Merge branch 'main' of github.com:orthanc-team/python-orthanc-api-client

orthanc_api_client/helpers.py

@@ -5,8 +5,6 @@
 import random
 from typing import Union, Optional
 from .helpers_internal import write_dataset_to_bytes
-from pydicom.dataset import Dataset, FileDataset
-from pydicom.uid import ExplicitVRLittleEndian
 import pydicom.uid
 from urllib3.filepost import encode_multipart_formdata, choose_boundary
 
@@ -108,16 +106,13 @@ def generate_test_dicom_file(
         ) -> bytes:
     buffer = bytearray(height * width * 2)
 
-    meta = pydicom.Dataset()
-    meta.MediaStorageSOPClassUID = pydicom.uid.MRImageStorage
-    meta.MediaStorageSOPInstanceUID = pydicom.uid.generate_uid()
-    meta.TransferSyntaxUID = pydicom.uid.ExplicitVRLittleEndian  
+    file_meta = pydicom.dataset.FileMetaDataset()
+    file_meta.MediaStorageSOPClassUID = pydicom.uid.MRImageStorage
+    file_meta.MediaStorageSOPInstanceUID = pydicom.uid.generate_uid()
+    file_meta.TransferSyntaxUID = pydicom.uid.ExplicitVRLittleEndian
 
-    ds = Dataset()
-    ds.file_meta = meta
-
-    ds.is_little_endian = True
-    ds.is_implicit_VR = False
+    ds = pydicom.dataset.Dataset()
+    ds.file_meta = file_meta
 
     ds.Modality = "MR"
     ds.SOPInstanceUID = pydicom.uid.generate_uid()
@@ -184,7 +179,7 @@ def encode_multipart_related(fields, boundary=None):
 
 
 def is_version_at_least(version_string: str, expected_major: int, expected_minor: int, expected_patch: Optional[int] = None) -> bool:
-    if version_string == "mainline":
+    if version_string.startswith("mainline"):
         return True
 
     split_version = version_string.split(".")

orthanc_api_client/helpers_internal.py

@@ -1,16 +1,8 @@
-from pydicom.filebase import DicomFileLike
-from pydicom import dcmwrite
 from io import BytesIO
 
 
 def write_dataset_to_bytes(dataset) -> bytes:
     # create a buffer
     with BytesIO() as buffer:
-        # create a DicomFileLike object that has some properties of DataSet
-        memory_dataset = DicomFileLike(buffer)
-        # write the dataset to the DicomFileLike object
-        dcmwrite(memory_dataset, dataset)
-        # to read from the object, you have to rewind it
-        memory_dataset.seek(0)
-        # read the contents as bytes
-        return memory_dataset.read()
+        dataset.save_as(buffer)
+        return buffer.getvalue()

orthanc_api_client/resources/resources.py

@@ -301,6 +301,59 @@ def modify_bulk(self, orthanc_ids: List[str] = [], replace_tags: Any = {}, remov
 
 
     def modify_bulk_async(self, orthanc_ids: List[str] = [], replace_tags: Any = {}, remove_tags: List[str] = [], keep_tags: List[str] = [], delete_original: bool = True, force: bool = False, transcode: Optional[str] = None, permissive: bool = False) -> Job:
+        return self._modify_bulk_async(
+            operation="modify",
+            orthanc_ids=orthanc_ids,
+            replace_tags=replace_tags,
+            remove_tags=remove_tags,
+            keep_tags=keep_tags,
+            delete_original=delete_original,
+            force=force,
+            transcode=transcode,
+            permissive=permissive)
+
+    def _modify_bulk(self, operation: str, orthanc_ids: List[str] = [], replace_tags: Any = {}, remove_tags: List[str] = [], keep_tags: List[str] = [], delete_original: bool = True, force: bool = False, transcode: Optional[str] = None, permissive: bool = False)  -> Tuple[List[str], List[str], List[str], List[str]]:
+        """
+        returns a tuple with:
+        - the list of modified instances ids
+        - the list of modified series ids
+        - the list of modified studies ids
+        - the list of modified patients ids
+        """
+        modified_instances_ids = []
+        modified_series_ids = []
+        modified_studies_ids = []
+        modified_patients_ids = []
+
+        job = self._modify_bulk_async(
+            operation=operation,
+            orthanc_ids=orthanc_ids,
+            replace_tags=replace_tags,
+            remove_tags=remove_tags,
+            keep_tags=keep_tags,
+            delete_original=delete_original,
+            force=force,
+            transcode=transcode
+        )
+
+        job.wait_completed()
+
+        if job.info.status == JobStatus.SUCCESS and "Resources" in job.content:
+            # extract the list of modified instances ids from the job content
+            for r in job.content.get("Resources"):
+                if r.get("Type") == "Instance":
+                    modified_instances_ids.append(r.get("ID"))
+                elif r.get("Type") == "Series":
+                    modified_series_ids.append(r.get("ID"))
+                elif r.get("Type") == "Study":
+                    modified_studies_ids.append(r.get("ID"))
+                elif r.get("Type") == "Patient":
+                    modified_patients_ids.append(r.get("ID"))
+            return modified_instances_ids, modified_series_ids, modified_studies_ids, modified_patients_ids
+        else:
+            raise api_exceptions.OrthancApiException(msg=f"Error while {'modifying' if operation == 'modify' else 'anonymizing'} bulk {self._get_level()}, job failed {json.dumps(job.info.content)}")
+
+    def _modify_bulk_async(self, operation: str, orthanc_ids: List[str] = [], replace_tags: Any = {}, remove_tags: List[str] = [], keep_tags: List[str] = [], delete_original: bool = True, force: bool = False, transcode: Optional[str] = None, permissive: bool = False) -> Job:
         query = {
             "Force": force,
             "Level": self._get_level(),
@@ -321,14 +374,45 @@ def modify_bulk_async(self, orthanc_ids: List[str] = [], replace_tags: Any = {},
             query['KeepSource'] = False
 
         r = self._api_client.post(
-            endpoint=f"/tools/bulk-modify",
+            endpoint=f"/tools/bulk-{operation}",
             json=query)
 
         if r.status_code == 200 and "ID" in r.json():
             return Job(api_client=self._api_client, orthanc_id=r.json()['ID'])
         else:
-            raise HttpError(http_status_code=r.status_code, msg="Error in bulk-modify", url=r.url, request_response=r)
+            raise HttpError(http_status_code=r.status_code, msg=f"Error in bulk-{operation}", url=r.url, request_response=r)
+
+
+    def anonymize_bulk(self, orthanc_ids: List[str] = [], replace_tags: Any = {}, remove_tags: List[str] = [], keep_tags: List[str] = [], delete_original: bool = False, force: bool = False, transcode: Optional[str] = None, permissive: bool = False)  -> Tuple[List[str], List[str], List[str], List[str]]:
+        """
+        returns a tuple with:
+        - the list of anonymized instances ids
+        - the list of anonymized series ids
+        - the list of anonymized studies ids
+        - the list of anonymized patients ids
+        """
+        return self._modify_bulk(
+            operation="anonymize",
+            orthanc_ids=orthanc_ids,
+            replace_tags=replace_tags,
+            remove_tags=remove_tags,
+            keep_tags=keep_tags,
+            delete_original=delete_original,
+            force=force,
+            transcode=transcode,
+            permissive=permissive)
 
+    def anonymize_bulk_async(self, orthanc_ids: List[str] = [], replace_tags: Any = {}, remove_tags: List[str] = [], keep_tags: List[str] = [], delete_original: bool = False, force: bool = False, transcode: Optional[str] = None, permissive: bool = False) -> Job:
+        return self._modify_bulk_async(
+            operation="anonymize",
+            orthanc_ids=orthanc_ids,
+            replace_tags=replace_tags,
+            remove_tags=remove_tags,
+            keep_tags=keep_tags,
+            delete_original=delete_original,
+            force=force,
+            transcode=transcode,
+            permissive=permissive)
 
     def print_daily_stats(self, from_date: datetime.date = None, to_date: datetime.date = None):
         if self._url_segment == "patients":

release-notes.md

@@ -1,3 +1,18 @@
+v 0.17.0
+========
+
+- Added `OrthancApiClient.instances.anonymize_bulk` and `OrthancApiClient.instances.anonymize_bulk_async`
+
+v 0.16.3
+========
+
+- Avoid pydicom warning when generating test files
+
+v 0.16.2
+========
+
+- `o.is_orthanc_version_at_least()` and `o.is_plugin_version_at_least()` now support "mainline-commitId" patterns
+
 v 0.16.1
 ========
 

setup.py

@@ -28,7 +28,7 @@
     # For a discussion on single-sourcing the version across setup.py and the
     # project code, see
     # https://packaging.python.org/guides/single-sourcing-package-version/
-    version='0.16.1',  # Required
+    version='0.17.0',  # Required
 
     # This is a one-line description or tagline of what your project does. This
     # corresponds to the "Summary" metadata field:
@@ -134,7 +134,7 @@
     # https://packaging.python.org/discussions/install-requires-vs-requirements/
     install_requires=[
         'requests>=2.31.0',
-        'pydicom>=3.0.0',
+        'pydicom>=3.0.1',
         'StrEnum>=0.4.15'
     ],
 

tests/docker-setup/docker-compose.yml

@@ -1,7 +1,7 @@
 version: "3"
 services:
   orthanc-a:
-    image: orthancteam/orthanc:24.2.3
+    image: orthancteam/orthanc:24.10.1
     ports: ["10042:8042"]
     environment:
       VERBOSE_STARTUP: "true"

tests/docker-setup/orthanc-b/Dockerfile

@@ -1,6 +1,6 @@
-FROM orthancteam/orthanc:24.2.3
+FROM orthancteam/orthanc:24.10.1
 
-RUN pip install pydicom
+RUN pip install  --break-system-packages pydicom==3.0.1
 
 RUN mkdir /scripts
 COPY plugin.py /scripts

tests/docker-setup/orthanc-b/plugin.py

@@ -6,25 +6,15 @@
 
 from io import BytesIO
 
-from pydicom import dcmread, dcmwrite
-from pydicom.filebase import DicomFileLike
-
 
 TOKEN = orthanc.GenerateRestApiAuthorizationToken()
 
 
 def write_dataset_to_bytes(dataset):
     # create a buffer
     with BytesIO() as buffer:
-        # create a DicomFileLike object that has some properties of DataSet
-        memory_dataset = DicomFileLike(buffer)
-        # write the dataset to the DicomFileLike object
-        dcmwrite(memory_dataset, dataset)
-        # to read from the object, you have to rewind it
-        memory_dataset.seek(0)
-        # read the contents as bytes
-        return memory_dataset.read()
-
+        dataset.save_as(buffer)
+        return buffer.getvalue()
 
 
 def get_api_token(output, uri, **request):
@@ -43,15 +33,26 @@ def worklist_callback(answers, query, issuerAet, calledAet):
     json_tags = json.loads(orthanc.DicomBufferToJson(dicom, orthanc.DicomToJsonFormat.FULL, orthanc.DicomToJsonFlags.NONE, 0))
 
     dataset = pydicom.dataset.Dataset()
-    dataset.is_little_endian = True
-    dataset.is_implicit_VR = False
+
+    file_meta = pydicom.dataset.FileMetaDataset()
+
+    # Set the FileMeta attributes
+    file_meta.MediaStorageSOPClassUID = '1.2.840.10008.5.1.4.31'
+    file_meta.MediaStorageSOPInstanceUID = generate_uid()
+    file_meta.ImplementationClassUID = '1.2.840.10008.5.1.4.1.1.2'
+    file_meta.TransferSyntaxUID = pydicom.uid.ExplicitVRLittleEndian
+    dataset.file_meta = file_meta
+
+    # dataset.TransferSyntaxUID = pydicom.uid.ExplicitVRLittleEndian
     dataset.AccessionNumber = 'A123456'
     dataset.StudyInstanceUID = '1.2.3.4'
     dataset.PatientName = 'PatientName'
     dataset.PatientID = 'PatientID'
     dataset.PatientBirthDate = '20220208'
     dataset.PatientSex = 'O'
-
+    # dataset.is_little_endian = True
+    # dataset.is_implicit_VR = False
+    # dataset.file_meta.TransferSyntaxUID = pydicom.uid.ExplicitVRLittleEndian
     dataset_bytes = write_dataset_to_bytes(dataset)
 
     answers.WorklistAddAnswer(query, dataset_bytes)

tests/test_api_client.py

@@ -3,6 +3,8 @@
 import subprocess
 import logging
 import datetime
+import uuid
+
 from orthanc_api_client import OrthancApiClient, generate_test_dicom_file, ChangeType, ResourceType, Study, Job, JobStatus, JobType, InstancesSet, LabelsConstraint, LogLevel, RemoteJob
 from orthanc_api_client.helpers import *
 import orthanc_api_client.exceptions as api_exceptions
@@ -937,6 +939,59 @@ def test_modify_bulk_instances(self):
         self.assertNotIn(instances_ids[0], modified_instances_ids)  # make sure the new ids are different from the original ones
 
 
+    def test_anonymize_bulk_series(self):
+        self.oa.delete_all_content()
+
+        self.oa.upload_folder(here / "stimuli/MR/Brain")
+        series_ids = self.oa.series.get_all_ids()
+
+        self.assertEqual(1, len(self.oa.studies.get_all_ids()))
+
+        _, modified_series_ids, __, ___ = self.oa.series.anonymize_bulk(
+            orthanc_ids=series_ids,
+            delete_original=False,
+            keep_tags=["SeriesDescription", "StudyDescription"]
+        )
+
+        self.assertEqual(2, len(modified_series_ids))
+        tags1 = self.oa.series.get_tags(modified_series_ids[0])
+        tags2 = self.oa.series.get_tags(modified_series_ids[1])
+        self.assertEqual(tags1.get("PatientName"), tags2.get("PatientName"))
+        self.assertEqual(tags1.get("StudyDescription"), tags2.get("StudyDescription"))
+        self.assertEqual(tags1.get("StudyInstanceUID"), tags2.get("StudyInstanceUID"))
+        self.assertIn(tags1.get("SeriesDescription"), ["sT2W/FLAIR", "T1/3D/FFE/C"])
+        # make sure both series are in the same anonymized study (the original study is still in Orthanc)
+        self.assertEqual(2, len(self.oa.studies.get_all_ids()))
+
+
+    def test_anonymize_bulk_study(self):
+        self.oa.delete_all_content()
+
+        self.oa.upload_folder(here / "stimuli/MR/Brain")
+        self.oa.upload_file(here / "stimuli/CT_small.dcm")
+
+        self.assertEqual(2, len(self.oa.studies.get_all_ids()))
+
+        _, __, modified_studies_ids, ___ = self.oa.series.anonymize_bulk(
+            orthanc_ids=self.oa.studies.get_all_ids(),
+            delete_original=False,
+            keep_tags=["SeriesDescription", "StudyDescription"],
+            replace_tags={
+                "PatientID": str(uuid.uuid4()),                    # orthanc does not put all studies in the same patient -> you must do it manually
+                "PatientName": f"Anonymized " + str(uuid.uuid4())
+            },
+            force=True
+        )
+
+        self.assertEqual(2, len(modified_studies_ids))
+        tags1 = self.oa.studies.get_tags(modified_studies_ids[0])
+        tags2 = self.oa.studies.get_tags(modified_studies_ids[1])
+        self.assertEqual(tags1.get("PatientName"), tags2.get("PatientName"))
+        self.assertNotEqual(tags1.get("StudyDescription"), tags2.get("StudyDescription"))
+        self.assertNotEqual(tags1.get("StudyInstanceUID"), tags2.get("StudyInstanceUID"))
+        # make sure both studies are in the same anonymized patient (the original patients are still in Orthanc)
+        self.assertEqual(3, len(self.oa.patients.get_all_ids()))
+
     def test_asyncio(self):
         self.oa.delete_all_content()
 
@@ -1490,6 +1545,7 @@ def test_version(self):
         self.assertFalse(is_version_at_least("1.1", 1, 2, 3))
 
         self.assertTrue(is_version_at_least("mainline", 1, 2, 3))  # mainline is always bigger than any version number !!!
+        self.assertTrue(is_version_at_least("mainline-548748", 1, 2, 3))  # mainline is always bigger than any version number !!!
 
         self.assertTrue(self.oa.is_orthanc_version_at_least(1, 9, 0))
         self.assertTrue(self.oa.is_plugin_version_at_least("dicom-web", 1, 5))