test(sipnet): cover JSON segmentation and restart chaining

CHANGELOG.md

@@ -9,12 +9,14 @@ For more information about this file see also [Keep a Changelog](http://keepacha
 ## Unreleased
 
 ### Added
+
+- Added regression coverage for SIPNET event-JSON segmentation and segmented restart chaining (#4021).
 - Added a downscaling-error diagnostic to `inst/ilamb/` in PEcAn.benchmark: compares the downscaling random forest's out-of-bag RMSE against the between-member ensemble spread at the SDA sites, quantifying downscaling error the ensemble spread does not represent. (#4070)
 - Added `metric_Bias()` standalone function and registered `"BIAS"` in `PEcAn.benchmark` metric registry (#4062, #4059).
 - Added `metric_CRPS()` function and registry entry to `PEcAn.benchmark` for Continuous Ranked Probability Score (CRPS) ensemble skill evaluation against observations (#4062).
 - Added `efi_long_to_array()` helper function to `PEcAn.benchmark` for reshaping EFI long format model outputs into wide ensemble member matrices (#4062).
 - Added ensemble member spaghetti line rendering under confidence ribbons in `metric_timeseries_plot()` (#4062).
-- Added `examples/benchmarks/salinas_soc_ensemble/` demonstrating multi-site SOC ensemble validation against Salinas observations out-of-the-box (#4062).
+- Added `examples/benchmarks/salinas_soc_ensemble/` demonstrating multi-site SOC ensemble validation against Salinas observations out-of-the-box (#4062)
 - Added ensemble calibration diagnostics to `inst/ilamb/` in PEcAn.benchmark (rank histogram, spread-skill ratio, coverage, reliability) for assessing whether an ensemble's spread is well calibrated against observations, complementing the ensemble-mean benchmarking.
 - Added regional calibration diagnostics to `inst/ilamb/` in PEcAn.benchmark: break the ensemble calibration assessment down by land cover class and by EPA/CEC ecoregion, reusing the ensemble calibration diagnostics, with figures.
 - Added an ESA CCI biomass benchmark extension to `inst/ilamb/` in PEcAn.benchmark: a converter for the ESACCI Biomass v7.0 product (2015-2024, with per-pixel uncertainty) to ILAMB-compatible netCDF, plus an observation-error test showing the ensemble overconfidence persists against a modern benchmark and after accounting for observation uncertainty.

models/sipnet/NEWS.md

@@ -1,5 +1,6 @@
 # PEcAn.SIPNET 1.10.0.9000
 
+* Added regression coverage for SIPNET event-JSON segmentation and segmented restart chaining (#4021).
 * `write.config.SIPNET` now maps the `leafNResorptionFrac` trait to its v2
     parameter, so PFT supplied values reach the param file instead of
     silently keeping the template default. Completes the nitrogen cycle

models/sipnet/tests/testthat/test-write_segmented_configs.R

@@ -92,13 +92,84 @@ test_that("write_segmented_configs", {
     )
   }
 
+  # Each segment writes its own restart and later segments read the previous one.
+  segment_rundirs <- file.path(
+    run_path, "segments", sprintf("segment_%03d", 1:3), "run"
+  )
+  for (seg in 1:3) {
+    config <- readLines(file.path(segment_rundirs[seg], "1", "sipnet.in"))
+    restart_in <- grep("^[[:space:]]*RESTART_IN[[:space:]]*=", config, value = TRUE)
+    restart_out <- grep("^[[:space:]]*RESTART_OUT[[:space:]]*=", config, value = TRUE)
+
+    expect_identical(
+      restart_out,
+      paste("RESTART_OUT =", file.path(segment_rundirs[seg], "restart.out"))
+    )
+    if (seg == 1) {
+      expect_length(restart_in, 0)
+    } else {
+      expect_identical(
+        restart_in,
+        paste("RESTART_IN =", file.path(segment_rundirs[seg - 1], "restart.out"))
+      )
+    }
+  }
+
   # job.sh includes calls to segment scripts
   jobsh <- readLines(file.path(run_path, "job.sh"))
   expect_match(jobsh, "bash .*segment_001/run/1/job.sh", all = FALSE)
   expect_match(jobsh, "bash .*segment_002/run/1/job.sh", all = FALSE)
   expect_match(jobsh, "bash .*segment_003/run/1/job.sh", all = FALSE)
 })
 
+test_that("segment_dataframe falls back to event_json for the configured site", {
+  pth <- withr::local_tempdir()
+  events_path <- file.path(pth, "events.json")
+  jsonlite::write_json(
+    list(
+      list(
+        site_id = "a",
+        events = list(
+          list(event_type = "planting", date = "2025-01-02", crop_code = "D12"),
+          list(event_type = "planting", date = "2025-01-05", crop_code = "G6")
+        )
+      ),
+      list(
+        site_id = "b",
+        events = list(
+          list(event_type = "planting", date = "2025-01-03", crop_code = "P1")
+        )
+      )
+    ),
+    path = events_path,
+    auto_unbox = TRUE
+  )
+  run_settings <- PEcAn.settings::as.Settings(list(
+    run = list(
+      site = list(id = "a", site.pft = list(veg = "pft1")),
+      inputs = list(event_json = list(path = events_path)),
+      start.date = "2025-01-01",
+      end.date = "2025-01-10"
+    )
+  ))
+
+  result <- PEcAn.SIPNET:::segment_dataframe(run_settings)
+
+  expect_equal(nrow(result), 3)
+  expect_identical(result$site_id, rep("a", 3))
+  expect_identical(result$segment_id, c("001", "002", "003"))
+  expect_identical(
+    result$start_date,
+    as.Date(c("2025-01-01", "2025-01-02", "2025-01-05"))
+  )
+  expect_identical(
+    result$end_date,
+    as.Date(c("2025-01-01", "2025-01-04", "2025-01-10"))
+  )
+  expect_identical(result$crop_code, c(NA_character_, "D12", "G6"))
+  expect_identical(result$pft[1], "pft1")
+})
+
 test_that("segment_dataframe returns empty when run start is after all crop cycles", {
   pth <- withr::local_tempdir()
   crp_chg_path <- file.path(pth, "cycles-a.csv")