[
https://issues.apache.org/jira/browse/CAMEL-7460?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Charles Moulliard deleted CAMEL-7460:
-------------------------------------
> Condition tested by URISupport.parseQuery() is not correct
> ----------------------------------------------------------
>
> Key: CAMEL-7460
> URL: https://issues.apache.org/jira/browse/CAMEL-7460
> Project: Camel
> Issue Type: Bug
> Reporter: Charles Moulliard
> Assignee: Charles Moulliard
>
> org.apache.camel.util.URISupport
> if (i < uri.length() - 2) {
> This instead should be:
> if (i <= uri.length() - 2) {
> Why :
> The reason is that you are trying to calculate the next character.
> So for example if we take the string "abc"
> When i = 0 we are at 'a' and we calculate next as 'b'
> But when i = 1 we are at 'b' and next should be 'c', but since the if
> statement fails the else statement runs and next gets set to '\u0000'
> "abc" is a simple example. Let's try something more "real world" now.
> The problem occurs when you have this example string:
> "flatten=false&recursive=false&delete=true&include=RAW(%5E.*%5B.%5D(xml))"
> If you try that string, you can see based on the logic when we get to the
> "closed paren" after the l in xml... next gets set to '\u0000' instead of
> ')' like it should. And then when we get to line 177
> boolean end = ch == RAW_TOKEN_END.charAt(0) && (next == '&' || next ==
> '\u0000');
> end gets set to true, when really we are not at the end. We had 1 more
> "closed paren" to process.
> Thus my last value gets set to RAW(%5E.*%5B.%5D(xml)
> When it really should be RAW(%5E.*%5B.%5D(xml))
--
This message was sent by Atlassian JIRA
(v6.2#6252)
