Fix setting authors in co-change-based author networks

This patch consists of three related fix and adaptations:

First, the method 'ProjectData$get.authors.by.data.source' does not
correct the column names of the returned data.frame anymore. This
establishes compatibility with the method 'ProjectData$get.authors'.
Additionally, the returned data.frame only contains unique entries. The
documentation is tidied.

Second, the method 'NetworkBuilder$get.author.network.cochange' is fixed
by adding the missing 'private$' prefix when accessing the project data.

Third, the assignment of author vertices is corrected to use only author
names with the correct vertex attribute (i.e., "name"). This adapts the
code with respect to the first change mentioned above.

This change fixes all failing tests in PR se-sic#149.

Signed-off-by: Claus Hunsen <[email protected]>

util-data.R

@@ -948,23 +948,27 @@ ProjectData = R6::R6Class("ProjectData",
             return(mylist)
         },
 
-        #' Get the list of authors by only looking at the specified data source. The constant
-        #' \code{DATASOURCE.TO.ARTIFACT.FUNCTION} describes the mapping between data source and the method which is
-        #' retrieving the data for each data source.
+        #' Get the list of authors by only looking only at the specified data source.
         #'
-        #' @param data.source the data source which can be either \code{"commits"}, \code{"mails"} or \code{"issues"}
+        #' *Note*: The constant \code{DATASOURCE.TO.ARTIFACT.FUNCTION} denotes the mapping between
+        #' data source and the method which is retrieving the data for each data source.
         #'
-        #' @return the list of authors extracted from the specified data source
+        #' @param data.source the data source which can be either \code{"commits"}, \code{"mails"},
+        #'                    or \code{"issues"}
+        #'
+        #' @return a data.frame of unique author names (columns \code{name} and \code{author.email}),
+        #'         extracted from the specified data source
         get.authors.by.data.source = function(data.source = c("commits", "mails", "issues")) {
             if (is.null(data.source)) {
                 stop ("Data source can't be null.")
             }
 
             data.source = match.arg(data.source)
             data.source.func = DATASOURCE.TO.ARTIFACT.FUNCTION[[data.source]]
-
             data = self[[data.source.func]]()[c("author.name", "author.email")]
-            names(data) = c("name", "email")
+
+            ## remove duplicates
+            data = unique(data)
 
             return (data)
         }

util-networks.R

@@ -175,9 +175,15 @@ NetworkBuilder = R6::R6Class("NetworkBuilder",
             ## Add author vertices back into the graph. Previously, commit information on untracked files
             ## ('UNTRACKED.FILE') and, if configured, the base artifact ('BASE.ARTIFACTS') has been removed and, hence,
             ## also corresponding author information. Re-add author vertices back to the network now by accessing the
-            ## complete author list.
-            authors = proj.data$get.authors.by.data.source(data.source = "commits")
-            author.net.data[["vertices"]] = authors["name"]
+            ## complete author list:
+            ## 1) get all authors on commits
+            authors = private$proj.data$get.authors.by.data.source(data.source = "commits")
+            ## 2) only select author names
+            authors = authors["author.name"]
+            ## 3) rename single column to "name" to correct mapping to vertex attribute "name"
+            colnames(authors) = "name"
+            ## 4) set author list as vertices
+            author.net.data[["vertices"]] = authors
 
             ## construct network from obtained data
             author.net = construct.network.from.edge.list(