Hopefully fixed.
At least I cannot experience this on in-CVS code under both
JDK 1.4.2 and JDK 1.5/5.0 on a pretty large project I got
these errors before.
I'd like to ask you, Henry and JC, to update from CVS and
work on a project you experienced the problems with.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've been working with this fix since 8pm last night (yes, I
did sleep in between), and perceived no problems. I'd say
this and related issues are fixed.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Well, here is my problem, and I am pretty sure it was not here before RC7:
I loaded a project (3000 segments), jump to next untranslated, shift to a
terminal app to look at top, when I go back the display is gone, the whole
page is all white. THen it comes back, I click on a segment at random and the
the display starts to mix lines and eventually disapears again and I have no
way to get it back...
I attach the error messages I got.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think I have something.
I loaded the same project. And started to move around using enter/next
untranslated. I click back and forth etc. And then, without mistake, I click one
last time on the first segment of the first file (although I doubt it is limited to
that first segment) and I loose reaction from OmegaT.
OmegaT is still supposed to be indexing, but reaction was pretty snappy that
far. Then when I click that "faulty" segment it starts running the CPU like
crazy, I shift to the terminal window to check top once or twice and at the
second time OmegaT is not able to refresh the display and gives up
displaying anything.
Also, that behavior seems to affect the TM since the last clicked segment gets
erased from the project TM. If I play with OmegaT when it is in that state it
will eventually do very weird things to the project TM (I attach one example).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
FWIW: OmegaT is still indexing, even thoug there is absolutely to way to get the
display back. I click on the project files but none gets displayed. Just in case that
matters, here is my "alias" for launching OmegaT:
cd ~/Documents/_travail/OmegaT/application/cvs/26_02/dist/
/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Commands/
java -server -Xmx512m -jar OmegaT.jar
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Ok, I did let my project rest all night to make sure that it indexed all
segments before I started working on it and here comes the ugly bug again: I
translate a few segments, build target, decide to revise a segment, click on it
and here I get the display locked and the Java error... Now funny think. I
switch back to the project, the display is still here, I switch to project files and
the display diseappears. I won't attach anything here until requested since it
looks like the problem is very much the same. I think I'll shift to RC6 for a
little while, with a boosted memory and -server, to make sure that I can work
:)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
All I can say is What the ...!
One thing is that a call to Document.insertString() should
not throw IllegalArgumentException!!! It can throw only
BadLocationException by contract...
Next it seems that after OmegaT somehow (possibly wrongly)
changes the document, Swing goes crazy by its own. Only the
first stack trace contains any OmegaT code, next Swing
throws them by itself...
JC, I have some questions to you:
1. What JVM do you use?
2. May you send me the project?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> One thing is that a call to Document.insertString() should
> not throw IllegalArgumentException!!!
>
Yes, it can. IllegalArgumentException is a subclass of
RuntimeException, which is not checked (i.e. you don't have
to declare it in a throws clause to be able to throw it).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=545103
By the way, this problem already appears when opening a
project or file. OmegaT positions the text such that only
the target segment is shown.
Logged In: YES
user_id=488500
I confirm.
Working on it...
Logged In: YES
user_id=488500
Fixed (but I won't include this to changes of RC7, because
it's IMO a regression)
Actually fixing this bug also implemented RFE
http://sourceforge.net/support/tracker.php?aid=1228857
fixed in CVS:
Checking in omegat/src/org/omegat/gui/main/MainWindow.java
new revision: 1.18; previous revision: 1.17
Logged In: YES
user_id=488500
I'm changing summary because of many errors here and
increasing priority. Please use RC7 (not CVS) so far,
because it's less affected.
Duplicates:
http://sourceforge.net/support/tracker.php?aid=1437455
http://sourceforge.net/support/tracker.php?aid=1437658
http://sourceforge.net/support/tracker.php?aid=1437733
Related RFE:
http://sourceforge.net/support/tracker.php?aid=1228857
Logged In: YES
user_id=488500
Hopefully fixed, but I'll do more testing to ensure it's gone.
Checking in omegat/src/org/omegat/gui/main/MainWindow.java
new revision: 1.20
Logged In: YES
user_id=488500
Hopefully fixed.
At least I cannot experience this on in-CVS code under both
JDK 1.4.2 and JDK 1.5/5.0 on a pretty large project I got
these errors before.
I'd like to ask you, Henry and JC, to update from CVS and
work on a project you experienced the problems with.
Logged In: YES
user_id=545103
I've been working with this fix since 8pm last night (yes, I
did sleep in between), and perceived no problems. I'd say
this and related issues are fixed.
Logged In: YES
user_id=915082
RC8-cvs
Well, here is my problem, and I am pretty sure it was not here before RC7:
I loaded a project (3000 segments), jump to next untranslated, shift to a
terminal app to look at top, when I go back the display is gone, the whole
page is all white. THen it comes back, I click on a segment at random and the
the display starts to mix lines and eventually disapears again and I have no
way to get it back...
I attach the error messages I got.
error messages
Logged In: YES
user_id=915082
And I attach the relevant log part too.
relevant log part "should not happen"
Logged In: YES
user_id=545103
I've had a similar experience with the Project Files window,
but I only experienced it once or twice, and can't reproduce it.
Logged In: YES
user_id=915082
I think I have something.
I loaded the same project. And started to move around using enter/next
untranslated. I click back and forth etc. And then, without mistake, I click one
last time on the first segment of the first file (although I doubt it is limited to
that first segment) and I loose reaction from OmegaT.
OmegaT is still supposed to be indexing, but reaction was pretty snappy that
far. Then when I click that "faulty" segment it starts running the CPU like
crazy, I shift to the terminal window to check top once or twice and at the
second time OmegaT is not able to refresh the display and gives up
displaying anything.
Also, that behavior seems to affect the TM since the last clicked segment gets
erased from the project TM. If I play with OmegaT when it is in that state it
will eventually do very weird things to the project TM (I attach one example).
log+Java errors+messed up TM
Logged In: YES
user_id=915082
FWIW: OmegaT is still indexing, even thoug there is absolutely to way to get the
display back. I click on the project files but none gets displayed. Just in case that
matters, here is my "alias" for launching OmegaT:
cd ~/Documents/_travail/OmegaT/application/cvs/26_02/dist/
/System/Library/Frameworks/JavaVM.framework/Versions/1.5.0/Commands/
java -server -Xmx512m -jar OmegaT.jar
Logged In: YES
user_id=915082
RC8_cvs
Ok, I did let my project rest all night to make sure that it indexed all
segments before I started working on it and here comes the ugly bug again: I
translate a few segments, build target, decide to revise a segment, click on it
and here I get the display locked and the Java error... Now funny think. I
switch back to the project, the display is still here, I switch to project files and
the display diseappears. I won't attach anything here until requested since it
looks like the problem is very much the same. I think I'll shift to RC6 for a
little while, with a boosted memory and -server, to make sure that I can work
:)
Logged In: YES
user_id=488500
All I can say is What the ...!
One thing is that a call to Document.insertString() should
not throw IllegalArgumentException!!! It can throw only
BadLocationException by contract...
Next it seems that after OmegaT somehow (possibly wrongly)
changes the document, Swing goes crazy by its own. Only the
first stack trace contains any OmegaT code, next Swing
throws them by itself...
JC, I have some questions to you:
1. What JVM do you use?
2. May you send me the project?
Logged In: YES
user_id=545103
> One thing is that a call to Document.insertString() should
> not throw IllegalArgumentException!!!
>
Yes, it can. IllegalArgumentException is a subclass of
RuntimeException, which is not checked (i.e. you don't have
to declare it in a throws clause to be able to throw it).
Logged In: YES
user_id=488500
And:
3. What are your options in Options -> Editing Field... dialog?
Logged In: YES
user_id=488500
Thanks a lot, Henry.
Still they could have mentioned that in the manual... :-(