summaryrefslogtreecommitdiff
path: root/unscrambler.py
diff options
context:
space:
mode:
authorScott Gasch <[email protected]>2022-05-31 22:12:19 -0700
committerScott Gasch <[email protected]>2022-05-31 22:12:19 -0700
commit2a6d7810f85b88fb89d11d3e6be64f974ab77f97 (patch)
tree8b77a19a93cf573276ea9caca3baf9bea055f2f4 /unscrambler.py
parentec7c26dad2c26e4e4c33a69252654e3b1155bc7b (diff)
More doc updates.
Diffstat (limited to 'unscrambler.py')
-rw-r--r--unscrambler.py49
1 files changed, 44 insertions, 5 deletions
diff --git a/unscrambler.py b/unscrambler.py
index a40213e..3ccddbb 100644
--- a/unscrambler.py
+++ b/unscrambler.py
@@ -102,10 +102,16 @@ class Unscrambler(object):
lookup methods support a "fuzzy match" argument that can be set to
request similar words that do not match exactly in addition to any
exact matches.
-
"""
def __init__(self, indexfile: Optional[str] = None):
+ """
+ Constructs an unscrambler.
+
+ Args:
+ indexfile: overrides the default indexfile location if provided
+ """
+
# Cached index per instance.
self.sigs = []
self.words = []
@@ -122,6 +128,7 @@ class Unscrambler(object):
@staticmethod
def get_indexfile(indexfile: Optional[str]) -> str:
+ """Returns the current indexfile location."""
if indexfile is None:
if 'unscrambler_default_indexfile' in config.config:
indexfile = config.config['unscramble_indexfile']
@@ -177,6 +184,12 @@ class Unscrambler(object):
the word and their frequencies. We try to cluster "similar"
words close to each other in the signature space.
+ Args:
+ word: the word to compute a signature for
+
+ Returns:
+ The word's signature.
+
>>> train = Unscrambler.compute_word_sig('train')
>>> train
23178969883741
@@ -201,9 +214,14 @@ class Unscrambler(object):
dictfile: str = '/usr/share/dict/words',
indexfile: str = '/usr/share/dict/sparse_index',
) -> None:
- """Before calling this method, change letter_sigs from the default above
- unless you want to populate the same exact files.
+ """
+ Repopulates the indexfile.
+ .. warning::
+
+ Before calling this method, change letter_sigs from the
+ default above unless you want to populate the same exact
+ files.
"""
words_by_sigs: Dict[int, str] = {}
seen = set()
@@ -229,10 +247,20 @@ class Unscrambler(object):
"""Looks up a potentially scrambled word optionally including near
"fuzzy" matches.
+ Args:
+ word: the word to lookup
+ window_size: the number of nearby fuzzy matches to return
+
+ Returns:
+ A dict of word -> bool containing unscrambled words with (close
+ to or precisely) the same letters as the input word. The bool
+ values in this dict indicate whether the key word is an exact
+ or near match. The count of entries in this dict is controlled
+ by the window_size param.
+
>>> u = Unscrambler()
>>> u.lookup('eanycleocipd', window_size=0)
{'encyclopedia': True}
-
"""
sig = Unscrambler.compute_word_sig(word)
return self.lookup_by_sig(sig, window_size=window_size)
@@ -242,6 +270,18 @@ class Unscrambler(object):
a previous call to Unscrambler.compute_word_sig. Optionally returns
near "fuzzy" matches.
+ Args:
+ sig: the signature of the word to lookup (see :meth:`compute_word_sig`
+ to generate these signatures).
+ window_size: the number of nearby fuzzy matches to return
+
+ Returns:
+ A dict of word -> bool containing unscrambled words with (close
+ to or precisely) the same letters as the input word. The bool
+ values in this dict indicate whether the key word is an exact
+ or near match. The count of entries in this dict is controlled
+ by the window_size param.
+
>>> sig = Unscrambler.compute_word_sig('sunepsapetuargiarin')
>>> sig
18491949645300288339
@@ -249,7 +289,6 @@ class Unscrambler(object):
>>> u = Unscrambler()
>>> u.lookup_by_sig(sig)
{'pupigerous': False, 'pupigenous': False, 'unpurposing': False, 'superpurgation': False, 'unsupporting': False, 'superseptuaginarian': True, 'purpurogallin': False, 'scuppaug': False, 'purpurigenous': False, 'purpurogenous': False, 'proppage': False}
-
"""
ret = {}
(_, location) = list_utils.binary_search(self.sigs, sig)