Improved codec: tap

dhondta · dhondta · commit 73b1534d88f3 · 2022-01-19T18:26:15.000+01:00
diff --git a/README.md b/README.md
@@ -288,8 +288,8 @@ o
 - [X] `radio`: aka NATO or radio phonetic alphabet
 - [X] `southpark`: converts letters to Kenny's language from Southpark (whitespace is also handled)
 - [X] `southpark-icase`: case insensitive variant of `southpark`
+- [X] `tap`: converts text to tap/knock code, commonly used by prisoners
 - [X] `tomtom`: similar to `morse`, using slashes and backslashes
-- [X] `tap` : converts tap/knock code commonly used by prisoners
 
 #### Others
 
diff --git a/codext/languages/__init__.py b/codext/languages/__init__.py
@@ -7,6 +7,6 @@
 from .navajo import *
 from .radio import *
 from .southpark import *
-from .tomtom import *
 from .tap import *
+from .tomtom import *
 
diff --git a/codext/languages/tap.py b/codext/languages/tap.py
@@ -11,46 +11,29 @@
 
 
 __examples__ = {
-    'enc(tap)': {'this is a test' : '.... ....  .. ...  .. ....  .... ...  .. ....  .... ...  . .  .... ....  . .....  .... ...  .... ....'},
-    'dec(tap)': {'.... ....  .. ...  .. ....  .... ...  .. ....  .... ...  . .  .... ....  . .....  .... ...  .... ....' : 'thisisatest'}
+    'enc(tap|knock-code|tap_code)': {'this is a test' : ".... ....⠀.. ...⠀.. ....⠀.... ...⠀ ⠀.. ....⠀.... ...⠀ ⠀. ."
+                                                        "⠀ ⠀.... ....⠀. .....⠀.... ...⠀.... ...."},
 }
+__guess__ = ["tap", "tap-inv"]
 
 
-def build_encmap(map) : 
-    dict = {}
-    i = 0
-    for col in range(1,6) : 
-        for row in range(1,6) : 
-            dict[map[i]] = "" + col * "." + " " + row * "."
+def __build_encmap(a):
+    d, i = {}, 0
+    for x in range(1,6): 
+        for y in range(1,6): 
+            d[a[i]] = x * "." + " " + y * "."
             i += 1
-    dict['k'] = dict['c']
-    return dict
-
-def encode_tap(text, errors = 'strict') : 
-    map = 'abcdefghijlmnopqrstuvwxyz'
-    ENCMAP = build_encmap(map)
-    encoded = ""
-    for i, letter in enumerate(text) :
-        try : 
-            encoded += ENCMAP[letter.lower()]
-        except KeyError : 
-            pass
-        if i != len(text) - 1 and letter != ' ': 
-            encoded += '  '
-    return encoded, len(text)    
-
-
-def decode_tap(text, errors = 'ignore') : 
-    map = 'abcdefghijlmnopqrstuvwxyz'
-    ENCMAP = build_encmap(map)
-    decoded = ""
-    for elem in text.split("  ") :
-        try : 
-            decoded += next(key for key, value in ENCMAP.items() if value == elem)
-        except StopIteration : 
-            print("Invalid character(s) in the input. This is what could be decoded :")
-    return decoded, len(text)    
-  
-
-add("tap", encode_tap, decode_tap, ignore_case="both")
+    d['k'], d[' '] = d['c'], " "
+    return d
+
+
+
+ENCMAP = {
+    '':    __build_encmap("abcdefghijlmnopqrstuvwxyz"),
+    'inv': __build_encmap("abcdefghijlmnopqrstuvwxyz"[::-1]),
+}
+
+
+if PY3:
+    add_map("tap", ENCMAP, ignore_case="both", sep="⠀", pattern=r"^(?:tap|knock)(?:[-_]code)?(|inv)$")
 
diff --git a/docs/enc/languages.md b/docs/enc/languages.md
@@ -166,34 +166,34 @@ This encodes text according to Kenny's language in Southpark.
 
 -----
 
-### Tom-Tom
+### Tap
 
-This codec is similar to morse. It converts text into slashes and backslashes.
+This codec implements the [tap/knock code](https://en.wikipedia.org/wiki/Tap_code) commonly used by prisoners. It uses 25 letters, "*k*" is encoded to the same token than "*c*".  
 
 **Codec** | **Conversions** | **Aliases** | **Comment**
 :---: | :---: | --- | ---
-`tomtom` | text <-> tom-tom encoded text | `tom-tom` | uses "`|`" as a separator
+`tap` | text <-> tap/knock encoded text | `knock`, `tap-code` | uses a large Unicode whitespace as a token separator ; Python 3 only
 
 ```python
->>> codext.encode("this is a test", "tom-tom")
-'\\\\/\\ /\\\\ /\\\\\\ \\/\\ | /\\\\\\ \\/\\ | / | \\\\/\\ /\\ \\/\\ \\\\/\\'
->>> codext.decode("\\\\/\\ /\\\\ /\\\\\\ \\/\\ | /\\\\\\ \\/\\ | / | \\\\/\\ /\\ \\/\\ \\\\/\\", "tomtom")
-'THIS IS A TEST'
+>>> codext.encode("this is a test", "tap")
+'.... ....⠀.. ...⠀.. ....⠀.... ...⠀ ⠀.. ....⠀.... ...⠀ ⠀. .⠀ ⠀.... ....⠀. .....⠀.... ...⠀.... ....'
+>>> codext.decode(".... ....⠀.. ...⠀.. ....⠀.... ...⠀ ⠀.. ....⠀.... ...⠀ ⠀. .⠀ ⠀.... ....⠀. .....⠀.... ...⠀.... ....", "knock")
+'this is a test'
 ```
 
 -----
 
-### Tap
+### Tom-Tom
 
-Converts tap/knock code [commonly used by prisoners](https://en.wikipedia.org/wiki/Tap_code). Uses 25 letters, 'k' codes as 'c'.  
+This codec is similar to morse. It converts text into slashes and backslashes.
 
 **Codec** | **Conversions** | **Aliases** | **Comment**
 :---: | :---: | --- | ---
-`tap` | text <-> tap/knock encoded text | `tap` | uses '&nbsp; &nbsp;' (double space) as a separator for letters. No spaces between words after decoding. 
+`tomtom` | text <-> tom-tom encoded text | `tom-tom` | uses "`|`" as a separator
 
 ```python
->>> codext.encode("this is a test", "tap")
-'.... ....  .. ...  .. ....  .... ...  .. ....  .... ...  . .  .... ....  . .....  .... ...  .... ....'
->>> codext.decode(".... ....  .. ...  .. ....  .... ...  .. ....  .... ...  . .  .... ....  . .....  .... ...  .... ....", "tap")
-'thisisatest'
+>>> codext.encode("this is a test", "tom-tom")
+'\\\\/\\ /\\\\ /\\\\\\ \\/\\ | /\\\\\\ \\/\\ | / | \\\\/\\ /\\ \\/\\ \\\\/\\'
+>>> codext.decode("\\\\/\\ /\\\\ /\\\\\\ \\/\\ | /\\\\\\ \\/\\ | / | \\\\/\\ /\\ \\/\\ \\\\/\\", "tomtom")
+'THIS IS A TEST'
 ```
