Tree Traverse

traverse_tree.R

if (!require("pacman")) install.packages("pacman")
pacman::p_load(data.tree, tidyverse)

dat <- data_frame(

    OrgUnitIdentifier = c(1, 6, 72, 147, 1471, 1553, 1771, 283, 3762, 1112, 31, 4353, 9374, 2612),
    Name = c(
        'Main University', 'Academic Affairs', 'College of Arts & Science', 'Biology', 'Biochemistry & Molecular Bio', 
        'Earth Sciences', 'Environmental Studies', 'Chemistry/Physics', 'Chemistry', 'Physics', 'College of Health Sciences', 
        'Health Sciences', 'Occupational Therapy', 'Social Work'
    ),
    ParentIdentifier = c(NA, 1, 6, 72, rep(147, 3), 72, rep(283, 2), 6, rep(31, 3))
    
)

##    OrgUnitIdentifier                         Name ParentIdentifier
##                <dbl>                        <chr>            <dbl>
##  1                 1              Main University               NA
##  2                 6             Academic Affairs                1
##  3                72    College of Arts & Science                6
##  4               147                      Biology               72
##  5              1471 Biochemistry & Molecular Bio              147
##  6              1553               Earth Sciences              147
##  7              1771        Environmental Studies              147
##  8               283            Chemistry/Physics               72
##  9              3762                    Chemistry              283
## 10              1112                      Physics              283
## 11                31   College of Health Sciences                6
## 12              4353              Health Sciences               31
## 13              9374         Occupational Therapy               31
## 14              2612                  Social Work               31

## Handling the NA/NULL/Missing for root
key <- dat %>%
    dplyr::select(OrgUnitIdentifier, Name) %>%
    stats::setNames(c('id', 'name'))

key2 <- dat %>%
    dplyr::select(OrgUnitIdentifier, ParentIdentifier)



## iterate up the tree
struct <- apply(key2, 1, function(x){

    if (is.na(x[['ParentIdentifier']])) return(x[['OrgUnitIdentifier']])
    par <- x[['ParentIdentifier']]
    path <- unlist(x)
    i <- length(path) + 1
    while(!is.na(par)){

        path[i] <- key2[['ParentIdentifier']][match(par, key2[['OrgUnitIdentifier']])]
        par <- path[i]
        i <- i + 1
  
    }

    return(rev(c(stats::na.omit(unname(path)))))
})


tree <- struct  %>%
    textshape::tidy_list('org', 'path') %>%
    dplyr::left_join(key, by = c('path' = 'id')) %>%
    dplyr::group_by(org) %>%
    dplyr::summarize(pathString = paste(unlist(name), collapse = '<>')) %>%
    data.tree::as.Node(pathDelimiter = '<>')


ptree <- gsub("\\G\\d", ' ', capture.output(tree), perl = TRUE)[-1]
rep <- paste('^', gsub('(^\\s+)(\\s\\S.+$)', '\\1', ptree[1]))

cat(paste(gsub(rep, '', ptree), collapse = '\n'), '\n')

## Main University                             
##  °--Academic Affairs                        
##      ¦--College of Arts & Science           
##      ¦   ¦--Chemistry/Physics               
##      ¦   ¦   ¦--Physics                     
##      ¦   ¦   °--Chemistry                   
##      ¦   °--Biology                         
##      ¦       ¦--Biochemistry & Molecular Bio
##      ¦       ¦--Earth Sciences              
##      ¦       °--Environmental Studies       
##      °--College of Health Sciences          
##          ¦--Health Sciences                 
##          ¦--Occupational Therapy            
##          °--Social Work