Merge pull request #9 from IdanHaim/RavenDB-6452

RavenDB-6452

pyravendb/store/session_query.py

@@ -5,6 +5,14 @@
 from datetime import timedelta
 import sys
 import time
+import re
+
+
+class EscapeQueryOptions(Enum):
+    EscapeAll = 0
+    AllowPostfixWildcard = 1
+    AllowAllWildcards = 2
+    RawQuery = 3
 
 
 class Query(object):
@@ -56,7 +64,20 @@ def select(self, *args):
             self.fetch = args
         return self
 
-    def _lucene_builder(self, value, action=None):
+    def _lucene_builder(self, value, action=None, escape_query_options=EscapeQueryOptions.EscapeAll):
+
+        if isinstance(value, str):
+            if escape_query_options == EscapeQueryOptions.EscapeAll:
+                value = Utils.escape(value, False, False)
+
+            elif escape_query_options == EscapeQueryOptions.AllowPostfixWildcard:
+                value = Utils.escape(value, False, False)
+            elif escape_query_options == EscapeQueryOptions.AllowAllWildcards:
+                value = Utils.escape(value, False, False)
+                value = re.sub(r'"\\\*(\s|$)"', "*${1}", value)
+            elif escape_query_options == EscapeQueryOptions.RawQuery:
+                value = Utils.escape(value, False, False).replace("\\*", "*")
+
         lucene_text = Utils.to_lucene(value, action=action)
 
         if len(self.query_builder) > 0 and not self.query_builder.endswith(' '):
@@ -70,13 +91,15 @@ def _lucene_builder(self, value, action=None):
     def __iter__(self):
         return self._execute_query().__iter__()
 
-    def where_equals(self, field_name, value):
+    def where_equals(self, field_name, value, escape_query_options=EscapeQueryOptions.EscapeAll):
         """
         To get all the document that equal to the value in the given field_name
 
         @param field_name:The field name in the index you want to query.
         :type str
         @param value: The value will be the fields value you want to query
+        @param escape_query_options: the way we should escape special characters
+        :type EscapeQueryOptions
         """
         if field_name is None:
             raise ValueError("None field_name is invalid")
@@ -89,7 +112,7 @@ def where_equals(self, field_name, value):
                         sort_hint = self.session.conventions.get_default_sort_option("long")
                 self._sort_hints.add("SortHint-{0}={1}".format(field_name, sort_hint))
 
-        lucene_text = self._lucene_builder(value, action="equal")
+        lucene_text = self._lucene_builder(value, action="equal", escape_query_options=escape_query_options)
         self.query_builder += "{0}:{1}".format(field_name, lucene_text)
         return self
 
@@ -108,17 +131,19 @@ def where(self, **kwargs):
                 self.where_equals(field_name, kwargs[field_name])
         return self
 
-    def search(self, field_name, search_terms):
+    def search(self, field_name, search_terms, escape_query_options=EscapeQueryOptions.RawQuery):
         """
         for more complex text searching
 
         @param field_name:The field name in the index you want to query.
         :type str
         @param search_terms: the terms you want to query
         :type str
+        @param escape_query_options: the way we should escape special characters
+        :type EscapeQueryOptions
         """
         search_terms = Utils.quote_key(str(search_terms))
-        search_terms = self._lucene_builder(search_terms, "search")
+        search_terms = self._lucene_builder(search_terms, "search", escape_query_options)
         self.query_builder += "{0}:{1}".format(field_name, search_terms)
         return self
 

pyravendb/tools/utils.py

@@ -166,8 +166,6 @@ def make_initialize_dict(document, entity_init):
     @staticmethod
     def to_lucene(value, action):
         query_text = ""
-        if isinstance(value, str):
-            value = re.escape(value).replace('\*', '*')
         if action == "in":
             if not value or len(value) == 0:
                 return None
@@ -281,3 +279,41 @@ def timedelta_to_str(timedelta_obj):
             if microseconds > 0:
                 timedelta_str += ".{0}".format(microseconds)
         return timedelta_str
+
+    @staticmethod
+    def escape(term, allow_wild_cards, make_phrase):
+        wild_cards = ['-', '&', '|', '!', '(', ')', '{', '}', '[', ']', '^', '"', '~', ':', '\\']
+        if not term:
+            return "\"\""
+        start = 0
+        length = len(term)
+        builder = ""
+        if length >= 2 and term[0] == '/' and term[1] == '/':
+            builder += "//"
+            start = 2
+        i = start
+        while i < length:
+            ch = term[i]
+            if ch == '*' or ch == '?':
+                if allow_wild_cards:
+                    i += 1
+                    continue
+
+            if ch in wild_cards:
+                if i > start:
+                    builder += term[start:i - start]
+
+                builder += '\\{0}'.format(ch)
+                start = i + 1
+                i += 1
+                continue
+
+            if ch == ' ' or ch == '\t':
+                if make_phrase:
+                    return "\"{0}\"".format(Utils.escape(term, allow_wild_cards, False))
+
+            i += 1
+        if length > start:
+            builder += term[start: length]
+
+        return builder

setup.py

@@ -3,7 +3,7 @@
 setup(
     name='pyravendb',
     packages=find_packages(),
-    version='1.3.1.2',
+    version='1.3.1.3',
     description='This is the official python client for RavenDB document database',
     author='Idan Haim Shalom',
     author_email='[email protected]',