Merge pull request #43 from brain-bican/schema_title

CAS json export list values fixes

Author: hkir-dev

setup.py

@@ -8,7 +8,7 @@
 
 setup(
     name="tdta",
-    version="0.1.0.dev13",
+    version="0.1.0.dev14",
     description="The aim of this project is to provide taxonomy development tools custom actions.",
     long_description=README,
     long_description_content_type="text/markdown",

src/tdta/tdt_export.py

@@ -2,6 +2,8 @@
 import sqlite3
 import ast
 import json
+import typing
+from typing import Union, List
 from contextlib import closing
 from pathlib import Path
 from datetime import datetime
@@ -193,6 +195,7 @@ def parse_review_data(cta, sqlite_db, table_name):
                             else:
                                 filtered_annotations[0].reviews = [ar]
 
+
 def get_table_names(sqlite_db):
     """
     Queries 'table' table to get all CAS related table names
@@ -222,7 +225,7 @@ def auto_fill_object_from_row(obj, columns, row):
         if hasattr(obj, column):
             value = row[columns.index(column)]
             if value:
-                if isinstance(type(getattr(obj, column)), list):
+                if is_list(obj, column):
                     if value.strip().startswith("\"") and value.strip().endswith("\""):
                         value = value.strip()[1:-1].strip()
                     elif value.strip().startswith("'") and value.strip().endswith("'"):
@@ -245,3 +248,24 @@ def auto_fill_object_from_row(obj, columns, row):
                 if msg["column"] in columns:
                     setattr(obj, msg["column"], msg["value"])
 
+
+def is_list(obj, field):
+    """
+    Checks if the field of the object is a list or has list typing.
+    Parameters:
+        obj: object
+        field: field name
+    Returns: True if the field is a list or has list typing, False otherwise
+    """
+    is_list_instance = isinstance(getattr(obj, field), list)
+    type_hint = typing.get_type_hints(obj).get(field)
+    if type_hint:
+        # is List
+        is_list_typing = typing.get_origin(type_hint) is list
+        # is Optional[List] or is Optional[List[str]]
+        is_optional_list_typed = typing.get_origin(type_hint) is Union and any(typing.get_origin(e) is list for e in typing.get_args(type_hint))
+
+        is_list_typing = is_list_typing or is_optional_list_typed
+    else:
+        is_list_typing = False
+    return is_list_instance or is_list_typing

src/test/tdt_export_test.py

@@ -1,11 +1,12 @@
 import unittest
 import os
-import shutil
-from tdta.tdt_export import export_cas_data
+from typing import List, Optional
+from tdta.tdt_export import export_cas_data, is_list
 
 TEST_DATA_FOLDER = os.path.join(os.path.dirname(os.path.realpath(__file__)), "./test_data/")
 TEST_OUTPUT = os.path.join(TEST_DATA_FOLDER, "cas_output.json")
 TEST_DB = os.path.join(TEST_DATA_FOLDER, "nanobot.db")
+TEST_DB_SILETTI = os.path.join(TEST_DATA_FOLDER, "nanobot_siletti_nn.db")
 
 
 class CASExportTests(unittest.TestCase):
@@ -63,3 +64,60 @@ def test_export(self):
         # self.assertEqual('incorrect', test_annotation["reviews"][0]['explanation'])
         # print(test_annotation["reviews"][0]['datestamp'])
         # self.assertEqual('2024-05-29T08:10:11.126Z', test_annotation["reviews"][0]['datestamp'])
+
+    def test_export_siletti_nn(self):
+        cas = export_cas_data(TEST_DB_SILETTI, TEST_OUTPUT, TEST_DATA_FOLDER)
+        self.assertTrue(os.path.exists(TEST_OUTPUT))
+
+        self.assertTrue(cas)
+        result = cas.to_dict()
+
+        self.assertTrue("annotations" in result)
+        test_annotation = [x for x in result["annotations"] if x["cell_label"] == "VendC_15"][0]
+        self.assertEqual("CS202210140_16", test_annotation["cell_set_accession"])
+        self.assertTrue("marker_gene_evidence" in test_annotation)
+        print(test_annotation["marker_gene_evidence"])
+        self.assertEqual(4, len(test_annotation["marker_gene_evidence"]))
+        self.assertTrue("FLT1" in test_annotation["marker_gene_evidence"])
+        self.assertTrue("CLDN5" in test_annotation["marker_gene_evidence"])
+        self.assertTrue("PECAM1" in test_annotation["marker_gene_evidence"])
+        self.assertTrue("MFSD2A" in test_annotation["marker_gene_evidence"])
+
+    def test_list_instance(self):
+        samples = AttributeSamples()
+
+        import inspect
+        members = inspect.getmembers(samples)
+
+        for name, value in members:
+            if not name.startswith('__'):  # Skip built-in attributes
+                print(name, value, type(value))
+
+        self.assertTrue(is_list(samples, "field1"))
+        self.assertTrue(is_list(samples, "field2"))
+        self.assertTrue(is_list(samples, "field3"))
+        self.assertTrue(is_list(samples, "field4"))
+        self.assertTrue(is_list(samples, "field5"))
+        self.assertTrue(is_list(samples, "field6"))
+        self.assertTrue(is_list(samples, "field7"))
+
+        self.assertFalse(is_list(samples, "field8"))
+        self.assertFalse(is_list(samples, "field9"))
+        self.assertFalse(is_list(samples, "field10"))
+
+
+class AttributeSamples:
+
+    field1 = list()
+    field2: Optional[List[str]] = None
+    field3: List[str] = []
+    field4: List[str] = ["a", "b", "c"]
+    field5: Optional[List[str]] = ["a", "b", "c"]
+    field6: Optional[List[int]] = None
+    field7: List = None
+
+    field8: str = None
+    field9: Optional[str] = None
+    field10 = None
+
+