Closes #2526 data raw data round/2 (#2539)

* Corrected code for data-raw/create_admiral_*.R functions. But throws errors. * Delete data-backup directory * chore: #2526 remove old admiral adlb script; push up newly run data * chore: #2526 rename file to be explicit in what it is doing; reran queries code and some fun things were found!! /s --------- Co-authored-by: Ben Straub <[email protected]>
pharmaverse · Oct 25, 2024 · b44fd3d · b44fd3d
1 parent ed25276
commit b44fd3d
Show file tree

Hide file tree

Showing 11 changed files with 37 additions and 624 deletions.
diff --git a/data-backup/admiral_adlb.rda b/data-backup/admiral_adlb.rda
diff --git a/data-backup/admiral_adsl.rda b/data-backup/admiral_adsl.rda
diff --git a/data-raw/admiral_adlb.R b/data-raw/admiral_adlb.R
diff --git a/data-raw/create_admiral_adlb.R b/data-raw/create_admiral_adlb.R
@@ -1,61 +1,13 @@
 #  Create dataset:   data/admiral_adlb.rda
 #  This script:  create_admiral_adlb.R creates dataset data/admiral_adlb.rda.
-#
 
-# Preliminary
-library(diffdf) # nolint
+# Create dataset data/admiral_adlb.rda
 
-# To clarify directories (can be removed)
-# nolint start: object_name_linter
-CACHE_DIR <- "~/.cache/R/admiral_templates_data/"
-DATA_DIR <- "data-dir"
-DATA_RAW <- "data-raw"
-TEMPLATE_DIR <- "inst/templates/"
+# Run template script to create adlb
+source("inst/templates/ad_adlb.R", echo = TRUE) # nolint
 
-# clean CACHE_DIR
-THE_FILE <- paste0(CACHE_DIR, "/adlb.rda")
-THE_FILE <- paste0(CACHE_DIR, "/adsl.rda")
-if (file.exists(THE_FILE)) file.remove(THE_FILE)
-# nolint end
-
-#
-# STEPS
-#
-# First, use template to create the R script (in data-raw/admiral_adlb.R).
-# Next, source this script and create the data (~/.cache/R/admiral_template_data/admiral_adlb.rda)
-# Finally, shorten this data (now ~ 1.2 MB) by selecting only certain USUBJID
-
-# orignal method  - OMIT
-if (FALSE) {
-  # ### original mehtod (method 1)
-  # # First,  create the R script (from a template)
-  adam_name <- "adlb"
-  save_path <- paste0("./data-raw/admiral_", adam_name, ".R")
-  use_ad_template(
-    adam_name = adam_name,
-    save_path = save_path,
-    open = FALSE,
-    overwrite = TRUE
-  )
-  # Second, source the script and save data in .cache
-  source("data-raw/admiral_adlb.R") # nolint
-  load("~/.cache/R/admiral_templates_data/adlb.rda")
-}
-
-#
-# Instead, USE template, as recommened by Buzz
-#
-source(paste0(TEMPLATE_DIR, "/ad_adlb.R")) # nolint
-load(paste0(CACHE_DIR, "adlb.rda"))
-
-#
-# limit rows, by selecting only these USUBJID
-#
-#' 01-701-1015, 01-701-1023, 01-701-1028, 01-701-1033,
-#' 01-701-1034, 01-701-1047, 01-701-1097, 01-705-1186,
-#' 01-705-1292, 01-705-1310, 01-708-1286
-
-usubjid <-
+# Limit rows by selecting only these USUBJIDs
+usubjids <-
   c(
     "01-701-1015",
     "01-701-1023",
@@ -70,42 +22,17 @@ usubjid <-
     "01-708-1286"
   )
 
-#  prepare for inner join
-user <- tibble(
-  USUBJID = usubjid
-)
-result <- inner_join(adlb, user)
-admiral_adlb <- result
-admiral_adlb
-
-#
-# Finally, save reduced ds
-#
-use_data(admiral_adlb, overwrite = TRUE)
-
-#
-#  TEST - is dataset identical to .... backup of unaltered dataset
-#
-e1 <- new.env()
-e2 <- new.env()
-load("data/admiral_adlb.rda", e1)
-
-# CHANGE to YOUR location of original dataset
-load("data-backup/admiral_adlb.rda", e2)
-
-# compare field names
-t <- tibble(e1 = names(e1$admiral_adlb), e2 = names(e2$admiral_adlb))
-t |> print(n = 111)
+admiral_adlb <- filter(adlb, USUBJID %in% usubjids)
 
-identical(e1$admiral_adlb, e2$admiral_adlb)
-diffdf(e1$admiral_adlb, e2$admiral_adlb)
+# Get previous dataset for comparison
+adlb_old <- admiral::admiral_adlb
 
-## Capture diffdf to file
+# Finally, save reduced dataset
+usethis::use_data(admiral_adlb, overwrite = TRUE)
 
-capture.output(
-  diffdf(
-    compare = e1$admiral_adlb, base = e2$admiral_adlb,
-    keys = c("STUDYID", "DOMAIN", "USUBJID", "AVAL", "VISIT")
-  ),
-  file = "data-raw/diffdf_adlb_23SEPT"
+# Compare with previous version
+diffdf::diffdf(
+  base = adlb_old,
+  compare = admiral_adlb,
+  keys = c("USUBJID", "PARAMCD", "AVISIT", "ADT")
 )
diff --git a/data-raw/create_admiral_adsl.R b/data-raw/create_admiral_adsl.R
@@ -1,78 +1,23 @@
 #  Create dataset:   data/admiral_adsl.rda
-#  This script:  create_admiral_adsl.R creates dataset data/admiral_adsl.rda.
+#  This script, create_admiral_adsl.R, creates dataset data/admiral_adsl.rda.
 #
 
-# Preliminary
-library(diffdf) # nolint
+# Create dataset data/admiral_adsl.rda
 
+# Run template script to create adsl
+source("inst/templates/ad_adsl.R", echo = TRUE) # nolint
 
-# nolint start: object_name_linter.
-# To clarify directories (can be removed)
-CACHE_DIR <- "~/.cache/R/admiral_templates_data/"
-DATA_DIR <- "data-dir"
-DATA_RAW <- "data-raw"
-TEMPLATE_DIR <- "inst/templates/"
+admiral_adsl <- adsl # use correct name
 
-# clean CACHE_DIR
-THE_FILE <- paste0(CACHE_DIR, "/adlb.rda")
-THE_FILE <- paste0(CACHE_DIR, "/adsl.rda")
-if (file.exists(THE_FILE)) file.remove(THE_FILE)
-# nolint end
-
-#
-# STEPS
-#
-# First, use template to create the R script (in data-raw/admiral_adlb.R).
-# Next, source this script and create the data (~/.cache/R/admiral_template_data/admiral_adlb.rda)
-# Finally, shorten this data (now ~ 1.2 MB) by selecting only certain USUBJID
-
-# OMIT -- orignal method  - OMIT
-if (FALSE) {
-  # ### original mehtod (method 1)
-  # # First,  create the R script (from a template)
-  adam_name <- "adlb"
-  save_path <- paste0("./data-raw/admiral_", adam_name, ".R")
-  use_ad_template(
-    adam_name = adam_name,
-    save_path = save_path,
-    open = FALSE,
-    overwrite = TRUE
-  )
-  # Second, source the script and save data in .cache
-  source("data-raw/admiral_adlb.R") # nolint
-  load("~/.cache/R/admiral_templates_data/adlb.rda")
-}
-
-#
-#
-# Instead, USE template, as recommened by Buzz
-#
-source(paste0(TEMPLATE_DIR, "/ad_adsl.R")) # nolint
-load(paste0(CACHE_DIR, "adsl.rda"))
-admiral_adsl <- adsl
-
-#
-# Finally, save reduced ds
-#
-use_data(admiral_adsl, overwrite = TRUE)
-
-#
-#  TEST - is dataset identical to .... backup of unaltered dataset
-#
-e1 <- new.env()
-e2 <- new.env()
-load("data/admiral_adsl.rda", e1)
-
-# CHANGE to YOUR location of original dataset
-load("data-backup/admiral_adsl.rda", e2)
+# Get previous dataset for comparison
+adsl_old <- admiral::admiral_adsl
 
+# Finally, save reduced dataset
+usethis::use_data(admiral_adsl, overwrite = TRUE)
 
-identical(e1$admiral_adsl, e2$admiral_adsl)
-diffdf(compare = e1$admiral_adsl, base = e2$admiral_adsl, keys = c("STUDYID", "USUBJID"))
-capture.output(
-  diffdf(
-    compare = e1$admiral_adsl, base = e2$admiral_adsl,
-    keys = c("STUDYID", "USUBJID")
-  ),
-  file = "data-raw/diffdf_23SEPT"
+# Compare with previous version
+diffdf::diffdf(
+  base = adsl_old,
+  compare = admiral_adsl,
+  keys = c("STUDYID", "USUBJID")
 )
diff --git a/data-raw/derive_single_dose.R → data-raw/create_ex_single.R b/data-raw/derive_single_dose.R → data-raw/create_ex_single.R
diff --git a/data-raw/example_qs.R → data-raw/create_example_qs.R b/data-raw/example_qs.R → data-raw/create_example_qs.R
diff --git a/data-raw/example_query_source.R → data-raw/create_queries_and_queries_mh.R b/data-raw/example_query_source.R → data-raw/create_queries_and_queries_mh.R
@@ -54,13 +54,20 @@ adae <- tibble::tribble(
   "APPLICATION SITE ERYTHEMA", "Application site erythema", 1
 )
 
+#  create queries.rda in data/
+usethis::use_data(queries, overwrite = TRUE)
+
 # try below:
 derive_vars_query(adae, queries)
 
-
 # example to use for ADMH:
 queries_mh <- queries %>%
   filter(SRCVAR %in% c("AELLT", "AEDECOD")) %>%
   mutate(SRCVAR = ifelse(SRCVAR == "AELLT", "MHLLT", "MHDECOD"))
 
+usethis::use_data(queries_mh, overwrite = TRUE)
+
+source("inst/templates/ad_admh.R", echo = TRUE) # nolint
+
+# Currently producing an error
 derive_vars_query(admh, queries_mh)
diff --git a/data/admiral_adlb.rda b/data/admiral_adlb.rda
diff --git a/data/queries.rda b/data/queries.rda
diff --git a/data/queries_mh.rda b/data/queries_mh.rda