Treenode export: add more metadata to treenode archive exports

The ZIP files generated by the treenode archive export contain now two
metadata files: nodes.csv and metadata.json. The first one was called
metadata.csv before. The second one is new and contains information on
the project, stacks and exporting user.

Author: tomka

CHANGELOG.md

@@ -213,6 +213,11 @@ Import/Export widget:
 - The treenode archive creation dialog allows now also to specify the Z radius.
   This allows to export an image block for each treenode.
 
+- The ZIP files generated by the treenode archive export contain now two
+  metadata files: nodes.csv and metadata.json. The first one was called
+  metadata.csv before. The second one is new and contains information on the
+  project, stacks and exporting user.
+
 Miscellaneous:
 
 - The built-in API docs (/apis endpoint) now supports deep links, which allows

django/applications/catmaid/control/treenodeexport.py

@@ -15,8 +15,8 @@
 from catmaid.control.common import get_relation_to_id_map, id_generator, get_request_bool
 from catmaid.control.cropping import (collect_stack_mirros, CropJob,
                                       extract_substack, ImageRetrievalError)
-from catmaid.models import ClassInstanceClassInstance, TreenodeConnector, \
-        Message, User, UserRole, Stack, Treenode
+from catmaid.models import (ClassInstanceClassInstance, TreenodeConnector,
+                            Message, Project, User, UserRole, Stack, Treenode)
 
 from celery import shared_task
 
@@ -196,8 +196,12 @@ def export_single_node(self, treenode) -> None:
 
     def post_process(self, nodes) -> None:
         """ Create a meta data file for all the nodes passed (usually all of the
-        ones queries before). This file is a table with the following columns:
+        ones queries before). This file is called nodes.csv and is a table with
+        the following columns:
         <treenode id> <parent id> <#presynaptic sites> <#postsynaptic sites> <x> <y> <z>
+
+        Additionally, a file metadata.json is created, which contains general
+        information on the referenced dataset.
         """
         # Get pre- and post synaptic sites
         presynaptic_to_rel = self.relation_map['presynaptic_to']
@@ -236,15 +240,26 @@ def post_process(self, nodes) -> None:
             ls.append(line)
 
         # Save metdata for each skeleton to files
+        project = Project.objects.get(id=self.job.project_id)
+        image_stacks = list(map(lambda s: s.title, project.stacks.all()))
         for skid, metadata in skid_to_metadata.items():
             path = self.skid_to_neuron_folder[skid]
-            with open(os.path.join(path, 'metadata.csv'), 'w') as f:
+            with open(os.path.join(path, 'nodes.csv'), 'w') as f:
                 f.write("This CSV file contains meta data for CATMAID skeleton " \
-                        "%s. The columns represent the following data:\n" % skid)
+                        f"{skid}. The columns represent the following data:\n")
                 f.write("treenode-id, parent-id, # presynaptic sites, " \
                         "# postsynaptic sites, x, y, z\n")
                 for line in metadata:
-                    f.write("%s\n" % line)
+                    f.write(f"{line}\n")
+            with open(os.path.join(path, 'metadata.json'), 'w') as f:
+                f.write(json.dumps({
+                    'project_id': project.id,
+                    'project_name': project.title,
+                    'skeleton_id': skid,
+                    'user_id': self.job.user.id,
+                    'user_name': self.job.user.username,
+                    'image_stacks': image_stacks,
+                }, indent=4))
 
 class ConnectorExporter(TreenodeExporter):
     """ Most of the infrastructure can be used for both treenodes and

django/applications/catmaid/static/js/widgets/import-export-widget.js

@@ -73,12 +73,14 @@ the selected neurons.
     The generated <em>tar.gz</em> archive contains one folder for every
     selected neuron, named after its ID. Such a folder contains image files for
     every treenode of the neuron's skeleton(s), named <em>treenode-id.tiff</em>.
-    Along those files a meta data file, named <em>metadata.csv</em>, is created.
-    It contains a table with meta data for every treenode ID (first column). The
-    remaining columns are <em>parent-id</em>, <em># presynaptic sites</em>,
-    <em># postsynaptic sites</em>, <em>x</em>, <em>y</em> and <em>z</em>. The
-    root node has no parent and its entry will have <em>null</em> in the
-    corresponding field in the meta data file.
+    Along those files two meta data files are created, named <em>nodes.csv</em>
+    and <em>metadata.json</em>. The former contains a table with meta data for
+    every treenode ID (first column). The remaining columns are
+    <em>parent-id</em>, <em># presynaptic sites</em>, <em># postsynaptic
+    sites</em>, <em>x</em>, <em>y</em> and <em>z</em>. The root node has no
+    parent and its entry will have <em>null</em> in the corresponding field in
+    the meta data file. The latter metadata file contains information on the
+    project, image stacks and exporting user.
   </li>
   <li>
     <a id='export-connector-archive' href='#'>