wip: tcr compression

regex flags do not need "u" when the search string does not make use of the feature

In python3, re.U is the default for unicode strings. For byte strings,
which is what we use, it is a fatal error.

src/calibre/ebooks/compression/tcr.py

@@ -1,5 +1,4 @@
 # -*- coding: utf-8 -*-
-
 from __future__ import absolute_import, division, print_function, unicode_literals
 
 __license__ = 'GPL 3'
@@ -35,10 +34,10 @@ class TCRCompressor(object):
         The intent is to create more unused codes.
         '''
         possible_codes = []
-        a_code = set(re.findall(b'(?msu).', self.coded_txt))
+        a_code = set(re.findall(b'(?ms).', self.coded_txt))
 
         for code in a_code:
-            single_code = set(re.findall(b'(?msu)%s.' % re.escape(code), self.coded_txt))
+            single_code = set(re.findall(b'(?ms)%s.' % re.escape(code), self.coded_txt))
             if len(single_code) == 1:
                 possible_codes.append(single_code.pop())
 
@@ -60,7 +59,7 @@ class TCRCompressor(object):
         '''
         Create new codes from codes that occur in pairs often.
         '''
-        possible_new_codes = list(set(re.findall(b'(?msu)..', self.coded_txt)))
+        possible_new_codes = list(set(re.findall(b'(?ms)..', self.coded_txt)))
         new_codes_count = []
 
         for c in possible_new_codes:
@@ -77,7 +76,7 @@ class TCRCompressor(object):
     def compress(self, txt):
         self._reset()
 
-        self.codes = list(set(re.findall(b'(?msu).', txt)))
+        self.codes = list(set(re.findall(b'(?ms).', txt)))
 
         # Replace the text with their corresponding code
         for c in txt: