From ca5e8b6a34ddd944397a12319128dec855ddbce0 Mon Sep 17 00:00:00 2001
From: JulioV <JulioV@users.noreply.github.com>
Date: Thu, 24 Oct 2019 16:27:43 -0400
Subject: [PATCH] Add communication sms metrics

---
 Snakefile                                |  8 +++++++-
 config.yaml                              | 12 +++++++++++-
 rules/features.snakefile                 | 11 +++++++++++
 src/features/communication_sms_metrics.R | 23 +++++++++++++++++++++++
 4 files changed, 52 insertions(+), 2 deletions(-)
 create mode 100644 rules/features.snakefile
 create mode 100644 src/features/communication_sms_metrics.R

diff --git a/Snakefile b/Snakefile
index 75719997..bd50a655 100644
--- a/Snakefile
+++ b/Snakefile
@@ -1,10 +1,16 @@
 configfile: "config.yaml"
 include: "rules/preprocessing.snakefile"
+include: "rules/features.snakefile"
 
 rule all:
     input:
         expand("data/raw/{pid}/{sensor}_raw.csv", pid=config["PIDS"], sensor=config["SENSORS"]),
-        expand("data/raw/{pid}/{sensor}_with_datetime.csv", pid=config["PIDS"], sensor=config["SENSORS"])
+        expand("data/raw/{pid}/{sensor}_with_datetime.csv", pid=config["PIDS"], sensor=config["SENSORS"]),
+        expand("data/processed/{pid}/com_sms_{sms_type}_{day_segment}_{metric}.csv",
+                            pid=config["PIDS"],
+                            sms_type = config["COM_SMS"]["SMS_TYPES"],
+                            day_segment = config["COM_SMS"]["DAY_SEGMENTS"],
+                            metric = config["COM_SMS"]["METRICS"]),
 
 # --- Packrat Rules --- #
 ## Taken from https://github.com/lachlandeer/snakemake-econ-r
diff --git a/config.yaml b/config.yaml
index fff3a17b..15433dd1 100644
--- a/config.yaml
+++ b/config.yaml
@@ -4,4 +4,14 @@ SENSORS: [messages]
 # Participants to include in the analysis
 # You must create a file for each participant
 # named pXXX containing their device_id
-PIDS: [p01, p02]
\ No newline at end of file
+PIDS: [p01, p02]
+
+# Global var with common day segments
+DAY_SEGMENTS: &day_segments
+  [daily, morning, afternoon, evening, night]
+
+# Communication SMS features config
+COM_SMS:
+  SMS_TYPES : [received, sent]
+  DAY_SEGMENTS: *day_segments
+  METRICS: [count, distinctcontacts]
\ No newline at end of file
diff --git a/rules/features.snakefile b/rules/features.snakefile
new file mode 100644
index 00000000..188513a7
--- /dev/null
+++ b/rules/features.snakefile
@@ -0,0 +1,11 @@
+rule communication_sms_metrics:
+    input: 
+        "data/raw/{pid}/messages_with_datetime.csv"
+    params:
+        sms_type = "{sms_type}",
+        day_segment = "{day_segment}",
+        metric = "{metric}"
+    output:
+        "data/processed/{pid}/com_sms_{sms_type}_{day_segment}_{metric}.csv"
+    script:
+        "../src/features/communication_sms_metrics.R"
\ No newline at end of file
diff --git a/src/features/communication_sms_metrics.R b/src/features/communication_sms_metrics.R
new file mode 100644
index 00000000..9aac5a85
--- /dev/null
+++ b/src/features/communication_sms_metrics.R
@@ -0,0 +1,23 @@
+source("packrat/init.R")
+
+library(dplyr)
+
+sms <-  read.csv(snakemake@input[[1]])
+day_segment <- snakemake@params[["day_segment"]]
+metric <-  snakemake@params[["metric"]]
+sms_type <-  snakemake@params[["sms_type"]]
+output_file <- snakemake@output[[1]]
+
+metrics <- sms %>% filter(message_type == ifelse(sms_type == "received", "1", "2"))
+
+if(day_segment == "daily"){
+  metrics <- metrics %>% group_by(local_date)
+} else {
+  metrics <- metrics %>% filter(day_segment == local_day_segment) %>% group_by(local_date)
+}
+
+metrics <- switch(metric,
+       "count" = metrics %>% summarise(!!paste("com", "sms", sms_type, day_segment, metric, sep = "_") := n()),
+       "distinctcontacts" = metrics %>% summarise(!!paste("com", "sms", sms_type, day_segment, metric, sep = "_") := n_distinct(trace)))
+
+write.csv(na.omit(metrics), output_file, row.names = F)
\ No newline at end of file