Add download_participants functionality

config.yaml

@@ -19,9 +19,16 @@ DAY_SEGMENTS: &day_segments
 TIMEZONE: &timezone
   America/New_York
 
+DATABASE_GROUP: &database_group
+  AAPECS
+
+DOWNLOAD_PARTICIPANTS:
+  IGNORED_DEVICE_IDS: [] # for example "5a1dd68c-6cd1-48fe-ae1e-14344ac5215f"
+  GROUP: *database_group
+
 # Download data config
 DOWNLOAD_DATASET:
-  GROUP: AAPECS
+  GROUP: *database_group
 
 # Readable datetime config
 READABLE_DATETIME:
@@ -112,4 +119,4 @@ STEP:
     ALL_STEPS: ["sumallsteps", "maxallsteps", "minallsteps", "avgallsteps", "stdallsteps"]
     SEDENTARY_BOUT: ["countsedentarybout", "maxdurationsedentarybout", "mindurationsedentarybout", "avgdurationsedentarybout", "stddurationsedentarybout"]
     ACTIVE_BOUT: ["countactivebout", "maxdurationactivebout", "mindurationactivebout", "avgdurationactivebout", "stddurationactivebout"]
-  THRESHOLD_ACTIVE_BOUT: 10 # steps
+  THRESHOLD_ACTIVE_BOUT: 10 # steps

rules/preprocessing.snakefile

@@ -1,3 +1,10 @@
+rule download_participants:
+    params:
+        group = config["DOWNLOAD_PARTICIPANTS"]["GROUP"],
+        ignored_device_ids = config["DOWNLOAD_PARTICIPANTS"]["IGNORED_DEVICE_IDS"]
+    script:
+        "../src/data/download_participants.R"
+
 rule download_dataset:
     input:
         "data/external/{pid}"

src/data/download_participants.R

@@ -0,0 +1,32 @@
+source("packrat/init.R")
+
+library(RMySQL)
+
+group <- snakemake@params[["group"]]
+ignored_device_ids <- snakemake@params[["ignored_device_ids"]]
+rmysql.settingsfile <- "./.env"
+
+stopDB <- dbConnect(MySQL(), default.file = rmysql.settingsfile, group = group)
+query <- paste0("SELECT device_id, brand FROM aware_device order by timestamp asc")
+participants <- dbGetQuery(stopDB, query)
+pids <- c()
+
+for(id in 1:nrow(participants)){
+    device_id <- participants$device_id[[id]]
+    brand <- ifelse(participants$brand[[id]] == "iPhone", "ios", "android")
+    if(!(device_id %in% ignored_device_ids)){
+        pid <- paste0("p", ifelse(id < 10, paste0("0", id), id))
+        pids <- append(pids, pid)
+        file_connection <- file(paste0("./data/external/", pid))
+        writeLines(c(device_id, brand), file_connection)
+        close(file_connection)
+    }
+}
+
+file_lines <-readLines("./config.yaml")
+for (i in 1:length(file_lines)){
+  if(startsWith(file_lines[i], "PIDS:")){
+    file_lines[i] <- paste0("PIDS: [", paste(pids, collapse = ", "), "]")
+  }
+}
+writeLines(file_lines, con = "./config.yaml")