[prev in list] [next in list] [prev in thread] [next in thread] 

List:       jedit-devel
Subject:    [ jEdit-devel ] [ jedit-Merge Requests-2984022 ] fix blocking
From:       "SourceForge.net" <noreply () sourceforge ! net>
Date:       2010-08-26 9:37:21
Message-ID: E1OoYtd-0007Gm-H7 () sfs-web-11 ! v29 ! ch3 ! sourceforge ! com
[Download RAW message or body]

Merge Requests item #2984022, was opened at 2010-04-08 19:28
Message generated for change (Comment added) made by kerik-sf
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=1235750&aid=2984022&group_id=588

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: for 4.3.x
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Eric Le Lay (kerik-sf)
Assigned to: Kazutoshi Satoda (k_satoda)
Summary: fix blocking HelpViewer

Initial Comment:
r17544
http://jedit.svn.sourceforge.net/viewvc/jedit?view=rev&revision=17544
r17546
http://jedit.svn.sourceforge.net/viewvc/jedit?view=rev&revision=17546
r17549
http://jedit.svn.sourceforge.net/viewvc/jedit?view=rev&revision=17549

see bug #1231914 - jEdit not responsible when waiting for unavailable input
https://sourceforge.net/tracker/index.php?func=detail&aid=1231914&group_id=588&atid=100588


r17544 targets slow throughput websites
r17549 targets unaccessible websites

I can only confirm that r17549 works for sure : I tried it when disabling the proxy, \
behind a corporate firewall (network timeout after a LONG time). r17544 seems to make \
page load slower on average, but the UI seems more responsive.


----------------------------------------------------------------------

> Comment By: Eric Le Lay (kerik-sf)
Date: 2010-08-26 11:37

Message:
I call everything in the EDT after setPage() and updated the comment in
r18419
http://jedit.svn.sourceforge.net/jedit/?rev=18419&view=rev
So could you please merge r17549, r17753, r18419 ?

----------------------------------------------------------------------

Comment By: Eric Le Lay (kerik-sf)
Date: 2010-07-19 21:50

Message:
shlomy committed an fix in r18207
http://jedit.svn.sourceforge.net/jedit/?rev=18207&view=rev
I'll update the comment this week.

----------------------------------------------------------------------

Comment By: Kazutoshi Satoda (k_satoda)
Date: 2010-05-08 10:11

Message:
Thanks for the refinement. Two concerns remain in my mind.

(1)
Is it really only NPE that is caused by this violation of
"The single-thread rule"?

The NPE mentioned in the JVM bug 4714674 seems local to NetBeans which
does doc.putProperty(Document.StreamDescriptionProperty, null) as a
workaround for their own problem.

If you are not sure about the actual problem which will happen, I
propose the following change of wording:
- Calling setPage outside from the EDT may lead to NPE :
+ Calling setPage outside from the EDT violates
+ "The single-tread rule" of Swing, but it's an
+ experienced workaround:

(2)
The comment justifies only the call of setPage(). How about other calls?

I think the application of invokeLater(), which now surrounds only
setValue(scrollPosition), should be widen for all the rest of operations
after setPage().

----------------------------------------------------------------------

Comment By: Eric Le Lay (kerik-sf)
Date: 2010-05-07 09:39

Message:
Good catch, but the whole point of the change is to call setPage() outside
of the EDT...
I added a note as you requested in r17753

----------------------------------------------------------------------

Comment By: Kazutoshi Satoda (k_satoda)
Date: 2010-05-05 12:52

Message:
Thank you for testing again. Now I reviewed the mix of all revisions;
r17544,17546,17549,17725. The behavior seems OK as before.

But now, I'm afraid of violation of "The single-thread rule" of Swing.
http://java.sun.com/products/jfc/tsc/articles/threads/threads1.html#single_thread_rule
 Sorry for not noticing this at the first review.

In the newly introduced thread, some methods are called and only one is
wrapped in SwingUtilities.invokeLater(). But based on the above rule,
I think most (all?) of them should be wrapped, shouldn't they?

I found that running JEditorPane#setPage() in non-AWT thread is used in
NetBeans was a workaround of the following problem.
http://bugs.sun.com/view_bug.do?bug_id=4714674
... And unfortunately, it says "Release Fixed: 7(b27)" which means the
fix is not available in Java 5 and 6. If we decide to call setPage() as
proposed here, I think a comment about the above problem should be put
there.

----------------------------------------------------------------------

Comment By: Eric Le Lay (kerik-sf)
Date: 2010-05-03 09:21

Message:
I reverted r17544 & r17546 to trunk in r17725.
Could you still apply r17549?
Thanks,

----------------------------------------------------------------------

Comment By: Eric Le Lay (kerik-sf)
Date: 2010-04-26 21:11

Message:
You seem to be right.
I inserted a link in some user guide to http://guide.macports.org, which
is quite a big HTML page and throttled my connection using ipfw :
sudo ipfw pipe 1 config bw 200Byte/s
sudo ipfw add 1 pipe 1 dst-port 80

The jEdit UI was completely frozen until the whole page was loaded.
I'm still experimenting with it. I'll let you know if I find a workaround.

----------------------------------------------------------------------

Comment By: Kazutoshi Satoda (k_satoda)
Date: 2010-04-24 19:20

Message:
It seems that r17549 is enough, and r17544 doesn't make any change over
r17549. Do you have any test case that r17544 makes difference?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=1235750&aid=2984022&group_id=588

------------------------------------------------------------------------------
Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
Be part of this innovative community and reach millions of netbook users 
worldwide. Take advantage of special opportunities to increase revenue and 
speed time-to-market. Join now, and jumpstart your future.
http://p.sf.net/sfu/intel-atom-d2d
-- 
-----------------------------------------------
jEdit Developers' List
jEdit-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jedit-devel


[prev in list] [next in list] [prev in thread] [next in thread]