From a015fbde70ec916066def3608b1d4452b45fb810 Mon Sep 17 00:00:00 2001
From: Ben Anderson <dataknut@icloud.com>
Date: Tue, 23 May 2017 13:38:03 +0100
Subject: [PATCH] updated function to create uniq pid & diarypid using a hash
function to ensure unique ids across all years (including UK TU 2014)
---
mtusFunctions.R | 34 +++++++++++++++++++++++++++++++---
1 file changed, 31 insertions(+), 3 deletions(-)
diff --git a/mtusFunctions.R b/mtusFunctions.R
index 4299ab4..6c30475 100644
--- a/mtusFunctions.R
+++ b/mtusFunctions.R
@@ -17,7 +17,8 @@ lb_myRequiredPackages <- function(x,y){
}
# Use the function to load the libraries required by this code
-reqLibs <- c("data.table" # fast data munching
+reqLibs <- c("data.table", # fast data munching
+ "openssl" # for hashing ids
)
print(paste0("Loading the following libraries using lb_myRequiredPackages: ", reqLibs))
@@ -30,6 +31,33 @@ ba_tidyNum <- function(number) {
format(number, big.mark=",", scientific=FALSE)
}
+###
+ba_MTUScreateIds <- function(dt){
+ # create a unique pid & unique diary pid that will be unique across all years
+ dt <- dt[, ba_pidChar := paste0(survey, swave, msamp, hldid,persid,sep= "_")] # force character
+ dt <- dt[, ba_pid := md5(ba_pidChar)] # hash
+ dt <- dt[, ba_diarypid := paste0(ba_pid, id, sep= "_")] # force character
+}
+
+# # diarypid
+# mtusUKEpsDT$ba_diarypid <-
+# group_indices(mtusUKEpsDT, survey,
+# swave,
+# msamp,
+# hldid,
+# persid,
+# id
+# )
+#
+# # pid
+# mtusUKEpsDT$ba_pid <-
+# group_indices(mtusUKEpsDT, survey,
+# swave,
+# msamp,
+# hldid,
+# persid
+# )
+
###
ba_MTUScreateEpisodeStartEndDateTimes <- function(dt){
# Setting up corrected start and end timestamps
@@ -53,8 +81,8 @@ ba_MTUScreateAllEpisodeStartTimesAsString <- function(dt){
# start time as string
# Use the fake start time we created earlier which includes all cases
dt <- dt[, st_hour := as.POSIXlt(dt$r_epStartDateTime)$hour]
- dt <- dt[, st_mins := as.POSIXlt(mtusUKEpsDT$r_epStartDateTime)$min]
- dt <- dt[, str_epStartTime := paste0(dt$st_hour,
+ dt <- dt[, st_mins := as.POSIXlt(dt$r_epStartDateTime)$min]
+ dt <- dt[, r_epStartTimeStr := paste0(dt$st_hour,
":",
dt$st_mins
)
--
GitLab