Re: PageCatalog

2011-09-30 Thread Howard Lewis Ship
Well, in production mode, the "dangerous" functionality is turned off, such as running GC, or forcing the component class loader to reset. So you can see a list of page names (boring, I would hope) or you could force all pages to load (not likely to cause much trouble, since all pages will eventual

Re: PageCatalog

2011-09-30 Thread Kalle Korhonen
Agree. How about: in production mode available from localhost only by default, overridable with a symbol value to some known ip/pattern? Kalle On Fri, Sep 30, 2011 at 6:00 PM, Bob Harner wrote: > I've been playing with the latest version just now. It's very cool to > have this. Any thoughts abo

Re: PageCatalog

2011-09-30 Thread Bob Harner
I've been playing with the latest version just now. It's very cool to have this. Any thoughts about how to secure this page in production? I think it's too sensitive to leave wide open by default, and yet some of this page would be very handy to have in production. On Fri, Sep 30, 2011 at 2:53 PM,

Build failed in Jenkins: tapestry-trunk-freestyle #557

2011-09-30 Thread Apache Jenkins Server
See Changes: [hlship] TAP5-1678: Have more of the links on PageCatalog update the page listing zone [hlship] TAP5-1678: Add ability, in development mode, to force a reload of component classes [hlship] TAP5-1678: Create an i

PageCatalog

2011-09-30 Thread Howard Lewis Ship
Just checked in some further improvements to PageCatalog, including an internal Layout component for decent L&F (shared with PageCatalog and ServiceStatus). I'm finding this component very useful helping me upgrade an existing app from 5.2 to 5.3: it makes it easy to spot the kind of problems that

Re: Bug/Improvement in JavascriptSupport PartialMarkupRendererFilter : millis vs. nanos

2011-09-30 Thread Howard Lewis Ship
We should be using an AtomicLong to generate the values instead; that would ensure no duplications. On Fri, Sep 30, 2011 at 3:56 AM, Christian Zulehner wrote: > Hi, > > after changing to nanos, the problem still occured, although not very > often... > So I changed it locally to use Apache's Rand

Ability to run only one test

2011-09-30 Thread Massimo Lusetti
I'm reading through gradle's doc and see that it's possible to run only one test with: gradle -Dtest.single=FullyQualifiendClassNameOfTheTest test but that it seems to not be the case for our builds. Does anyone here is able to run a single test from gradle ? Cheers -- Massimo http://meridio.blo

Re: Bug/Improvement in JavascriptSupport PartialMarkupRendererFilter : millis vs. nanos

2011-09-30 Thread Christian Zulehner
Hi, after changing to nanos, the problem still occured, although not very often... So I changed it locally to use Apache's RandomUtils.nextLong() instead of System.nanoTime(). Anyone had the same problems or a better idea how to solve this? Although I think it's not a 100% safe implementati

Re: Which state for fixed issues?

2011-09-30 Thread Taha Hafeez
In one of the setups I am using, 'Fixed' is when the developer thinks the issue is resolved. The fix is then ACCEPTED or REJECTED by the user/developer who created the issue. In case the fix is ACCEPTED, issue gets CLOSED. On Fri, Sep 30, 2011 at 2:45 AM, Massimo Lusetti wrote: > On Thu, Sep 29,