index.Rmd

---
title: ""
subtitle: " "
output: html_document
---

```{r, include = FALSE}
#need these maybe to create lists of attributes about the repos:
#r-on-github - to list functions within repos....etc.
#gh - to list repos
#maybe just search websites?
```


```{r, include = FALSE, eval = FALSE}
library(tidyverse)
library(magrittr)
#getting the webpage
#URL_OCS <- c("https://github.com/opencasestudies?tab=repositories") #old version
URL_OCS <- c("https://github.com/orgs/opencasestudies/repositories")
webpage <- URL_OCS %>%
  xml2::read_html() 
webpage
#selecting mostly just the repoinfo
webpage_element <- webpage %>%
  rvest::html_nodes(xpath='//*[contains(concat( " ", @class, " " ), concat( " ", "d-inline-block", " " ))]')
attrs_ocs <- webpage_element %>%
  rvest::html_attrs()
text_ocs <-webpage_element %>%
  rvest::html_text()
#getting the names of the repos from the GitHUB page
repo_text<-map_df(attrs_ocs, as_tibble) 
repo_text %<>% 
  mutate(is_case_study = str_detect(value, "opencasestudies/ocs")) %>%
  filter(is_case_study == "TRUE") %>%
  mutate(is_hovercard = str_detect(value, "hover")) %>%
  filter(is_hovercard == "FALSE") %>%
  mutate(case_study_name = str_remove(value, "/opencasestudies/")) %>%
  mutate(interactive = str_detect(value, "interactive")) %>%
  filter(interactive == FALSE) %>%
    select(repo_name = value, case_study_name)
repo_non_bp <- repo_text %>% filter(!str_detect(repo_name, "-bp"))
repo_bp <- repo_text %>% filter(str_detect(repo_name, "-bp")) %>%
                          filter(!str_detect(repo_name, "template|firearm|flex|v2"))
  
# run the fist chunk first!
repo_non_bp
repo_non_bp <- repo_non_bp %>% filter(!str_detect(repo_name, "firearm")) # get just the health expenditure case study for now - not firearm legislation
# add the health expenditure case study to the list
repo_bp <-rbind(repo_bp, repo_non_bp)
```


```{r, include = FALSE, eval = FALSE}
# create table about repo and html location and names for case studies
OCS_info <- repo_bp %>%
  tibble(
    repo_url =  paste0("https://github.com/", repo_name), 
    case_study_url =  paste0("https://www.opencasestudies.org/", case_study_name))
```


```{r, include = FALSE, eval = FALSE}
#Get Readmes - one example
URL_readme <-c('https://github.com/opencasestudies/ocs-bp-co2-emissions')
readme_ex <-xml2::read_html(URL_readme)
webpage_element <- readme_ex %>%
        rvest::html_nodes(xpath='//*[contains(concat( " ", @class, " " ), concat( " ", "container-lg", " " ))]')
text_ocs <- webpage_element %>%
  rvest::html_text()
text_ocs %<>% str_split("\n") %>%
             as_tibble(unlist(.), .name_repair = ~ c("Text"))
text_ocs %>% mutate(near_title = str_detect(string = Text, pattern = "Title"))
text_ocs
```


```{r, include = FALSE, eval = FALSE}
##in case we need something else from the readme
# get_readme <- function(URL){
#   xml2::read_html(URL) %>%
#   rvest::html_nodes(xpath='//*[contains(concat( " ", @class, " " ), concat( " ", "container-lg", " " ))]') %>%
#   rvest::html_text()}
# 
# URL <- OCS_info$repo_url
# Readmes <-pmap(list(URL), get_readme)
```

```{r, include = FALSE, eval = FALSE}
get_title <-function(URL){
  xml2::read_html(URL) %>%
  rvest::html_nodes(xpath='//*[contains(concat( " ", @class, " " ), concat( " ", "toc-ignore", " " ))]') %>%
  rvest::html_text()
}
URL <- OCS_info$case_study_url
Titles <-pmap(list(URL), get_title)
Titles %<>% str_remove("Open Case Studies: ") %>%
  str_remove("Open Case Studies : ") %>%
  as_tibble()
```

```{r, include = FALSE, eval = FALSE}
  URL_readme <-c('https://github.com/opencasestudies/ocs-bp-RTC-Wrangling')
Packages_used <-  xml2::read_html(URL_readme) %>%
        rvest::html_nodes(xpath='//td | //th') %>%
        rvest::html_text() %>%
  as_tibble(unlist(.), .name_repair = ~c("Packages")) %>%
  mutate(is_package = !str_detect(Packages, " |Package")) %>% # if there is a space or the word package = FALSE
  filter(is_package == "TRUE") %>%
  select(Packages) %>%
  filter(Packages != "Data")
Get_packages <-function(URL){
  xml2::read_html(URL) %>%
  rvest::html_nodes(xpath='//td | //th') %>%
  rvest::html_text() %>%
  as_tibble(unlist(.), .name_repair = ~c("Packages")) %>%
  mutate(is_package = !str_detect(Packages, " |Package")) %>% # if there is a space or the word package = FALSE
  filter(is_package == "TRUE") %>%
  select(Packages) %>%
  filter(Packages != "Data") %>%
  filter(Packages != "Link") %>%
  filter(Packages != "link") %>%
  filter(Packages != "Use") %>% 
  arrange(Packages)
}
URL <- OCS_info$repo_url
Packages<-pmap(list(URL), Get_packages)
  
```

```{r, include = FALSE, eval = FALSE}
#get Data Science objectives
  URL_readme <-c('https://github.com/opencasestudies/ocs-bp-co2-emissions')
Objectives<-  xml2::read_html(URL_readme) %>%
        rvest::html_nodes(xpath='//li') %>%
        rvest::html_text()
Obj_and_quest <- Objectives %>% as_tibble() %>%
  filter(!str_detect(value, "(?<= )\n{2,}|(\n(?=[:space:]))|→")) %>%
  filter(str_detect(value, "[:space:]")) %>%
  filter(!str_detect(value, "GitHub|Sales|plans|management|guides|Code review|forum|Lab"))
Questions <- Obj_and_quest %>% filter(str_detect(value, "\\?"))
Objectives <- Obj_and_quest %>% filter(!str_detect(value, "\\?"))
#create functions to do this:
get_questions <- function(URL){
  xml2::read_html(URL) %>%
        rvest::html_nodes(xpath='//li') %>%
        rvest::html_text() %>%
        as_tibble() %>%
        filter(!str_detect(value, "(?<= )\n{2,}|(\n(?=[:space:]))|→")) %>%
        filter(str_detect(value, "[:space:]")) %>%
        filter(!str_detect(value, "GitHub|Sales|plans|management|guides|Code review|forum|Lab"))%>%
        filter(str_detect(value, "\\?"))%>%
        mutate(value = str_remove(value, "\n"))
}
get_objectives <- function(URL){
  xml2::read_html(URL) %>%
        rvest::html_nodes(xpath='//li') %>%
        rvest::html_text() %>%
        as_tibble() %>%
        filter(!str_detect(value, "(?<= )\n{2,}|(\n(?=[:space:]))|→")) %>%
        filter(str_detect(value, "[:space:]")) %>%
        filter(!str_detect(value, "GitHub|Sales|plans|management|guides|Code review|forum|Lab"))%>%
        filter(!str_detect(value, "\\?")) %>%
        mutate(value = str_remove(value, "\n")) %>%
        filter(!str_detect(value, "Donohue|Lott|Census|Kristen|themagick|compositionsfor|Bureau|A |guide|discussion|Greatvideo|More examples|consciousness|erratic|bluish|limp|snore|clammy|purplish|modifythese|acrossyears|unable to talk|Anexplanation|Alist|cheatsheets|Acourse|http|Health Insurance|Health Care"))
}
```

```{r, include = FALSE, eval = FALSE}
library(magrittr)
URL <- OCS_info$repo_url
Questions<-pmap(list(URL), get_questions)
Objectives <-pmap(list(URL), get_objectives)
#Collapse the individual tibbles into strings
make_string <- function(vector){
  toString(pull(vector))
}
Packages_again <- pmap(list(Packages), make_string) %>% unlist()
     Obj_again <- pmap(list(Objectives), make_string) %>% unlist()
Questions_again <- pmap(list(Questions), make_string) %>% unlist()
#Combine all the info and make the links with the appropriate names and open in a new tab (target="_blank")
OCS_info %<>%
  mutate(Packages = Packages_again,
         Title = Titles$value,
         Questions = Questions_again,
         Objectives = Obj_again)
OCS_info %<>%
  # mutate(`GitHub Repository` =  paste0('<a href="',repo_url,'"',' target="_blank"','>',repo_name,"</a>"), # older version... before just showing "static version as text for link"
  mutate(`GitHub Repository` =  paste0('<a href="', repo_url,'"',' target="_blank"','>',"Static Version Repository","</a>"), 
         `Case Study` =  paste0('<a href="',case_study_url,'"',' target="_blank"','>',Title,"</a>"))
OCS_info %<>% 
  mutate(Category = case_when(str_detect(string = case_study_name, pattern = "bp") ~ 'Bloomberg American Health Initiative <br></br><img src="https://opencasestudies.github.io/img/icon-bahi.png" height="52"> </img>', TRUE ~ 'General <br></br><img src="https://opencasestudies.github.io/img/icon.png" height="52"> </img>' ))
save(OCS_info, file = here::here( "ocs_info2.rda"))
```


```{r table-display-prep, include = FALSE}
#Start here!
library(tidyverse)
library(widgetframe)
library(magrittr)
```
   
```{r, include = FALSE}
library(RCurl)
load(file = here::here("ocs_info2.rda"))
#First create url for interactive version
OCS_info %<>%
  mutate(Interactive_url = paste0(str_replace( string = case_study_url, pattern = "https://www.opencasestudies.org/", replacement = "https://rsconnect.biostat.jhsph.edu/"), "-interactive"))
# Then add image for link
OCS_info %<>%
  mutate(Interactive_version = paste0('<br></br>', '<a href="', Interactive_url,'"',' target="_blank"','>', 'Interactive Version:<br></br><img src=\"https://rsconnect.biostat.jhsph.edu/ocs-bp-opioid-rural-urban-interactive/_w_fa6f6915/www/img/interactive_logo.png" height=\"52\"> </img>', '</a>'))
# add bloomberg website link
OCS_info %<>% mutate(Category = paste0('<br></br> <a href=\"https://americanhealth.jhu.edu/open-case-studies" target=\"_blank\"> <br></br>', Category))
```



```{r, include = FALSE}
#Adding static version text
OCS_info %<>% mutate(Static_version =  paste0('<br></br>', '<a href="', case_study_url,'"',' target="_blank"','>', 'Static Version:<br></br><img src=\"https://opencasestudies.github.io/img/icon-bahi.png" height=\"52\"></img></a>'))
OCS_info %<>% mutate('Case Study' =paste0(OCS_info$Title, OCS_info$Static_version))
#Only add interactive version if url exists
OCS_info %<>% mutate(`Case Study` = case_when(
  url.exists(Interactive_url) == TRUE ~ paste0(`Case Study`, Interactive_version), 
  url.exists(Interactive_url) == FALSE ~ `Case Study`))
```

```{r, include = FALSE}
save(OCS_info, file = here::here( "ocs_info_int.rda"))
```


```{r, include = FALSE}
# adding interactive repo link if separate repo
load(file = here::here("ocs_info_int.rda"))
OCS_info %<>% mutate(int_repo_url = paste0(repo_url, "-interactive"))
OCS_info %<>% mutate(int_repo_version =  paste0('<a href="', int_repo_url,'"',' target="_blank"','>',"<br></br>Interactive Version Repository","</a>"))
      
#Only add interactive version repo if url exists               
OCS_info %<>% mutate(`GitHub Repository` = case_when(
  url.exists(int_repo_url) == TRUE & str_detect(pattern = "Interactive", OCS_info$`Case Study`) == TRUE ~ paste0(`GitHub Repository`, int_repo_version), 
  TRUE ~ `GitHub Repository`))
```

```{r, include = FALSE}
#adding interactive repo link if branch and already not independent int repo
OCS_info %<>% mutate(int_repo_branch = paste0(repo_url, "/tree/interactive"))
OCS_info %<>% mutate(int_repo_version =  paste0('<a href="', int_repo_branch,'"',' target="_blank"','>',"<br></br>Interactive Version Repository","</a>"))
      
#Only add interactive version repo if url exists               
OCS_info %<>% mutate(`GitHub Repository` = case_when(
  url.exists(int_repo_branch) == TRUE & str_detect(pattern = "Interactive", OCS_info$`Case Study`) == TRUE & str_detect(pattern = "Interactive", OCS_info$`GitHub Repository`) == FALSE ~ paste0(`GitHub Repository`, int_repo_version),
  TRUE ~ `GitHub Repository`))
```

```{r, include = FALSE}
#change health expenditure icon for static version
OCS_info %<>% mutate('Case Study' = case_when(
    str_detect(Category, pattern = "General") == TRUE ~ str_replace(string =OCS_info$`Case Study`, pattern = "icon-bahi.png", replacement = "icon.png"), 
    TRUE ~OCS_info$'Case Study'))
save(OCS_info, file = here::here( "ocs_info_int_with_repo.rda"))
```

```{r table-display, echo = FALSE}
library(DT)
library(dplyr)
load(file = here::here("ocs_info_int_with_repo.rda"))
OCS_info_table <- OCS_info %>% 
  select(`Case Study`, `GitHub Repository`, Packages, Objectives, Category)%>%
  DT::datatable(
    rownames = FALSE,
    #caption = 'Table 1: This is a simple caption for the table.',
    class = 'cell-border stripe',
    escape = FALSE,
    filter = "top",
    extensions = list("Scroller"),
  #   options = list(autoWidth = FALSE,
  # columnDefs = list(width = '300px', width = '200px', width = '90px', width = '200px', width = '100px'),
  
  options = list(autoWidth = TRUE,
              columnDefs = list(list(targets=c(0), visible=TRUE, width='120'),
                                list(targets=c(1), visible=TRUE, width='120'),
                                list(targets=c(2), visible=TRUE, width='110'),
                                list(targets=c(3), visible=TRUE, width='260'),
                                list(targets=c(4), visible=TRUE, width='100'),
                                list(targets='_all', visible=FALSE)),
                                
  pageLength = 10, scrollY = TRUE, scrollX= TRUE, scrollCollapse = TRUE, fillContainer = TRUE,
                   initComplete = JS(
    "function(settings, json) {",
    "$(this.api().table().header()).css({'background-color': '#888888', 'color': '#fff'});",
    "}"))
  )
  
 
```

<div style = "width:100%; height:auto; margin: 0; padding:10px">

``` {r fig1, echo = FALSE, include=TRUE}
 OCS_info_table
```
</div>