Adding content type to file repository and clarifying version of NIFTI for files (#53)

* Adding content type to filerepository

* dealing with gz files

* one more item in openminds

* using try except

* using expect only where it is needed

* deleting unwanted space

Author: Peyman-N

bids2openminds/main.py

@@ -10,7 +10,7 @@
 import openminds.latest.controlled_terms as controlled_terms
 from openminds import IRI
 
-from .utility import table_filter, pd_table_value, file_hash, file_storage_size
+from .utility import table_filter, pd_table_value, file_hash, file_storage_size, detect_nifti_version
 from .mapping import bids2openminds_instance
 
 
@@ -315,6 +315,7 @@ def create_subjects(subject_id, layout_df, layout, collection):
 def create_file(layout_df, BIDS_path, collection):
 
     file_repository = omcore.FileRepository(
+        format=omcore.ContentType.by_name("application/vnd.bids"),
         iri=IRI(pathlib.Path(BIDS_path).absolute().as_uri()))
     collection.add(file_repository)
     files_list = []
@@ -327,7 +328,7 @@ def create_file(layout_df, BIDS_path, collection):
         iri = IRI(pathlib.Path(path).absolute().as_uri())
         name = os.path.basename(path)
         hashes = file_hash(path)
-        storage_size = file_storage_size(path)
+        storage_size_obj, file_size = file_storage_size(path)
         if pd.isna(file["subject"]):
             if file["suffix"] == "participants":
                 if extension == ".json":
@@ -350,7 +351,7 @@ def create_file(layout_df, BIDS_path, collection):
             elif extension in [".nii", ".nii.gz"]:
                 content_description = f"Data file for {file['suffix']} of subject {file['subject']}"
                 data_types = controlled_terms.DataType.by_name("voxel data")
-                # file_format=omcore.ContentType.by_name("nifti")
+                file_format = detect_nifti_version(path, extension, file_size)
             elif extension == [".tsv"]:
                 if file["suffix"] == "events":
                     content_description = f"Event file for {file['suffix']} of subject {file['subject']}"
@@ -368,7 +369,7 @@ def create_file(layout_df, BIDS_path, collection):
             # is_part_of=file_bundels
             name=name,
             # special_usage_role
-            storage_size=storage_size,
+            storage_size=storage_size_obj,
         )
         collection.add(file)
         files_list.append(file)

bids2openminds/utility.py

@@ -2,16 +2,16 @@
 import json
 import os
 import re
+import gzip
 from warnings import warn
 
 import pandas as pd
 
 import openminds.latest.controlled_terms as controlled_terms
-from openminds.latest.core import Hash, QuantitativeValue
+from openminds.latest.core import Hash, QuantitativeValue, ContentType
 from openminds.latest.controlled_terms import UnitOfMeasurement
 
 
-
 def camel_to_snake(name):
     name = re.sub("(.)([A-Z][a-z]+)", r"\1_\2", name)
     return re.sub("([a-z0-9])([A-Z])", r"\1_\2", name).lower()
@@ -75,10 +75,11 @@ def openminds_instance(list: list, Terminologie: str = None, is_list: bool = Tru
     for item in list:
         if item.replace(" ", "")[0:32] == "@id:https://openminds.ebrains.eu":
             slash_location = item.rfind("/")
-            item_name = item[slash_location + 1 :]
+            item_name = item[slash_location + 1:]
             item_name_snake = camel_to_snake(item_name)
             if not (Terminologie):
-                Terminologie = item[item[:slash_location].rfind("/") + 1 : slash_location]
+                Terminologie = item[item[:slash_location].rfind(
+                    "/") + 1: slash_location]
                 Terminologie = Terminologie[0].upper() + Terminologie[1:]
             controlled_class = getattr(controlled_terms, Terminologie)
             openminds_item = getattr(controlled_class, item_name_snake)
@@ -140,5 +141,66 @@ def file_hash(file_path: str, algorithm: str = "MD5"):
 
 def file_storage_size(file_path: str):
     file_stats = os.stat(file_path)
-    file_size = QuantitativeValue(value=file_stats.st_size, unit=UnitOfMeasurement.by_name("byte"))
-    return file_size
+    file_size = QuantitativeValue(
+        value=file_stats.st_size, unit=UnitOfMeasurement.by_name("byte"))
+    return file_size, file_stats.st_size
+
+
+def detect_nifti_version(file_name, extension, file_size):
+
+    nii1_sizeof_hdr = 348
+    nii2_sizeof_hdr = 540
+
+    if extension == ".nii":
+
+        with open(file_name, 'rb') as fp:
+            byte_data = fp.read(4)
+
+        sizeof_hdr = int.from_bytes(byte_data, byteorder='little')
+
+        if sizeof_hdr == 0:
+            return None
+
+        if sizeof_hdr == nii1_sizeof_hdr:
+            return ContentType.by_name("application/vnd.nifti.1")
+
+        elif sizeof_hdr == nii2_sizeof_hdr:
+            return ContentType.by_name("application/vnd.nifti.2")
+
+        else:  # big endian
+            sizeof_hdr = int.from_bytes(byte_data, byteorder='big')
+
+            if sizeof_hdr == nii1_sizeof_hdr:
+                return ContentType.by_name("application/vnd.nifti.1")
+
+            elif sizeof_hdr == nii2_sizeof_hdr:
+                return ContentType.by_name("application/vnd.nifti.2")
+
+    if extension == ".nii.gz":
+        try:
+            with gzip.open(file_name, 'rb') as fp:
+                byte_data = fp.read(4)
+        except gzip.BadGzipFile:
+            return None
+
+        sizeof_hdr = int.from_bytes(byte_data, byteorder='little')
+
+        if sizeof_hdr == 0:
+            return None
+
+        if sizeof_hdr == nii1_sizeof_hdr:
+            return ContentType.by_name("application/vnd.nifti.1")
+
+        elif sizeof_hdr == nii2_sizeof_hdr:
+            return ContentType.by_name("application/vnd.nifti.2")
+
+        else:  # big endian
+            sizeof_hdr = int.from_bytes(byte_data, byteorder='big')
+
+            if sizeof_hdr == nii1_sizeof_hdr:
+                return ContentType.by_name("application/vnd.nifti.1")
+
+            elif sizeof_hdr == nii2_sizeof_hdr:
+                return ContentType.by_name("application/vnd.nifti.2")
+
+    return None

test/test_example_datasets_click.py

@@ -3,7 +3,7 @@
 from bids2openminds.converter import convert_click
 from click.testing import CliRunner
 
-(test_data_set, number_of_openminds_files) = ("ds003", 100)
+(test_data_set, number_of_openminds_files) = ("ds003", 101)
 
 
 def test_example_datasets_click():