Merge pull request #114 from IDEA-Research/dev

release: 0.9.2
IDEA-Research · Dec 13, 2023 · 4b475a6 · 4b475a6
2 parents 8a7a014 + 6112526
commit 4b475a6
Show file tree

Hide file tree

Showing 4 changed files with 35 additions and 31 deletions.
diff --git a/deepdataspace/model/dataset.py b/deepdataspace/model/dataset.py
@@ -439,3 +439,26 @@ def eval_description(self):
             except:
                 logger.warning(f"Failed to eval description_func[{self.description_func}] for dataset[{self.id}]")
                 return self.description or self.path
+
+    @staticmethod
+    def cascade_delete(dataset: "DataSet"):
+        """
+        Cascade delete the dataset, along with all its images, labels, categories and objects.
+        """
+        if dataset is None:
+            return
+
+        dataset_id = dataset.id
+        print(f"dataset [{dataset_id}] is found, deleting...")
+
+        print(f"dataset [{dataset_id}] is found, deleting categories...")
+        Category.delete_many({"dataset_id": dataset_id})
+
+        print(f"dataset [{dataset_id}] is found, deleting labels...")
+        Label.delete_many({"dataset_id": dataset_id})
+
+        print(f"dataset [{dataset_id}] is found, deleting images...")
+        Image(dataset_id).get_collection().drop()
+
+        DataSet.delete_many({"id": dataset_id})
+        print(f"dataset [{dataset_id}] is deleted.")
diff --git a/deepdataspace/scripts/dataset_cmds.py b/deepdataspace/scripts/dataset_cmds.py
@@ -5,7 +5,6 @@
 """
 
 import os
-import shutil
 
 import pkg_resources
 
@@ -17,10 +16,7 @@
 @ddsop.command("delete_one", help="Delete a dataset.")
 @click.argument("dataset_dir")
 def delete_one(dataset_dir):
-    from deepdataspace.model import Image
-    from deepdataspace.model import Label
     from deepdataspace.model import DataSet
-    from deepdataspace.model import Category
     from deepdataspace.utils.string import get_str_md5
 
     dataset_dir = os.path.abspath(dataset_dir)
@@ -30,20 +26,7 @@ def delete_one(dataset_dir):
         print(f"dataset [{dataset_dir}] is not imported before, skip...")
         return
 
-    dataset_id = dataset.id
-    print(f"dataset [{dataset_id}] is found, deleting...")
-
-    print(f"dataset [{dataset_id}] is found, deleting categories...")
-    Category.delete_many({"dataset_id": dataset_id})
-
-    print(f"dataset [{dataset_id}] is found, deleting labels...")
-    Label.delete_many({"dataset_id": dataset_id})
-
-    print(f"dataset [{dataset_id}] is found, deleting images...")
-    Image(dataset_id).get_collection().drop()
-
-    DataSet.delete_many({"id": dataset_id})
-    print(f"dataset [{dataset_id}] is deleted.")
+    DataSet.cascade_delete(dataset)
 
 
 @ddsop.command("import_all", help="Trigger a background task of importing all datasets in a data dir.")

diff --git a/deepdataspace/services/dds.py b/deepdataspace/services/dds.py
@@ -170,6 +170,7 @@ def _init_shared_files_and_dirs(self):
         if self.quickstart is True:
             if self.data_dir is None:
                 self.data_dir = os.path.join(config.RUNTIME_DIR, "datasets")
+                os.makedirs(self.data_dir, exist_ok=True)
             self.init_samples()
         config.DATA_DIR = self.data_dir
 
@@ -315,20 +316,17 @@ def _init_shared_libs(self):
             url = f"{self.dl_prefix}/lib/libcurl.so.4.8.0"
             download_by_requests(url, config.SHARED_CURL_LIB)
 
-        if config.LD_LIBRARY_PATH is None:
-            config.LD_LIBRARY_PATH = config.SHARED_LIB_DIR
-        elif config.SHARED_LIB_DIR is not None and config.LD_LIBRARY_PATH != config.SHARED_LIB_DIR:
-            config.LD_LIBRARY_PATH = f"{config.SHARED_LIB_DIR}:{config.LD_LIBRARY_PATH}"
+        paths = {config.SHARED_LIB_DIR}
+        if config.LD_LIBRARY_PATH:
+            for path in config.LD_LIBRARY_PATH.split(":"):
+                paths.add(path)
 
-        ld_path = os.environ.get("LD_LIBRARY_PATH", None)
-        if ld_path is not None:
-            config.LD_LIBRARY_PATH = f"{ld_path}:{config.LD_LIBRARY_PATH}"
-
-        current_lib_paths = find_shared_dirs_on_ubuntu()
-        current_lib_paths = ":".join(current_lib_paths)
-        if current_lib_paths:
-            config.LD_LIBRARY_PATH = f"{current_lib_paths}:{config.LD_LIBRARY_PATH}"
+        lib_paths = find_shared_dirs_on_ubuntu()
+        for path in lib_paths:
+            paths.add(path)
 
+        ld_library_path = ":".join(paths)
+        config.LD_LIBRARY_PATH = ld_library_path
         os.environ["LD_LIBRARY_PATH"] = config.LD_LIBRARY_PATH
 
     def _init_sentry(self):

diff --git a/setup.py b/setup.py
@@ -2,7 +2,7 @@
 
 from setuptools import setup
 
-version = "0.9.1"
+version = "0.9.2"
 
 description = "A tool for CV dataset labeling, visualizing and analysing"
 with open("README.md", "r", encoding="utf8") as fp: