Hi Ercan and Steffen,
I think
http://sourceforge.net/tracker/?func=detail&aid=2706995&group_id=167540&atid=843359
is caused by the Activity Sequencer when it merges (optimizes)
TextActivites. Instead of sending
a TextActivity like: user entered A, then B, then C ... it sends the
TextActivity: user entered ABC at once.
Option 2 is a no go. If some insert a large block per C&P, the CPU load
will go mad if you
do not use the bulk method for inserting annotations into the annotation
model. The
option will also suffer from the fact that you are discarding
information when the
C&P text is larger than 20 characters.
Franz your opinion ?
BR,
Stefan
Am 14.11.2012 19:57, schrieb Steffen Pade:
Hi,
thank you for pointing us in the right direction! We believe that we
have been able to isolate the issue:
If some characters are put inbetween a larger chunk of text that has
been annotated by one large annotation (as c&p'ed text is), this large
annotation will not be split, but rather 'overridden' by the latest 20
chars. Because of this 20 annotation per user limit, those newly input
chars will start to be 'reclaimed' by that one large annotation, as soon
as the maximum amount of annotations is exceeded. We thought about two
solution approaches, we would like to discuss, one of which we favour
(the first one):
1. As it is done with another users' annotations, when something is
written
locally, split those annotations as well because of a third party's
update. We believe that in de.fu_berlin.inf.dpp.editor.replaceText(...)
the ContributionAnnotationManager's splitAnnotation()-function has to be
called before the remote annotation is inserted.
2. Instead of annotating a larger chunk of text, annotate each character
individually (this approach would probably kill performance-wise and
remove some desired functionality). Meaning only the last 20 chars of
a pasted text would be annotated.
Another approach would be
http://sourceforge.net/tracker/?func=detail&aid=2706995&group_id=167540&atid=843359,
of course. If you favor that one, however, we would like you to
elaborate, whether that behavior should apply to c&p'ed text having < 20
characters as well, etc.
Your feedback on this issue is appreciated.
Best regards,
Ercan & Steffen
On Nov 11, 2012, at 11:57 PM, Stefan Rossbach <srossb...@arcor.de
<mailto:srossb...@arcor.de>> wrote:
Hi,
the class you are looking for is:
de.fu_berlin.inf.dpp.editor.internal.ContributionAnnotationManager
Your task: (re)split the annotations and not just remove / add them.
As for performance: calculate the annotations first and use the bulk
method (just
browse through the Annotation API. It would be nice if we can improve
the performance
here (just open the task manager and you will see that CPU consumption
is going mad
when you writing / just press and hold a single key an your keyboard).
https://sourceforge.net/tracker/?func=detail&aid=3494852&group_id=167540&atid=843359
And maybe you can check if
https://sourceforge.net/tracker/?func=detail&aid=2706995&group_id=167540&atid=843359
is still an issue.
BR,
Stefan
Am 11.11.2012 19:02, schrieb Ercan Kücükkaraca:
Hi everyone,
At first, sorry for not complying with your rules, we'll do our best to
comply in the future.
Thank you very much for all your feedback on this issue. After taking
this into account, we have decided to settle for the suggested bug
#3544370 [1]
We started with the installation of a third eclipse/saros to accurately
perform the procedure as described in [1]. After successful reproduction
of the bug, we took an in-depth look into the code and the log files to
obtain a deeper understanding of the problem. So far we got the
impression, that the problem originates somewhere in package
de.fu_berlin.inf.dpp.editor. However, we have yet to narrow it down...
While trying to figure out how the annotations work, we came to the
question, where Saros removes/resets/repaints the text-highlighting in
the editor. When the MAX_HISTORY_LENGTH is reached (standard 20
annotations), the annotation is removed from history and the annotation
model, thus to our understanding for these older annotations they
shouldn't be highlighted at all, since they do not exist anymore.
We will continue working on this problem this week and would appreciate
your input greatly!
Thanks in advance for your help.
Best Regards,
Ercan & Steffen
------
[1]
http://sourceforge.net/tracker/?func=detail&aid=3544370&group_id=167540&atid=843359
On 08.11.2012 13:08, Stefan Rossbach wrote:
I think bug entry
http://sourceforge.net/tracker/?func=detail&aid=3544370&group_id=167540&atid=843359
should be fine for two people.
BR,
Stefan
Am 07.11.2012 19:37, schrieb Zieris, Franz:
Dear Steffen and Ercan,
I appreciate your enthusiasm, but please follow the conventions of
our project and write your mails in English.
To sum your mail up:
* You decided to fix bug #3572498 (see [1])
* You found a spot in the sourcecode that seems somehow related to
the topic.
* You presented your solution approach.
* You're asking for feedback.
First of all: Please register on Sourceforge and add a comment to
this bug to claim it, i.e. tell the world that you plan to work on it.
OK, now there are two more aspects:
1. The whole idea of "presenting a solution approach outline to the
community" only works if there *is* an approach to discuss. The
information you already gathered and the time you invested in writing
this mail are enough to simply try your idea yourself: If it works,
you're done and may submit your patch; if it doesn't you have
something to think about. What I'm trying to say: What you presented
does not sound like an approach, it already looks like the solution!
So what type of answer were you expecting from the community? This
leads us to point 2.
2. Since finding the corresponding spot in the code only requires
two or three full-text searches in Eclipse, and the solution might be
as simple as it looks like, this bug is not really the kind of issue
that two developers work on for more than one week. So please browse
the tracker again, in order to find a "medium-sized" problem.
Please don't get me wrong! I have no objections to fixing even this
bugs (I even created this tracker item, as you might have noticed),
but I doubt this one is appropriate for the purpose of your
university course. Maybe fixing a bunch of tiny bugs is also
acceptable, please talk to your tutor, if you insist to solve this
bug.
Alternative bugs, that are probably medium-sized:
* #3514625 (see [2])
* #3573784 (see [3])
* #3528461 (see [4], quite unimportant)
I'm pretty sure that there are many more. Don't bother to ask
something like "I looked at bug #XXXXXXX, browsed the sourcecode for
at least 1 hour and I think this one is not trivial, but also small
enough to handle it -- do you agree?" on this mailing list.
Best Regards,
Franz
Links:
[1]
http://sourceforge.net/tracker/?func=detail&aid=3572498&group_id=167540&atid=843359
[2]
http://sourceforge.net/tracker/?func=detail&group_id=167540&atid=843359&aid=3514625
[3]
http://sourceforge.net/tracker/?func=detail&group_id=167540&atid=843359&aid=3573784
[4]
http://sourceforge.net/tracker/?func=detail&group_id=167540&atid=843359&aid=3528461
PS: Whenever you refer to content that is available online, please
provide a URL and don't copy the content itself (copying might be OK
if you refer to specific portion of the content). This holds
especially if the content you're talking about might change over time.
-----Original Message-----
From: Ercan Kücükkaraca [mailto:ercan.kuecuekkar...@fu-berlin.de]
Sent: Wednesday, November 07, 2012 4:01 PM
To: dpp-devel@lists.sourceforge.net
Subject: [DPP-Devel] Bug 3572498
Liebe Saros-Community,
im Rahmen der Veranstaltung Softwareprozesse an der FU Berlin haben
wir die Aufgabe, einen geeigneten Bug eines OpenSource Projektes zu
beheben.
Wir (Steffen Pade & Ercan Kücükkaraca) haben uns wegen seiner
Großartigkeit für Saros entschieden. Nach ausgedehnter Suche im
Saros-Bugtracker haben wir uns für die Bearbeitung des Bugs mit der ID
3572498 entschieden. Der Wortlaut ist wie folgt:
Deactivate "Add to Saros Session" for read-only users - ID: 3572498
Last Update: Franz Zieris ( franzzieris ) - 2012-09-27 12:16:27 PDT
Priority: 3
Status: Open
Resolution: None
Assigned: Nobody/Anonymous
Category: User Experience
This results from the fix of #3455399. Since this fix, read-only
users are no longer able to add new projects to a running session.
But the corresponding entry in the context menu is still active. For
a better user experience this entry should be deactivated for
read-only users.
Nachdem wir versucht haben, uns einen groben Überblick über die
Projektsourcen geschafft haben, haben wir uns für folgenden
Lösungsansatz entschlossen, den wir gerne zur Diskussion stellen
würden:
An geeigneter Stelle in der plugin.xml wird eine Abfrage
eingerichtet, die kontrolliert, ob der User WriteAccess hat oder
nicht. Genauer müsste
<reference
definitionId="de.fu_berlin.inf.dpp.ui.definitions.participantsHasWriteAccess">
</reference>
eingefügt werden.
Über eine kurzfristige Rückmeldung würden wir uns freuen!
Beste Grüße
Steffen & Ercan
------------------------------------------------------------------------------
LogMeIn Central: Instant, anywhere, Remote PC access and management.
Stay in control, update software, and manage PCs from one command
center Diagnose problems and improve visibility into emerging IT
issues Automate, monitor and manage. Do more in less time with
Central http://p.sf.net/sfu/logmein12331_d2d
_______________________________________________
DPP-Devel mailing list
DPP-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dpp-devel
------------------------------------------------------------------------------
LogMeIn Central: Instant, anywhere, Remote PC access and management.
Stay in control, update software, and manage PCs from one command
center
Diagnose problems and improve visibility into emerging IT issues
Automate, monitor and manage. Do more in less time with Central
http://p.sf.net/sfu/logmein12331_d2d
_______________________________________________
DPP-Devel mailing list
DPP-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dpp-devel
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_nov
_______________________________________________
DPP-Devel mailing list
DPP-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dpp-devel
------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov
_______________________________________________
DPP-Devel mailing list
DPP-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/dpp-devel
