X-Git-Url: http://git.veekun.com/zzz-pokedex.git/blobdiff_plain/517e947f5ab92e233677f95286af9feabeb68fbc..c87f91105dbce6befb0d4b68c1d6b9bb2a69e805:/pokedex/db/__init__.py

diff --git a/pokedex/db/__init__.py b/pokedex/db/__init__.py
index c4013f4..99988fb 100644
--- a/pokedex/db/__init__.py
+++ b/pokedex/db/__init__.py
@@ -1,3 +1,6 @@
+# encoding: utf-8
+import re
+
 from sqlalchemy import engine_from_config, orm
 
 from ..defaults import get_default_db_uri
@@ -21,7 +24,7 @@ def connect(uri=None, session_args={}, engine_args={}, engine_prefix=''):
         uri = get_default_db_uri()
 
     ### Do some fixery for MySQL
-    if uri[0:5] == 'mysql':
+    if uri.startswith('mysql:'):
         # MySQL uses latin1 for connections by default even if the server is
         # otherwise oozing with utf8; charset fixes this
         if 'charset' not in uri:
@@ -45,9 +48,39 @@ def connect(uri=None, session_args={}, engine_args={}, engine_prefix=''):
         **all_session_args)
     session = MultilangScopedSession(sm)
 
-    # Default to English.  Warning, magic constant, blah blah.  Trying to fetch
-    # English here would kinda break on new databases.  TODO still not an ideal
-    # solution, I guess.
-    session._default_language_id = 9
+    # Default to English.  Warning, magic constant, messing with internals,
+    # blah blah.  Trying to fetch English here would kinda break on new
+    # databases.  TODO still not an ideal solution, I guess.
+    session.registry()._default_language_id = 9
 
     return session
+
+def identifier_from_name(name):
+    """Make a string safe to use as an identifier.
+
+    Valid characters are lowercase alphanumerics and "-". This function may
+    raise ValueError if it can't come up with a suitable identifier.
+
+    This function is useful for scripts which add things with names.
+    """
+    if isinstance(name, str):
+        identifier = name.decode('utf-8')
+    else:
+        identifier = name
+    identifier = identifier.lower()
+    identifier = identifier.replace(u'+', u' plus ')
+    identifier = re.sub(u'[ _–]+', u'-', identifier)
+    identifier = re.sub(u"['./;’(),:]", u'', identifier)
+    identifier = identifier.replace(u'é', u'e')
+    identifier = identifier.replace(u'♀', u'-f')
+    identifier = identifier.replace(u'♂', u'-m')
+    if identifier in (u'???', u'????'):
+        identifier = u'unknown'
+    elif identifier == u'!':
+        identifier = u'exclamation'
+    elif identifier == u'?':
+        identifier = u'question'
+
+    if not identifier.replace(u"-", u"").isalnum():
+        raise ValueError(identifier)
+    return identifier