study manipulation fixes

brukerapi/dataset.py

@@ -175,12 +175,14 @@ def __init__(self, path, **state):
             raise FileNotFoundError(self.path)
 
         # directory constructor
-        if self.path.is_dir() and state.get("load"):
+        if self.path.is_dir():
             content = os.listdir(self.path)
             if "fid" in content:
                 self.path = self.path / "fid"
             elif "2dseq" in content:
                 self.path = self.path / "2dseq"
+            elif state.get("load") is LOAD_STAGES["empty"] and self.path.stem in DEFAULT_STATES:
+                pass
             else:
                 raise NotADatasetDir(self.path)
 

brukerapi/folders.py

@@ -352,18 +352,19 @@ def get_dataset(self, exp_id: str | None = None, proc_id: str | None = None) ->
         :param proc_id: name of the processing folder
         :return: fid, or 2dseq :obj:`.Dataset`
         """
-        if exp_id:
-            exp = self._get_exp(exp_id)
+        if exp_id is None:
+            raise ValueError("exp_id is required")
 
-        if proc_id:
+        exp = self._get_exp(exp_id)
+        if proc_id is not None:
             return exp._get_proc(proc_id)["2dseq"]
         return exp["fid"]
 
     def _get_exp(self, exp_id):
-        for exp in self.experiment_list:
+        for exp in self.get_experiment_list():
             if exp.path.name == exp_id:
                 return exp
-        return None
+        raise KeyError(f"Experiment '{exp_id}' not found in {self.path}")
 
 
 class Experiment(Folder):
@@ -414,10 +415,10 @@ def validate(self):
             raise NotExperimentFolder
 
     def _get_proc(self, proc_id):
-        for proc in self.processing_list:
+        for proc in self.get_processing_list():
             if proc.path.name == proc_id:
                 return proc
-        return None
+        raise KeyError(f"Processing '{proc_id}' not found in {self.path}")
 
 
 class Processing(Folder):

docs/source/tutorials/how-to-study.rst

@@ -3,24 +3,24 @@ How to work with Bruker study?
 
 .. code-block:: python
 
-   from brukerapi.study import Study
+   from brukerapi.folders import Study
 
-    study = Study('path_to_study')
+   study = Study('path_to_study')
 
-    #get list of scans (fid data sets) contained in the study
-    study.scans
+   # get list of experiments contained in the study
+   study.get_experiment_list()
 
-    #get list of recos (2dseq data sets) contained in the study
-    study.recos
+   # get list of processing folders contained in the study
+   study.get_processing_list()
 
-    #get data set from the study hierarchy
-    study.get_dataset(scan_id='2', reco_id='1')
+   # get a data set from the study hierarchy
+   study.get_dataset(exp_id='2', proc_id='1')
 
 Data set obtained from ``Study`` object are empty by default, to access its content, the data set needs to be loaded. Either using the load function.
 
 .. code-block:: python
 
-    dataset = study.get_dataset(scan_id='2', reco_id='1')
+    dataset = study.get_dataset(exp_id='2', proc_id='1')
 
     dataset.load()
     dataset.data
@@ -30,9 +30,8 @@ Or using context manager.
 
 .. code-block:: python
 
-    with study.get_dataset(scan_id='2', reco_id='1') as dataset:
+    with study.get_dataset(exp_id='2', proc_id='1') as dataset:
         dataset.data
         dataset.get_value('VisuCoreSize')
 
 
-

test/test_dataset.py

@@ -20,6 +20,20 @@ def test_unsupported_dataset_type(tmp_path):
         Dataset(path)
 
 
+@pytest.mark.parametrize(
+    ("path", "dataset_type"),
+    [
+        ("test/test_data/PV51/0.2H2/10", "fid"),
+        ("test/test_data/PV51/0.2H2/10/pdata/1", "2dseq"),
+    ],
+)
+def test_directory_constructor_uses_default_load(path, dataset_type):
+    dataset = Dataset(path)
+
+    assert dataset.type == dataset_type
+    assert dataset.data.size > 0
+
+
 @pytest.mark.skip(reason="in progress")
 def test_parameters(test_parameters):
     dataset = Dataset(test_parameters[0], load=False)

test/test_folders.py

@@ -1,5 +1,7 @@
+from pathlib import Path
+
 from brukerapi.dataset import Dataset
-from brukerapi.folders import Folder, Processing
+from brukerapi.folders import Folder, Processing, Study
 
 
 def test_folder_traversal_skips_processed_spectra(tmp_path):
@@ -31,3 +33,20 @@ def test_folder_traversal_skips_processed_spectra(tmp_path):
     processing = next(child for child in experiment.get_processing_list() if isinstance(child, Processing))
     processing_datasets = {child.path.name for child in processing.children if isinstance(child, Dataset)}
     assert processing_datasets == {"2dseq"}
+
+
+def test_study_get_dataset_returns_fid_and_2dseq():
+    study = Study(
+        Path("test/test_data/PV51/0.2H2"),
+        dataset_state={"parameter_files": [], "property_files": [], "load": 0},
+    )
+
+    fid = study.get_dataset(exp_id="10")
+    reconstructed = study.get_dataset(exp_id="10", proc_id="1")
+
+    assert fid.path.name == "fid"
+    assert reconstructed.path.name == "2dseq"
+
+    with fid, reconstructed:
+        assert fid.data.size > 0
+        assert reconstructed.data.size > 0