Several fixes. Multiple namespaces for 'rep'

ingestor/ingestor/browse_report.py

@@ -34,7 +34,8 @@ from .util import pairwise
 class BrowseReportParserError(ValueError):
     pass
 
-NS_REP = 'http://ngeo.eo.esa.int/schema/browseReport'
+NS_REP = 'http://ngeo.eo.esa.int/ngEO/browseReport/1.0'
+NS_REP_OLD = 'http://ngeo.eo.esa.int/schema/browseReport'
 NS_BSI = 'http://ngeo.eo.esa.int/schema/browse/ingestion'
 
 
@@ -47,11 +48,16 @@ nsmap = {
 def rep(tag):
     return f'{{{NS_REP}}}{tag}'
 
+def rep_old(tag):
+    return f'{{{NS_REP_OLD}}}{tag}'
 
 def bsi(tag):
     return f'{{{NS_BSI}}}{tag}'
 
 
+ALLOWED_ROOT_TAGS = {rep('browseReport'), rep_old('browseReport'), bsi('ingestBrowse')}
+
+
 def parse_browse_report(input_file):
     """
         :returns: list of browses
@@ -62,72 +68,77 @@ def parse_browse_report(input_file):
         raise BrowseReportParserError('Failed to parse XML.') from e
     root = tree.getroot()
 
-    if not root.tag in [rep('browseReport'), bsi('ingestBrowse')]:
+    if not root.tag in ALLOWED_ROOT_TAGS:
         raise BrowseReportParserError(
             'Document is not a browse report or an ingest browse instruction.'
         )
 
+    if root.tag == rep_old('browseReport'):
+        used_rep = rep_old
+    else:
+        used_rep = rep
+
     return {
-        'responsible_or_name': root.find(rep('responsibleOrgName')),
-        'date_time': dateutil.parser.parse(root.find(rep('dateTime'))),
-        'browse_type': root.findtext(rep('browseType')),
+        'responsible_org_name': root.findtext(used_rep('responsibleOrgName')),
+        'date_time': dateutil.parser.parse(root.findtext(used_rep('dateTime'))),
+        'browse_type': root.findtext(used_rep('browseType')),
         'browses': [
-            parse_browse(elem)
-            for elem in root.iterfind(rep('browse'))
+            parse_browse(elem, used_rep)
+            for elem in root.iterfind(used_rep('browse'))
         ],
     }
 
 
-def parse_browse(elem):
+def parse_browse(elem, used_rep):
     browse = {
         'type': '',
-        'filename': elem.findtext(rep('fileName')),
-        'image_type': elem.findtext(rep('imageType')),
+        'browse_identifier': elem.findtext(used_rep('browseIdentifier')),
+        'filename': elem.findtext(used_rep('fileName')),
+        'image_type': elem.findtext(used_rep('imageType')),
         'reference_system_identifier': elem.findtext(
-            rep('referenceSystemIdentifier')
+            used_rep('referenceSystemIdentifier')
         ),
-        'start_time': dateutil.parser.parse(elem.findtext(rep('startTime'))),
-        'end_time': dateutil.parser.parse(elem.findtext(rep('endTime'))),
+        'start_time': dateutil.parser.parse(elem.findtext(used_rep('startTime'))),
+        'end_time': dateutil.parser.parse(elem.findtext(used_rep('endTime'))),
     }
 
-    rectified_elem = elem.find(rep('rectifiedBrowse'))
-    footprint_elem = elem.find(rep('footprint'))
-    geotiff_elem = elem.find(rep('modelInGeotiff'))
-    regular_grid_browse = elem.find(rep('modelInGeotiff'))
+    rectified_elem = elem.find(used_rep('rectifiedBrowse'))
+    footprint_elem = elem.find(used_rep('footprint'))
+    geotiff_elem = elem.find(used_rep('modelInGeotiff'))
+    regular_grid_browse = elem.find(used_rep('modelInGeotiff'))
 
-    if rectified_elem:
+    if rectified_elem is not None:
         browse['type'] = 'rectified_browse'
         browse['rectified'] = {
             'coord_list': [
                 (float(x), float(y))
                 for x, y in pairwise(
-                    footprint_elem.findtext(rep('coordList')).split()
+                    rectified_elem.findtext(used_rep('coordList')).split()
                 )
             ],
         }
 
-    elif footprint_elem:
+    elif footprint_elem is not None:
         browse['type'] = 'footprint_browse'
         browse['footprint'] = {
             'col_row_list': [
                 (int(x), int(y))
                 for x, y in pairwise(
-                    footprint_elem.findtext(rep('colRowList')).split()
+                    footprint_elem.findtext(used_rep('colRowList')).split()
                 )
             ],
             'coord_list': [
                 (float(x), float(y))
                 for x, y in pairwise(
-                    footprint_elem.findtext(rep('coordList')).split()
+                    footprint_elem.findtext(used_rep('coordList')).split()
                 )
             ],
         }
 
-    elif geotiff_elem:
+    elif geotiff_elem is not None:
         browse['type'] = 'model_in_geotiff_browse'
-        browse['model_in_geotiff'] = True
 
-    elif regular_grid_browse:
+    elif regular_grid_browse is not None:
         # TODO: other browse types
         browse['type'] = 'regular_grid_browse'
         raise NotImplementedError('Regular grid browses are not supported')