Re: [Freemind-developer] 0.9.0 for testing
A premier mind-mapping software written in Java
Brought to you by:
christianfoltin,
danielpolansky
From: Christian F. (GMX) <chr...@gm...> - 2010-12-04 21:38:14
|
Hi Dan, recently, I published 0.9.0 RC12 which solves every known issue (IMHO). Even the string conversion has been moved from XSLT to a java method without memory problems. Moreover, commons-lang was removed. I hope, that you are able to test it, soon. Best regards, Chris Am 01.12.10 09:05, schrieb Dan Polansky: > Hello Chris, > > scripting: I have updated the bug report with a further comment: > > Further comment to my previous comment: It seems that the thing with > setting the security manager two times to get it removed is intentional; > there is a note to that effect at the top of FreeMindSecurityManager class. > > The key thing is that, after the recent bug fix, the security manager is > being removed twice (as it should not): once in the method "evaluate", and > once before the line "System.setOut(oldOut)". It seems that it should > suffice that the line > "securityManager.setFinalSecurityManager(scriptingSecurityManager);" before > "System.setOut(oldOut)" is commented out. > > http://sourceforge.net/tracker/index.php?func=detail&aid=2789907&group_id=7118&atid=107118 > > Best regards, > Dan > > > On Wed, Dec 1, 2010 at 12:24 AM, Dan Polansky <dan...@gm...> wrote: >> Hello Chris, >> >> I have had a look at the scripting bug, and have proposed a bug fix, >> one that you need to review, though. I have posted the fix to the bug >> report: >> >> Scripts do not work if file operations are not permitted >> http://sourceforge.net/tracker/index.php?func=detail&aid=2789907&group_id=7118&atid=107118 >> >> The last comment to the bug report, for your convenience: >> >> I get the undesirable behavior described with 0.9.0 RC11 removed by >> performing the following change in ScriptingEngine.executeScript method: >> >> // setting the same security manager the second time causes it to be >> // removed. >> //securityManager.setFinalSecurityManager(scriptingSecurityManager); // >> Commented out. --Dan >> securityManager.setFinalSecurityManager(null); // Inserted: be explicit >> about setting it to null. --Dan >> >> At that location, the security manager should be disabled; if it is not, >> it causes problems with loading of some classes later. Setting an object >> two times to disable something seems like a bug-prone idea. I am setting >> the final security manager as null, as that is the intended effect. I am >> not the author of the code, so I am not sure whether it matches the >> intention. >> >> As regards the method securityManager.setFinalSecurityManager, I do not >> understand why it is implemented the way it is: >> >> public void setFinalSecurityManager(SecurityManager >> pFinalSecurityManager) { >> if(pFinalSecurityManager == mFinalSecurityManager) { >> mFinalSecurityManager = null; >> return; >> } >> if(mFinalSecurityManager != null) { >> throw new SecurityException("There is a SecurityManager installed >> already."); >> } >> mFinalSecurityManager = pFinalSecurityManager; >> } >> >> I would implement the method as follows: >> >> public void setFinalSecurityManager(SecurityManager pFinalSecurityManager) >> { >> mFinalSecurityManager = pFinalSecurityManager; >> } >> >> The method should not worry about whether there is already a security >> manager set. When the caller wants to get the security manager removed, the >> caller should be very explicit about it, by passing "null". >> >> --Dan >> > > ------------------------------------------------------------------------------ > Increase Visibility of Your 3D Game App & Earn a Chance To Win $500! > Tap into the largest installed PC base & get more eyes on your game by > optimizing for Intel(R) Graphics Technology. Get started today with the > Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs. > http://p.sf.net/sfu/intelisp-dev2dev > _______________________________________________ > Freemind-developer mailing list > Fre...@li... > https://lists.sourceforge.net/lists/listinfo/freemind-developer > |