Consistency checks for CDS classes

cds_consistency.R

library(didactr)
library(pdftools)
library(dplyr)
library(httr)
library(googledrive)
library(broom)
library(tidyr)
n_pdf_pages = function(file) {
  if (length(file) == 0) {
    return(NA)
  }
  pdf_info(file)$pages
}

res_path = file.path("manuscript", "resources")
# get manuscript md files and check names of 
manuscript_files = list.files(pattern = ".md$", path = "manuscript",
                              full.names = TRUE)
man_stubs = sub("[.]md$", "", basename(manuscript_files))
# md file has highest precedence
df = data_frame(lesson = man_stubs, md_file = manuscript_files)

df$id = sapply(df$md_file, function(fname) {
  x = readLines(fname, warn = FALSE)
  x = grep(x, pattern = "\\[(S|s)lides\\]", value = TRUE)
  x = sub(".*\\((http.*)\\).*", "\\1", x)
  x = unlist(sapply(x, function(r) parse_url(r)$path))
  x = sub("/edit$", "", x)
  x = basename(x)
  x = unique(x)
  if (length(x) > 1) {
    warning(paste0("Multiple sheets identified!  Please check ", 
                   fname))
  }
  if (length(x) == 0) {
    return(NA)
  }
  return(x)
})


if (anyDuplicated(df$id)) {
  dup_df = df %>% 
    group_by(id) %>% 
    add_tally() %>% 
    filter(n > 1)
  warning("Duplicated IDs are present!  MD files are off")
  print(dup_df)
}

image_links = lapply(df$md_file, function(fname) {
  x = readLines(fname, warn = FALSE)
  x = grep(x, pattern = "!\\[.*\\]\\((images.*)\\)", value = TRUE)
  x = sub(x, pattern = "!\\[(.*)\\]\\((images.*)\\)", replacement = "\\1")
  return(x)
})

images = lapply(df$md_file, function(fname) {
  x = readLines(fname, warn = FALSE)
  x = grep(x, pattern = "!\\[.*\\]\\((images.*)\\)", value = TRUE)
  x = sub(x, pattern = "!\\[.*\\]\\((images.*)\\)", replacement = "\\1")
  return(x)
})

df$all_images_exist = sapply(images, function(x) {
  all(file.exists(file.path(res_path, x)))
})

drive_info = drive_get(id = df$id)
if (nrow(drive_info) > 0) {
  drive_info = drive_info %>% 
    rename(gs_name = name)
  mod_time = sapply(drive_info$drive_resource, 
                    function(x) {
                      x$modifiedTime
                    })
  drive_info$mod_time = mod_time
  drive_info = drive_info %>% 
    select(-drive_resource)
  df = left_join(df, drive_info, by = "id")
  df = distinct(df)
}

path = file.path("manuscript", "resources", "images")

if (!dir.exists(path)) {
  dir.create(path, recursive = TRUE, showWarnings = FALSE)
}

script_path = "scripts"

if (!dir.exists(script_path)) {
  dir.create(script_path, recursive = TRUE, showWarnings = FALSE)
}

df = df %>% 
  mutate(img_dir = file.path(path, lesson))

# naming conventions for the images folders
img_dirs = list.dirs(path = path, recursive = FALSE, 
                     full.names = TRUE)
names(img_dirs) = img_dirs

df = df %>% 
  mutate(has_img_dir = img_dir %in% img_dirs)

# if img_dir doesn't exist, then create one
bad_img_dir = !df$has_img_dir
if (any(bad_img_dir)) {
  sapply(df$img_dir[bad_img_dir], dir.create, recursive = TRUE,
         showWarnings = FALSE)
}
df$has_img_dir = NULL

# check if image directories exist but don't have MD file
bad_img_dir = !(img_dirs %in% df$img_dir)
if (any(bad_img_dir)) {
  warning(paste0("An image directory exists but doesn't correspond to a ", 
                 "lesson.  Possible naming inconsistency.  Possible:"))
  cat(img_dirs[bad_img_dir], sep = "\n")
}



# Check if a image folder has a PDF
df$pdf = sapply(df$img_dir,
                function(x) {
                  pdfs = list.files(pattern = "[.]pdf",
                                    path = x,
                                    full.names = TRUE)
                  if (length(pdfs) > 1) {
                    warning(paste0(path, " had more than one PDF! ", 
                                   "Only grabbing first"))
                    pdfs = pdfs[1]
                  }
                  if (length(pdfs) == 0) {
                    return(NA)
                  }
                  return(pdfs)
                })

# Check the number of pages of the pdf to cross-ref with the pngs
df$pdf_pages = sapply(df$pdf, n_pdf_pages)

# list out the pngs of the folder
png_names = lapply(df$img_dir,
                   function(x) {
                     pngs = list.files(pattern = "[.]png",
                                       path = x)
                     pngs
                   })
df$n_pngs = sapply(png_names, length)