Many tests done, still more to come.

[szyfrow.git] / szyfrow / support / language_models.py

diff --git a/szyfrow/support/language_models.py b/szyfrow/support/language_models.py
index 271b8dba023f93f35ab9c6b93d51e10d36fcdcb4..95dbeb277b8c828f62cc6e8ac83938e683491ed0 100644 (file)
--- a/szyfrow/support/language_models.py
+++ b/szyfrow/support/language_models.py
@@ -4,14 +4,18 @@ import collections
 import itertools
 from math import log10
 import os 
+import importlib.resources as pkg_resources
 
 import szyfrow.support.norms
-from szyfrow.support.utilities import sanitise
+from szyfrow.support.utilities import sanitise, deduplicate
+from szyfrow import language_model_files
+
 
 def datafile(name, sep='\t'):
     """Read key,value pairs from file.
     """
-    with open(os.path.join(os.path.dirname(os.path.realpath(__file__)), name), 'r') as f:
+    with pkg_resources.open_text(language_model_files, name) as f:
+    # with open(p name), 'r') as f:
         for line in f:
             splits = line.split(sep)
             yield [splits[0], int(splits[1])]
@@ -25,10 +29,37 @@ normalised_english_bigram_counts = szyfrow.support.norms.normalise(english_bigra
 english_trigram_counts = collections.Counter(dict(datafile('count_3l.txt')))
 normalised_english_trigram_counts = szyfrow.support.norms.normalise(english_trigram_counts)
 
-with open(os.path.join(os.path.dirname(os.path.realpath(__file__)), 'words.txt'), 'r') as f:
+with pkg_resources.open_text(language_model_files, 'words.txt') as f:
     keywords = [line.rstrip() for line in f]
 
 
+def transpositions_of(keyword):
+    """Finds the transpostions given by a keyword. For instance, the keyword
+    'clever' rearranges to 'celrv', so the first column (0) stays first, the
+    second column (1) moves to third, the third column (2) moves to second, 
+    and so on.
+
+    If passed a tuple, assume it's already a transposition and just return it.
+
+    >>> transpositions_of('clever')
+    (0, 2, 1, 4, 3)
+    >>> transpositions_of('fred')
+    (3, 2, 0, 1)
+    >>> transpositions_of((3, 2, 0, 1))
+    (3, 2, 0, 1)
+    """
+    if isinstance(keyword, tuple):
+        return keyword
+    else:
+        key = deduplicate(keyword)
+        transpositions = tuple(key.index(l) for l in sorted(key))
+        return transpositions
+
+transpositions = collections.defaultdict(list)
+for word in keywords:
+    transpositions[transpositions_of(word)] += [word]
+
+
 def weighted_choice(d):
     """Generate random item from a dictionary of item counts
     """