Fix #61

CITATION.cff

@@ -8,7 +8,7 @@ message: 'To cite package "wowi" in publications use:'
 type: software
 license: GPL-3.0-or-later
 title: 'wowi: Detect Spatial Clusters of High Acute Malnutrition Rates'
-version: 1.0.1
+version: 1.0.2
 abstract: Utilities for detecting statistically significant spatial clusters of high
   acute malnutrition rates using SaTScan's Bernoulli spatial-scan model.
 authors:
@@ -23,7 +23,7 @@ preferred-citation:
   authors:
   - name: Tomás Zaba
   year: '2025'
-  notes: R package version 1.0.1
+  notes: R package version 1.0.2
   url: https://tiwowi.github.io/wowi/
 repository-code: https://github.com/tiwowi/wowi
 url: https://tiwowi.github.io/wowi/

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: wowi
 Title: Detect Spatial Clusters of High Acute Malnutrition Rates
-Version: 1.0.1
+Version: 1.0.2
 Authors@R: 
     person(given = "Tomás", 
             family = "Zaba",

NEWS.md

@@ -1,3 +1,10 @@
+# wowi 1.0.2
+
+## Bug fixes
+
+* Resolved issue [#61](https://github.com/tiwowi/wowi/issues/61).
+In the previous version, the regular expressions used to match text patterns into tables would break whenever the list of location IDs in a cluster wrapped onto a new line, altering the expected match order. This caused several columns to remain empty. The parser now handles these cases programmatically. 
+
 # wowi 1.0.1
 
 ## General updates 

R/satscan-runner.R

@@ -127,7 +127,8 @@
 #' 
 #' 
 
-# no# nocov start
+# nocov start
+
 ww_run_satscan <- function(
     .data,
     filename = NULL,

R/utils.R

@@ -2,58 +2,91 @@
 #'
 #' @keywords internal
 #'
+#' 
 skip_if_no_satscan <- function(ss_path = "/Applications/SaTScan.app/Contents/app/satscan") {
   testthat::skip_if_not(file.exists(ss_path), message = "SaTScan is not installed or not found")
 }
 
 #'
-#'
+#' Extract results from SaTScan-text-based output 
+#' 
+#' @param file SaTScan-text-based output result given as "main" to be parsed
+#' 
 #' @keywords internal
-#'
-#'
+#' 
+#' 
+
+
+# nocov start
+
 parse_clusters <- function(file) {
-  ## Access the txt-based results ----
+
+  ## Subset SaTScan-text-based output file ----
   txt <- file$main
 
-  ## Find line indices with "Location IDs included" ----
+  ## Find line indices for cluster starts and coordinates
   cluster_start <- stringr::str_which(txt, "^\\d+\\.Location IDs included\\.:")
+  coords <- stringr::str_which(txt, "^\\s+Coordinates / radius")
+
+  ## Extract the name of the survey area ----
   area_name <- stringr::str_which(txt, "^[Case File]+\\:")
 
-  ## Define line offsets to grab per cluster ----
-  cluster_blocks <- lapply(cluster_start, function(start_idx) {
-    txt[start_idx:(start_idx + 10)]
-  })
-
-  ## Parse each cluster block ----
-  parsed_clusters <- lapply(cluster_blocks, function(block) {
-    tibble::tibble(
-      survey_area = stringr::str_extract_all(basename(txt[[area_name]]), "^[^.]+") |>
-        as.character(),
-      nr_EAs = stringr::str_extract(txt[[18]], "\\d+") |> as.integer(),
-      total_children = stringr::str_extract(txt[[19]], "\\d+") |> as.integer(),
-      total_cases = stringr::str_extract(txt[[20]], "\\d+") |> as.integer(),
-      `%_cases` = stringr::str_extract(txt[[21]], "\\d+") |> as.double(),
-      location_ids = stringr::str_extract_all(block[1], "[0-9]+")[[1]][-1] |>
-        paste(collapse = ","),
-      geo = stringr::str_extract(block[2], "\\d+\\.\\d+\\s+\\w\\,\\s+\\d+\\.\\d+\\s+\\w"),
-      radius = stringr::str_extract(block[2], "\\d+\\.\\d+\\s*\\w+$"),
-      span = stringr::str_extract(block[3], "[0-9]+[.]+[0-9]+\\s+\\w+$"),
-      children = stringr::str_extract(block[4], "[0-9]+") |> as.integer(),
-      n_cases = stringr::str_extract(block[5], "[0-9]+") |> as.integer(),
-      expected_cases = stringr::str_extract(block[6], "[0-9]+[.]+[0-9]+") |> as.double(),
-      observedExpected = stringr::str_extract(block[7], "[0-9]+[.]+[0-9]+") |> as.double(),
-      relative_risk = stringr::str_extract(block[8], "[0-9]+\\.\\d+") |> as.double(),
-      `%_cases_in_area` = stringr::str_extract(block[9], "[0-9]+\\.\\d+") |> as.double(),
-      log_lik_ratio = stringr::str_extract(block[10], "[0-9]+\\.\\d+") |> as.double(),
-      pvalue = stringr::str_extract(block[11], "[0-9]+[.]+[0-9]+") |> as.double(),
+  ## Guard: align pairs if counts differ
+  n <- min(length(cluster_start), length(coords))
+  if (n == 0) return(tibble::tibble())
+
+  out <- vector("list", n)
+
+  for (j in seq_len(n)) {
+    idx <- cluster_start[j]
+    coord <- coords[j]
+
+    ## Collect all lines containing IDs up to the coordinates line
+    id_block <- txt[idx:(coord - 1)]
+    ids_vec <- stringr::str_extract_all(id_block, "[0-9]+") |> unlist()
+
+    ## Drop the leading cluster number (e.g., "1", "2") if present
+    location_ids <- if (length(ids_vec) > 1) {
+      paste(ids_vec[-1], collapse = ",")
+    } else {
+      NA_character_
+    }
+
+    ## Build tibble for this cluster (offsets are stable in SaTScan output)
+    out[[j]] <- tibble::tibble(
+
+      ### Summary metadata (from fixed lines in your example) ----
+      survey_area = as.character(stringr::str_extract_all(basename(txt[[area_name]]), "^[^.]+")),
+      nr_EAs = as.integer(stringr::str_extract(txt[18], "\\d+")),
+      total_children = as.integer(stringr::str_extract(txt[19], "\\d+")),
+      total_cases = as.integer(stringr::str_extract(txt[20], "\\d+")),
+      `%_cases` = as.double(stringr::str_extract(txt[21], "\\d+\\.?\\d*")),
+
+      ### Cluster-specific ----
+      location_ids = location_ids,
+      geo = stringr::str_extract(txt[coord],
+        "\\d+\\.\\d+\\s+\\w\\,\\s+\\d+\\.\\d+\\s+\\w"),
+      radius = stringr::str_extract(txt[coord],"[0-9]+\\.[0-9]+\\s*km"),
+      span = stringr::str_extract(txt[coord + 1], "[0-9]+\\.[0-9]+\\s*km"),
+      children = as.integer(stringr::str_extract(txt[coord + 2], "\\d+")),
+      n_cases = as.integer(stringr::str_extract(txt[coord + 3], "\\d+")),
+      expected_cases = as.double(stringr::str_extract(txt[coord + 4], "[0-9]+\\.[0-9]+")),
+      observedExpected = as.double(stringr::str_extract(txt[coord + 5], "[0-9]+\\.[0-9]+")),
+      relative_risk = as.double(stringr::str_extract(txt[coord + 6], "[0-9]+\\.[0-9]+")),
+      `%_cases_in_area` = as.double(stringr::str_extract(txt[coord + 7], "[0-9]+\\.?[0-9]*")),
+      log_lik_ratio = as.double(stringr::str_extract(txt[coord + 8], "[0-9]+\\.[0-9]+")),
+      pvalue = as.double(stringr::str_extract(txt[coord + 9], "[0-9]+\\.?[0-9]*")),
+
+      ### Check if IPC AMN reqs for survey disaggregation is met ----
       ipc_amn = ifelse(
-        test = length(strsplit(.data$location_ids, ",\\s*")[[1]]) >= 5 & as.numeric(.data$children) > 100,
-        yes = "yes",
-        no = "no"
+        length(strsplit(location_ids, ",\\s*")[[1]]) >= 5 & !is.na(children) & children >= 100,
+        "yes", "no"
       )
-    )
-  })
+    ) 
+  }
 
-  ## Combine all into one data frame ----
-  dplyr::bind_rows(parsed_clusters)
+  ## Return binded results ----
+  dplyr::bind_rows(out)
 }
+
+# nocov end

README.md

@@ -11,6 +11,8 @@ public.](https://www.repostatus.org/badges/latest/wip.svg)](https://www.repostat
 [![Lifecycle:
 experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental)
 [![R-CMD-check](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml)
+[![Test
+app](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml)
 [![Codecov test
 coverage](https://codecov.io/gh/tiwowi/wowi/graph/badge.svg)](https://app.codecov.io/gh/tiwowi/wowi)  
 <!-- badges: end -->
@@ -107,7 +109,7 @@ citation("wowi")
 
       Tomás Zaba (2025). _wowi: Utilities for detecting statistically
       significant spatial clusters of high acute malnutrition rates using
-      SaTScan's Bernoulli spatial-scan model_. R package version 1.0.1,
+      SaTScan's Bernoulli spatial-scan model_. R package version 1.0.2,
       <https://tiwowi.github.io/wowi/>.
 
     A BibTeX entry for LaTeX users is
@@ -116,7 +118,7 @@ citation("wowi")
         title = {wowi: Utilities for detecting statistically significant spatial clusters of high acute malnutrition rates using SaTScan's Bernoulli spatial-scan model},
         author = {{Tomás Zaba}},
         year = {2025},
-        note = {R package version 1.0.1},
+        note = {R package version 1.0.2},
         url = {https://tiwowi.github.io/wowi/},
       }
 

README.qmd

@@ -10,6 +10,7 @@ format: gfm
 [![Project Status: WIP – Initial development is in progress, but there has not yet been a stable, usable release suitable for the public.](https://www.repostatus.org/badges/latest/wip.svg)](https://www.repostatus.org/#wip)
 [![Lifecycle: experimental](https://img.shields.io/badge/lifecycle-experimental-orange.svg)](https://lifecycle.r-lib.org/articles/stages.html#experimental)
 [![R-CMD-check](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml)
+[![Test app](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml/badge.svg)](https://github.com/tiwowi/wowi/actions/workflows/R-CMD-check.yaml)
 [![Codecov test coverage](https://codecov.io/gh/tiwowi/wowi/graph/badge.svg)](https://app.codecov.io/gh/tiwowi/wowi)  
  <!-- badges: end -->
 

inst/CITATION

@@ -4,6 +4,6 @@ bibentry(
   title    = "wowi: Utilities for detecting statistically significant spatial clusters of high acute malnutrition rates using SaTScan's Bernoulli spatial-scan model",
   author   = person("Tomás Zaba"),
   year     = 2025,
-  note     = "R package version 1.0.1",
+  note     = "R package version 1.0.2",
   url      = "https://tiwowi.github.io/wowi/"
 )

inst/app/ui.R

@@ -43,7 +43,7 @@ ui <- tagList(
       ),
 
       ### Right side of the page navigation bar ----
-      tags$span("v1.0.1",
+      tags$span("v1.0.2",
         id = "app-version",
         style = "font-size: 12.5px; color:  rgba(31, 42, 68, 0.58);
         position: fixed; top: 40px; right: 20px;"

tests/testthat/test-utils.R

@@ -25,6 +25,7 @@ testthat::test_that(
         .by = "zscores",
         oedema = oedema
       )
+
 
     ### Create a temporary directory ----
     tmp <- withr::local_tempdir() # ensures cleanup after test
@@ -80,5 +81,23 @@ testthat::test_that(
     testthat::expect_true(is.double(dplyr::pull(r[[1]][16])))
     testthat::expect_true(is.double(dplyr::pull(r[[1]][17])))
     testthat::expect_true(is.character(dplyr::pull(r[[1]][18])))
+
+  # ## Check if results are in the tibble are correct ----
+    df <- r$.df
+    testthat::expect_equal(df$nr_EAs[1], 36)
+    testthat::expect_equal(df$total_cases[1], 26)
+    testthat::expect_equal(df$"%_cases"[1], 7.8)
+    testthat::expect_equal(df$location_ids[1], "10,9")
+    testthat::expect_equal(df$geo[1], "34.113909 N, 3.087933 E")
+    testthat::expect_equal(df$radius[1], "1.20 km")
+    testthat::expect_equal(df$span[1], "1.20 km")
+    testthat::expect_equal(df$children[1], 25)
+    testthat::expect_equal(df$n_cases[1], 6)
+    testthat::expect_equal(df$expected_cases[1], 1.95)
+    testthat::expect_equal(df$observedExpected[1], 3.07)
+    testthat::expect_equal(df$relative_risk[1], 3.70)
+    testthat::expect_equal(df$"%_cases_in_area"[1], 24.0)
+    testthat::expect_equal(df$log_lik_ratio[1], 3.458213)
+    testthat::expect_equal(df$pvalue[1], 0.55)
   }
 )