URGENT! Corrupted saved RoiSet.zip of hand modified ROIs

[Ruszkai Ákos]

Hello!

Sorry for the title, but I'm stuck with this problem now for hours,
and I can't save my datas, I need to present in 8 hours or so :(
I've defined some ROIs with particle analyzer plugin, and some of them
needed adjustment by hand. I've updated the ROI, so the new defined
shape is saved.
When I've saved all my ROI's into a RoiSet.zip file, and later tried
to open, it turns out that the zip file is corrupted, and ImageJ can't
open it.
Or to be precise, it can only open them until the first modified ROI.

Here's the error message I get from ImageJ in a popup window:

java.io.EOFException: Unexpected end of ZLIB input stream

Please, help me, I'm in a really big trouble

Regards,

Ákos Ruszkai

[Ruszkai Ákos]

Hello

On 12 September 2011 00:30, Bill Mohler <[hidden email]> wrote:
> How many ROIs in your zip file?  What OS are you running ?

It should be exactly 200. I know ImageJ can handle that much, as I've
used an older version two years ago for my thesis (too bad my first
thing was to update, so I don't even know what version I've used years
ago :( )
OS is Debian GNU/Linux testing,
akion@quadra:~$ uname -a
Linux quadra 3.0.0-1-686-pae #1 SMP Sun Jul 24 14:27:32 UTC 2011 i686 GNU/Linux

Java is:
akion@quadra:~$ java -version
java version "1.6.0_18"
OpenJDK Runtime Environment (IcedTea6 1.8.9) (6b18-1.8.9-0.1)
OpenJDK Server VM (build 16.0-b13, mixed mode)

> Are you sure that no ROIs are selected when you save the zip?

Yes, I haven't selected any ROIs. Also, ImageJ can save and reload
individual modified ROIs one by one flawlessly.
Also, it's not so urgent now. I've decided not to save the ROIs, only
the results. I know I won't be able to replroduce the same
measurements anymore, but chances are low I have to remeasure them
again, so I'm not bothering with that anymore.
But the bug still applies.

Regards,

Ákos Ruszkai

[Rasband, Wayne (NIH/NIMH) [E]]

On Sep 11, 2011, at 5:16 PM, Ruszkai Ákos wrote:

> Hello!
>
> Sorry for the title, but I'm stuck with this problem now for hours,
> and I can't save my datas, I need to present in 8 hours or so :(
> I've defined some ROIs with particle analyzer plugin, and some of them
> needed adjustment by hand. I've updated the ROI, so the new defined
> shape is saved.
> When I've saved all my ROI's into a RoiSet.zip file, and later tried
> to open, it turns out that the zip file is corrupted, and ImageJ can't
> open it.
> Or to be precise, it can only open them until the first modified ROI.
>
> Here's the error message I get from ImageJ in a popup window:
>
> java.io.EOFException: Unexpected end of ZLIB input stream
>
> Please, help me, I'm in a really big trouble

This bug should be fixed in the 1.45o daily build.

-wayne

[Ruszkai Ákos]

Hello

Thank you very much, the compression error is gone.
However, here's another: on the hand modified ROIs ImageJ daily build
145o11 can't measure:

Circularity (results window returns constant 1)
FeretX/Y/Angle (results window returns constant 0)
Solidity (results window returns constant NaN)

Regards,

Ákos

On 12 September 2011 07:17, Rasband, Wayne (NIH/NIMH) [E]
<[hidden email]> wrote:
>
> This bug should be fixed in the 1.45o daily build.
>
> -wayne
>

[Ruszkai Ákos]

Hello

No, I alter my ROIs with the freehand selection tool (I press left ALT, so
it removes my selected parts from the ROI).
Sure, I'll be able to send them later today.

Regards,

Ákos

On 12 September 2011 08:49, Rasband, Wayne (NIH/NIMH) [E] <
[hidden email]> wrote:

>
> On Sep 12, 2011, at 1:39 AM, Ruszkai Ákos wrote:
>
>> Hello
>>
>> Thank you very much, the compression error is gone.
>> However, here's another: on the hand modified ROIs ImageJ daily build
>> 145o11 can't measure:
>>
>> Circularity (results window returns constant 1)
>> FeretX/Y/Angle (results window returns constant 0)
>> Solidity (results window returns constant NaN)
>
> Hi,
>
> Are you using the ROI brush tool to hand modify the ROIs? Could you send

me an example of one of your hand modified ROIs? I should be able to fix
these problem tomorrow.
>
> Best regards,
>
> -wayne
>
>

[Ruszkai Ákos]

Hello

On 12 September 2011 17:37, Rasband, Wayne (NIH/NIMH) [E] <
[hidden email]> wrote:

> On Sep 12, 2011, at 2:54 AM, Ruszkai Ákos wrote:Dear Ákos,
>
> The ImageJ 1.45o11 daily build fixes the bug that caused FeretX/Y/Angle and
> Solidity values of edited ROIs to not be calculated.
>
>
Thank you very much, I'll update my version. Do you still need the corrupted
files?

Best regards, and thank you again!

Ákos