implemented ecoding/decoding in commonchars.py

text_codecs/commonchars.py

@@ -0,0 +1,93 @@
+from easybits import Bits
+import codecs
+
+def encode(text):
+    "An encoder which only handles ASCII: non-ASCII characters are replaced with '?'. "
+    "this will remove the most common characters in english texts [e, t, a, o, i, n, s, h] and by writing a 1 and followed by 3 bits"
+    "other charcaters will be written with stadard 8 bit code"
+    "this ecoding has about a 77 percent compression rate"
+    "there is some loss in quality since unknown characters are replaced with ? and not decoded"
+   
+    result = Bits()
+    commonchars = ['e', 't', 'a', 'o', 'i', 'n', 's', 'h']
+    for char in text:
+        try:
+            b = Bits(char, encoding='ascii')
+        except UnicodeEncodeError:
+            b = Bits('?', encoding='ascii')
+            char = '?'
+        if char in commonchars:
+            result = result.concat(Bits('1'))
+            i = commonchars.index(char)
+            if i == 0: 
+                result = result.concat(Bits('000'))
+            elif i == 1: 
+                result = result.concat(Bits("001"))
+            elif i == 2: 
+                result = result.concat(Bits("010"))
+            elif i == 3: 
+                result = result.concat(Bits("011"))
+            elif i == 4: 
+                result = result.concat(Bits("100"))
+            elif i == 5: 
+                result = result.concat(Bits("101"))
+            elif i == 6: 
+                result = result.concat(Bits("110"))
+            elif i == 7: 
+                result = result.concat(Bits("111"))
+        else:
+            result = result.concat(b)
+    return result.bytes
+
+def decode(data):
+    "Matching decoder. Checks if first bit is 1 or 0. If it is 1, then it changes character out of encoded form, which is 3 bits long. "
+    "If not, it just converts it using standard ascii which is 8 bits."
+    bits = Bits(bytes(data))
+    text = ""
+    commonchars = ['e', 't', 'a', 'o', 'i', 'n', 's', 'h']
+    i = 0
+    while i < len(bits):
+        if bits[i] == 1:
+            byte = bits[i+1:i+4]
+            if byte == Bits('000'):
+                text += commonchars[0]
+            elif byte == Bits('001'):
+                text += commonchars[1]
+            elif byte == Bits('010'):
+                text += commonchars[2]
+            elif byte == Bits('011'):
+                text += commonchars[3]
+            elif byte == Bits('100'):
+                text += commonchars[4]
+            elif byte == Bits('101'):
+                text += commonchars[5]
+            elif byte == Bits('110'):
+                text += commonchars[6]
+            elif byte == Bits('111'):
+                text += commonchars[7]
+            i += 4
+        
+        else:
+            b = bits[i:i+8]
+            text += b.ascii
+            i += 8
+    
+    return text
+
+def register_codec(encode, decode, name):
+    """Registers a codec so that it can later be used to encode
+    or decode strings and bytes.
+    """
+    def encode_wrapper(text):
+        return encode(text), len(text)
+
+    def decode_wrapper(data):
+        return decode(data), len(data)
+
+    def search_for_codec(query):
+        if query == name:
+            return codecs.CodecInfo(encode_wrapper, decode_wrapper, name=name)
+
+    codecs.register(search_for_codec)
+
+register_codec(encode, decode, "commonchars")