Added codec: base45

dhondta · dhondta · commit 134a7c647264 · 2021-10-16T22:12:14.000+02:00
diff --git a/codext/base/__init__.py b/codext/base/__init__.py
@@ -1,5 +1,6 @@
 # -*- coding: UTF-8 -*-
 from .ascii85 import *
+from .base45 import *
 from .base85 import *
 from .base91 import *
 from .base100 import *
diff --git a/codext/base/base45.py b/codext/base/base45.py
@@ -0,0 +1,84 @@
+# -*- coding: UTF-8 -*-
+"""Base45 Codec - base45 content encoding.
+
+This codec:
+- en/decodes strings from str to str
+- en/decodes strings from bytes to bytes
+- decodes file content to str (read)
+- encodes file content from str to bytes (write)
+"""
+from ._base import digits, lower, upper
+from ..__common__ import *
+
+
+__examples__ = {
+    'enc(base45|base-45|base_45)': {'this is a test!': "AWE+EDH44.OEOCC7WE QEX0"},
+    'dec(base45)':                 {'BAD STRING\00': None},
+}
+__guess__ = ["base45", "base45-inv"]
+
+
+B45 = {
+    '':    digits + upper + " $%*+-./:",
+    'inv': upper + digits + " $%*+-./:",
+}
+
+
+__chr = lambda c: chr(c) if isinstance(c, int) else c
+__ord = lambda c: ord(c) if not isinstance(c, int) else c
+
+
+class Base45DecodeError(ValueError):
+    pass
+
+
+def base45_encode(mode):
+    mode = mode.replace("inverted", "inv").replace("_", "-").lstrip("-")
+    b45 = B45[['inv', ''][mode == ""]]
+    def encode(text, errors="strict"):
+        t, s = b(text), ""
+        for i in range(0, len(text), 2):
+            n = 256 * __ord(t[i])
+            try:
+                n += __ord(t[i+1])
+            except IndexError:
+                n = __ord(t[i])
+                s += b45[n % 45] + b45[n // 45]
+                break
+            m = n // 45**2
+            n -= m * 45**2
+            s += b45[n % 45] + b45[n // 45] + b45[m]
+        return s, len(text)
+    return encode
+
+
+def base45_decode(mode):
+    mode = mode.replace("inverted", "inv").replace("_", "-").lstrip("-")
+    b45 = {c: i for i, c in enumerate(B45[['inv', ''][mode == ""]])}
+    def decode(text, errors="strict"):
+        t, s, err = b(text), "", "'base45' codec can't decode character '%s' in position %d"
+        for i in range(0, len(text), 3):
+            try:
+                n = b45[__chr(t[i])]
+            except KeyError:
+                raise Base45DecodeError(err % (__chr(t[i]), i))
+            try:
+                j = i + 1
+                n += 45 * b45[__chr(t[j])]
+            except KeyError:
+                raise Base45DecodeError(err % (__chr(t[j]), j))
+            try:
+                k = i + 2
+                n += 45 ** 2 * b45[__chr(t[k])]
+            except KeyError:
+                raise Base45DecodeError(err % (__chr(t[k]), k))
+            except IndexError:
+                s += __chr(n)
+                continue
+            s += __chr(n // 256) + __chr(n % 256)
+        return s, len(text)
+    return decode
+
+
+add("base45", base45_encode, base45_decode, r"^base[-_]?45(|[-_]inv(?:erted)?)$")
+
diff --git a/tests/test_generated.py b/tests/test_generated.py
@@ -84,7 +84,7 @@ def _template(self):
                             with codecs.open(tfile, 'wb', encoding=ename) as f:
                                 f.write(b(s))
                             with codecs.open(tfile, 'rb', encoding=ename) as f:
-                                s2 = f.read().strip(b("\x00"))
+                                s2 = f.read() if PY3 else f.read().rstrip("\x00")
                             self.assertEqual(b(icdec(s2)), b(icdec(s)))
                             os.remove(tfile)
                     else:
