[Debian-med-packaging] Bug#753809: ginkgocadx: will not display studies

Karsten Hilbert Karsten.Hilbert at gmx.net
Sun Sep 21 12:03:57 UTC 2014


On Sun, Sep 21, 2014 at 10:41:54AM +1000, Dmitry Smirnov wrote:

> Could you please help me to reproduce this problem?
> I can successfully open bunch of dicom files I have as well as anything from 
> Ginkgo cloud. I'm not as familiar with ginkgocadx interface as you are so I 
> will need a detailed how-to instructions what to do to hit the problem.

No problem. It only seems to appear when using a DICOM server
(say, Orthanc). This is my workflow to reproduce the problem:

- open Ginkgo
- open a DICOM (.dcm) file which has not been viewed
  in Ginkgo before
- Ginkgo will happily display this file
- in Ginkgo upload the file to the DICOM server

At this stage, other DICOM viewers (say, Aeskulap) will
happily retrieve the file from the DICOM server _and_
gladly display it.

- in Gingko remove the file from the viewing history
- in Ginkgo query the DICOM server for the study (the DICOM file)
  you just uploaded
- in Ginkgo download the DICOM file from the DICOM server
- Ginkgo now fails to create a thumbnail and also fails
  to display the DICOM file after download despite having
  properly displayed it before

Note that it IS possible to download the very same file from
the very same DICOM server and properly display it using
_another_ DICOM viewer (say, Aeskulap). In other words, the
failure does NOT depend on the DICOM server.

I also tried the following:

- at the filesystem level copy the downloaded-and-failing-to-display
  DICOM file to another directory
- in Ginkgo remove the failing file from the history
- in Ginkgo open the copied file (which previously failed
  after the download) as if it was a new file
- Ginkgo _still_ fails to display the file

This means, that something permanent is being done to the
file when Ginkgo first downloads it from the DICOM server.

Note that other DICOM viewers (Aeskulap) gladly display the
copy (which, again, still fails in Ginkgo).

_Very_ strange. And frustrating :-)

Karsten
-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346



More information about the Debian-med-packaging mailing list