Mercurial > hg > orthanc-tests
changeset 217:a491d5ec370a
test encodings
| author | Sebastien Jodogne <s.jodogne@gmail.com> |
|---|---|
| date | Thu, 14 Feb 2019 11:32:44 +0100 |
| parents | 6c6411d34951 |
| children | 70fba41776be |
| files | Database/Encodings/DavidClunie/DICOMDIR Database/Encodings/DavidClunie/SCSARAB Database/Encodings/DavidClunie/SCSFREN Database/Encodings/DavidClunie/SCSGERM Database/Encodings/DavidClunie/SCSGREEK Database/Encodings/DavidClunie/SCSH31 Database/Encodings/DavidClunie/SCSH32 Database/Encodings/DavidClunie/SCSHBRW Database/Encodings/DavidClunie/SCSI2 Database/Encodings/DavidClunie/SCSRUSS Database/Encodings/DavidClunie/SCSX1 Database/Encodings/DavidClunie/SCSX2 Database/Encodings/DavidClunie/charsettests.screenshot.png Tests/Tests.py |
| diffstat | 14 files changed, 57 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- a/Tests/Tests.py Fri Feb 08 16:34:07 2019 +0100 +++ b/Tests/Tests.py Thu Feb 14 11:32:44 2019 +0100 @@ -4444,3 +4444,60 @@ Compare('MarekLatin2.dcm', 'MarekLatin2.json') Compare('HierarchicalAnonymization/StructuredReports/IM0', 'HierarchicalAnonymization/StructuredReports/IM0.json') + + + def test_issue_95_encodings(self): + # https://bitbucket.org/sjodogne/orthanc/issues/95/server-cant-support-chinese + # Check out image: "../Database/Encodings/DavidClunie/charsettests.screenshot.png" + + # Very useful tool: "file2" from package "file-kanji" + + def GetPatientName(dicom): + i = UploadInstance(_REMOTE, dicom) ['ID'] + j = DoGet(_REMOTE, '/instances/%s/tags?simplify' % i) + return j['PatientName'] + + def ComparePatientName(name, dicom): + self.assertEqual(name, GetPatientName(dicom)) + + # gdcmraw -t 10,10 -i SCSFREN -o /tmp/tag && uconv -f ISO-IR-100 -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Buc^Jérôme', 'Encodings/DavidClunie/SCSFREN') + + # gdcmraw -t 10,10 -i SCSI2 -o /tmp/tag && uconv -f KOREAN -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Hong^Gildong=洪^吉洞=홍^길동', 'Encodings/DavidClunie/SCSI2') # Since Orthanc 1.5.5 + + # gdcmraw -t 10,10 -i SCSX2 -o /tmp/tag && uconv -f GB18030 -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Wang^XiaoDong=王^小东=', 'Encodings/DavidClunie/SCSX2') + + # gdcmraw -t 10,10 -i SCSX1 -o /tmp/tag && cat /tmp/tag && echo + ComparePatientName(u'Wang^XiaoDong=王^小東=', 'Encodings/DavidClunie/SCSX1') + + # gdcmraw -t 10,10 -i SCSH31 -o /tmp/tag && uconv -f JIS -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Yamada^Tarou=山田^太郎=やまだ^たろう', 'Encodings/DavidClunie/SCSH31') + + # gdcmraw -t 10,10 -i SCSGERM -o /tmp/tag && uconv -f ISO-IR-100 -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Äneas^Rüdiger', 'Encodings/DavidClunie/SCSGERM') + + # gdcmraw -t 10,10 -i SCSGREEK -o /tmp/tag && uconv -f ISO-IR-126 -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Διονυσιος', 'Encodings/DavidClunie/SCSGREEK') + + # gdcmraw -t 10,10 -i SCSRUSS -o /tmp/tag && uconv -f ISO-IR-144 -t UTF-8 /tmp/tag && echo + ComparePatientName(u'Люкceмбypг', 'Encodings/DavidClunie/SCSRUSS') + + # gdcmraw -t 10,10 -i SCSHBRW -o /tmp/tag && uconv -f ISO-IR-138 -t UTF-8 /tmp/tag && echo + # NB: Hebrew is a right-to-left encoding, copying/pasting from + # Linux console into Emacs automatically reverse the string + ComparePatientName(u'שרון^דבורה', 'Encodings/DavidClunie/SCSHBRW') + + # gdcmraw -t 10,10 -i SCSARAB -o /tmp/tag && uconv -f ISO-IR-127 -t UTF-8 /tmp/tag && echo + # NB: Right-to-left as for Hebrew (SCSHBRW), and the Ubuntu console can't display such + # characters by default, but copy/paste works with Emacs + ComparePatientName(u'قباني^لنزار', 'Encodings/DavidClunie/SCSARAB') + + # SCSH32: This SpecificCharacterSet is composed of 2 + # codepages: "ISO 2022 IR 13" (i.e. "SHIFT_JIS") until the + # first equal, then "ISO 2022 IR 87" (i.e. "JIS") for the + # remainer. Orthanc only takes into consideration the first + # codepage: This is a limitation. + # gdcmraw -t 10,10 -i SCSH32 -o /tmp/tag && cut -d '=' -f 1 /tmp/tag | uconv -f SHIFT_JIS -t UTF-8 + self.assertTrue(GetPatientName('Encodings/DavidClunie/SCSH32').startswith(u'ﾔﾏﾀﾞ^ﾀﾛｳ='))