On Aug 30, 2010, at 5:15 PM, abusenius (SVN) wrote:
> Author: abusenius
> Date: 2010-08-30 17:15:09 +0200 (Mon, 30 Aug 2010)
> New Revision: 30796
>
> Modified:
> platform/web/trunk/standard/src/main/webapp/WEB-INF/web.xml
> Log:
> Updated SetCharacterEncodingFilter class in web.xml after moving
>
> Modified: platform/web/trunk/standard/src/main/webapp/WEB-INF/web.xml
> ===================================================================
> --- platform/web/trunk/standard/src/main/webapp/WEB-INF/web.xml 2010-08-30 14:31:45 UTC (rev 30795)
> +++ platform/web/trunk/standard/src/main/webapp/WEB-INF/web.xml 2010-08-30 15:15:09 UTC (rev 30796)
> @@ -22,7 +22,7 @@
> without affecting the whole container (and the other applications hosted). -->
> <filter>
> <filter-name>Set Character Encoding</filter-name>
> - <filter-class>com.xpn.xwiki.web.SetCharacterEncodingFilter</filter-class>
> + <filter-class>org.xwiki.container.servlet.filters.internal.SetCharacterEncodingFilter</filter-class>
This is going to cause problems to users when they upgrade...
We need to make it as easy as possible:
- deprecate the old class but keep it
- add what needs to be done in the release notes.
WDYT?
Thanks
-Vincent
> <!-- The encoding to use. This must be the same as the one in xwiki.cfg (hopefully only one
> encoding will be used later). -->
> <init-param>
Cool to have a test for this, thanks Alex.
However I think that in the near future we need to find a way to write these type of tests as unit tests instead of functional tests since otherwise we're going to have too many functional tests, which will lead to the following problems:
- take a lot of time to execute (and hence are not executed often + need more machines, etc)
- are hard to debug when they fail (since they're not unit tests)
- do not make us improve the design of the code
WDYT?
Thanks
-Vincent
PS: I haven't looked at this example in particular and my comment is general. Maybe this is testing old code which would need some heavy refactoring before we can write a unit test for it.
On Aug 27, 2010, at 11:33 AM, abusenius (SVN) wrote:
> Author: abusenius
> Date: 2010-08-27 11:33:57 +0200 (Fri, 27 Aug 2010)
> New Revision: 30768
>
> Modified:
> enterprise/trunk/distribution-test/ui-tests/src/test/it/org/xwiki/it/ui/LoginTest.java
> Log:
> XWIKI-5434: Added a functional test
>
> Modified: enterprise/trunk/distribution-test/ui-tests/src/test/it/org/xwiki/it/ui/LoginTest.java
> ===================================================================
> --- enterprise/trunk/distribution-test/ui-tests/src/test/it/org/xwiki/it/ui/LoginTest.java 2010-08-27 09:30:38 UTC (rev 30767)
> +++ enterprise/trunk/distribution-test/ui-tests/src/test/it/org/xwiki/it/ui/LoginTest.java 2010-08-27 09:33:57 UTC (rev 30768)
> @@ -26,6 +26,7 @@
> import org.xwiki.it.ui.administration.elements.GlobalRightsPage;
> import org.xwiki.it.ui.framework.AbstractTest;
> import org.xwiki.it.ui.framework.elements.LoginPage;
> +import org.xwiki.it.ui.framework.elements.editor.WikiEditPage;
> import org.xwiki.it.ui.xe.elements.HomePage;
>
> /**
> @@ -114,4 +115,31 @@
> rights.unforceAuthenticatedView();
> }
> }
> +
> + @Test
> + public void testRedirectPreservesPOSTParameters() throws InterruptedException
> + {
> + String test = "Test string " + System.currentTimeMillis();
> + final String space = "Main";
> + final String page = "POSTTest";
> + LoginPage loginPage = this.homePage.clickLogin();
> + loginPage.loginAsAdmin();
> + // start editing a page
> + WikiEditPage editPage = new WikiEditPage();
> + editPage.switchToEdit(space, page);
> + editPage.setTitle(test);
> + editPage.setContent(test);
> + // emulate expired session: delete the cookies
> + getDriver().manage().deleteAllCookies();
> + // try to save
> + editPage.clickSaveAndView();
> + // we should have been redirected to login
> + Assert.assertTrue(getDriver().getCurrentUrl().startsWith(getUtil().getURL("XWiki", "XWikiLogin", "login")));
> + loginPage.loginAsAdmin();
> + // we should have been redirected back to view, and the page should have been saved
> + Assert.assertTrue(getDriver().getCurrentUrl().startsWith(getUtil().getURL(space, page)));
> + editPage.switchToEdit(space, page);
> + Assert.assertEquals(test, editPage.getTitle());
> + Assert.assertEquals(test, editPage.getContent());
> + }
> }
>
> _______________________________________________
> notifications mailing list
> notifications(a)xwiki.org
> http://lists.xwiki.org/mailman/listinfo/notifications
Hi devs,
I would like to contribute on svn an application i have developed that logs
messages from velocity and groovy code blocks. It uses XWiki objects and
does not require changes in XE installation to work. Only requires PR rights
for groovy scripts to work.
Currently the application can be found at
http://flavius.myxwiki.org/xwiki/bin/view/Log/.
I only require commit rights to contrib sandbox to get started.
SVN username: flavius
Thanks,
--
XWiki SAS
Web Developer
Flavius Olaru
I still strongly believe that user photos should appear in the stream before any click.
Especially for user updates instead of the "user.gif" icon.
Thanks
Jerome.
----- "Ecaterina Valica" <valicac(a)gmail.com> wrote:
> Added:
> - Application + User Changes
> http://incubator.myxwiki.org/xwiki/bin/download/Improvements/ActivityStream…
>
> - Feed + Filters
> http://incubator.myxwiki.org/xwiki/bin/download/Improvements/ActivityStream…
>
> Prototype:
> http://incubator.myxwiki.org/xwiki/bin/view/Improvements/ActivityStreamP3D
>
> Thanks,
> Caty
>
>
> On Tue, Aug 31, 2010 at 11:35, Fabio Mancinelli
> <fabio.mancinelli(a)xwiki.com>wrote:
>
> > On 08/30/2010 07:16 PM, Jerome Velociter wrote:
> >
> > >> The "user centric" view, imho, is also useful because a "user"
> could
> > >> also be an application that sends events (e.g., meeting or
> deadlines
> > >> reminders, etc.) that can be shown in a more natural way in a
> user
> > >> centric view. Basically even applications might share their
> thoughts
> > >> through the activity stream ;)
> > >
> > > I completely agree with the last sentence. BUT I think it's not
> > incompatible with a document/user mixed-approched (I don't think
> either full
> > document-centric or full user-centric can embrace the versatility of
> XWiki)
> > >
> > > I see two types of entries :
> > >
> > > * Document updates. Basically what is currently in latest
> proposal
> > (though I would add little user pictures here as well, even before
> the
> > click). This is KM, collaboration, etc.
> > > * Users/applications updates. Those are less structured, more
> a-la
> > facebook.
> > >
> > > I see something like
> >
> http://incubator.myxwiki.org/xwiki/bin/view/Improvements/ActivityStreamMixe…
> picture, would have to be refined)
> > > (Note I also have right-padded a bit more the
> > comment/annotation/pencil/etc icons that are on top of user photos,
> I think
> > they were a bit too masked by them)
> > >
> > I would have called that version "user centric" as well since
> documents
> > updates might be seen as posts done by the "XWiki system" user :)
> >
> > Anyway, I think that the mixed approach is fine.
> >
> > -Fabio
> > _______________________________________________
> > devs mailing list
> > devs(a)xwiki.org
> > http://lists.xwiki.org/mailman/listinfo/devs
> >
> _______________________________________________
> devs mailing list
> devs(a)xwiki.org
> http://lists.xwiki.org/mailman/listinfo/devs
Hu guys,
there is any way to supress content parts from the pdf report?
Here is the problem:
Most of pages have a toc macro with this code
{{box cssClass="floatinginfobox"}}
{{toc depth="3"/}}
{{/box}}
As we know, the pdf report also puts a toc page.
My reports are getting two tocs, one from report and one from content.
But i do not want to remove de pdf toc because some wiki pages do not have
the content toc.
There is a way to supress the content toc from appearing in the pdf report?
thanks
--
L. Marcelo Serique
Hi guys,
Short story:
1/ glorify macro categories and use them for more than presentational
purposes (e.g. using a category to differentiate "gadgets" macros from
other macros)
2/ allow a macro to be part of multiple categories at the same time
WDYT?
Long story:
I have read http://markmail.org/thread/wwv56pojo6rix5zv about the
current implementation for macro categories and I noticed that the
discussion / approach there focuses on the presentational use of
categories for macros, how to display them grouped to the user. I would
say there's an important additional usecase, the usage of categories as
macros metadata, describing their semantic, to allow some apps to
implement different behaviour for macros in a category or another. For
example, in the case of gadgets / dashboard, if a gadget would be
implemented as a macro, then we'd need some sort of method to
differentiate the macros that can be used as gadgets. The most
'semantic' way would be to use a gadgets category to mark the gadgets,
and the dashboard implementation will allow only macros from this
category to be added in the dashboard. However this approach would mean
that macro category gains importance, and we need to think if it's still
ok that an admin can change the category of a macro and the macro
category declared by the author can be completely overwritten.
WDYT about using the macro categories for much more than grouping for
presentation?
I'm +1, and I think that ftm there's no need to strategy change wrt to
who establishes the category of a macro.
Also, because of this, it's very possible that we might need a macro to
be part of more than one category, because, to continue the example, we
might want gadgets to also be grouped in several categories (of
gadgets), or a gadget to also be included in the, say, "presentation"
category of macros to be used in plain xwiki documents.
WDYT?
I am +1 for this as well, and ready to start building the patch (modulo
some macros API changes) as soon as we all agree.
Thanks,
Anca
Hello,
I have been working with Alex on security related code and have been impressed with his abilities as well as
dedication to consistent improvement of the project and his own code.
Alex has worked on some complex changes to difficult code such as:
XWIKI-5275: Prevent nested script macros.
http://jira.xwiki.org/jira/browse/XWIKI-5275
As well as a large quantity of cross site script related patches such as:
XWIKI-5161: Using XML symbols (<, >, &, ") inside the document title/name/space breaks various parts of the UI and causes the PDF export to throw exceptions
http://jira.xwiki.org/jira/browse/XWIKI-5161
A full list is here, most are not viewable by the public.
http://jira.xwiki.org/jira/secure/ViewProfile.jspa?name=nickless#selectedTa…
Alex has developed an automated cross site script fuzzing test which can be seen here:
http://svn.xwiki.org/svnroot/xwiki/contrib/sandbox/xwiki-enterprise-test-es…
He is also responsible for parts of xwiki-crypto including PKCS7 signing and validation and X509 certificate handling.
see:
http://svn.xwiki.org/svnroot/xwiki/platform/core/trunk/xwiki-crypto/src/mai…http://svn.xwiki.org/svnroot/xwiki/platform/core/trunk/xwiki-crypto/src/mai…http://svn.xwiki.org/svnroot/xwiki/platform/core/trunk/xwiki-crypto/src/mai…
As far as write access to the main repository, I think there is no question that he will be a good committer.
It should be noted that I would be +1 for giving write access to anyone who demonstrates a need.
The remaining issue is the responsibility of being on the "steering committee", having the power to -1 a vote.
Alex and I agree on a lot of issues where we hold the minority position so my opinion should be taken with the salt it deserves.
That said, he has the courage to voice disagreement when he has it,
he takes an interest in decisions regarding the core rather than being concentrated on the user interface,
and he has the ability to spot problems in design which will not become obvious until later.
I think Alex Busenius will be a great committer.
+1
Caleb