Improve weekly aggregation in outbreak_model()

[joshwlambert]

This PR addresses #168 but goes beyond fixing the typo in the if statement and rewrites a section of outbreak_model() to more efficiently create the empty weeks in the output data.table.

The complete outbreak data, with missing weeks is now created using a non-equi join. The need for the original if statement is removed.

This update also changes the type of the columns, from double to integer, resulting in a smaller memory footprint for the outbreak data.

Test expectations are updated to reflect the change in type for the week, weekly_cases and cumulative columns returned from outbreak_model() (and therefore also scenario_sim()).

Summary by CodeRabbit

Release Notes

• Chores
  • Updated internal data type handling for weekly case calculations to ensure consistent integer representation across output columns.
  • Refined weekly case computation logic for improved data processing consistency.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

This PR refactors the weekly_cases calculation in the outbreak model to pre-initialise a zero-filled container for all weeks and overlay actual outbreak data, whilst updating numeric columns to integer types throughout the codebase and tests.

Changes

Cohort / File(s)	Summary
Global variable declaration R/globals.R	Added "i.weekly_cases" to the global variables list recognised by utils::globalVariables() for the outbreak_model group.
Core logic refactoring R/outbreak_model.R	Refactored weekly_cases construction to pre-initialise a zero-filled container for weeks 0 to max_week, then join actual outbreak counts from weekly_cases_outbreak. Removed manual missing_weeks calculation and rbindlist step.
Test snapshots and expectations tests/testthat/_snaps/scenario_sim.md, tests/testthat/test-outbreak_model.R, tests/testthat/test-scenario_sim.R	Updated type annotations and test expectations from numeric to integer for sim, week, weekly_cases, and cumulative columns; replaced numeric literals with integer literals (1L, 0L, 25L, etc.) in test assertions.

Possibly related PRs

• Improve package test suite #160: Updates tests and snapshots asserting weekly_cases structure and types, directly aligned with the main PR's changes to weekly_cases construction and global variable additions.

Suggested labels

testing

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarises the main change—improving weekly aggregation in outbreak_model() by replacing a faulty conditional with a non-equi join approach.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 0

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

R/outbreak_model.R (1)

10-12: Documentation does not reflect the new integer column types.

The @return documentation states that $week, $weekly_cases, and $cumulative are numeric, but the implementation now produces integer columns. Consider updating the documentation to match.

📝 Suggested documentation update

 #' `@return` `data.table` of cases by week, cumulative cases, and the effective
 #' reproduction number of the outbreak. `data.table` columns are:
-#' * `$week`: `numeric`
-#' * `$weekly_cases`: `numeric`
-#' * `$cumulative`: `numeric`
+#' * `$week`: `integer`
+#' * `$weekly_cases`: `integer`
+#' * `$cumulative`: `integer`
 #' * `$effective_r0`: `numeric`
 #' * `$cases_per_gen`: `list`

🧹 Nitpick comments (1)

R/outbreak_model.R (1)

108-116: Potential type inconsistency: floor() returns double, but week column is integer.

The floor(onset / 7) expression on line 108 returns a double, whilst the week column in weekly_cases (line 114) is created as integer via 0:max_week. Although the data.table join will still function correctly, this creates a subtle type mismatch between weekly_cases_outbreak$week (double) and weekly_cases$week (integer).

Consider wrapping in as.integer() for explicit type consistency:

♻️ Suggested refinement

-  weekly_cases_outbreak <- case_data[, week := floor(onset / 7)
+  weekly_cases_outbreak <- case_data[, week := as.integer(floor(onset / 7))
                            ][, list(weekly_cases = .N), by = week
                              ]