X-Git-Url: https://wannabe.guru.org/gitweb/?a=blobdiff_plain;f=unscrambler.py;h=9df82f61a516b78541feddf864ca49f9ce31668b;hb=90dc74b2ff8f36dd2495632a8150ce37dab61f1c;hp=d3686d67ed763d9b240b58ee9d43b5151ed31ea8;hpb=4c86d98e337703016afc7122cd165bb2498b88bc;p=python_utils.git

diff --git a/unscrambler.py b/unscrambler.py
index d3686d6..9df82f6 100644
--- a/unscrambler.py
+++ b/unscrambler.py
@@ -1,17 +1,18 @@
 #!/usr/bin/env python3
 
 import logging
-from typing import Dict, Mapping
+from typing import Dict, Mapping, Optional
 
 import config
 import decorator_utils
+import file_utils
 import list_utils
 
 cfg = config.add_commandline_args(
-    f'Unscramble! ({__file__})', 'A fast word unscrambler.'
+    f'Unscrambler base library ({__file__})', 'A fast word unscrambler.'
 )
 cfg.add_argument(
-    "--unscramble_indexfile",
+    "--unscrambler_default_indexfile",
     help="Path to a file of signature -> word index.",
     metavar="FILENAME",
     default="/usr/share/dict/sparse_index",
@@ -100,25 +101,31 @@ class Unscrambler(object):
 
     """
 
-    def __init__(self):
+    def __init__(self, indexfile: Optional[str] = None):
         # Cached index per instance.
         self.sigs = []
         self.words = []
 
-        if 'unscramble_indexfile' in config.config:
-            indexfile = config.config['unscramble_indexfile']
-        else:
-            indexfile = "/usr/share/dict/sparse_index"
-
-        with open(indexfile, 'r') as rf:
+        filename = self.get_indexfile(indexfile)
+        with open(filename, 'r') as rf:
             lines = rf.readlines()
         for line in lines:
             line = line[:-1]
             (fsig, word) = line.split('+')
-            fsig = int(fsig, 16)
-            self.sigs.append(fsig)
+            isig = int(fsig, 16)
+            self.sigs.append(isig)
             self.words.append(word)
 
+    def get_indexfile(self, indexfile: Optional[str]) -> str:
+        if indexfile is None:
+            if 'unscrambler_default_indexfile' in config.config:
+                indexfile = config.config['unscramble_indexfile']
+            else:
+                indexfile = "/usr/share/dict/sparse_index"
+        else:
+            assert file_utils.file_is_readable(indexfile), f"Can't read {indexfile}"
+        return indexfile
+
     # 52 bits
     @staticmethod
     def _compute_word_fingerprint(word: str, population: Mapping[str, int]) -> int:
@@ -198,7 +205,7 @@ class Unscrambler(object):
         unless you want to populate the same exact files.
 
         """
-        words_by_sigs = {}
+        words_by_sigs: Dict[int, str] = {}
         seen = set()
         with open(dictfile, "r") as f:
             for word in f:
@@ -218,23 +225,19 @@ class Unscrambler(object):
                 word = words_by_sigs[sig]
                 print(f'0x{sig:x}+{word}', file=f)
 
-    def lookup(
-        self, word: str, *, include_fuzzy_matches: bool = False
-    ) -> Dict[str, bool]:
+    def lookup(self, word: str, *, window_size: int = 5) -> Dict[str, bool]:
         """Looks up a potentially scrambled word optionally including near
         "fuzzy" matches.
 
         >>> u = Unscrambler()
-        >>> u.lookup('eanycleocipd', include_fuzzy_matches=False)
+        >>> u.lookup('eanycleocipd', window_size=0)
         {'encyclopedia': True}
 
         """
         sig = Unscrambler.compute_word_sig(word)
-        return self.lookup_by_sig(sig, include_fuzzy_matches=include_fuzzy_matches)
+        return self.lookup_by_sig(sig, window_size=window_size)
 
-    def lookup_by_sig(
-        self, sig: int, *, include_fuzzy_matches: bool = False
-    ) -> Dict[str, bool]:
+    def lookup_by_sig(self, sig: int, *, window_size: int = 5) -> Dict[str, bool]:
         """Looks up a word that has already been translated into a signature by
         a previous call to Unscrambler.compute_word_sig.  Optionally returns
         near "fuzzy" matches.
@@ -244,23 +247,23 @@ class Unscrambler(object):
         18491949645300288339
 
         >>> u = Unscrambler()
-        >>> u.lookup_by_sig(sig, include_fuzzy_matches=True)
+        >>> 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 = {}
         (exact, location) = list_utils.binary_search(self.sigs, sig)
-        start = location - 5
+        start = location - window_size
         if start < 0:
             start = 0
-        end = location + 6
+        end = location + 1 + window_size
         if end > len(self.words):
             end = len(self.words)
 
         for x in range(start, end):
             word = self.words[x]
             fsig = self.sigs[x]
-            if include_fuzzy_matches is True or (fsig == sig):
+            if window_size > 0 or (fsig == sig):
                 ret[word] = fsig == sig
         return ret