This script allows to harvest all the records from the GeoCat CSW service

GeoCat_CSW.py

"""
CSW harvesting script for the GeoCat CSW service.
This script allows to harvest all the records from the GeoCat CSW service
Author: [Maxime Collombin]
Date: [14/08/2023]
"""

# -*- coding: utf-8 -*-


from owslib.csw import CatalogueServiceWeb
from owslib.fes import PropertyIsLike
import pandas as pd
import requests

# Define the base URL of the CSW service

url = 'https://www.geocat.ch/geonetwork/srv/fre/csw'

# Define the query parameters

params = {
    'service': 'CSW',
    'version': '2.0.2',
    'request': 'GetRecords',
    'constraintLanguage': 'CQL_TEXT',
    'constraint': "dc:subject = 'modèle de géodonnées minimal MGDM'",
    'constraint_language_version': '1.1.0',
    'typeNames': 'csw:Record',
    'resultType': 'hits'
}

# Send the request to the CSW service and store the response

response = requests.get(url, params=params)

# Extract the value of numberOfRecordsMatched from the response content

response_content = response.content.decode('utf-8')
start_index = response_content.index('numberOfRecordsMatched') + len('numberOfRecordsMatched') + 2
end_index = response_content.index('"', start_index)
num_matched = int(response_content[start_index:end_index])

# Define the CSW service URL with the params defined above

csw_url = url + '?' + '&'.join(['='.join(i) for i in list(params.items())[:-1]]) + '&resultType=results&maxRecords=' + str(num_matched) + '&lang=fr'

# Define the CSW service URL 

csw_url = url + '?service=CSW&version=2.0.2&request=GetRecords&constraintLanguage=CQL_TEXT&constraint=dc:type%20=%20\'model\'&constraint_language_version=1.1.0&typeNames=csw:Record&resultType=results&maxRecords=' + str(num_matched) + '&lang=fr'

# get the records attributes

csw = CatalogueServiceWeb(csw_url)


csw.getrecords2(maxrecords=num_matched)
records = csw.records
 
df = pd.DataFrame(columns=['identifier', 'title', 'abstract', 'url'])

for key, value in records.items():
    df = df.append({'identifier': value.identifier, 'title': value.title, 'abstract': value.abstract, 'url': 'https://www.geocat.ch/geonetwork/srv/fre/catalog.search#/metadata/' + str(value.identifier)}, ignore_index=True)
    
# save the dataframe to an xlsx file with utf-8 encoding

df.to_excel('csw_records.xlsx', encoding='utf-8')

There seem to be several records on geocat.ch for the same MGDM. One for the geoservices and/or bulk data and one for the models without linking the 2 sources.

Example for the alignment of national roads:

• Service and data
• Data model

What could be the explanations/reasons?