[
https://issues.apache.org/jira/browse/WICKET-4008?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15508046#comment-15508046
]
Pedro Santos commented on WICKET-4008:
--------------------------------------
Hi [~mgrigorov],
I agree that shouldParsePropertyExpressionsWithSpaceInMethod should pass and
that the exception message when parsing "per son" was misleading, thanks.
I added the failing test shouldParsePropertyExpressionStartingWithDigits
because one can have getter/setters for a property starting with a digit even
though it isn't a valid java identifier. My idea is to break the "property" in
the grammar into multiple production rules. So: "bean.1property" will be a
valid expression having a "bean property" but "bean.1property()" will fail.
Proposed grammar:
bean property = property name, [ index ]
java property = java identifier , [ index | method sign ]
map property = index
property expression = [ bean property | java property | map property ], {
"." , property expression }
> Improve PropertyResolver to use a parser
> ----------------------------------------
>
> Key: WICKET-4008
> URL: https://issues.apache.org/jira/browse/WICKET-4008
> Project: Wicket
> Issue Type: Improvement
> Components: wicket
> Affects Versions: 6.0.0-beta1
> Reporter: Pedro Santos
> Assignee: Pedro Santos
> Priority: Trivial
> Attachments: WICKET-4004.patch
>
>
> The usage of a parser is a improvement since it will be able to feedback
> erroneous property expressions with meaningful exception messages.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
