[#151] Basic XML to spreadsheet conversion

Author: Bjwebb

flattentool/__init__.py

@@ -41,7 +41,8 @@ def spreadsheet_output(spreadsheet_output_class, name):
         raise Exception('The requested format is not available')
 
 
-def flatten(input_name, schema=None, output_name='flattened', output_format='all', main_sheet_name='main', root_list_path='main', rollup=False, root_id=None, use_titles=False, **_):
+def flatten(input_name, schema=None, output_name='flattened', output_format='all', main_sheet_name='main',
+            root_list_path='main', rollup=False, root_id=None, use_titles=False, xml=False, id_name='id',  **_):
     """
     Flatten a nested structure (JSON) to a flat structure (spreadsheet - csv or xlsx).
 
@@ -61,7 +62,9 @@ def flatten(input_name, schema=None, output_name='flattened', output_format='all
         root_list_path=root_list_path,
         schema_parser=schema_parser,
         root_id=root_id,
-        use_titles=use_titles)
+        use_titles=use_titles,
+        xml=xml,
+        id_name=id_name)
     parser.parse()
 
     def spreadsheet_output(spreadsheet_output_class, name):

flattentool/cli.py

@@ -68,6 +68,13 @@ def create_parser():
     parser_flatten.add_argument(
         "-f", "--output-format",
         help="Type of template you want to create. Defaults to all available options")
+    parser_flatten.add_argument(
+        "--xml",
+        action='store_true',
+        help="Use XML as the output format")
+    parser_flatten.add_argument(
+        "--id-name",
+        help="String to use for the identifier key, defaults to 'id'")
     parser_flatten.add_argument(
         "-m", "--main-sheet-name",
         help="The name of the main sheet, as seen in the first tab of the spreadsheet for example. Defaults to main")

flattentool/json_input.py

@@ -15,6 +15,7 @@
 from flattentool.sheet import Sheet
 from warnings import warn
 import codecs
+import xmltodict
 
 BASIC_TYPES = [six.text_type, bool, int, Decimal, type(None)]
 
@@ -45,12 +46,14 @@ class JSONParser(object):
     # Named for consistency with schema.SchemaParser, but not sure it's the most appropriate name.
     # Similarily with methods like parse_json_dict
 
-    def __init__(self, json_filename=None, root_json_dict=None, schema_parser=None, root_list_path=None, root_id='ocid', use_titles=False):
+    def __init__(self, json_filename=None, root_json_dict=None, schema_parser=None, root_list_path=None,
+                 root_id='ocid', use_titles=False, xml=False, id_name='id'):
         self.sub_sheets = {}
         self.main_sheet = Sheet()
         self.root_list_path = root_list_path
         self.root_id = root_id
         self.use_titles = use_titles
+        self.id_name = id_name
         if schema_parser:
             self.main_sheet = schema_parser.main_sheet
             self.sub_sheets = schema_parser.sub_sheets
@@ -60,6 +63,11 @@ def __init__(self, json_filename=None, root_json_dict=None, schema_parser=None,
         else:
             self.rollup = False
 
+        if xml:
+            with codecs.open(json_filename, 'rb') as xml_file:
+                root_json_dict = xmltodict.parse(xml_file, cdata_key='')['iati-activities']
+            json_filename = None
+
         if json_filename is None and root_json_dict is None:
             raise ValueError('Etiher json_filename or root_json_dict must be supplied')
 
@@ -114,8 +122,8 @@ def parse_json_dict(self, json_dict, sheet, json_key=None, parent_name='', flatt
         if self.root_id and self.root_id in json_dict:
             parent_id_fields[sheet_key(sheet, self.root_id)] = json_dict[self.root_id]
 
-        if 'id' in json_dict:
-            parent_id_fields[sheet_key(sheet, parent_name+'id')] = json_dict['id']
+        if self.id_name in json_dict:
+            parent_id_fields[sheet_key(sheet, parent_name+self.id_name)] = json_dict[self.id_name]
 
 
         for key, value in json_dict.items():

setup.py

@@ -1,7 +1,7 @@
 from setuptools import setup
 import sys
 
-install_requires = ['jsonref', 'schema', 'openpyxl>=2,<2.4', 'six', 'pytz']
+install_requires = ['jsonref', 'schema', 'openpyxl>=2,<2.4', 'six', 'pytz', 'xmltodict']
 
 if sys.version < '3':
     install_requires.append('unicodecsv')