dicim reader test...

[kerautret]

The merge of the dicom reader was perhaps to fast:

kerautre:~/EnCours/DGtal/build[master⚡]$ ./tests/io/readers/testDicomReader New Block [Testing class DicomReader] Args: ./tests/io/readers/testDicomReader
[Image - STLVector] size=2130048 valuetype=1bytes Domain=[HyperRectDomain] = [[PointVector] {0, 0, 0}]x[[PointVector] {128, 127, 128}] Number of points with (val>0) = 2130048 WARNING: In /Users/kerautre/Dev/InsightToolkit-4.4.2/Modules/IO/GDCM/src/itkGDCMSeriesFileNames.cxx, line 114 GDCMSeriesFileNames (0x7fb854b00990): No Series can be found, make sure your restrictions are not too strong

[ERR]DicomReader: can't read /Users/kerautre/EnCours/DGtal/tests/samples Exception catched. Message : DGtal IO error Passed. EndBlock [Testing class DicomReader](1227.89 ms)

[dcoeurjo]

arg. I agree... the test returns a "0" but there is an exception fail.

I just ran a "make test" which didn't complain..

[dcoeurjo]

wait... it is normal. the test just checks that the exception is raised... no problem

[kerautret]

ok so just a warning, but the [ERR] is a little confusing

[dcoeurjo]

maybe.. but the test checks that the error is raised;)

[kerautret]

I see, it should be from: std::string filename = testPath + "samples/null.dcm"; Is it useful or null.dcm is missing ?

[dcoeurjo]

yes it is : if you want to test that the exception is raised, try to load a fake dicom image (non existing one). So the test is relevant.

[kerautret]

ha ok, with the filename print it could be less ambiguous if we don't check the source ;)

[kerautret]

I add it should be better now.

[kerautret]

I close it