[
https://issues.apache.org/jira/browse/JCR-2732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13016293#comment-13016293
]
Benjamin Papez edited comment on JCR-2732 at 4/6/11 8:01 AM:
-------------------------------------------------------------
I submitted a patch to fix the exception. The problem in the code is that in
case of '~' and ':' the escaped flag is not propertly handled. So if the
predecesessor of these two characters is the escaping backslash setting escaped
to true, the successor character after '~' or ':' will be escaped, which is not
intended.
was (Author: bpapez):
Index: JackrabbitQueryParser.java
===================================================================
--- JackrabbitQueryParser.java (revision 1089329)
+++ JackrabbitQueryParser.java (working copy)
@@ -87,10 +87,17 @@
// escape tilde so we can use it for similarity query
rewritten.append("\\");
}
+ if (escaped) {
+ rewritten.append('\\');
+ escaped = false;
+ }
rewritten.append('~');
} else if (textsearch.charAt(i) == ':') {
// fields as known in lucene are not supported
rewritten.append("\\:");
+ if (escaped) {
+ escaped = false;
+ }
} else {
if (escaped) {
rewritten.append('\\');
> ParseException in xpath query using an escaped string in jackrabbit 2.x
> (works in 1.6)
> --------------------------------------------------------------------------------------
>
> Key: JCR-2732
> URL: https://issues.apache.org/jira/browse/JCR-2732
> Project: Jackrabbit Content Repository
> Issue Type: Bug
> Components: query, xpath
> Affects Versions: 2.2.4
> Reporter: fabrizio giustina
> Attachments: ColonBracketSearchTest.java, JackrabbitQueryParser.patch
>
>
> For a particular sequence of chars, ":)" os ":(", jackrabbit 2.x seems to
> break also when the xpath statement is properly escaped
> Looks like the way I escape the ":)" os ":(" sequence was used to work fine
> in jackrabbit 1.6, but produces a parsing error in jackrabbit 2.x.
> The following query, with a space in between ":" and ")" works fine in any
> version of jackrabbit:
> {code}
> //*[jcr:contains(@title, '\: \)')]
> {code}
> This one, without any space, works only in jackrabbit 1.6:
> {code}
> //*[jcr:contains(@title, '\:\)')]
> {code}
> in 2.x the result is a ParseException: Cannot parse '\:\\)': Encountered "
> ")" ")
> Is anything changed in how xpath queries must be escaped in 2.x or Is this a
> bug?
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
