Support for ISO 19115 Part 3 XML (#933)

* Added initial code to ingest ISO 19115-3 XML

* Add ISO 19115-3 XML plugin

* Remove old files

* Rename profile

* Add support for output of vertical extent and funder

* Fix bugs in output code

* Fix incorrect namespace issue

* Get address info from contacts when GetCapabilities not supplied

* Add funder parsing, fix vert extent path

* Add initial tests

* Fix numerous path bugs

* Fix errors in constraints and spatial resolution

* Allowed anchor tag for identifiers

* Updated processing of image description and bands

* Add functional tests

* Add funder column to db

* Fix error im gml namespace path

* Add mdb unit test

* Cleaning up and adding comments

* Add mdb function comments

* Add db columns for vert extent

* Add function comments to profile code

* Fix vert extent and gml paths in tests

* Fix services bugs and add tests

* Update functional tests for new mdb XML test record

* Update comments and readme for ISO 19115 Part 3 XML

* Add mdb transaction tests

* Fixes:

- Missing contact name in converted records
- Add more tests

* Update due to functionality moved to OWSLib

* Updated tests and modifications for changed config format

* Update expected test responses after OWSLib funder changes

* Restore correct version of cite.db

* Remove cite.db from pull request

* Remove unnecessary trailing comma in setup.py

* Renamed func. test suites from 'mdb' to 'iso19115p3'

* Remove 'Funder' field and tests

* Correct the profile name in the iso19115p3 test suite

* Update iso19115p3 func. tests for profile name change

Author: vjf

pycsw/core/config.py

@@ -82,6 +82,7 @@ def __init__(self, prefix='csw30'):
             'gmd': 'http://www.isotc211.org/2005/gmd',
             'gml': 'http://www.opengis.net/gml',
             'gml32': 'http://www.opengis.net/gml/3.2',
+            'mdb': 'http://standards.iso.org/iso/19115/-3/mdb/2.0',
             'ogc': 'http://www.opengis.net/ogc',
             'os': 'http://a9.com/-/spec/opensearch/1.1/',
             'ows': 'http://www.opengis.net/ows',
@@ -118,7 +119,7 @@ def __init__(self, prefix='csw30'):
                 'pycsw:Metadata': 'metadata',
                 # raw metadata payload type, xml as default for now
                 'pycsw:MetadataType': 'metadata_type',
-                # bag of metadata element and attributes ONLY, no XML tages
+                # bag of metadata element and attributes ONLY, no XML tags
                 'pycsw:AnyText': 'anytext',
                 'pycsw:Language': 'language',
                 'pycsw:Title': 'title',
@@ -134,6 +135,8 @@ def __init__(self, prefix='csw30'):
                 'pycsw:Type': 'type',
                 # geometry, specified in OGC WKT
                 'pycsw:BoundingBox': 'wkt_geometry',
+                'pycsw:VertExtentMin': 'vert_extent_min',
+                'pycsw:VertExtentMax': 'vert_extent_max',
                 'pycsw:CRS': 'crs',
                 'pycsw:AlternateTitle': 'title_alternate',
                 'pycsw:RevisionDate': 'date_revision',

pycsw/core/metadata.py

@@ -162,6 +162,9 @@ def _parse_metadata(context, repos, record):
         return [_parse_dc(context, repos, exml)]
     elif root == '{%s}DIF' % context.namespaces['dif']:  # DIF
         pass  # TODO
+    elif root == '{%s}MD_Metadata' % context.namespaces['mdb']:
+        # ISO 19115p3 XML
+        return [_parse_iso(context, repos, exml)]
     else:
         raise RuntimeError('Unsupported metadata format')
 
@@ -1364,23 +1367,29 @@ def get_value_by_language(pt_group, language, pt_type='text'):
     return recobj
 
 def _parse_iso(context, repos, exml):
+    """ Parses ISO 19139, ISO 19115p3 """
 
     from owslib.iso import MD_ImageDescription, MD_Metadata, SV_ServiceIdentification
     from owslib.iso_che import CHE_MD_Metadata
 
     recobj = repos.dataset()
     bbox = None
     links = []
+    mdmeta_ns = 'gmd'
 
     if exml.tag == '{http://www.geocat.ch/2008/che}CHE_MD_Metadata':
         md = CHE_MD_Metadata(exml)
+    elif exml.tag == '{http://standards.iso.org/iso/19115/-3/mdb/2.0}MD_Metadata':
+        from owslib.iso3 import MD_Metadata
+        md = MD_Metadata(exml)
+        mdmeta_ns = 'mdb'
     else:
         md = MD_Metadata(exml)
 
     md_identification = md.identification[0]
 
     _set(context, recobj, 'pycsw:Identifier', md.identifier)
-    _set(context, recobj, 'pycsw:Typename', 'gmd:MD_Metadata')
+    _set(context, recobj, 'pycsw:Typename', f'{mdmeta_ns}:MD_Metadata')
     _set(context, recobj, 'pycsw:Schema', context.namespaces['gmd'])
     _set(context, recobj, 'pycsw:MdSource', 'local')
     _set(context, recobj, 'pycsw:InsertDate', util.get_today_and_now())
@@ -1394,7 +1403,7 @@ def _parse_iso(context, repos, exml):
     _set(context, recobj, 'pycsw:Modified', md.datestamp)
     _set(context, recobj, 'pycsw:Source', md.dataseturi)
 
-    if md.referencesystem is not None:
+    if md.referencesystem is not None and md.referencesystem.code is not None:
         try:
             code_ = 'urn:ogc:def:crs:EPSG::%d' % int(md.referencesystem.code)
         except ValueError:
@@ -1421,11 +1430,21 @@ def _parse_iso(context, repos, exml):
         elif len(md_identification.resourcelanguagecode) > 0:
             _set(context, recobj, 'pycsw:ResourceLanguage', md_identification.resourcelanguagecode[0])
 
+        # Geographic bounding box
         if hasattr(md_identification, 'bbox'):
             bbox = md_identification.bbox
         else:
             bbox = None
 
+        # Vertical extent of a bounding box
+        if hasattr(md_identification, 'extent'):
+            if hasattr(md_identification.extent, 'vertExtMin') and \
+                md_identification.extent.vertExtMin is not None:
+                _set(context, recobj, 'pycsw:VertExtentMin', md_identification.extent.vertExtMin)
+            if hasattr(md_identification.extent, 'vertExtMax') and \
+                md_identification.extent.vertExtMax is not None:
+                _set(context, recobj, 'pycsw:VertExtentMax', md_identification.extent.vertExtMax)
+
         if (hasattr(md_identification, 'keywords') and
             len(md_identification.keywords) > 0):
             all_keywords = [item for sublist in md_identification.keywords for item in sublist.keywords if item is not None]
@@ -1435,14 +1454,17 @@ def _parse_iso(context, repos, exml):
                  json.dumps([t for t in md_identification.keywords if t.thesaurus is not None], 
                             default=lambda o: o.__dict__))
 
+        # Creator
         if (hasattr(md_identification, 'creator') and
             len(md_identification.creator) > 0):
             all_orgs = set([item.organization for item in md_identification.creator if hasattr(item, 'organization') and item.organization is not None])
             _set(context, recobj, 'pycsw:Creator', ';'.join(all_orgs))
+        # Publisher
         if (hasattr(md_identification, 'publisher') and
             len(md_identification.publisher) > 0):
             all_orgs = set([item.organization for item in md_identification.publisher if hasattr(item, 'organization') and item.organization is not None])
             _set(context, recobj, 'pycsw:Publisher', ';'.join(all_orgs))
+        # Contributor
         if (hasattr(md_identification, 'contributor') and
             len(md_identification.contributor) > 0):
             all_orgs = set([item.organization for item in md_identification.contributor if hasattr(item, 'organization') and item.organization is not None])
@@ -1527,18 +1549,18 @@ def _parse_iso(context, repos, exml):
 
                     _set(context, recobj, 'pycsw:Keywords', keywords)
 
-            bands = []
-            for band in ci.bands:
-                band_info = {
-                    'id': band.id,
-                    'units': band.units,
-                    'min': band.min,
-                    'max': band.max
-                }
-                bands.append(band_info)
+                bands = []
+                for band in ci.bands:
+                    band_info = {
+                        'id': band.id,
+                        'units': band.units,
+                        'min': band.min,
+                        'max': band.max
+                    }
+                    bands.append(band_info)
 
-            if len(bands) > 0:
-                _set(context, recobj, 'pycsw:Bands', json.dumps(bands))
+                if len(bands) > 0:
+                    _set(context, recobj, 'pycsw:Bands', json.dumps(bands))
 
     if hasattr(md, 'acquisition') and md.acquisition is not None:
         platform = md.acquisition.platforms[0]
@@ -1557,10 +1579,10 @@ def _parse_iso(context, repos, exml):
     if hasattr(md, 'distribution'):
         dist_links = []
         if hasattr(md.distribution, 'online'):
-            LOGGER.debug('Scanning for gmd:transferOptions element(s)')
+            LOGGER.debug(f'Scanning for {mdmeta_ns}:transferOptions element(s)')
             dist_links.extend(md.distribution.online)
         if hasattr(md.distribution, 'distributor'):
-            LOGGER.debug('Scanning for gmd:distributorTransferOptions element(s)')
+            LOGGER.debug(f'Scanning for {mdmeta_ns}:distributorTransferOptions element(s)')
             for dist_member in md.distribution.distributor:
                 dist_links.extend(dist_member.online)
         for link in dist_links: