Format titles using metadata

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: lovecraftr
 Type: Package
 Title: H. P. Lovecraft's works, for text analysis
-Version: 0.2.1
+Version: 0.2.2
 Authors@R:
     person("Jordan", "Bradford", email = "[email protected]", role = c("aut", "cre"))
 Description: A collection of H. P. Lovecraft's works as R datasets.

_pkgdown.yml

@@ -5,13 +5,21 @@ template:
 
 reference:
 - title: Data
-  desc: >
-    {`lovecraftr`} contains the following datasets.
+
+- subtitle: Data Frames
+- contents:
+  - lovecraft
+
+- subtitle: Novels
+  desc: Character vectors of novels
+- contents:
+  - has_keyword("novella")
+
+- subtitle: Short Stories
+  desc: Character vectors of short stories
   contents:
-  - has_keyword("datasets")
+  - has_keyword("short")
 
 - title: Functions
-  desc: >
-    {`lovecraftr`} contains the following functions.
   contents:
   - lovecraft_corpus

data-raw/generate_data.R

@@ -5,10 +5,18 @@
 # the raw txt files in data-raw/corpus
 
 
+title_case <- function(x) {
+  stopwords <- c("and", "of", "the", "on", "at", "in", "with", "to", "that") # Add more stopwords if needed
+  words <- stringr::str_split(tolower(x), " ")[[1]]
+  words <- ifelse(words %in% stopwords & seq_along(words) != 1, words, stringr::str_to_title(words))
+  return(stringr::str_c(words, collapse = " "))
+}
+
 raw_txt_path <- file.path("data-raw", "corpus")
 corpus <- list.files(raw_txt_path)
 raw_txt_root <- "https://raw.githubusercontent.com/jrdnbradford/lovecraftr/main/data-raw/corpus/"
 rda_root <- "https://github.com/jrdnbradford/lovecraftr/raw/main/data/"
+metadata <- lovecraftr:::metadata
 
 data_docs_file <- file.path("R", "data.R")
 file.remove(data_docs_file)
@@ -19,12 +27,12 @@ header <- "# This documentation is generated by a script, do not edit by hand
 "
 
 docs_template <- "{header}
-#' @title Dataset of \"{title}\"
-#' @description A dataset containing text of H. P. Lovecraft's \"{title}\"
+#' @title Dataset of {formatted_title}
+#' @description A dataset containing text of H. P. Lovecraft's {formatted_title}
 #' @usage data({dataset_name})
 #' @format A character vector with {num_vectors} elements, with roughly 80 characters per line.
 #' @docType data
-#' @keywords datasets
+#' @keywords datasets {type}
 #' @source Public domain.
 #'   See the {raw_txt_link}
 #'   or download the {rda_link}.
@@ -43,9 +51,11 @@ for (txt in corpus) {
   assign(txt, text)
   do.call(usethis::use_data, list(dataset_name, overwrite = TRUE))
 
-  title <- head(text, 1)
-  raw_txt_link <- paste0("[raw text file", "]", "(", raw_txt_root, dataset_name, ")")
-  rda_link <- paste0("[RDA file", "]", "(", rda_root, dataset_name, ".rda)")
+  type <- subset(metadata, title == dataset_name)$type
+  title <- title_case(head(text, 1))
+  formatted_title <-ifelse(type == "novella", paste0("_", title, "_"), paste0("\"", title, "\""))
+  raw_txt_link <- paste0("[raw text file](", raw_txt_root, dataset_name, ")")
+  rda_link <- paste0("[RDA file](", rda_root, dataset_name, ".rda)")
   dataset_docs <- glue::glue(docs_template)
   docs <- c(docs, dataset_docs)