incubator-jspwiki-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Janne Jalkanen <>
Subject Re: Parsing fun
Date Tue, 13 Oct 2009 06:25:43 GMT

Yes, we've allowed spaces in wiki page names since 2.6.

However, we do also check - for backwards compatibility - also for the  
mashedtogethername if the wiki page with spaces does not exist. So

[Test page] => first check for the existence of "Test page" => if not  
exist, then try to find "TestPage".

There's a fairly hairy problem which is that we kind of assume that  
the comparison can be done in a case insensitive manner.  So on some  
operating systems (like Linux) [test page] might actually not match  
"Test page", but it would match "TestPage" (because mashing also  
normalizes the page name).  For 3.0 this is not going to be a problem,  
since all pages - also the ones with spaces - are normalized (or  
should be normalized; if they're not, then there's a bug.)

This is why you probably have not noticed this, since the hop to page  
names with spaces was fairly transparent.


On Oct 13, 2009, at 04:45 , Andrew Jaquith wrote:

> Ok -- I think I am actually CONFUSED. The unit tests for MarkupParser
> show a test where the output of a cleaned-link is expected to have
> spaces.
> So, which is it? Do we allow spaces in wiki page names? If so, that
> would suggest the failing JSPWikiMarkupParser test is wrong.
> Also, if spaces are allowed, how do you resolve page names? Are pages
> "Test Page" and "TestPage" the same page?
> Maybe I need to RTFM... just not sure where the M is...
> Andrew
> On Mon, Oct 12, 2009 at 7:26 PM, Andrew Jaquith
> <> wrote:
>> This is fairly amusing...
>> It turns out that one of the failing JSPWikiMarkupParser unit tests,
>> testHyperlinks3, is failing because MarkupParser.cleanLink() isn't  
>> cleaning.
>> The link markup "hyperlink too" should result in "HyperlinkToo",  
>> no? It
>> results instead in "Hyperlink too".
>> What's even more amusing is that MarkupParserTest doesn't even test  
>> for this
>> type of (common?) whitespace condition. Clearly we need to test for  
>> this --
>> and I'll fix the bug too while I'm at it. :)
>> Andrew

View raw message