should have saved already

Author: libjohn

transform_and_merge_01.Rmd

@@ -5,19 +5,24 @@ output: html_notebook
 
 **Goal:** identify countries on opposite sides
 
+## Disclaimers
+
+- User should verify the results
+- There is a potential to generate more concise code.  Meanwhile, this appears to work just fine.
+
 ## Load Libraries
 
-```{r}
+```{r loadpackages}
 library(tidyverse)
 ```
 
 
 ## Example Data
 
-```{r}
-country <- c("A","B","C","D","A","E","F","A","C","D","G","H")
+```{r sampledata}
+country <- c("A","B","C","D","A","E","F","X","Y","Z","Q","R")
 dispute <- c(1,1,1,1,2,2,2,3,3,3,3,3)
-side_a <- c(0,0,1,1,0,1,1,0,1,0,1,0)
+side_a <- c(0,0,1,1,0,1,1,0,0,0,1,1)
 side_b <- c(1,1,0,0,1,0,0,1,1,1,0,0)
 
 country_df <- data_frame(country, dispute, side_a, side_b)
@@ -30,7 +35,7 @@ country_df
 
 ### Side B Countries
 
-```{r}
+```{r sidea}
 side_a <- country_df %>% 
   filter(side_a == 0)
   
@@ -39,7 +44,7 @@ side_a
 
 ### Side A Countries
 
-```{r}
+```{r sideb}
 side_b <- country_df %>% 
   filter(side_a == 1)
   
@@ -49,41 +54,43 @@ side_b
 
 ## Transform Data ; Build Data Frame
 
-```{r}
+```{r for_df}
 
 # declare Tibble (data frame)
 df_total = tibble()
 
 for (i in 1:max(country_df$dispute)) {
-  # Build vector of disputants for each Dispute Category
+  # Build vector of disputants (multi-valued cell: country_y) 
   country_y <- country_df %>% 
     filter(dispute == i) %>% 
     select(-dispute) %>% 
-    mutate(side_a = if_else(side_a == 0, country, NA_character_),
-           side_b = if_else(side_b == 0, country, NA_character_)) %>% 
+    mutate(side_b = if_else(side_b == 0, country, NA_character_)) %>% 
     select(side_b) %>% 
-    na.omit() %>% 
+    drop_na() %>% 
     unique() %>% 
-    .[[1]] %>% 
-    paste(., collapse = "|") 
+    as_vector() %>% 
+    paste0(collapse = "|") 
   
   # Build a data frame that adds the disputants vector as a column-variable; 
-  # drops blank rows ; then split the multi-valued cells (disputants)
+  # drops blank rows ; separate the multi-valued cells (disputants/country_y)
+  # into rows
   country_df_temp <- country_df %>% 
     filter(dispute == i) %>% 
-    mutate(side_a = if_else(side_a == 0, country, NA_character_),
-           country_disputes = if_else(side_b == 1, country_y, NA_character_)) %>% 
-    select(country, country_disputes) %>% 
+    mutate(country_disputes = if_else(side_b == 1, country_y, NA_character_)) %>% 
     drop_na() %>% 
-    separate_rows(country_disputes, convert = TRUE) %>% 
-    arrange(country_disputes) %>% 
-    mutate(Dispute = i)
+    separate_rows(country_disputes, convert = TRUE) 
   
   # Append big Dataframe from from the iterated dataframes
   df_total <- bind_rows(df_total, country_df_temp)
   
 }
 
-df_total
+df_total%>% 
+  arrange(country_disputes) %>% 
+  rename(country_x = country,
+         counrty_y = country_disputes) %>% 
+  select(dispute, counrty_y, country_x)
+
+```
+
 
-```