A function demonstrating tokenized pagination of graphQL for R, (querying the github API to get vz-risk/VCDB issue IDs to compare to a verisr object)

crosswalk_github_issues.R

#' A function to find github issues which may have been closed w/o being added
#' 
#' NOTE: Many github issues have 'na' for their issue #, leading to false positives
#' NOTE: Requires ghql package, currently only at https://github.com/ropensci/ghql
#'       `devtools::install_github("ropensci/ghql")`
#' 
#' @param veris a verisr dataframe
#' @param gh_token a github user token for api access
#' @return a list of github issue numbers not in veris
#' @export
crosswalk_github_issues <- function(veris, gh_token) {
  # initialize API
  cli$exec(qry$queries$myquery)
  cli <- ghql::GraphqlClient$new(
    url = "https://api.github.com/graphql",
    headers = httr::add_headers(Authorization = paste0("token ", gh_token))
  )
  cli$load_schema()
  
  # iterator to stop while loop
  qry <- ghql::Query$new()
  qry$query('initialIssueCursor', '{
    repository(owner: "vz-risk", name: "VCDB") {
      issues(first: 100,  states: CLOSED) {
        totalCount,
        edges {
          node {
            number
          }
        },
        pageInfo {
          startCursor
          endCursor
          hasNextPage
        }
      }
    }
  }')
  getIssues <- '{
    repository(owner:"vz-risk", name:"VCDB") {
      issues (first:100, after:\"%s\", states:CLOSED) {
        totalCount,
        edges {
          node {
            number
          }
        },
        pageInfo {
          endCursor,
          hasNextPage
        }
      }
    }
  }'
  qry$query('getissues', getIssues)
  q <- cli$exec(qry$queries$initialIssueCursor)
  cursor <- q$data$repository$issues$pageInfo$endCursor
  issues <- q$data$repository$issues$edges$node$number
  while (q$data$repository$issues$pageInfo$hasNextPage) {
    # message(paste0("At offset: ", cursor)) # DEBUG
    qry$queries$getissues$query <- sprintf(getIssues, cursor)
    q <- cli$exec(qry$queries$getissues)
    cursor <- q$data$repository$issues$pageInfo$endCursor
    issues <- c(issues, q$data$repository$issues$edges$node$number)
  }
  
  setdiff(issues, veris$plus.github)
}

This uses https://github.com/ropensci/ghql to query the graph. Replacement of the 'after' cursor value is kind of rough but works.