more magic types and better docs

DESCRIPTION

@@ -11,7 +11,8 @@ Maintainer: Bob Rudis <[email protected]>
 Description: 'MIME' types are shorthand descriptors for file contents and can be
     determined from "magic" bytes in file headers, file contents or intuited from
     file extensions. Tools are provided to perform limited "magic" tests as well
-    as mapping 'MIME' types from a database of over 1,500 extension mappings.
+    as mapping 'MIME' types from a database of over 1,500 extension mappings. It is
+    useful as a more portable alternative to the 'wand' package.
 URL: https://github.com/hrbrmstr/simplemagic
 BugReports: https://github.com/hrbrmstr/simplemagic/issues
 Encoding: UTF-8

NAMESPACE

@@ -2,4 +2,5 @@
 
 export(get_content_type)
 export(guess_content_type)
+export(simplemagic_mime_db)
 importFrom(tools,file_ext)

R/aaa.R

@@ -1,3 +1,7 @@
+#' File extension-to-MIME mapping data frame
+#'
+#' @docType data
+#' @export
 structure(list(extension = c("pyc", "dwg", "ez", "aw", "arj",
 "atom", "xml", "atomcat", "atomsvc", "mm", "mme", "hqx", "hqx",
 "boo", "book", "ccxml", "cdf", "cdmia", "cdmic", "cdmid", "cdmio",
@@ -813,4 +817,4 @@ structure(list(extension = c("pyc", "dwg", "ez", "aw", "arj",
 "text/x-pascal", "video/jpm", "video/mj2", "video/mp4", "video/mpeg",
 "video/quicktime", "video/vnd.mpegurl", "video/x-ms-asf")), row.names = c(NA,
 -1763L), class = c("tbl_df", "tbl", "data.frame"), .Names = c("extension",
-"mime_type")) -> .ext_to_mime
+"mime_type")) -> simplemagic_mime_db

R/get-content-type.R

@@ -1,18 +1,26 @@
 #' Discover MIME type of a file based on contents
 #'
+#' There are a limited number of header "magic" bytes checked directly by
+#' this function but cover quite a bit of ground. After that, [guess_content_type()] is called which uses
+#' file extension-to-MIME mappings. File an issue or PR if more magic-byte-level
+#' comparisons are required/desired. If no match is found, `???` is returned.
+#'
 #' @details
 #' Initial in-R header mapping logic borrowed from `MimeTypes.java` from
 #' [`servoy-client`](https://github.com/Servoy/servoy-client)
 #'
 #' @md
 #' @param path path to a file
+#' @return character vector
 #' @export
+#' @examples
+#' get_content_type(system.file("extdat", "test.pdf", package="simplemagic"))
 get_content_type <- function(path) {
 
   path <- path.expand(path)
   if (!file.exists(path)) stop("File not found.", call.=FALSE)
 
-  hdr <- readBin(path, "raw", n=11)
+  hdr <- readBin(path, "raw", n=1024)
 
   if (all(c(0xCA,0xFE,0xBA,0xBE) == hdr[1:4])) return("application/java-vm")
 
@@ -23,6 +31,7 @@ get_content_type <- function(path) {
   }
 
   if (all(c(0x25,0x50,0x44,0x46,0x2d,0x31,0x2e) == hdr[1:7])) return("application/pdf")
+  if (all(c(0x25,0x50,0x44,0x46) == hdr[1:4])) return("application/x-pdf")
 
   if (all(c(0x38,0x42,0x50,0x53,0x00,0x01) == hdr[1:6])) return("image/photoshop")
 
@@ -54,6 +63,19 @@ get_content_type <- function(path) {
   if (all(c(0x21,0x20,0x58,0x50,0x4d,0x32) == hdr[1:6])) return("image/x-pixmap")
   if (all(c(137,80,78,71,13,10,26,10) == hdr[1:8])) return("image/png")
 
+  if (all(c(0x23,0x21,0x2f,0x62,0x69,0x6e,0x2f,0x6e,0x6f,0x64,0x65) == hdr[1:11]))
+    return("application/javascript")
+  if (all(c(0x23,0x21,0x2f,0x62,0x69,0x6e,0x2f,0x6e,0x6f,0x64,0x65,0x6a,0x73) == hdr[1:13]))
+    return("application/javascript")
+  if (all(c(0x23,0x21,0x2f,0x75,0x73,0x72,0x2f,0x62,0x69,0x6e,0x2f,0x6e,0x6f,0x64,0x65) == hdr[1:15]))
+    return("application/javascript")
+  if (all(c(0x23,0x21,0x2f,0x75,0x73,0x72,0x2f,0x62,0x69,0x6e,0x2f,0x6e,0x6f,0x64,0x65,0x6a,0x73) == hdr[1:17]))
+    return("application/javascript")
+  if (all(c(0x23,0x21,0x2f,0x75,0x73,0x72,0x2f,0x62,0x69,0x6e,0x2f,0x65,0x6e,0x76,0x20,0x6e,0x6f,0x64,0x65) == hdr[1:19]))
+    return("application/javascript")
+  if (all(c(0x23,0x21,0x2f,0x75,0x73,0x72,0x2f,0x62,0x69,0x6e,0x2f,0x65,0x6e,0x76,0x20,0x6e,0x6f,0x64,0x65,0x6a,0x73) == hdr[1:21]))
+    return("application/javascript")
+
   if (all(c(0xFF,0xD8,0xFF) == hdr[1:3])) {
     if (0xE0 == hdr[4]) return("image/jpeg")
     if (0xE1 == hdr[4]) {
@@ -75,6 +97,14 @@ get_content_type <- function(path) {
     return("application/zip")
   }
 
+  if (all(c(0x5a,0x4d) == hdr[1:2])) return("x-system/exe")
+
+  if (all(c(0x75,0x73,0x74,0x61,0x72) == hdr[258:262])) return("application/pax")
+
+  if (all(c(0x00,0x00,0x01,0xBA) == hdr[1:4])) return("video/mpeg")
+  if (all(c(0x00,0x00,0x01,0xB3) == hdr[1:4])) return("video/mpeg")
+
+
   return(guess_content_type(path))
 
 }

R/guess-content-type.R

@@ -1,5 +1,9 @@
 #' Guess MIME type from filename (extension)
 #'
+#' Uses an internal database of over 1,500 file extension-to-MIME mappings to
+#' return one or more associated types for a given input path. If no match is
+#' found, `???` is returned.
+#'
 #' @details
 #' Incorporates standard IANA MIME extension mappings and those from
 #' [`servoy-client`](https://github.com/Servoy/servoy-client) and
@@ -8,15 +12,18 @@
 #'
 #' @md
 #' @param path path to file
+#' @return character vector
 #' @export
+#' @examples
+#' guess_content_type(system.file("extdat", "test.pdf", package="simplemagic"))
 guess_content_type <- function(path) {
 
   path <- path.expand(path)
   if (!file.exists(path)) stop("File not found.", call.=FALSE)
 
   extension <- trimws(tolower(tools::file_ext(path)))
 
-  res <- .ext_to_mime[(.ext_to_mime$extension == extension),]$mime_type
+  res <- simplemagic_mime_db[(simplemagic_mime_db$extension == extension),]$mime_type
 
   if (length(res) == 0) return("???")
 

R/simplemagic-package.R

@@ -1,5 +1,15 @@
 #' Lightweight File 'MIME' Type Detection Based On Contents or Extension
 #'
+#' Provides a more portable/ligtweight alternative to the `wand` package.
+#'
+#' @section Some important details:
+#'
+#' The header checking is minimal (i.e. nowhere near as comprehensive as `libmagic`) but
+#' covers quite a bit of ground. If there are content-check types from
+#' [`magic sources`](https://github.com/threatstack/libmagic/tree/master/magic/)
+#' that you would like coded into the package, please file an issue and
+#' _include the full line(s)_ from that linked `magic.tab` that you would like mapped.
+#'
 #' @md
 #' @name simplemagic
 #' @docType package

README.Rmd

@@ -13,15 +13,17 @@ determined from "magic" bytes in file headers, file contents or intuited from
 file extensions. Tools are provided to perform limited "magic" tests as well
 as mapping 'MIME' types from a database of over 1,500 extension mappings.
 
+Provides a more portable/ligtweight alternative to the `wand` package.
+
 ## SOME IMPORTANT DETAILS
 
-The header checking is minimal (i.e. nowhere near as comprehensive as `libmagic`).
-If there are content-check types from [`magic.tab`](https://opensource.apple.com/source/ksh/ksh-13/ksh/src/lib/libast/misc/magic.tab.auto.html) that you would like coded into the package, please file an issue and _include the full line(s)_ from that linked `magic.tab` that you would like mapped.
+The header checking is minimal (i.e. nowhere near as comprehensive as `libmagic`) but covers quite a bit of ground. If there are content-check types from [`magic sources`](https://github.com/threatstack/libmagic/tree/master/magic/) that you would like coded into the package, please file an issue and _include the full line(s)_ from that linked `magic.tab` that you would like mapped.
 
 ## What's Inside The Tin
 
 - `get_content_type`:	Discover MIME type of a file based on contents
 - `guess_content_type`:	Guess MIME type from filename (extension)
+- `simplemagic_mime_db`:	File extension-to-MIME mapping data frame
 
 The following functions are implemented:
 

README.md

@@ -11,11 +11,14 @@ from file extensions. Tools are provided to perform limited “magic”
 tests as well as mapping ‘MIME’ types from a database of over 1,500
 extension mappings.
 
+Provides a more portable/ligtweight alternative to the `wand` package.
+
 ## SOME IMPORTANT DETAILS
 
 The header checking is minimal (i.e. nowhere near as comprehensive as
-`libmagic`). If there are content-check types from
-[`magic.tab`](https://opensource.apple.com/source/ksh/ksh-13/ksh/src/lib/libast/misc/magic.tab.auto.html)
+`libmagic`) but covers quite a bit of ground. If there are content-check
+types from [`magic
+sources`](https://github.com/threatstack/libmagic/tree/master/magic/)
 that you would like coded into the package, please file an issue and
 *include the full line(s)* from that linked `magic.tab` that you would
 like mapped.
@@ -24,6 +27,7 @@ like mapped.
 
   - `get_content_type`: Discover MIME type of a file based on contents
   - `guess_content_type`: Guess MIME type from filename (extension)
+  - `simplemagic_mime_db`: File extension-to-MIME mapping data frame
 
 The following functions are implemented:
 
@@ -124,7 +128,7 @@ list.files(system.file("extdat", package="simplemagic"), full.names=TRUE) %>%
     ## 62 test.sh                    application/x-shar                                                       
     ## 63 test.sh                    text/x-script.sh                                                         
     ## 64 test.sh                    text/x-sh                                                                
-    ## 65 test.tar                   application/x-tar                                                        
+    ## 65 test.tar                   application/pax                                                          
     ## 66 test.tar.gz                application/octet-stream                                                 
     ## 67 test.tar.gz                application/x-compressed                                                 
     ## 68 test.tar.gz                application/x-gzip