feat: add circle archiving wizard GUI

Author: paulmueller

CHANGELOG

@@ -1,5 +1,6 @@
 1.1.0
  - feat: add circle archiving in the BagIt (RFC 8493) format
+ - feat: add circle archiving wizard GUI
  - fix: handle resource downloads whose target directory got deleted
  - fix: check for existence of persistent job list in upload
  - enh: allow to programmatically abort download jobs

dcoraid/bagit/archive.py

@@ -60,12 +60,14 @@ def bag_circle(api: CKANAPI,
     # and if yes, compute the MD5 hash and compare it. If the comparison
     # fails, then the user has to choose a different `target_path`, because
     # we cannot guarantee data integrity.
+    target_path.mkdir(parents=True, exist_ok=True)
     circle_jsonlines_path = target_path / "circle.jsonlines"
     if circle_jsonlines_path.exists():
         lines = circle_jsonlines_path.read_text().split("\n")
         hasher2 = hashlib.sha256()
         for line in lines:
-            hasher2.update(json.loads(line)["id"].encode(encoding="utf-8"))
+            if line.strip():
+                hasher2.update(json.loads(line)["id"].encode(encoding="utf-8"))
         sha256_hash2 = hasher2.hexdigest()
         if sha256_hash != sha256_hash2:
             raise ValueError(
@@ -87,8 +89,8 @@ def bag_circle(api: CKANAPI,
         if callback:
             callback(ii / num_datasets)
 
-        dataset_index = ii+1
-        prefix = str(dataset_index).zfill(max_digits)
+        dataset_index = ii + 1
+        prefix = str(dataset_index).zfill(max_digits+1)
         bag_path = target_path / f"{prefix}_{ds_dict['name']}"
 
         if not manifest.is_bagged(bag_path):
@@ -148,14 +150,24 @@ def bag_dataset(api: CKANAPI,
         download_resource(api=api,
                           bag_path=bag_path,
                           res_dict=res_dict,
-                          condensed=False)
+                          condensed=False,
+                          abort_event=abort_event,
+                          )
+
+        if abort_event is not None and abort_event.is_set():
+            return
 
         # condensed resource
         if res_dict["name"].endswith(".rtdc"):
             download_resource(api=api,
                               bag_path=bag_path,
                               res_dict=res_dict,
-                              condensed=True)
+                              condensed=True,
+                              abort_event=abort_event,
+                              )
+
+    if abort_event is not None and abort_event.is_set():
+        return
 
     # create BagIt files
     info.write_bag_info(bag_path=bag_path,
@@ -171,7 +183,9 @@ def bag_dataset(api: CKANAPI,
 def download_resource(api: CKANAPI,
                       bag_path: pathlib.Path,
                       res_dict: dict,
-                      condensed: bool):
+                      condensed: bool,
+                      abort_event: threading.Event = None,
+                      ):
     """Download and verify a resource from DCOR
 
     Parameters
@@ -183,6 +197,8 @@ def download_resource(api: CKANAPI,
         CKAN resource dictionary
     condensed
         Whether to download the condensed resource (or the original resource)
+    abort_event
+        For stopping the download process prematurely
     """
     data_path = bag_path / "data"
     data_path.mkdir(parents=True, exist_ok=True)
@@ -192,7 +208,14 @@ def download_resource(api: CKANAPI,
     dj = DownloadJob(api=api,
                      resource_id=res_dict["id"],
                      download_path=dl_path,
-                     condensed=condensed
+                     condensed=condensed,
                      )
-    dj.task_download_resource()
+    if abort_event is not None and abort_event.is_set():
+        return
+
+    dj.task_download_resource(abort_event=abort_event)
+
+    if abort_event is not None and abort_event.is_set():
+        return
+
     dj.task_verify_resource()

dcoraid/gui/main.py

@@ -25,7 +25,8 @@
 from .preferences import PreferencesDialog
 from .status_widget import StatusWidget
 from . import updater
-from .wizard_init import SetupWizard
+from .wizard_bagit import BagItWizard
+from .wizard_init import InitWizard
 
 
 file_manager = ExitStack()
@@ -100,7 +101,8 @@ def __init__(self, *args, **kwargs):
         self.menubar.setNativeMenuBar(False)
         # File menu
         self.actionPreferences.triggered.connect(self.dlg_pref.show)
-        self.actionSetupWizard.triggered.connect(self.on_wizard)
+        self.actionInitWizard.triggered.connect(self.on_wizard_init)
+        self.actionBagItWizard.triggered.connect(self.on_wizard_bagit)
         # Help menu
         self.actionSoftware.triggered.connect(self.on_action_software)
         self.actionAbout.triggered.connect(self.on_action_about)
@@ -361,8 +363,13 @@ def on_progress_db_update(self, data):
             self.logger.error("Progress dialog not defined")
 
     @QtCore.pyqtSlot()
-    def on_wizard(self):
-        self.wizard_init = SetupWizard(self)
+    def on_wizard_bagit(self):
+        self.wizard_bagit = BagItWizard(self)
+        self.wizard_bagit.exec()
+
+    @QtCore.pyqtSlot()
+    def on_wizard_init(self):
+        self.wizard_init = InitWizard(self)
         self.wizard_init.exec()
 
 

dcoraid/gui/main.ui

@@ -181,7 +181,8 @@
     <property name="title">
      <string>File</string>
     </property>
-    <addaction name="actionSetupWizard"/>
+    <addaction name="actionInitWizard"/>
+    <addaction name="actionBagItWizard"/>
     <addaction name="separator"/>
     <addaction name="actionPreferences"/>
     <addaction name="separator"/>
@@ -207,8 +208,7 @@
   </action>
   <action name="actionQuit">
    <property name="icon">
-    <iconset theme="times-circle">
-     <normaloff>../../../../.designer/backup</normaloff>../../../../.designer/backup</iconset>
+    <iconset theme="times-circle"/>
    </property>
    <property name="text">
     <string>Quit</string>
@@ -221,8 +221,7 @@
   </action>
   <action name="actionPreferences">
    <property name="icon">
-    <iconset theme="cogs">
-     <normaloff>../../../../.designer/backup</normaloff>../../../../.designer/backup</iconset>
+    <iconset theme="cogs"/>
    </property>
    <property name="text">
     <string>Preferences...</string>
@@ -237,15 +236,22 @@
     <string>About</string>
    </property>
   </action>
-  <action name="actionSetupWizard">
+  <action name="actionInitWizard">
    <property name="icon">
-    <iconset theme="hat-wizard">
-     <normaloff>../../../../.designer/backup</normaloff>../../../../.designer/backup</iconset>
+    <iconset theme="hat-wizard"/>
    </property>
    <property name="text">
     <string>Setup wizard...</string>
    </property>
   </action>
+  <action name="actionBagItWizard">
+   <property name="icon">
+    <iconset theme="folder-tree"/>
+   </property>
+   <property name="text">
+    <string>Archive to BagIt format...</string>
+   </property>
+  </action>
  </widget>
  <customwidgets>
   <customwidget>

dcoraid/gui/settings.py

@@ -0,0 +1,41 @@
+import pathlib
+from PyQt6 import QtCore
+
+
+def get_dir(topic: str,
+            settings: QtCore.QSettings
+            ) -> str:
+    """Given a QSettings instance, return a directory
+
+    If writable is `True`, return a writable directory
+    """
+    location = settings.value(f"paths/{topic}", "")
+
+    if not location:
+        # use default value
+        location = QtCore.QStandardPaths.standardLocations(
+            QtCore.QStandardPaths.StandardLocation.HomeLocation)[0]
+    else:
+        # check whether the location exists
+        path = pathlib.Path(location)
+        if not path.is_dir():
+            for pp in path.parents:
+                if pp.is_dir():
+                    location = str(pp)
+                    break
+            else:
+                location = "."
+
+    return location or "."
+
+
+def set_dir(topic: str,
+            path: pathlib.Path | str,
+            settings: QtCore.QSettings,
+            ):
+    """Given a QSettings instance, save a directory topic"""
+    path = pathlib.Path(path)
+    if path.exists():
+        if not path.is_dir():
+            path = path.parent
+        settings.setValue(f"paths/{topic}", str(path))