I created a text codec ascii5 similar to ascii7

but with differnt characters included. This reads/encodes
with 5 bits instead of the original 7.

.envrc

@@ -0,0 +1 @@
+source .venv/bin/activate

text_codecs/alphanumeric.py

@@ -1,10 +1,28 @@
 import string
 import codecs
-from custom_codecs.register import register_codec
+#from text_codecs.register import register_codec
 from easybits import Bits
 
 allowed_characters = string.ascii_letters + string.digits
 
+import codecs
+
+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)
+
 def encode(text):
     """A (very) lossy encoder which only saves ASCII letters, numbers, and spaces.
     Everything else is discarded. All whitespace (e.g. tabs) is converted into spaces.

text_codecs/ascii5.py

@@ -0,0 +1,46 @@
+from easybits import Bits
+
+import codecs
+
+def register_codec(encode, decode, name):
+    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)
+
+characters=['A','B','C','D','E','F','G','H','I','J','K','L','M', 'N', 'O','P','Q','R','S','T','U','V','W','X','Y','Z',' ','.',',','?','!']
+
+
+def encode(text):
+    """This encoder consists of the 26 uppercase letters and select punctuation. 
+    Characters that are not defined are removed, which makes the encoding less accurate than ascii7. 
+    There are 5 bits for each byte, so the function ascii5 compresses text into 5/8 of its original size."""
+    result= Bits()
+    text = text.upper()
+    for character in text:
+        if character in characters:
+            index = characters.index(character)
+            b=Bits(bin(index)[2:].zfill(5))
+            result = result.concat(b)
+    return result.bytes
+
+def decode(data):
+    """This function reads 5 bits and returns the text to match."""
+    bits=Bits(bytes(data))
+    text = ''
+    for i in range(0,len(bits),5):
+        s=bits[i:i+5]
+        index=int(s.bin,2)
+        text+= characters[index]
+    return text
+
+
+register_codec(encode, decode, "ascii5")
+

text_codecs/ascii7.py

@@ -1,6 +1,24 @@
-from custom_codecs.register import register_codec
+#from text_codecs.register import register_codec
 from easybits import Bits
 
+import codecs
+
+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)
+
 def encode(text):
     """An encoder which only handles ASCII: non-ASCII characters
     are replaced with '?'. Once all the characters are ASCII, this encoder

text_codecs/hello.txt

@@ -0,0 +1 @@
+hello!