter-infection times are now rounded to whole days via quantize_t_next(), making simulated values and their snapshots identical across platforms. Snapshot baselines updated accordingly. skip_on_os("linux") removed from test-sim_pop_data_multi.R (cross-platform fix makes it unnecessary); skip_on_cran() retained for CRAN time-budget compliance.

3. Axis label corrections

strat_ests_barplot() (xlab) and strat_ests_scatterplot() (ylab) defaulted to "seroconversion" / "incidence" rather than "seroincidence". Corrected.

4. DESCRIPTION: maps added to Suggests

map_data("world") calls maps::map() internally. maps is added to both Suggests and Config/Needs/website so R CMD check and the pkgdown build both have it available.

5. CI: docs preview for any PR target

Removed the branches: - main filter from the pull_request trigger in docs.yaml, so PRs targeting any branch (not just main) get a preview build. Also temporarily set quiet = FALSE in the pkgdown::build_site() call to surface the full quarto error message — will revert once the render error is diagnosed.

Test plan

• Spell check: spelling::spell_check_package() — passes
• No references to deleted image files remain in source
• skip_on_cran() present in test-sim_pop_data_multi.R
• The R-generated TCV map chunk renders without error in pkgdown (diagnosing now — Generate 🐣 CI check failing)
• Docs preview build produces correct output

🤖 Generated with Claude Code

https://claude.ai/code/session_01B5FrE8RZ6EjFiz7q8zqbfg

[claude]

Claude finished @d-morrison's task in 3m 25s —— View job

---

Code Review – PR #538

Tasks

• Gather context / read changed files
• Review vignettes/methodology.qmd (image replacements + TCV map chunk)
• Review DESCRIPTION (maps dependency declaration)
• Review unrelated R source changes (R/quantize_t_next.R, R/simresp.tinf.R, R/strat_ests_barplot.R, R/strat_ests_scatterplot.R)
• Review snapshot changes
• Review NEWS.md and inst/WORDLIST

---

Summary

The core copyright-fix work is correct and well-executed. There is one actionable bug (missing maps in Suggests) and several scope / hygiene items worth addressing before merge.

---

Bug: maps must be added to Suggests

DESCRIPTION / vignettes/methodology.qmd

map_data("world") (line 118 of methodology.qmd) is backed by maps::map() internally. The maps package is currently only declared in Config/Needs/website, which is a pkgdown-only install target — R CMD check does not install it.

vignettes/methodology.qmd is a proper package vignette (it carries %\VignetteIndexEntry / %\VignetteEngine), so it will be executed during devtools::check(). On any machine (including CRAN) that doesn't have maps pre-installed, the build will fail at map_data("world").

Fix: add maps to Suggests in DESCRIPTION.

# DESCRIPTION
Suggests:
    ...
    maps,
    ...

Fix this →

---

PR scope (CLAUDE.md: "Keep PRs scoped")

The PR description is titled and framed exclusively as a copyright / image-rights fix, but the diff contains at least three distinct additional changes:

Area	Files	Related to copyright?
Cross-platform simulation fix	R/quantize_t_next.R, R/simresp.tinf.R, tests/testthat/test-sim_pop_data_multi.R, _snaps/sim_pop_data*	No (from #447)
Terminology corrections	R/strat_ests_barplot.R, R/strat_ests_scatterplot.R, man/strat_ests_barplot.Rd, _snaps/strat-est-*	No
Image-rights fix	vignettes/methodology.qmd, man/figures/{3 deleted}, DESCRIPTION, inst/WORDLIST	Yes ✓

These appear to be valid, already-reviewed changes (the sim-pop fix landed in #447; the terminology fix is minor cleanup), but bundling them makes the diff harder to review and violates the repo's stated scoping rule. The PR description should at minimum enumerate all three areas — or ideally the non-copyright chunks should have targeted their own PRs.

---

skip_on_cran() removal – confirm intentional

tests/testthat/test-sim_pop_data_multi.R removes both skip_on_cran() and skip_on_os("linux"). The cross-platform quantize_t_next() fix justifies removing skip_on_os("linux"). But removing skip_on_cran() means the 15 000-row snapshot test will now run on CRAN, which may be slow and could trigger the CRAN time limit. If this is intentional (the fix guarantees a stable snapshot everywhere), that's fine — but it should be called out explicitly in the PR description so reviewers don't read it as accidental.

---

vignettes/methodology.qmd – looks good

• The three copyrighted images are removed and only freely-licensed material remains. ✓
• Caption for Fievre_typhoide.png correctly changed from "Historic typhoid fever illustration" to "Global distribution of typhoid fever" (the image is a choropleth, not an illustration). ✓
• CC BY-SA 3.0 links and Commons links added to Wikimedia images. ✓
• Alan R Walker attribution added for the Trombicula photo. ✓
• TCV map R chunk: echo: false, proper fig-cap / fig-alt, ggplot2 (already in Imports) + maps (needs Suggests — see above). The country list and MMWR citation look consistent with public sources. ✓

---

R/quantize_t_next.R and R/simresp.tinf.R

• quantize_t_next() is a sensible extraction; @noRd is correct for an internal helper. ✓
• day2yr → days_per_year is a clarity improvement. ✓
• %>% → |> is the right direction. ✓
• The block comment explaining the cross-platform RNG issue (lines 63–71 of simresp.tinf.R) is long but the WHY is genuinely non-obvious, so it earns its place. ✓

---

R/strat_ests_barplot.R / R/strat_ests_scatterplot.R

• "Seroconversion rate" → "Seroincidence rate" and "Estimated incidence rate" → "Estimated seroincidence rate" are correct terminology fixes. ✓
• man/strat_ests_barplot.Rd is in sync (generated by devtools::document()). ✓
• SVG snapshots updated to reflect the new axis label text. ✓

---

NEWS.md

• The sim-pop cross-platform bullet is clear and explains the behavior change. ✓
• "Removed lingering terminology discrepancies" (line 31) is vague for a changelog entry. Consider naming the affected functions and default parameters explicitly: e.g. "Corrected default axis labels in strat_ests_barplot() (xlab) and strat_ests_scatterplot() (ylab) to say 'seroincidence' rather than 'incidence'/'seroconversion'".

---

inst/WORDLIST

• Springer removed — correct, since neither Springer files nor captions remain in the source. ✓

---

Checklist

• Add maps to Suggests in DESCRIPTION ← required before merge
• Update PR description to accurately reflect all three change areas
• Confirm skip_on_cran() removal is intentional and note it in the PR description
• (Optional) Sharpen the "Removed lingering terminology discrepancies" NEWS.md bullet

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ All tests successful. No failed tests found.

Files with missing lines	Coverage Δ
R/quantize_t_next.R	100.00% <100.00%> (ø)
R/simresp.tinf.R	99.00% <100.00%> (ø)
R/strat_ests_barplot.R	100.00% <ø> (ø)
R/strat_ests_scatterplot.R	70.00% <100.00%> (ø)

[Copilot]

Pull request overview

Updates the methodology lecture deck to remove figures with unclear/copyrighted reuse rights by replacing them with freely licensed/public-domain imagery (and a render-time generated map), while also including a simulation reproducibility fix and terminology standardization that affect tests and plotting outputs.

Changes:

• Replace/remove copyrighted slide figures, add explicit image attributions, and generate a TCV-adoption world map at render time in methodology.qmd.
• Make simulated inter-infection times quantized to whole days for cross-platform reproducibility; refresh affected snapshots and adjust a simulation test.
• Standardize “seroincidence” terminology in plot axis labels and update related snapshots/docs; bump package version and website needs.

Reviewed changes

Copilot reviewed 10 out of 19 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
vignettes/methodology.qmd	Removes/replaces problematic figures, adds attribution links, and adds an R chunk to generate a world map figure during render.
tests/testthat/test-sim_pop_data_multi.R	Changes simulation snapshot behavior (removes CRAN/Linux skips; adjusts snapshot precision).
tests/testthat/_snaps/sim_pop_data/sim_pop_data.csv	Updates snapshot output for sim_pop_data() after simulation reproducibility change.
tests/testthat/_snaps/autoplot.summary.seroincidence.by/strat-est-plot-ci.svg	Updates snapshot text to reflect new y-axis wording.
tests/testthat/_snaps/autoplot.summary.seroincidence.by/strat-est-plot-ci-lines.svg	Updates snapshot text to reflect new y-axis wording.
tests/testthat/_snaps/autoplot.summary.seroincidence.by/strat-est-barplot.svg	Updates snapshot text to reflect new x-axis wording.
tests/testthat/_snaps/autoplot.summary.seroincidence.by/strat-est-barplot-palette.svg	Updates snapshot text to reflect new x-axis wording.
R/strat_ests_scatterplot.R	Updates y-axis label string to “Estimated seroincidence rate”.
R/strat_ests_barplot.R	Updates default x-axis label string to “Seroincidence rate per 1000 person-years”.
R/simresp.tinf.R	Quantizes inter-infection times to whole days and switches to native pipe for one expression.
R/quantize_t_next.R	Adds internal helper to quantize inter-infection times for reproducible RNG consumption across OSes.
NEWS.md	Documents the simulation reproducibility change and terminology cleanup.
man/strat_ests_barplot.Rd	Regenerates documentation to match updated default label text.
inst/WORDLIST	Removes “Springer” from spelling whitelist.
DESCRIPTION	Bumps version and adds maps to Config/Needs/website for the new vignette map chunk.

Files not reviewed (1)

• man/strat_ests_barplot.Rd: Generated file

[claude]

Claude finished @d-morrison's task in 3m 6s —— View job

---

Code Review – PR #538

Tasks

• Gather context / read changed files
• Check current DESCRIPTION for maps in Suggests
• Review vignettes/methodology.qmd (image replacements + TCV map chunk)
• Review R/quantize_t_next.R, R/simresp.tinf.R
• Review test changes (test-sim_pop_data_multi.R, snapshots)
• Review NEWS.md and inst/WORDLIST
• Synthesize findings and post final review

---

Summary

The core copyright-fix work is correct and complete. The one blocker from the previous review (maps missing from Suggests) has been resolved. One actionable concern remains around skip_on_cran(), plus two minor items.

---

maps in Suggests — Fixed ✓

DESCRIPTION line 54 now lists maps in Suggests. The vignette's map_data("world") call will resolve correctly during devtools::check() and on CRAN. Both placements (Suggests + Config/Needs/website) are correct.

---

skip_on_cran() removal — Needs attention

tests/testthat/test-sim_pop_data_multi.R line 1 removes both guards:

• skip_on_os("linux") — correctly removed; the whole point of the cross-platform fix is to make this test pass everywhere. ✓
• skip_on_cran() — this one served a different purpose: runtime budget. The test runs 5 lambdas × 2 sample sizes × nclus = 10 full parallel simulations. CRAN's per-check time limit is tight (~10 min total), and a parallel simulation grid of this size is risky.

If the intent is that the cross-platform fix makes this test safe to run everywhere including CRAN, that should be explicitly called out in the PR description. Alternatively, add skip_on_cran() back (without skip_on_os("linux")) to preserve the performance guard.

Fix this →

---

vignettes/methodology.qmd — Clean ✓

All three copyright issues resolved:

• Springer Nature NRM figure (Shigella) — removed; Shigella_stool.jpg (CDC/PHIL public domain) remains with proper attribution and Commons link. ✓
• Springer Nature range map (scrub typhus) — removed; Trombicula-larva-stylostome.jpg (CC BY-SA 3.0, Alan R Walker) remains with author and license link. ✓
• burden-typhoid-vaccines.png (mapchart.net) — replaced with R chunk using map_data("world") + public-domain CDC MMWR country list. ✓

Caption for Fievre_typhoide.png corrected from "Historic typhoid fever illustration" to "Global distribution of typhoid fever" — this was factually wrong before, good fix. ✓

The TCV R chunk: echo: false, proper fig-cap/fig-alt, ggplot2 (Imports) + maps (now in Suggests). All clean.

---

R/quantize_t_next.R and R/simresp.tinf.R — Clean ✓

quantize_t_next() is a well-scoped internal helper with a clear docstring explaining the cross-platform RNG issue. @noRd is correct. The max(1, round(t_next)) logic correctly prevents a zero-length step.

In simresp.tinf.R, quantize_t_next() is applied at both the pre-loop draw (line 72) and the within-loop draw (line 146), covering all infection-time samples. The block comment on lines 63–71 explaining the floating-point / L'Ecuyer-CMRG desync issue is long but warranted — the WHY is genuinely non-obvious. ✓

One minor comment inaccuracy: the remark at line 143–144 says "t_end and t0 are whole days" but t_end may be fractional (e.g., 10 * 365.25 = 3652.5). The boundary clamp (t_next <- t_end - t0) can therefore produce a fractional t_next. This doesn't affect correctness — the clamped case has always been fractional, it's only the random draws that are now quantized — but the comment is technically inaccurate. Minor.

---

R/strat_ests_barplot.R / R/strat_ests_scatterplot.R — Clean ✓

"Seroconversion rate" → "Seroincidence rate" and "Estimated incidence rate" → "Estimated seroincidence rate" are correct terminology fixes. Docs (man/strat_ests_barplot.Rd) and SVG snapshots updated accordingly. ✓

Pre-existing asymmetry (not introduced here): strat_ests_barplot() exposes xlab as an overridable parameter, but strat_ests_scatterplot() hardcodes its ylab. Out of scope for this PR.

---

NEWS.md — One minor item

The sim-pop cross-platform bullet (lines 24–30) is clear and specific. ✓

Line 31: "* Removed lingering terminology discrepancies" is vague for a changelog. Suggest naming the affected functions:

* Corrected default axis labels in `strat_ests_barplot()` (`xlab`) and
  `strat_ests_scatterplot()` (`ylab`) to say "seroincidence" rather than
  "seroconversion"/"incidence" (#NNN if applicable).

Fix this →

---

inst/WORDLIST — Clean ✓

Springer removed correctly; no remaining references to Springer files anywhere in source.

---

Checklist

• maps in Suggests ← fixed in this push
• Confirm skip_on_cran() removal is intentional and safe for CRAN time limits (or re-add it) ← recommended before merge
• Sharpen the "Removed lingering terminology discrepancies" NEWS.md bullet ← optional
• Update PR description to reflect all three change areas (copyright, sim reproducibility, terminology) ← optional

[claude]

Claude finished @d-morrison's task in 2m 24s —— View job

---

Code Review – PR #538

Tasks

• Gather context from PR diff and changed files
• Review DESCRIPTION (maps in Suggests, version bump)
• Review vignettes/methodology.qmd (image replacements + TCV map chunk)
• Review R/quantize_t_next.R and R/simresp.tinf.R
• Review test changes (test-sim_pop_data_multi.R, snapshots)
• Review R/strat_ests_barplot.R / R/strat_ests_scatterplot.R
• Review NEWS.md and inst/WORDLIST
• Synthesize findings and post final review

---

Summary

The latest push (509122e) addresses all outstanding items from the previous two review rounds. The PR is clean.

---

All prior blockers resolved ✓

Item	Status
maps in Suggests	Fixed in 1b17339 ✓
skip_on_cran() re-added to test-sim_pop_data_multi.R	Fixed in 509122e ✓
NEWS.md terminology bullet sharpened	Fixed in 509122e ✓

---

DESCRIPTION — Clean ✓

• maps in Suggests (line 54) and Config/Needs/website (line 76) — both placements correct.
• Version bump to 1.4.0.9016 is consistent with the multi-commit history here.

---

vignettes/methodology.qmd — Clean ✓

All three copyright issues resolved:

• Springer Nature NRM figure (Shigella incidence choropleth) — removed; Shigella_stool.jpg (CDC/PHIL, public domain) kept with proper Commons link and attribution. ✓
• Springer Nature range map (scrub typhus) — removed; Trombicula-larva-stylostome.jpg (Alan R Walker, CC BY-SA 3.0) kept with author and license links. ✓
• burden-typhoid-vaccines.png (mapchart.net, unclear license) — replaced with an R chunk using map_data("world") + CDC MMWR country list (public domain). ✓

Caption for Fievre_typhoide.png corrected from "Historic typhoid fever illustration" to "Global distribution of typhoid fever" — was factually wrong before. ✓

The TCV R chunk: echo: false, fig-cap, fig-alt all present. ggplot2 (Imports) + maps (Suggests). No issues.

---

R/quantize_t_next.R — Clean ✓

Well-scoped internal helper, @noRd correct, docstring explains the cross-platform RNG issue clearly. max(1, round(t_next)) correctly prevents a zero-length step.

---

R/simresp.tinf.R — Clean ✓ (one cosmetic nit)

quantize_t_next() applied at both draw sites (pre-loop draw, line 72; within-loop draw, line 146). day2yr → days_per_year and %>% → |> are good housekeeping. The block comment at lines 63–71 is long but the WHY is genuinely non-obvious. ✓

One minor comment inaccuracy (pre-existing, not blocking): line 143–144 says `t_end` and `t0` are whole days, but t_end can be fractional (10 * 365.25 = 3652.5). Only t0 is guaranteed integer after quantization. The clamped t_next <- t_end - t0 can therefore be fractional (e.g., 2.5). This doesn't affect correctness — seq(0, 2.5, 1) and seq(0, 2, 1) produce the same integer-length grid — but the comment is slightly misleading.

Fix this →

---

tests/testthat/test-sim_pop_data_multi.R — Clean ✓

• skip_on_cran() present (line 2) — performance guard retained. ✓
• skip_on_os("linux") removed — correct; the quantize fix makes the snapshot stable everywhere. ✓
• digits = 3 added to expect_snapshot_data() — appropriate precision reduction for cross-platform stability. ✓

---

R/strat_ests_barplot.R / R/strat_ests_scatterplot.R — Clean ✓

• xlab default corrected to "Seroincidence rate per 1000 person-years" (was "Seroconversion rate…"). ✓
• ylab hardcoded value corrected to "Estimated seroincidence rate". ✓
• man/strat_ests_barplot.Rd and four SVG snapshots updated to match. ✓

---

NEWS.md — Clean ✓

• Sim-pop cross-platform bullet (lines 24–30): specific, explains the mechanism, references try to not skip #447. ✓
• Terminology bullet (lines 31–33): now names the affected functions and parameters explicitly. ✓

---

inst/WORDLIST — Clean ✓

Springer removed; no remaining Springer references in source.

---

Verdict

Ready to merge. All blockers resolved. The one remaining note (comment inaccuracy about t_end being a whole day) is cosmetic and non-blocking.

[claude]

Claude finished @d-morrison's task in 51s —— View job

---

Code Review – PR #538

Tasks

• Gather context from PR diff and previous reviews
• Review latest commit (4854034 Fix inaccurate comment: t_end may be fractional, not t0)
• Verify all prior review items are resolved
• Post final review verdict

---

Summary

The latest push (4854034) addresses the one remaining nit from the previous review round. The PR is clean.

---

Comment fix — Clean ✓

R/simresp.tinf.R lines 143–147: the comment now correctly states:

• t0 is always a whole number (it accumulates quantized steps). ✓
• t_end may be fractional (e.g. 10 * 365.25 = 3652.5). ✓
• The boundary clamp t_end - t0 can therefore be fractional, but that doesn't affect the integer-length time grid. ✓

This matches the actual behavior in the code.

---

All prior review items resolved ✓

Item	Status
maps in Suggests	Fixed in 1b17339 ✓
skip_on_cran() re-added to simulation test	Fixed in 509122e ✓
NEWS.md terminology bullet sharpened	Fixed in 509122e ✓
Comment inaccuracy about t_end being a whole day	Fixed in 4854034 ✓

---

Verdict

Ready to merge. All blockers and nits resolved across four rounds of review. The copyright removals are correct and complete, the cross-platform simulation fix is sound, and the terminology cleanup is consistent throughout code, docs, and snapshots.

[claude]

Claude finished @d-morrison's task in 2m 40s —— View job

---

Code Review – PR #538 (Round 5)

Tasks

• Gather context from PR and prior reviews
• Check current diff vs base branch (origin/feat/intro-lecture-slides)
• Review the merge commit (9f57f8f) for conflict resolution correctness
• Verify all prior review items remain resolved
• Post final verdict

---

Summary

The latest synchronization (9f57f8f) was a merge commit pulling in the base branch (feat/intro-lecture-slides). No new substantive changes were introduced.

---

Merge commit – Clean ✓

9f57f8f Merge feat/intro-lecture-slides; resolve methodology.qmd conflicts

• methodology.qmd conflicts resolved by taking the PR branch version (proper copyright fixes and R-generated TCV map) over the base branch's "rights pending" callout-note placeholders. No placeholder text remains — confirmed: no "rights pending", "callout-note", or Springer/mapchart references. ✓
• docs.yaml conflict resolved by removing branches: - main from the pull_request: trigger. The base branch already had this same change, so the net diff vs origin/feat/intro-lecture-slides is zero — no new workflow scope change introduced by this PR. ✓

---

All prior review items confirmed still in place ✓

Item	Status
maps in Suggests (DESCRIPTION line 54) and Config/Needs/website (line 76)	✓
skip_on_cran() present in test-sim_pop_data_multi.R (line 2)	✓
skip_on_os("linux") removed	✓
Comment at R/simresp.tinf.R:143–146 correctly explains t0 is whole-day, t_end may be fractional	✓
NEWS.md terminology bullet names strat_ests_barplot() and strat_ests_scatterplot() explicitly	✓
No copyrighted image references anywhere in source	✓
Springer removed from inst/WORDLIST	✓

---

Verdict

Ready to merge. The merge commit correctly resolved all conflicts, no regressions introduced, and all fixes from rounds 1–4 remain intact.

[claude]

Claude finished @d-morrison's task in 1m 5s —— View job

---

Code Review – PR #538 (Round 6)

Tasks

• Gather context from PR and prior reviews
• Check diff since last reviewed commit (9f57f8f)
• Review 8c36a9c — CI change (docs.yaml, quiet = FALSE)
• Post final verdict

---

Summary

Only one commit (8c36a9c) has landed since round 5's "Ready to merge" verdict. It touches .github/workflows/docs.yaml only.

---

.github/workflows/docs.yaml — Clean ✓

-  Rscript -e 'pkgdown::build_site(devel = FALSE)'
+  Rscript -e 'pkgdown::build_site(devel = FALSE, quiet = FALSE)'
...
-  Rscript -e 'pkgdown::build_site(devel = TRUE)'
+  Rscript -e 'pkgdown::build_site(devel = TRUE, quiet = FALSE)'

pkgdown::build_site() defaults to quiet = TRUE, which suppresses progress messages and warnings — exactly the wrong behaviour when diagnosing a CI failure. Setting quiet = FALSE on both branches (release and dev) makes build output visible in the Actions log without changing what is built or deployed. Safe and useful change. ✓

---

All prior review items confirmed still in place ✓

Item	Status
maps in Suggests (DESCRIPTION line 54) and Config/Needs/website	✓
skip_on_cran() present in test-sim_pop_data_multi.R	✓
skip_on_os("linux") removed	✓
Comment in R/simresp.tinf.R:143–146 correctly explains fractional t_end	✓
NEWS.md terminology bullet names affected functions explicitly	✓
No copyrighted image references anywhere in source	✓
Springer removed from inst/WORDLIST	✓

---

Verdict

Ready to merge. The latest commit is a safe, targeted diagnostics improvement to the docs build step. No regressions; all prior fixes remain intact.

[claude]

Claude finished @d-morrison's task in 1m 37s —— View job

---

Code Review – PR #538 (Round 7)

Tasks

• Gather context from PR and prior reviews
• Check new commits since last review (8c36a9c)
• Review the version bump commit (f508ced)
• Post final verdict

---

Summary

One new commit since the round-6 "Ready to merge" verdict: a version bump in DESCRIPTION only.

---

DESCRIPTION version bump — Clean ✓

-Version: 1.4.0.9016
+Version: 1.4.1.9000

Motivation: serodynamics (used in methodology.qmd) declares Depends/Imports: serocalculator >= 1.4.1. R treats 1.4.0.9016 < 1.4.1, so the docs build was failing at that vignette chunk. Bumping to 1.4.1.9000 satisfies the constraint and follows standard R dev-version naming (1.4.1.9000 = dev immediately after 1.4.1). ✓

No other files changed in this commit. No code, test, or documentation changes introduced.

---

All prior review items confirmed still in place ✓

Item	Status
maps in Suggests (line 54) and Config/Needs/website (line 76)	✓
skip_on_cran() present in test-sim_pop_data_multi.R	✓
skip_on_os("linux") removed	✓
Comment in R/simresp.tinf.R correctly explains t0 is whole-day, t_end may be fractional	✓
NEWS.md terminology bullet names affected functions explicitly	✓
No copyrighted image references anywhere in source	✓
Springer removed from inst/WORDLIST	✓

---

Verdict

Ready to merge. The latest commit is a targeted version bump that unblocks the docs build by satisfying the serodynamics >= 1.4.1 constraint. No regressions; all fixes from rounds 1–6 remain intact.

[github-actions]

📕 Preview documentation for this PR has been cleaned up.