Base64 decoding from Teemu Antti-Poika; refs #84

2 files changed, 117 insertions, 14 deletions

diff --git a/modules/base64/src/main/clojure/clojure/contrib/base64.clj b/modules/base64/src/main/clojure/clojure/contrib/base64.clj
index 2556487c..cef22c50 100644
--- a/modules/base64/src/main/clojure/clojure/contrib/base64.clj
+++ b/modules/base64/src/main/clojure/clojure/contrib/base64.clj
@@ -1,7 +1,9 @@
-;;; base64.clj: Experimental Base-64 encoding and (later) decoding
+;;; base64.clj: Experimental Base-64 encoding and decoding
 
-;; by Stuart Sierra, http://stuartsierra.com/
+;; by Stuart Sierra, http://stuartsierra.com/ - encode
 ;; August 19, 2009
+;; by Teemu Antti-Poika (anttipoi@gmail.com) - decode
+;; May 12, 2010
 
 ;; Copyright (c) Stuart Sierra, 2009. All rights reserved.  The use
 ;; and distribution terms for this software are covered by the Eclipse
@@ -12,18 +14,25 @@
 ;; remove this notice, or any other, from this software.
 
 
-(ns ^{:doc "Base-64 encoding and (maybe later) decoding.  
+(ns ^{:doc "Base-64 encoding and decoding.  
 
   This is mainly here as an example.  It is much slower than the
   Apache Commons Codec implementation or sun.misc.BASE64Encoder."
        :author "Stuart Sierra"}
     clojure.contrib.base64
-  (:import (java.io InputStream Writer ByteArrayInputStream
-                    StringWriter)))
+  (:import (java.io InputStream Writer ByteArrayInputStream 
+                    ByteArrayOutputStream StringReader StringWriter)))
 
 (def *base64-alphabet*
      "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/=")
 
+(defn- char-to-index-map 
+  []
+  (into {} 
+        (map #(vec [(int %1) %2])
+             *base64-alphabet* 
+             (iterate inc 0))))
+
 (defn encode
   "Encodes bytes of input, writing Base 64 text on output.  alphabet
   is a 65-character String containing the 64 characters to use in the
@@ -86,14 +95,52 @@
                output *base64-alphabet* line-length)
        (.toString output))))
 
+(defn- get-next-char
+  "Consume and return next character from reader. Ignore and eat end-of-lines characters. Return -1 on end."
+  [reader]
+  (let [c (.read reader)]
+    (if (or (= c 10) (= c 13))
+      (recur reader)
+      c)))
 
-;;; tests 
+(defn decode
+  "Decodes base64-encoded content from str-reader. Writes resulting bytes to out."
+  [^StringReader str-reader ^ByteArrayOutputStream out]
+  (let [next-char (get-next-char str-reader)]
+    (when (not (= next-char -1))
+      (let [c-to-int (char-to-index-map)
+            content-char? (fn [i] (not (= i 64))) ; 64 is index for the pad character =
+            sb0 (c-to-int next-char)
+            sb1 (c-to-int (get-next-char str-reader))
+            sb2 (c-to-int (get-next-char str-reader))
+            sb3 (c-to-int (get-next-char str-reader))
+            _ (when (not (and sb0 sb1 sb2 sb3)) 
+                (throw (IllegalArgumentException. "Illegal Base64-encoded input: illegal characters or missing padding")))
+            _ (when (not (and (content-char? sb0) (content-char? sb1)))
+                (throw (IllegalArgumentException. "Illegal Base64-encoded input: padding char at illegl position")))
+            b0 (bit-or
+                (bit-shift-left sb0 2)
+                (bit-shift-right (bit-and 0x30 sb1) 4))]
+        (.write out b0)
+        (when (content-char? sb2)
+          (let [b1 (bit-or
+                    (bit-shift-left (bit-and 0xF sb1) 4)
+                    (bit-shift-right (bit-and 0x3C sb2) 2))]
+            (.write out b1)
+            (when (content-char? sb3)
+              (let [b2 (bit-or
+                        (bit-shift-left (bit-and 0x3 sb2) 6)
+                        sb3)]
+                (.write out b2)
+                (recur str-reader out)))))))))
+  
 
-;; (deftest t-encode-str
-;;   (is (= (encode-str "") ""))
-;;   (is (= (encode-str "f") "Zg=="))
-;;   (is (= (encode-str "fo") "Zm8="))
-;;   (is (= (encode-str "foo") "Zm9v"))
-;;   (is (= (encode-str "foob") "Zm9vYg=="))
-;;   (is (= (encode-str "fooba") "Zm9vYmE="))
-;;   (is (= (encode-str "foobar") "Zm9vYmFy")))
+(defn decode-str
+  "Decodes base64-encoded String using encoding. Encoding defaults to UTF-8."
+  ([s] (decode-str s "UTF-8"))
+  ([^String s ^String encoding]
+     (when s
+       (let [baos (ByteArrayOutputStream.)
+             str-reader (StringReader. s)]
+         (decode str-reader baos)
+         (String. (.toByteArray baos) encoding)))))
\ No newline at end of file
diff --git a/modules/base64/src/main/clojure/clojure/contrib/test_base64.clj b/modules/base64/src/main/clojure/clojure/contrib/test_base64.clj
new file mode 100644
index 00000000..8ae63792
--- /dev/null
+++ b/modules/base64/src/main/clojure/clojure/contrib/test_base64.clj
@@ -0,0 +1,56 @@
+;   Copyright (c) Teemu Antti-Poika, May 2010. All rights reserved.
+;   The use and distribution terms for this software are covered by the
+;   Eclipse Public License 1.0 (http://opensource.org/licenses/eclipse-1.0.php)
+;   which can be found in the file epl-v10.html at the root of this 
+;   distribution.
+;   By using this software in any fashion, you are agreeing to be bound by
+;   the terms of this license.
+;   You must not remove this notice, or any other, from this software.
+
+;; test namespace for clojure.contrib.base64
+
+
+(ns clojure.contrib.test-base64
+  (:use [clojure.test]
+        [clojure.contrib.base64]))
+
+
+(deftest t-encode-str
+  (is (= (encode-str "") ""))
+  (is (= (encode-str "f") "Zg=="))
+  (is (= (encode-str "fo") "Zm8="))
+  (is (= (encode-str "foo") "Zm9v"))
+  (is (= (encode-str "foob") "Zm9vYg=="))
+  (is (= (encode-str "fooba") "Zm9vYmE="))
+  (is (= (encode-str "foobar") "Zm9vYmFy")))
+
+(deftest t-encode-multiline-str
+  (is (= (encode-str "This fits on one line" "UTF-8" 72) "VGhpcyBmaXRzIG9uIG9uZSBsaW5l"))
+  (is (= (encode-str "This is written on multiple lines" "UTF-8" 10) "VGhpcyBpcyB3\ncml0dGVuIG9u\nIG11bHRpcGxl\nIGxpbmVz")))
+
+(deftest t-decode-str
+  (is (nil? (decode-str nil)))
+  (is (= (decode-str "") ""))
+  (is (= (decode-str "Zg==") "f"))
+  (is (= (decode-str "Zm8=") "fo"))
+  (is (= (decode-str "Zm9v") "foo"))
+  (is (= (decode-str "Zm9vYg==") "foob"))
+  (is (= (decode-str "Zm9vYmE=") "fooba"))
+  (is (= (decode-str "Zm9vYmFy") "foobar")))
+
+(deftest t-decode-multiline-str
+  (is (= (decode-str "VGhpcyBpcyB3\ncml0dGVuIG9u\nIG11bHRpcGxl\nIGxpbmVz") "This is written on multiple lines"))
+  (is (= (decode-str "VGhpcyBpcyB3\r\ncml0dGVuIG9u\r\nIG11bHRpcGxl\r\nIGxpbmVz") "This is written on multiple lines")))
+
+(deftest t-decode-str-fails-on-incorrect-charcters-in-input
+  (is (thrown? IllegalArgumentException (decode-str ";AB=")))
+  (is (thrown? IllegalArgumentException (decode-str "A;B=")))
+  (is (thrown? IllegalArgumentException (decode-str "AB;=")))
+  (is (thrown? IllegalArgumentException (decode-str "ABC;"))))
+
+(deftest t-decode-str-fails-when-padding-character-occurs-at-illegal-positions
+  (is (thrown? IllegalArgumentException (decode-str "=ABC")))
+  (is (thrown? IllegalArgumentException (decode-str "A=BC"))))
+
+(deftest t-decode-str-fails-when-input-length-is-not-divisble-with-four
+  (is (thrown? IllegalArgumentException (decode-str "ABCDE"))))
\ No newline at end of file