diff --git a/inst/function_inventory/function_dependency_graph.html b/inst/function_inventory/function_dependency_graph.html new file mode 100644 index 00000000..246106e6 --- /dev/null +++ b/inst/function_inventory/function_dependency_graph.html @@ -0,0 +1,5318 @@ + + + + +visNetwork + + + + + + + +
+
+
+ + + + diff --git a/inst/function_inventory/function_inventory.Rmd b/inst/function_inventory/function_inventory.Rmd new file mode 100644 index 00000000..0b151bfe --- /dev/null +++ b/inst/function_inventory/function_inventory.Rmd @@ -0,0 +1,130 @@ +--- +title: "Function Inventory" +author: "Awa Synthia" +date: "2024-10-07" +output: html_document +--- + +## Overview +This document generates an inventory of functions in the MolEvolvR package and +identifies dependencies between them, using the interactive network +visualizations to illustrate function relationships. + +```{r setup, include=FALSE} +# Load necessary libraries +library(stringr) +library(igraph) +library(visNetwork) +library(htmlwidgets) +``` + +Extract Function Names + +We start by extracting the names of all functions defined within in the R +directory. + +```{r} +# Function to extract function names from R files +extract_function_names <- function(file) { + content <- readLines(file) + function_lines <- grep("^[a-zA-Z0-9_]+\\s*<-\\s*function", content, + value = TRUE) + function_names <- str_extract(function_lines, "^[a-zA-Z0-9_]+") + return(na.omit(unique(function_names))) +} + +# List all R files in the package +function_files <- list.files(path = "/home/simple/awasyn/MolEvolvR/R", pattern = "*.R", + full.names = TRUE) +all_function_names <- unlist(lapply(function_files, extract_function_names)) +head(all_function_names) + +``` + + +Identify Dependencies + +Next, we check each function for calls to other functions within its definition, +enabling us to map out dependencies between functions. + +```{r} +# Function to find dependencies within function definitions in each file +find_dependencies <- function(file, all_function_names) { + content <- readLines(file) + dependencies <- list() + + for (func_name in all_function_names) { + func_start <- grep(paste0("^", func_name, "\\s*<-\\s*function"), content) + if (length(func_start) > 0) { + func_body <- content[func_start[1]:min(length(content), func_start[1] + 20)] + called_functions <- unique(unlist(str_extract_all(func_body, + paste0("\\b(", paste(all_function_names, collapse = "|"), ")\\b")))) + if (length(called_functions) > 0) { + dependencies[[func_name]] <- called_functions + } + } + } + return(dependencies) +} + +# Apply to all files and flatten the list +all_dependencies <- do.call(c, lapply(function_files, find_dependencies, + all_function_names)) +head(all_dependencies) + +``` + + +Interactive Visualization with visNetwork + +Here, we build an interactive network of function dependencies using visNetwork, +allowing for exploration of complex relationships between functions. + +```{r} +# Prepare data for visNetwork +create_vis_data <- function(dependencies) { + nodes <- data.frame(id = unique(c(names(dependencies), + unlist(dependencies))), + label = unique(c(names(dependencies), + unlist(dependencies))), + stringsAsFactors = FALSE) + edges <- do.call(rbind, lapply(names(dependencies), function(func) { + if (length(dependencies[[func]]) > 0) { + data.frame(from = func, to = dependencies[[func]], + stringsAsFactors = FALSE) + } else { + NULL + } + })) + return(list(nodes = nodes, edges = edges)) +} + +# Generate the dependency data +vis_data <- create_vis_data(all_dependencies) + +# Create the visNetwork graph +network <- visNetwork(vis_data$nodes, vis_data$edges, height = "700px", + width = "100%") %>% + visEdges(arrows = "to") %>% + visOptions(highlightNearest = TRUE, nodesIdSelection = TRUE) %>% + visInteraction(navigationButtons = TRUE, zoomView = TRUE) %>% + visLayout(randomSeed = 123) # Ensures layout consistency + +# Display the network graph in the RMarkdown output +network +``` + +Saving the Interactive Graph + +To save the interactive graph as an HTML file, we use saveWidget, which will +generate a self-contained HTML document for easy sharing and review. + +```{r} +# Save the network as an HTML file +# Uncomment the following line to sace gapht to interactive html file +saveWidget(network, "function_dependency_graph.html", selfcontained = TRUE) + +``` + + + diff --git a/inst/function_inventory/function_inventory.html b/inst/function_inventory/function_inventory.html new file mode 100644 index 00000000..f3e19f58 --- /dev/null +++ b/inst/function_inventory/function_inventory.html @@ -0,0 +1,5822 @@ + + + + + + + + + + + + + + + +Function Inventory + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + +
+

Overview

+

This document generates an inventory of functions in the MolEvolvR +package and identifies dependencies between them, using the interactive +network visualizations to illustrate function relationships.

+

Extract Function Names

+

We start by extracting the names of all functions defined within in +the R directory.

+
# Function to extract function names from R files
+extract_function_names <- function(file) {
+  content <- readLines(file)
+  function_lines <- grep("^[a-zA-Z0-9_]+\\s*<-\\s*function", content, 
+                         value = TRUE)
+  function_names <- str_extract(function_lines, "^[a-zA-Z0-9_]+")
+  return(na.omit(unique(function_names)))
+}
+
+# List all R files in the package
+function_files <- list.files(path = "/home/simple/awasyn/MolEvolvR/R", pattern = "*.R", 
+                             full.names = TRUE)
+all_function_names <- unlist(lapply(function_files, extract_function_names))
+head(all_function_names)
+
## [1] "sinkReset"         "addLineage"        "acc2Lineage"      
+## [4] "efetchIPG"         "IPG2Lineage"       "mapOption2Process"
+

Identify Dependencies

+

Next, we check each function for calls to other functions within its +definition, enabling us to map out dependencies between functions.

+
# Function to find dependencies within function definitions in each file
+find_dependencies <- function(file, all_function_names) {
+  content <- readLines(file)
+  dependencies <- list()
+
+  for (func_name in all_function_names) {
+    func_start <- grep(paste0("^", func_name, "\\s*<-\\s*function"), content)
+    if (length(func_start) > 0) {
+      func_body <- content[func_start[1]:min(length(content), func_start[1] + 20)]
+      called_functions <- unique(unlist(str_extract_all(func_body, 
+          paste0("\\b(", paste(all_function_names, collapse = "|"), ")\\b"))))
+      if (length(called_functions) > 0) {
+        dependencies[[func_name]] <- called_functions
+      }
+    }
+  }
+  return(dependencies)
+}
+
+# Apply to all files and flatten the list
+all_dependencies <- do.call(c, lapply(function_files, find_dependencies, 
+                                      all_function_names))
+head(all_dependencies)
+
## $sinkReset
+## [1] "sinkReset"  "addLineage"
+## 
+## $addLineage
+## [1] "addLineage"  "acc2Lineage"
+## 
+## $acc2Lineage
+## [1] "acc2Lineage" "efetchIPG"   "IPG2Lineage"
+## 
+## $efetchIPG
+## [1] "efetchIPG"
+## 
+## $IPG2Lineage
+## [1] "IPG2Lineage" "GCA2Lineage" "efetchIPG"  
+## 
+## $mapOption2Process
+## [1] "mapOption2Process"    "mapAdvOption2Process"
+

Interactive Visualization with visNetwork

+

Here, we build an interactive network of function dependencies using +visNetwork, allowing for exploration of complex relationships between +functions.

+
# Prepare data for visNetwork
+create_vis_data <- function(dependencies) {
+  nodes <- data.frame(id = unique(c(names(dependencies), 
+                                    unlist(dependencies))),
+                      label = unique(c(names(dependencies), 
+                                       unlist(dependencies))),
+                      stringsAsFactors = FALSE)
+  edges <- do.call(rbind, lapply(names(dependencies), function(func) {
+    if (length(dependencies[[func]]) > 0) {
+      data.frame(from = func, to = dependencies[[func]], 
+                 stringsAsFactors = FALSE)
+    } else {
+      NULL
+    }
+  }))
+  return(list(nodes = nodes, edges = edges))
+}
+
+# Generate the dependency data
+vis_data <- create_vis_data(all_dependencies)
+
+# Create the visNetwork graph
+network <- visNetwork(vis_data$nodes, vis_data$edges, height = "700px", 
+                      width = "100%") %>%
+  visEdges(arrows = "to") %>%
+  visOptions(highlightNearest = TRUE, nodesIdSelection = TRUE) %>%
+  visInteraction(navigationButtons = TRUE, zoomView = TRUE) %>%
+  visLayout(randomSeed = 123) # Ensures layout consistency
+
+# Display the network graph in the RMarkdown output
+network
+
+ +

Saving the Interactive Graph

+

To save the interactive graph as an HTML file, we use saveWidget, +which will generate a self-contained HTML document for easy sharing and +review.

+
# Save the network as an HTML file
+# Uncomment the following line to sace gapht to interactive html file
+saveWidget(network, "function_dependency_graph.html", selfcontained = TRUE)
+
+ + + + +
+ + + + + + + + + + + + + + +