1 Introduction
2 Package RCurl
- 2.1 Check existence
- 2.2 Get personal information
3 Package httr
- 3.1 Get personal information
4 Conclusion
- 4.1 Wrapper functions for GET and content
- 4.2 Use of wrapper functions

Available at RPubs

1 Introduction

I will try to connect to the FAIRDOM SEEK with the collection of their APIs.

The API description with examples is available here.

My final goal is to enable upload/download from pISA-tree to FAIRDOMHub. This feature will be a part of the R package pISAR that will enable use of pISA structure and metafile information for reproducible statistical analyses.

2 Package `RCurl`

My first attempt will be the RCurl package avaialbe at CRAN. Additional description is a part of Omegahat project.

library(RCurl)

## Loading required package: bitops

2.1 Check existence

First check if the base URL exists

(FAIRDOMExists = url.exists("https://www.fairdomhub.org/"))

## [1] TRUE

2.2 Get personal information

For this I will use the GET \people\{id} from documentation and my ID {808}.

person <- getURL("https://www.fairdomhub.org/people/808")
str(person)

##  chr "<!doctype html>\n\n<html lang=\"en\">\n<head>\n  <meta http-equiv=\"content-type\" content=\"text/html;charset=UTF-8\"/>\n  <me"| __truncated__

It returns the HTML content and might not be the optimal way.

3 Package `httr`

The use is described in LockeData blog.

This solution uses packages httr and jsonlite and is based on libcurl (the same is used by RCurl). Package httr has useful vignettes that were a good guide for the example below.

library(httr)
library(jsonlite)

3.1 Get personal information

r <- GET("https://www.fairdomhub.org/people/808",
         add_headers(Accept="application/json"))
names(r)

##  [1] "url"         "status_code" "headers"     "all_headers" "cookies"    
##  [6] "content"     "date"        "times"       "request"     "handle"

## Response [https://www.fairdomhub.org/people/808]
##   Date: 2018-06-11 18:17
##   Status: 200
##   Content-Type: application/vnd.api+json; charset=utf-8
##   Size: 1.36 kB
## <BINARY BODY>

The response is structured and can be investigated with the helper functions. The response has three important parts: status, headers, and body.

3.1.1 Status

Check status

http_status(r)

## $category
## [1] "Success"
## 
## $reason
## [1] "OK"
## 
## $message
## [1] "Success: (200) OK"

or just a status code

r$status_code

## [1] 200

3.1.2 Body

Body can be accessed with the function content

cont <- content(r,"text")
cont

## [1] "{\"data\":{\"id\":\"808\",\"type\":\"people\",\"attributes\":{\"avatar\":\"/people/808/avatars/473\",\"title\":\"Andrej Blejec\",\"description\":null,\"first_name\":\"Andrej\",\"last_name\":\"Blejec\",\"web_page\":\"http://ablejec.nib.si\",\"orcid\":\"http://orcid.org/0000-0001-7484-6031\",\"mbox_sha1sum\":\"e37350ada8b1af0c7fe7a6cb61a6a5d35d20f8bd\",\"phone\":\"+38659232789\",\"skype_name\":\"ablejec\",\"expertise\":[\"Bioinformatics\",\"Computational Statistics\",\"Mathematical and statistical modeling\",\"Programming\",\"R\",\"Statistics\",\"Visualization\"],\"tools\":[\"Computational Systems Biology\",\"Data Management\",\"Data integration\",\"R\",\"Statistical tools\",\"Statistics\"],\"project_positions\":[{\"project_id\":\"57\",\"position_id\":\"3\",\"position_name\":\"Project Coordinator\"}]},\"relationships\":{\"projects\":{\"data\":[{\"id\":\"56\",\"type\":\"projects\"},{\"id\":\"57\",\"type\":\"projects\"}]},\"institutions\":{\"data\":[{\"id\":\"189\",\"type\":\"institutions\"}]},\"investigations\":{\"data\":[]},\"studies\":{\"data\":[]},\"assays\":{\"data\":[]},\"data_files\":{\"data\":[]},\"models\":{\"data\":[]},\"sops\":{\"data\":[]},\"publications\":{\"data\":[]},\"presentations\":{\"data\":[]},\"events\":{\"data\":[]},\"documents\":{\"data\":[]}},\"links\":{\"self\":\"/people/808\"},\"meta\":{\"created\":\"2016-11-07T22:17:49.000Z\",\"modified\":\"2017-08-01T08:28:42.000Z\",\"api_version\":\"0.1\",\"uuid\":\"f1427a00-8765-0134-769f-549f350973c0\",\"base_url\":\"https://fairdomhub.org\"}},\"jsonapi\":{\"version\":\"1.0\"}}"

length(cont)

## [1] 1

This gives us a text version of the JSON object. We can use built-in parsers to get other forms of data. We can parse the content into the R list object

cont <- content(r,"parsed",type="application/json")
cont

## $data
## $data$id
## [1] "808"
## 
## $data$type
## [1] "people"
## 
## $data$attributes
## $data$attributes$avatar
## [1] "/people/808/avatars/473"
## 
## $data$attributes$title
## [1] "Andrej Blejec"
## 
## $data$attributes$description
## NULL
## 
## $data$attributes$first_name
## [1] "Andrej"
## 
## $data$attributes$last_name
## [1] "Blejec"
## 
## $data$attributes$web_page
## [1] "http://ablejec.nib.si"
## 
## $data$attributes$orcid
## [1] "http://orcid.org/0000-0001-7484-6031"
## 
## $data$attributes$mbox_sha1sum
## [1] "e37350ada8b1af0c7fe7a6cb61a6a5d35d20f8bd"
## 
## $data$attributes$phone
## [1] "+38659232789"
## 
## $data$attributes$skype_name
## [1] "ablejec"
## 
## $data$attributes$expertise
## $data$attributes$expertise[[1]]
## [1] "Bioinformatics"
## 
## $data$attributes$expertise[[2]]
## [1] "Computational Statistics"
## 
## $data$attributes$expertise[[3]]
## [1] "Mathematical and statistical modeling"
## 
## $data$attributes$expertise[[4]]
## [1] "Programming"
## 
## $data$attributes$expertise[[5]]
## [1] "R"
## 
## $data$attributes$expertise[[6]]
## [1] "Statistics"
## 
## $data$attributes$expertise[[7]]
## [1] "Visualization"
## 
## 
## $data$attributes$tools
## $data$attributes$tools[[1]]
## [1] "Computational Systems Biology"
## 
## $data$attributes$tools[[2]]
## [1] "Data Management"
## 
## $data$attributes$tools[[3]]
## [1] "Data integration"
## 
## $data$attributes$tools[[4]]
## [1] "R"
## 
## $data$attributes$tools[[5]]
## [1] "Statistical tools"
## 
## $data$attributes$tools[[6]]
## [1] "Statistics"
## 
## 
## $data$attributes$project_positions
## $data$attributes$project_positions[[1]]
## $data$attributes$project_positions[[1]]$project_id
## [1] "57"
## 
## $data$attributes$project_positions[[1]]$position_id
## [1] "3"
## 
## $data$attributes$project_positions[[1]]$position_name
## [1] "Project Coordinator"
## 
## 
## 
## 
## $data$relationships
## $data$relationships$projects
## $data$relationships$projects$data
## $data$relationships$projects$data[[1]]
## $data$relationships$projects$data[[1]]$id
## [1] "56"
## 
## $data$relationships$projects$data[[1]]$type
## [1] "projects"
## 
## 
## $data$relationships$projects$data[[2]]
## $data$relationships$projects$data[[2]]$id
## [1] "57"
## 
## $data$relationships$projects$data[[2]]$type
## [1] "projects"
## 
## 
## 
## 
## $data$relationships$institutions
## $data$relationships$institutions$data
## $data$relationships$institutions$data[[1]]
## $data$relationships$institutions$data[[1]]$id
## [1] "189"
## 
## $data$relationships$institutions$data[[1]]$type
## [1] "institutions"
## 
## 
## 
## 
## $data$relationships$investigations
## $data$relationships$investigations$data
## list()
## 
## 
## $data$relationships$studies
## $data$relationships$studies$data
## list()
## 
## 
## $data$relationships$assays
## $data$relationships$assays$data
## list()
## 
## 
## $data$relationships$data_files
## $data$relationships$data_files$data
## list()
## 
## 
## $data$relationships$models
## $data$relationships$models$data
## list()
## 
## 
## $data$relationships$sops
## $data$relationships$sops$data
## list()
## 
## 
## $data$relationships$publications
## $data$relationships$publications$data
## list()
## 
## 
## $data$relationships$presentations
## $data$relationships$presentations$data
## list()
## 
## 
## $data$relationships$events
## $data$relationships$events$data
## list()
## 
## 
## $data$relationships$documents
## $data$relationships$documents$data
## list()
## 
## 
## 
## $data$links
## $data$links$self
## [1] "/people/808"
## 
## 
## $data$meta
## $data$meta$created
## [1] "2016-11-07T22:17:49.000Z"
## 
## $data$meta$modified
## [1] "2017-08-01T08:28:42.000Z"
## 
## $data$meta$api_version
## [1] "0.1"
## 
## $data$meta$uuid
## [1] "f1427a00-8765-0134-769f-549f350973c0"
## 
## $data$meta$base_url
## [1] "https://fairdomhub.org"
## 
## 
## 
## $jsonapi
## $jsonapi$version
## [1] "1.0"

3.1.3 The structure of the parsed response

names(cont)

## [1] "data"    "jsonapi"

cont$jsonapi

## $version
## [1] "1.0"

names(cont$data)

## [1] "id"            "type"          "attributes"    "relationships"
## [5] "links"         "meta"

Response elements

cont$data$id

## [1] "808"

List the response content

d <- cont$data
for (nm in names(d)) {
  cat("\n",nm,"--------------\n\n")
  print(d[[nm]])
}

## 
##  id --------------
## 
## [1] "808"
## 
##  type --------------
## 
## [1] "people"
## 
##  attributes --------------
## 
## $avatar
## [1] "/people/808/avatars/473"
## 
## $title
## [1] "Andrej Blejec"
## 
## $description
## NULL
## 
## $first_name
## [1] "Andrej"
## 
## $last_name
## [1] "Blejec"
## 
## $web_page
## [1] "http://ablejec.nib.si"
## 
## $orcid
## [1] "http://orcid.org/0000-0001-7484-6031"
## 
## $mbox_sha1sum
## [1] "e37350ada8b1af0c7fe7a6cb61a6a5d35d20f8bd"
## 
## $phone
## [1] "+38659232789"
## 
## $skype_name
## [1] "ablejec"
## 
## $expertise
## $expertise[[1]]
## [1] "Bioinformatics"
## 
## $expertise[[2]]
## [1] "Computational Statistics"
## 
## $expertise[[3]]
## [1] "Mathematical and statistical modeling"
## 
## $expertise[[4]]
## [1] "Programming"
## 
## $expertise[[5]]
## [1] "R"
## 
## $expertise[[6]]
## [1] "Statistics"
## 
## $expertise[[7]]
## [1] "Visualization"
## 
## 
## $tools
## $tools[[1]]
## [1] "Computational Systems Biology"
## 
## $tools[[2]]
## [1] "Data Management"
## 
## $tools[[3]]
## [1] "Data integration"
## 
## $tools[[4]]
## [1] "R"
## 
## $tools[[5]]
## [1] "Statistical tools"
## 
## $tools[[6]]
## [1] "Statistics"
## 
## 
## $project_positions
## $project_positions[[1]]
## $project_positions[[1]]$project_id
## [1] "57"
## 
## $project_positions[[1]]$position_id
## [1] "3"
## 
## $project_positions[[1]]$position_name
## [1] "Project Coordinator"
## 
## 
## 
## 
##  relationships --------------
## 
## $projects
## $projects$data
## $projects$data[[1]]
## $projects$data[[1]]$id
## [1] "56"
## 
## $projects$data[[1]]$type
## [1] "projects"
## 
## 
## $projects$data[[2]]
## $projects$data[[2]]$id
## [1] "57"
## 
## $projects$data[[2]]$type
## [1] "projects"
## 
## 
## 
## 
## $institutions
## $institutions$data
## $institutions$data[[1]]
## $institutions$data[[1]]$id
## [1] "189"
## 
## $institutions$data[[1]]$type
## [1] "institutions"
## 
## 
## 
## 
## $investigations
## $investigations$data
## list()
## 
## 
## $studies
## $studies$data
## list()
## 
## 
## $assays
## $assays$data
## list()
## 
## 
## $data_files
## $data_files$data
## list()
## 
## 
## $models
## $models$data
## list()
## 
## 
## $sops
## $sops$data
## list()
## 
## 
## $publications
## $publications$data
## list()
## 
## 
## $presentations
## $presentations$data
## list()
## 
## 
## $events
## $events$data
## list()
## 
## 
## $documents
## $documents$data
## list()
## 
## 
## 
##  links --------------
## 
## $self
## [1] "/people/808"
## 
## 
##  meta --------------
## 
## $created
## [1] "2016-11-07T22:17:49.000Z"
## 
## $modified
## [1] "2017-08-01T08:28:42.000Z"
## 
## $api_version
## [1] "0.1"
## 
## $uuid
## [1] "f1427a00-8765-0134-769f-549f350973c0"
## 
## $base_url
## [1] "https://fairdomhub.org"

Person data are in the attributes part

Here is the first name of the person with id 808:

cont$data$attributes$first_name

## [1] "Andrej"

4 Conclusion

The rudimetary R read connection with FAIRDOMHub can be established using the package httr. From here it should be fairly possible to write the wrapper functions for easier usage.

4.1 Wrapper functions for `GET` and `content`

# this function is equivaalent to a call:
# curl -X GET "https://www.fairdomhub.org/what/id" -H  "accept: application/json"
#
myGET <- function(what, id,
                  uri="https://www.fairdomhub.org", ... ){
  if(!missing(what)) uri <- paste0(uri,"/",what)
  if(!missing(id)) uri <- paste0(uri,"/",id)
  r <- GET(uri,
         add_headers(Accept="application/json"))
  cat("Status code:",r$status_code,"\n")
  invisible(r)
}

myData <- function(r,
                  part="attributes",
                  type="application/json",  ...){
  invisible(content(r,"parsed",type=type)$data[[part]])
}

4.2 Use of wrapper functions

Get data for person 808

# https://www.fairdomhub.org/people/808
r <- myGET("people","808")

## Status code: 200

r$status_code

## [1] 200

Get the person 808 last name and tools

myData(r)$last_name

## [1] "Blejec"

myData(r)$tools

## [[1]]
## [1] "Computational Systems Biology"
## 
## [[2]]
## [1] "Data Management"
## 
## [[3]]
## [1] "Data integration"
## 
## [[4]]
## [1] "R"
## 
## [[5]]
## [1] "Statistical tools"
## 
## [[6]]
## [1] "Statistics"

Get person 808 institution id

myData(r,"relationships")$institutions

## $data
## $data[[1]]
## $data[[1]]$id
## [1] "189"
## 
## $data[[1]]$type
## [1] "institutions"

instID <- myData(r, "relationships")$institutions$data[[1]]$id
instID

## [1] "189"

Get institution instID` name. The details are in the institutions part.

# https://www.fairdomhub.org/institutions/189
rinst <- myGET("institutions",instID)

## Status code: 200

myData(rinst)$title

## [1] "National Institute of Biology"

Full institution description

catln <- function(...) cat(...,"\n")
d <- sapply(myData(rinst),catln)

## /institutions/189/avatars/568 
## National Institute of Biology 
## SI 
##  
## Ljubljana 
## Vecna pot 111
## SI-1000 Ljubljana
## Slovenia 
## http://www.nib.si

Test API connection to FAIRDOMHub

A. Blejec andrej.blejec@nib.si | National Institute of Biology | Ljubljana, Slovenia

10 junij 2018

1 Introduction

2 Package `RCurl`

2.1 Check existence

2.2 Get personal information

3 Package `httr`

3.1 Get personal information

3.1.1 Status

3.1.2 Body

3.1.3 The structure of the parsed response

4 Conclusion

4.1 Wrapper functions for `GET` and `content`

4.2 Use of wrapper functions

Test API connection to FAIRDOMHub

A. Blejec andrej.blejec@nib.si | National Institute of Biology | Ljubljana, Slovenia

10 junij 2018

1 Introduction

2 Package RCurl

2.1 Check existence

2.2 Get personal information

3 Package httr

3.1 Get personal information

3.1.1 Status

3.1.2 Body

3.1.3 The structure of the parsed response

4 Conclusion

4.1 Wrapper functions for GET and content

4.2 Use of wrapper functions

2 Package `RCurl`

3 Package `httr`

4.1 Wrapper functions for `GET` and `content`