Fixed schemingdcat and add custom_formats

README.md

@@ -134,7 +134,7 @@ pdm run python ogc2ckan/ogc2ckan.py
 ## Additional info
 ### CKAN Schemas
 The CKAN output schemas are located in the [`ogc2ckan/ckan_datasets`](./ogc2ckan/ckan_datasets) folder. The schemas are used to map the metadata fields from the different sources to the CKAN dataset fields. Now are available the following schemas:
-   * `geodcatap`: Schema based in [GeoDCAT-AP Schema for CKAN](https://github.com/mjanez/ckanext-scheming_dcat).
+   * `geodcatap`: Schema based in [GeoDCAT-AP Schema for CKAN](https://github.com/mjanez/ckanext-schemingdcat).
    * `base`: A DCAT schema with the basic fields. 
 
 You can create your own Schema.

ogc2ckan/harvesters/base.py

@@ -1088,7 +1088,7 @@ def _clean_name(name):
 
         # Replace accented and special characters with their unaccented equivalents or _
         name = ''.join(accent_map.get(c, c) for c in name)
-        name = re.sub(r'[^a-zñ0-9_.-]', '_', name.lower().strip())
+        name = re.sub(r'[^a-zñ0-9_.-]', '-', name.lower().strip())
 
         # Truncate the name to 40 characters
         name = name[:40]

ogc2ckan/harvesters/table.py

@@ -18,7 +18,7 @@
 
 # custom functions
 from config.ogc2ckan_config import get_log_module
-from mappings.default_ogc2ckan_config import OGC2CKAN_HARVESTER_CONFIG, OGC2CKAN_HARVESTER_MD_CONFIG, OGC2CKAN_CKANINFO_CONFIG
+from mappings.default_ogc2ckan_config import OGC2CKAN_HARVESTER_CONFIG, OGC2CKAN_HARVESTER_MD_CONFIG, OGC2CKAN_CKANINFO_CONFIG, CUSTOM_FORMAT_RULES
 from controller.mapping import get_df_mapping_json
 
 log_module = get_log_module(os.path.abspath(__file__))
@@ -409,7 +409,7 @@ def get_distribution(self, ckan_info: CKANInfo, dataset, distribution, datadicti
                     'id': distribution_id,
                     'url': r.get('url', ''),
                     'name': dist_name,
-                    'format': self._update_custom_formats(format_type, r.get('url', '')),
+                    'format': self._update_custom_format(format_type, r.get('url', '')),
                     'media_type': media_type,
                     'description': r.get('description', ''),
                     'license': r.get('license', ckan_info.default_license),
@@ -447,11 +447,12 @@ def get_datadictionary(self, datadictionary, datadictionaryfield, table_datadict
 
 
     @staticmethod
-    def _update_custom_formats(format, url=None, **args):
-        """Update the custom format.
+    def _update_custom_format(format, url=None, **args):
+        """Update the custom format based on custom rules.
         
-        If the format contains 'esri' or 'arcgis' (case-insensitive) or the URL contains 'viewer.html?url=',
-        the format is updated to 'HTML'.
+        The function checks the format and URL against a set of custom rules (CUSTOM_FORMAT_RULES). If a rule matches,
+        the format is updated according to that rule. This function is designed to be easily
+        extendable with new rules.
         
         Args:
             format (str): The custom format to update.
@@ -461,9 +462,14 @@ def _update_custom_formats(format, url=None, **args):
         Returns:
             str: The updated custom format.
         """
-        if isinstance(format, str) and (any(string in format.lower() for string in ['esri', 'arcgis']) or 'viewer.html?url=' in url):
-            format = 'HTML'
-
+        if isinstance(format, str):
+            format_lower = format.lower()
+
+            for rule in CUSTOM_FORMAT_RULES:
+                if any(string in format_lower for string in rule['format_strings']) or rule['url_string'] in url:
+                    format = rule['new_format']
+                    break
+
         return format
 
     @staticmethod

ogc2ckan/mappings/default_ogc2ckan_config.py

@@ -196,4 +196,13 @@
 }
 
 # loose definition of BCP47-like strings
-BCP_47_LANGUAGE = u'^[a-z]{2,8}(-[0-9a-zA-Z]{1,8})*$'
+BCP_47_LANGUAGE = u'^[a-z]{2,8}(-[0-9a-zA-Z]{1,8})*$'
+
+CUSTOM_FORMAT_RULES = [
+    {
+        'format_strings': ['esri', 'arcgis'],
+        'url_string': 'viewer.html?url=',
+        'new_format': 'HTML'
+    },
+    # Add more rules here as needed
+]

ogc2ckan/ogc2ckan.py

@@ -111,7 +111,7 @@ def start_harvesting(config_file):
             logging.warning(f"{log_module}:[INSECURE] SSL_UNVERIFIED_MODE:'{ckan_info.ssl_unverified_mode}'. Only if you trust the host.")    
 
         if ckan_info.metadata_distributions == True or ckan_info.metadata_distributions == "True":
-            logging.warning(f"{log_module}:METADATA_DISTRIBUTIONS:'{ckan_info.metadata_distributions}'. It is not necessary if you do not intend to generate distributions for geographic metadata (INSPIRE ISO19139) or Linked Open Data (GeoDCAT-AP). ckanext-scheming_dcat already links the most important metadata profiles (https://github.com/mjanez/ckanext-scheming_dcat).")
+            logging.warning(f"{log_module}:METADATA_DISTRIBUTIONS:'{ckan_info.metadata_distributions}'. It is not necessary if you do not intend to generate distributions for geographic metadata (INSPIRE ISO19139) or Linked Open Data (GeoDCAT-AP). ckanext-schemingdcat already links the most important metadata profiles (https://github.com/mjanez/ckanext-schemingdcat).")
 
         logging.info(f"{log_module}:Type of activated harvesters: {', '.join([f'{h.upper()}' for h in active_harvesters])}")               
         logging.info(f"{log_module}:CKAN_URL: {ckan_info.ckan_site_url}")