kelvint 2004/01/10 05:28:33
Modified: contributions/javascript/queryValidator
luceneQueryValidator.js
Log:
An almost complete rewrite of all RE patterns.
Supports a more complete set of possible characters and also
special characters referenced mentioned in
http://jakarta.apache.org/lucene/docs/queryparsersyntax.html.
Supports full query parser syntax except for AND NOT OR, but warns user of possible
errors for these queries.
Added alertUser option to disable window.alert boxes.
Revision Changes Path
1.5 +100 -14
jakarta-lucene-sandbox/contributions/javascript/queryValidator/luceneQueryValidator.js
Index: luceneQueryValidator.js
===================================================================
RCS file:
/home/cvs/jakarta-lucene-sandbox/contributions/javascript/queryValidator/luceneQueryValidator.js,v
retrieving revision 1.4
retrieving revision 1.5
diff -u -r1.4 -r1.5
--- luceneQueryValidator.js 16 Dec 2003 02:22:02 -0000 1.4
+++ luceneQueryValidator.js 10 Jan 2004 13:28:33 -0000 1.5
@@ -1,6 +1,10 @@
// Author: Kelvin Tan ([EMAIL PROTECTED])
// JavaScript Lucene Query Validator
// Version: $Id$
+// Tested: IE 6.0.2800 and Mozilla Firebird 0.7
+
+// Special characters are + - && || ! ( ) { } [ ] ^ " ~ * ? : \
+// Special words are (case-sensitive) AND NOT OR
// Makes wildcard queries case-insensitive if true.
// Refer to http://www.mail-archive.com/[EMAIL PROTECTED]/msg00646.html
@@ -13,6 +17,13 @@
wildcardCaseInsensitive = bool;
}
+var alertUser = true;
+
+function setAlertUser(bool)
+{
+ alertUser = bool;
+}
+
// validates a lucene query.
// @param Form field that contains the query
function doCheckLuceneQuery(queryField)
@@ -20,48 +31,123 @@
var query = queryField.value;
if(query != null && query.length > 0)
{
- var matches = query.match(/^(\*)|([^a-zA-Z0-9_]\*)/);
-
+ // check for allowed characters
+ var matches = query.match(/[^a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED] ]/);
+ if(matches != null && matches.length > 0)
+ {
+ if(alertUser) alert("Invalid search query! The allowed characters are a-z A-Z
0-9. _ + - : () \" & * ? | ! {} [ ] ^ ~ \\ @ #. Please try again.")
+ return false;
+ }
+
// check wildcards are used properly
+ matches = query.match(/^[\*]*$|([\s]\*)/);
if(matches != null && matches.length > 0)
{
- alert("Invalid search query! The wildcard (*) character must be preceded by
at least one alphabet or number. Please try again.")
+ if(alertUser) alert("Invalid search query! The wildcard (*) character must be
preceded by at least one alphabet or number. Please try again.")
return false;
}
-
+
+ // check for && usage
+ // NB: doesn't handle term1 && term2 && term3 in Firebird 0.7
+ matches = query.match(/[&]{2}/);
+ if(matches != null && matches.length > 0)
+ {
+ matches = query.match(/^([a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED]( &&
)?[a-zA-Z0-9_+\-:.()\"*?|[EMAIL PROTECTED] ]*)+$/); // note missing & in pattern
+ if(matches == null)
+ {
+ if(alertUser) alert("Invalid search query! Queries containing the special
characters && must be in the form: term1 && term2. Please try again.")
+ return false;
+ }
+ }
+
+ // check ^ is used properly
+ matches = query.match(/^[^\^]*$|^([a-zA-Z0-9_+\-:.()\"*?&|[EMAIL
PROTECTED](\^[\d]+)?[ ]*)+$/); // note missing ^ in pattern
+ if(matches == null)
+ {
+ if(alertUser) alert("Invalid search query! The caret (^) character must be
preceded by alphanumeric characters and followed by numbers. Please try again.")
+ return false;
+ }
+
+ // check ~ is used properly
+ matches = query.match(/^[^~]*$|^([a-zA-Z0-9_+\-:.()\"*?&|[EMAIL
PROTECTED](~[\d.]+)?[ ]*)+$/); // note missing ~in pattern
+ if(matches == null)
+ {
+ if(alertUser) alert("Invalid search query! The tilde (~) character must be
preceded by alphanumeric characters and followed by numbers. Please try again.")
+ return false;
+ }
+
+ // check ! is used properly
+ // NB: doesn't handle term1 ! term2 ! term3
+ matches = query.match(/^[^!]*$|^([a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED]( !
)?[a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED] ]*)+$/);
+ if(matches == null || matches.length == 0)
+ {
+ if(alertUser) alert("Invalid search query! Queries containing the special
character ! must be in the form: term1 ! term2. Please try again.")
+ return false;
+ }
+
+ // check question marks are used properly
+ matches = query.match(/^(\?)|([^a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED])/);
+ if(matches != null && matches.length > 0)
+ {
+ if(alertUser) alert("Invalid search query! The question mark (?) character
must be preceded by at least one alphabet or number. Please try again.")
+ return false;
+ }
+
// check parentheses are used properly
- matches = query.match(/^([^\n()]*|(\(([a-zA-Z0-9_+\-:()\" ]|\*)+\)))*$/);
+ matches = query.match(/^[^()]*$|^(\([a-zA-Z0-9_+\-:.()\"*?&|[EMAIL
PROTECTED]))*$/);
if(matches == null || matches.length == 0)
{
- alert("Invalid search query! Parentheses must contain at least one alphabet
or number. Please try again.")
+ if(alertUser) alert("Invalid search query! Parentheses must be closed and
contain at least one alphabet or number. Please try again.")
return false;
}
// check '+' and '-' are used properly
- matches = query.match(/^(([^\n+-]*|[+-]([a-zA-Z0-9_:()]|\*)+))*$/);
+ matches = query.match(/^[^\n+\-]*$|^([+-]?[a-zA-Z0-9_+\-:.()\"*?&|[EMAIL
PROTECTED] ]?)+$/);
if(matches == null || matches.length == 0)
{
- alert("Invalid search query! '+' and '-' modifiers must be followed by at
least one alphabet or number. Please try again.")
+ if(alertUser) alert("Invalid search query! '+' and '-' modifiers must be
followed by at least one alphabet or number. Please try again.")
return false;
- }
+ }
+
+ // check AND, OR and NOT are used properly
+ matches = query.match(/AND|OR|NOT/);
+ if(matches != null && matches.length > 0)
+ {
+ // I've no idea why the code below doesn't work since it's identical to the
exclamation mark and && RE
+ //matches = query.match(/^([a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED](?: AND
)?(?: OR )?(?: NOT )?[a-zA-Z0-9_+\-:.()\"*?&|[EMAIL PROTECTED] ]*)+$/);
+
+ // we'll notify the user that this query is not validated and an error may
result
+ if(alertUser)
+ {
+ return confirm("Validation for queries containing AND/OR/NOT is currently
not supported. If your query is not well-formed, an error may result.");
+ }
+ /*
+ if(matches == null || matches.length == 0)
+ {
+ if(alertUser) alert("Invalid search query! Queries containing AND/OR/NOT
must be in the form: term1 AND|OR|NOT term2 Please try again.")
+ return false;
+ }
+ */
+ }
+
// check that quote marks are closed
matches = query.match(/\"/g);
- if(matches != null)
+ if(matches != null && matches.length > 0)
{
var number = matches.length;
if((number % 2) > 0)
{
- alert("Invalid search query! Please close all quote (\") marks.");
+ if(alertUser) alert("Invalid search query! Please close all quote (\")
marks.");
return false;
}
}
// check ':' is used properly
- matches =
query.match(/^(([^\n:]*|([a-zA-Z0-9_]|\*)+[:]([a-zA-Z0-9_()"]|\*)+))*$/);
+ matches = query.match(/^[^:]*$|^([a-zA-Z0-9_+\-.()\"*?&|[EMAIL
PROTECTED](:[a-zA-Z0-9_+\-.()\"*?&|[EMAIL PROTECTED])?[ ]*)+$/); // note missing : in
pattern
if(matches == null || matches.length == 0)
{
- alert("Invalid search query! Field declarations (:) must be preceded by at
least one alphabet or number and followed by at least one alphabet or number. Please
try again.")
+ if(alertUser) alert("Invalid search query! Field declarations (:) must be
preceded by at least one alphabet or number and followed by at least one alphabet or
number. Please try again.")
return false;
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
