--- a/PSKManagement.py	Tue Nov 20 11:32:42 2018 +0100
+++ b/PSKManagement.py	Wed Nov 21 14:10:51 2018 +0100
@@ -7,7 +7,10 @@
 import os
 import time
 import json
+from zipfile import ZipFile
 
+# PSK Management Data model :
+# [[ID,Desc, LastKnownURI, LastConnect]]
 COL_ID,COL_URI,COL_DESC,COL_LAST = range(4)
 
 def _pskpath(project_path):
@@ -16,51 +19,52 @@
 def _mgtpath(project_path):
     return os.path.join(_pskpath(project_path), 'management.json')
 
-def _default():
-    return ['', # default description
+def _default(ID):
+    return [ID,
+            '', # default description
             None, # last known URI
             None]  # last connection date
 
+def _dataByID(data):
+    return {row[COL_ID]:row for row in data}
+
 def _LoadData(project_path):
+    """ load known keys metadata """
     if os.path.isdir(_pskpath(project_path)):
         _path = _mgtpath(project_path)
-        # load known keys metadata
-        # {ID:(Desc, LastKnownURI, LastConnect)}
-        return json.loads(open(_path).read()) \
-               if os.path.exists(_path) else {}
-    return {}
+        if os.path.exists(_path):
+            return json.loads(open(_path).read())
+    return []
 
-def GetData(project_path):
-    # [(ID, Desc, LastKnownURI, LastConnect)
-    data = []
-    loaded_data = _LoadData(project_path)
+def _filterData(psk_files, data_input):
+    input_by_ID = _dataByID(data_input)
+    output = []
     # go through all secret files available an build data
     # out of data recoverd from json and list of secret.
     # this implicitly filters IDs out of metadata who's
     # secret is missing
-    psk_files = os.listdir(_pskpath(project_path))
     for filename in psk_files:
        if filename.endswith('.secret'):
            ID = filename[:-7]  # strip filename extension
-           meta = loaded_data.get(ID,_default())                  
-           data.append([ID]+meta)
-    return data
+           output.append(input_by_ID.get(ID,_default(ID)))
+    return output
 
+def GetData(project_path):
+    loaded_data = _LoadData(project_path)
+    psk_files = os.listdir(_pskpath(project_path))
+    return _filterData(psk_files, loaded_data)
 
 def DeleteID(project_path, ID):
     secret_path = os.path.join(_pskpath(project_path), ID+'.secret')
     os.remove(secret_path)
 
-def _StoreData(project_path, data):
+def SaveData(project_path, data):
     pskpath = _pskpath(project_path)
     if not os.path.isdir(pskpath):
         os.mkdir(pskpath)
     with open(_mgtpath(project_path), 'w') as f:
         f.write(json.dumps(data))
 
-def SaveData(project_path, data):
-    to_store = {row[0]:row[1:] for row in data}
-    _StoreData(project_path, to_store)
 
 def UpdateID(project_path, ID, secret, URI):
     pskpath = _pskpath(project_path)
@@ -71,11 +75,49 @@
     with open(secpath, 'w') as f:
         f.write(ID+":"+secret)
 
+    # here we directly use _LoadData, avoiding filtering that could be long
     data = _LoadData(project_path)
-    dataForID = [ID] + (data.get(ID, _default()) if data else _default())
+    idata = _dataByID(data)
+    dataForID = idata.get(ID, _default(ID)) if data else _default(ID)
     dataForID[COL_URI] = URI
     # FIXME : could store time instead os a string and use DVC model's cmp 
     # then date display could be smarter, etc - sortable sting hack for now
     dataForID[COL_LAST] = time.strftime('%y/%M/%d-%H:%M:%S')
-    data[ID] = dataForID[1:]
-    _StoreData(project_path, data)
+    SaveData(project_path, data)
+
+def ExportIDs(project_path, export_zip):
+    with ZipFile(export_zip, 'w') as zf:
+        path = _pskpath(project_path)
+        for nm in os.listdir(path):
+            if nm.endswith('.secret') or nm == 'management.json':
+                zf.write(os.path.join(path, nm), nm)
+
+def ImportIDs(project_path, import_zip, should_I_replace_callback):
+    data = GetData(project_path)
+
+    zip_loaded_data = json.loads(zf.open('management.json').read())
+    name_list = zf.namelist()
+    zip_filtered_data = _filterData(name_list, loaded_data)
+
+    idata = _dataByID(data)
+
+    for imported_row in zip_filtered_data:
+        ID = imported_row[COL_ID]
+        existing_row = idata.get(ID, None)
+        if existing_row is None:
+            data.append(imported_row)
+        else:
+            # callback returns the selected list for merge or none if canceled
+            result = should_I_replace_callback(existing_row, imported_row)
+            if result is None:
+                break
+            
+            if result: 
+                # replace with imported
+                existing_row[:] = imported_row
+                # copy the key of selected
+                self.extract(ID+".secret", _pskpath(project_path))
+
+    SaveData(project_path, data)
+
+