only knit rmd files to md if they have changed

[rnaseq-cwl-training.git] / bin / generate_md_episodes.R

diff --git a/bin/generate_md_episodes.R b/bin/generate_md_episodes.R
index 4fecabe99ec35686302aee2d7ba24f1b6378c6f7..60d65e970be608b571e7a47e74e966f6aee7dd2a 100644 (file)
--- a/bin/generate_md_episodes.R
+++ b/bin/generate_md_episodes.R
@@ -1,41 +1,55 @@
 generate_md_episodes <- function() {
 
-    if (require("knitr")) {
-        if (packageVersion("knitr") < '1.9.19') {
-            stop("knitr must be version 1.9.20 or higher")
-        }
-    } else stop("knitr 1.9.20 or above is needed to build the lessons.")
+    library("methods")
+    
+    if (require("knitr") && packageVersion("knitr") < '1.9.19')
+        stop("knitr must be version 1.9.20 or higher")
 
     if (!require("stringr"))
         stop("The package stringr is required for generating the lessons.")
 
-    ## where the Rmd files and the datasets are located
-    rmd_path <- "_episodes_rmd"
-    rmd_data <- file.path(rmd_path, "data")
-
-    ## where the markdown files and the datasets will end up
-    dest_path <- "_episodes"
-    dest_data <- file.path(dest_path, "data")
-
-    ## find all the Rmd files, and generates the paths for their respective outputs
-    src_rmd <- list.files(pattern = "??-*.Rmd$", path = rmd_path, full.names = TRUE)
-    dest_md <- file.path(dest_path, gsub("Rmd$", "md", basename(src_rmd)))
+    if (require("checkpoint") && packageVersion("checkpoint") >=  '0.4.0') {
+        required_pkgs <-
+             checkpoint:::scanForPackages(project = "_episodes_rmd",
+                                          verbose=FALSE, use.knitr = TRUE)$pkgs
+    } else {
+        stop("The checkpoint package (>= 0.4.0) is required to build the lessons.")
+    }
+
+    missing_pkgs <- required_pkgs[!(required_pkgs %in% rownames(installed.packages()))]
+
+    if (length(missing_pkgs)) {
+        message("Installing missing required packages: ",
+                paste(missing_pkgs, collapse=", "))
+        install.packages(missing_pkgs)
+    }
+
+    ## get the Rmd file to process from the command line, and generate the path for their respective outputs
+    args  <- commandArgs(trailingOnly = TRUE)
+    if (length(args) != 2){
+      stop("input and output file must be passed to the script")
+    }
+    
+    src_rmd <- args[1]
+    dest_md <- args[2]
 
     ## knit the Rmd into markdown
-    mapply(function(x, y) {
-        knitr::knit(x, output = y)
-    }, src_rmd, dest_md)
-
-
-    ## copy the datasets from _episodes_rmd/data to _episodes/data
-    rmd_data_files <- list.files(path = rmd_data, full.names = TRUE)
-    dest_data_files <- file.path(dest_data, basename(rmd_data_files))
-
-    if (!dir.exists(file.path(dest_data)))
-        dir.create(file.path(dest_data))
-
-    apply(cbind(rmd_data_files, dest_data_files), 1,
-          function(x) file.copy(x[1], x[2]))
+    knitr::knit(src_rmd, output = dest_md)
+
+    # Read the generated md files and add comments advising not to edit them
+    vapply(dest_md, function(y) {
+      con <- file(y)
+      mdfile <- readLines(con)
+      if (mdfile[1] != "---")
+        stop("Input file does not have a valid header")
+      mdfile <- append(mdfile, "# Please do not edit this file directly; it is auto generated.", after = 1)
+      mdfile <- append(mdfile, paste("# Instead, please edit", 
+                                     basename(y), "in _episodes_rmd/"), after = 2)
+      writeLines(mdfile, con)
+      close(con)
+      return(paste("Warning added to YAML header of", y))
+    },
+    character(1))
 }
 
 generate_md_episodes()