From 434b5d44b07fa6b5dbac35c72ee222c4db7693de Mon Sep 17 00:00:00 2001
From: stroeder <stroeder>
Date: Wed, 29 Apr 2009 18:13:55 +0000
Subject: [PATCH] Fixed ldap.schema.tokenizer.split_tokens() to accept a single
 DOLLAR as separator

---
 Lib/ldap/schema/tokenizer.py            | 85 +++++++++++++++++++++++++
 Tests/Lib/ldap/schema/test_tokenizer.py | 30 +++++++++
 2 files changed, 115 insertions(+)
 create mode 100644 Lib/ldap/schema/tokenizer.py
 create mode 100644 Tests/Lib/ldap/schema/test_tokenizer.py

diff --git a/Lib/ldap/schema/tokenizer.py b/Lib/ldap/schema/tokenizer.py
new file mode 100644
index 0000000..469c326
--- /dev/null
+++ b/Lib/ldap/schema/tokenizer.py
@@ -0,0 +1,85 @@
+"""
+ldap.schema.tokenizer - Low-level parsing functions for schema element strings
+
+See http://www.python-ldap.org/ for details.
+
+\$Id: tokenizer.py,v 1.13 2009/04/29 18:13:55 stroeder Exp $
+"""
+
+
+def split_tokens(s,keywordDict):
+  """
+  Returns list of syntax elements with quotes and spaces
+  stripped.
+  """
+  result = []
+  result_append = result.append
+  s_len = len(s)
+  i = 0
+  while i<s_len:
+    start = i
+    while i<s_len and s[i]!="'":
+      if s[i]=="(" or s[i]==")":
+        if i>start:
+          result_append(s[start:i])
+        result_append(s[i])
+        i +=1 # Consume parentheses
+        start = i
+      elif s[i]==" " or s[i]=="$":
+        if i>start:
+          result_append(s[start:i])
+        i +=1
+        # Consume more space chars
+        while i<s_len and s[i]==" ":
+          i +=1
+        start = i
+      else:
+        i +=1
+    if i>start:
+      result_append(s[start:i])
+    i +=1
+    if i>=s_len:
+      break
+    start = i
+    while i<s_len and s[i]!="'":
+      i +=1
+    if i>=start:
+      result_append(s[start:i])
+    i +=1
+  return result # split_tokens()
+
+
+def extract_tokens(l,known_tokens):
+  """
+  Returns dictionary of known tokens with all values
+  """
+  assert l[0].strip()=="(" and l[-1].strip()==")",ValueError(l)
+  result = {}
+  result_has_key = result.has_key
+  result.update(known_tokens)
+  i = 0
+  l_len = len(l)
+  while i<l_len:
+    if result_has_key(l[i]):
+      token = l[i]
+      i += 1 # Consume token
+      if i<l_len:
+        if result_has_key(l[i]):
+          # non-valued
+          result[token] = (())
+        elif l[i]=="(":
+          # multi-valued
+          i += 1 # Consume left parentheses
+          start = i
+          while i<l_len and l[i]!=")":
+            i += 1
+          result[token] = tuple(filter(lambda v:v!='$',l[start:i]))
+          i += 1 # Consume right parentheses
+        else:
+          # single-valued
+          result[token] = l[i],
+          i += 1 # Consume single value
+    else:
+      i += 1 # Consume unrecognized item
+  return result
+
diff --git a/Tests/Lib/ldap/schema/test_tokenizer.py b/Tests/Lib/ldap/schema/test_tokenizer.py
new file mode 100644
index 0000000..8f66f8a
--- /dev/null
+++ b/Tests/Lib/ldap/schema/test_tokenizer.py
@@ -0,0 +1,30 @@
+import ldap.schema
+from ldap.schema.tokenizer import split_tokens,extract_tokens
+
+testcases_split_tokens = (
+  (" BLUBBER DI BLUBB ", ["BLUBBER", "DI", "BLUBB"]),
+  ("BLUBBER DI BLUBB",["BLUBBER","DI","BLUBB"]),
+  ("BLUBBER  DI   BLUBB  ",["BLUBBER","DI","BLUBB"]),
+  ("BLUBBER  DI  'BLUBB'   ",["BLUBBER","DI","BLUBB"]),
+  ("BLUBBER ( DI ) 'BLUBB'   ",["BLUBBER","(","DI",")","BLUBB"]),
+  ("BLUBBER(DI)",["BLUBBER","(","DI",")"]),
+  ("BLUBBER ( DI)",["BLUBBER","(","DI",")"]),
+  ("BLUBBER ''",["BLUBBER",""]),
+  ("( BLUBBER (DI 'BLUBB'))",["(","BLUBBER","(","DI","BLUBB",")",")"]),
+  ("BLUBB (DA$BLAH)",['BLUBB',"(","DA","BLAH",")"]),
+  ("BLUBB ( DA $  BLAH )",['BLUBB',"(","DA","BLAH",")"]),
+  ("BLUBB (DA$ BLAH)",['BLUBB',"(","DA","BLAH",")"]),
+  ("BLUBB (DA $BLAH)",['BLUBB',"(","DA","BLAH",")"]),
+  ("BLUBB 'DA$BLAH'",['BLUBB',"DA$BLAH"]),
+  ("BLUBB DI 'BLU B B ER' DA 'BLAH' ",['BLUBB','DI','BLU B B ER','DA','BLAH']),
+  ("BLUBB DI 'BLU B B ER' DA 'BLAH' LABER",['BLUBB','DI','BLU B B ER','DA','BLAH','LABER']),
+  ("BLUBBER DI 'BLU'BB ER' DA 'BLAH' ", ["BLUBBER", "DI", "BLU'BB ER", "DA", "BLAH"]), # for Oracle
+  ("BLUBB DI 'BLU B B ER'MUST 'BLAH' ",['BLUBB','DI','BLU B B ER','MUST','BLAH']) # for Oracle
+)
+
+for t,r in testcases_split_tokens:
+  l = ldap.schema.tokenizer.split_tokens(t,{'MUST':None})
+  if l!=r:
+    print 'String:',repr(t)
+    print '=>',l
+    print 'differs from',r