Posts by marcelvanherk

    Hi,


    I guess that explains, it should become:


    ExportFilter0 = StudyDate >= '20080815'


    So you are also saying that for versions of PostGres beyond 8.3 DoubleBackSlashToDB should be set to 0? Or should there be a third setting for DoubleBackSlashToDB?


    Regards,


    Marcel

    Mark,


    I cannot confirm this behavior. What version and exactly what query where you using? This is what I see for a daterange query:


    Code
    [MICROPLUS] 0008,0020 10 DA StudyDate "20080101- "
    ..
    [MICROPLUS] Query Distinct Tables: DICOMStudies
    [MICROPLUS] Columns : DICOMStudies.StudyDate, DICOMStudies.PatientNam, DICOMStudies.PatientID
    [MICROPLUS] Where : DICOMStudies.StudyDate >= '20080101'


    I believe that your problem may be in a ExportFilter, i.e., it is a problem with a query given in dicom.ini and not a problem with the server source.


    Marcel

    Hi Alberto,


    your fix in CodeSequence in vrtosql.cpp is correct: in the while it should indeed be

    Code
    if (Levels->Get(*Index)<=level) break;


    However, there is another bug as well, a line in ProcessQuery should be:

    Code
    int l = level; Levels->Add ( l );


    The current incorrect lines are:

    Code
    if (Levels->Get(*Index)<level) break;
    ..
    int l = level+1; Levels->Add ( l );



    Marcel

    Hi,


    you seem to have failed to redefine StandardStorage in all files: the first 2 error messages relate to StandardStorage and were the same that were listed before. The next message means that private symbol queue in dgate already exists in solaris. Rename queue to conquest_queue in all places in that file.


    Marcel

    Hi,


    this behaviour looks normal: MSSQL will use most available RAM to store indexes - it is not related to the dgate out of memory error (you see that it does not leak memory).


    The dgate out of memory occurs on a malformed DICOM image, where the length of an element is read faulty as a very big number causing "out of memory" when dgate tries to allocate that. If 1.4.14beta there is a flag to supress the out-of memory message - causing the image to fail loading without stopping the server.


    Marcel

    Hi,


    Simple, you can use hostnames in acrnema.map.


    If you have many hosts that view images you can use e.g. "Vhostname" as AE title. An entry "V* * 1234 un" will then map all of these hosts assuming they communicate uncompressed and on port 1234.


    Marcel

    Hi,


    please ignore the warning: multi-character character constant, it is in the core design and the code is OK for bigendian and littleendian machines.


    Both other errors indicate that StandardStorage is not defined. Can you check that file storage.hpp is compiling, are these the first ERROR messages?


    Marcel

    Hi Stephane,


    I looked into your problem (after I got home from vacation). The problem is not that conquest does not accept the message but that the error message on the conquest side is wrong.


    Storescu can send any data, but it can only do so with the transfer syntax that you specify on the command line. If that does not correspond with the transfer syntax of the file, it will not recomoress the file, but gives an error message and hang up the DICOM connection.


    Conquest then reports - incorrectly -: *** Association rejected: you may need to add the listed presentation context as sop to dgatesop.lst".


    This works, though:


    storescu -v --required --propose-lossless 127.0.0.1 5678 NMc.dcm


    I will add the incorrect error message to the bug list.


    Marcel


    here is your mail:


    Thank you for answering so quickly. Here is an example of the files I have to deal with (NM.dcm). It's a simple whole body image from NucMed. I forgot to mention that The original transfer syntax is 1.2.840.10008.1.2.4.57. After decompression (with dcmdjpeg.exe) it becomes obviously 1.2.840.10008.1.2.1 (that file I can transfer) and if I recompress (with dcmcjpeg.exe to obtain NMc.dcm) the syntax transfer is 1.2.840.10008.1.2.4.70 which I can't transfer either. I'm almost embarrassed to think it might be a simple config problem but I really tought I tried all angles with the configs.

    Louis,


    The GUI assumes dgate runs as a service when dgate runs when the GUI starts. So if the GUI crashes or if dgate fails to stop when you close the GUI, you need to manually kill process dgate prior to starting the GUI to get into normal operation again. The bind error is just that the server runs when it starts to start a new one. It seems that something like this is happening. Not sure what causes the disconnect though.


    Marcel

    Hi,


    these are two standard DICOM losless compression options - you should look into dicom literature for this issue. I tend to use J2, but I see no difference in performance: they both compress around 2-3 times.


    Marcel

    Hi Louis,


    the conquest browser stores a pointer to the current record it is at. Apparently the saved record is incorrect. Do you keep getting it when you navigate to an existing patient in the browser, and then change pages? Which database do you use?


    Marce;