merge main

Merge branch 'main' into use-delphidocs

# Conflicts:
#	DESCRIPTION

Author: dajmcdon

DESCRIPTION

@@ -2,32 +2,52 @@ Type: Package
 Package: epidatasets
 Title: Epidemiological Data for Delphi Tooling Examples
 Version: 0.0.1
-Authors@R: 
-    person(c("Daniel", "J."), "McDonald", , "daniel@stat.ubc.ca", role = c("cre", "aut"))
+Authors@R: c(
+    person(c("Daniel", "J."), "McDonald", , "daniel@stat.ubc.ca", role = "aut"),
+    person("Nat", "DeFries", , "ndefries@andrew.cmu.edu", role = c("cre", "aut")),
+    person("Johns Hopkins University Center for Systems Science and Engineering", role = "dtc",
+           comment = "Owner of COVID-19 cases and deaths data from the COVID-19 Data Repository"),
+    person("Johns Hopkins University", role = "cph",
+           comment = "Copyright holder of COVID-19 cases and deaths data from the COVID-19 Data Repository"),
+    person("Carnegie Mellon University Delphi Group", role = "dtc",
+           comment = "Owner of masking, social-distancing, and CLI data from the COVID-19 Trends and Impacts Survey. Owner of claims-based CLI data from the Delphi Epidata API"),
+    person("The COVID-19 Canada Open Data Working Group", role = "dtc",
+           comment = "Owner of Canadian COVID-19 cases rates from the Covid19Canada data repository"),
+    person("Statistics Canada", role = "dtc",
+           comment = "Owner of Canadian graduate employment income data from the Statistics Canada website"),
+    person("Google", role = "dtc",
+           comment = "Collaborator on CLI data from the Google symptom surveys")
+  )
 Description: This package contains data sets used to compile vignettes and
     other documentation in Delphi R Packages. The goal is to avoid calls
-    to the Delphi Epidata API, and deposit some examples here for easy
+    to the Delphi Epidata API, and to deposit some examples here for easy
     offline use.
 License: MIT + file LICENSE
-URL: https://cmu-delphi.github.io/epidatasets/
+URL: https://github.com/cmu-delphi/epidatasets,
+    https://cmu-delphi.github.io/epidatasets/
 Depends:
     R (>= 2.10)
 Suggests: 
     covidcast,
+    data.table,
     dplyr,
     epidatr,
-    epipredict,
     here,
+    httr,
+    jsonlite,
     lubridate,
     magrittr,
     purrr,
     readr
+Enhances:
+    epiprocess (>= 0.9.0),
+    tibble
 Remotes:
     cmu-delphi/delphidocs,
     cmu-delphi/epidatr,
-    cmu-delphi/epipredict,
     cmu-delphi/epiprocess
 Config/Needs/website: cmu-delphi/delphidocs
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 7.2.3
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.3.2

LICENSE

@@ -1,2 +1,2 @@
-YEAR: 2023
-COPYRIGHT HOLDER: epidatasets authors
+This contains a collection of data from different sources under different
+licenses; please see the documentation for each object for license information.

LICENSE.md

@@ -0,0 +1,26 @@
+library(here)
+
+internal_data_path <- here("R/sysdata.rda")
+
+save_to_sysdata <- function(obj, obj_name) {
+  # Create a new environment for storing sysdata contents
+  sysdata_env <- new.env(hash = FALSE, parent = emptyenv())
+
+  # Load current internal data into this new environment
+  if(file.exists(internal_data_path)) {
+    load(internal_data_path, envir = sysdata_env)
+  }
+
+  # Add or replace the object
+  sysdata_env[[obj_name]] <- obj
+
+  # Save the environment back to the internal package datafile
+  save(
+    list = names(sysdata_env),
+    file = internal_data_path,
+    envir = sysdata_env,
+    compress = "xz",
+    # For backwards compatibility with older R versions (<3.5)
+    version = 2
+  )
+}

R/epipredict-data.R

@@ -0,0 +1,16 @@
+library(here)
+
+internal_data_path <- here("data-raw")
+files <- list.files(
+  internal_data_path, pattern = ".*[.]R",
+  full.names = FALSE
+)
+for (file in files) {
+  if (startsWith(file, "_")) {
+    # File is a helper script and does not generate data.
+    next
+  }
+  path <- here(file.path("data-raw", file))
+  message("running ", path, " ...")
+  source(path)
+}

R/epiprocess-data.R

@@ -0,0 +1,49 @@
+library(dplyr)
+library(epidatr)
+library(epiprocess)
+
+source(here::here("data-raw/_helper.R"))
+
+dv_subset <- pub_covidcast(
+  source = "doctor-visits",
+  signals = "smoothed_adj_cli",
+  time_type = "day",
+  geo_type = "state",
+  time_values = epirange(20200601, 20211201),
+  geo_values = "*",
+  issues = epirange(20200601, 20211201)
+) %>%
+  select(geo_value, time_value, version = issue, percent_cli = value) %>%
+  # Drop DC and territories.
+  filter(!(geo_value %in% c("as", "gu", "dc", "mp", "pr", "vi"))) %>%
+  # We're using compactify=FALSE here and below to avoid some testthat test
+  # failures on tests that were based on a non-compactified version.
+  as_epi_archive(compactify = FALSE)
+
+case_rate_subset <- pub_covidcast(
+  source = "jhu-csse",
+  signals = "confirmed_7dav_incidence_prop",
+  time_type = "day",
+  geo_type = "state",
+  time_values = epirange(20200601, 20211201),
+  geo_values = "*",
+  issues = epirange(20200601, 20211201)
+) %>%
+  select(geo_value, time_value, version = issue, case_rate_7d_av = value) %>%
+  filter(!(geo_value %in% c("as", "gu", "dc", "mp", "pr", "vi"))) %>%
+  as_epi_archive(compactify = FALSE)
+
+# Use `epiprocess::epix_merge` to avoid having to reimplement `sync`ing
+# behavior. After merging, convert DT component back to tibble.
+archive_cases_dv_subset_all_states_tbl = epix_merge(
+  dv_subset, case_rate_subset,
+  sync = "locf",
+  compactify = TRUE)$DT %>%
+  as_tibble()
+
+# We're trying to do:
+#   usethis::use_data(archive_cases_dv_subset_all_states_tbl, internal = TRUE, overwrite = TRUE, compress = "xz")
+# but `usethis::use_data` can only store multiple objects if they're added in
+# the same call. This workaround is from
+# https://github.com/r-lib/usethis/issues/1512
+save_to_sysdata(archive_cases_dv_subset_all_states_tbl, "archive_cases_dv_subset_all_states_tbl")

R/sysdata.rda

@@ -1,40 +1,46 @@
-dv_subset <- covidcast(
-  data_source = "doctor-visits",
+library(dplyr)
+library(epidatr)
+library(epiprocess)
+
+source(here::here("data-raw/_helper.R"))
+
+dv_subset <- pub_covidcast(
+  source = "doctor-visits",
   signals = "smoothed_adj_cli",
   time_type = "day",
   geo_type = "state",
   time_values = epirange(20200601, 20211201),
   geo_values = "ca,fl,ny,tx",
   issues = epirange(20200601, 20211201)
 ) %>%
-  fetch() %>%
   select(geo_value, time_value, version = issue, percent_cli = value) %>%
   # We're using compactify=FALSE here and below to avoid some testthat test
   # failures on tests that were based on a non-compactified version.
   as_epi_archive(compactify = FALSE)
 
-case_rate_subset <- covidcast(
-  data_source = "jhu-csse",
+case_rate_subset <- pub_covidcast(
+  source = "jhu-csse",
   signals = "confirmed_7dav_incidence_prop",
   time_type = "day",
   geo_type = "state",
   time_values = epirange(20200601, 20211201),
   geo_values = "ca,fl,ny,tx",
   issues = epirange(20200601, 20211201)
 ) %>%
-  fetch() %>%
   select(geo_value, time_value, version = issue, case_rate_7d_av = value) %>%
   as_epi_archive(compactify = FALSE)
 
-archive_cases_dv_subset = epix_merge(
+# Use `epiprocess::epix_merge` to avoid having to reimplement `sync`ing
+# behavior. After merging, convert DT component back to tibble.
+archive_cases_dv_subset_tbl = epix_merge(
   dv_subset, case_rate_subset,
   sync = "locf",
-  compactify = FALSE)
-
-# If we directly store an epi_archive R6 object as data, it will store its class
-# implementation there as well. To prevent mismatches between these stored
-# implementations and the latest class definition, don't store them as R6
-# objects; store the DT and construct the R6 object on request.
-archive_cases_dv_subset_dt <- archive_cases_dv_subset$DT
+  compactify = FALSE)$DT %>%
+  as_tibble()
 
-usethis::use_data(archive_cases_dv_subset_dt, overwrite = TRUE)
+# We're trying to do:
+#   usethis::use_data(archive_cases_dv_subset_tbl, internal = TRUE, overwrite = TRUE, compress = "xz")
+# but `usethis::use_data` can only store multiple objects if they're added in
+# the same call. This workaround is from
+# https://github.com/r-lib/usethis/issues/1512
+save_to_sysdata(archive_cases_dv_subset_tbl, "archive_cases_dv_subset_tbl")