adjust the proxy logic based on the canvas

README.md

@@ -34,3 +34,6 @@ cp ca-signing-key.pem ~/ontology-time-machine/ca-signing-key.pem
 
 ### Not working: 
 - curl -x http://0.0.0.0:8899 -H "Accept: text/turtle" --cacert ca-cert.pem http://ontologi.es/days#
+
+
+# https://archivo.dbpedia.org/download?o=http://linked-web-apis.fit.cvut.cz/ns/core&v=2020.07.16-115638&versionMatching=timeStampClosest

ontologytimemachine/custom_proxy.py

@@ -2,15 +2,13 @@
 from proxy.http.parser import HttpParser, httpParserTypes
 from proxy.common.utils import build_http_response
 from proxy.http.methods import HttpMethods
-from ontologytimemachine.utils.utils import proxy_logic
+from ontologytimemachine.utils.utils import proxy_logic, parse_arguments
 from ontologytimemachine.utils.utils import check_if_archivo_ontology_requested
-from ontologytimemachine.utils.utils import get_headers_and_expected_type
-from ontologytimemachine.utils.utils import get_ontology_from_request
+from ontologytimemachine.utils.mock_responses import mock_response_403
 from requests.exceptions import SSLError, Timeout, ConnectionError, RequestException
 from http.client import responses
 import proxy
 import sys
-import requests
 import logging
 
 
@@ -24,19 +22,34 @@
 class OntologyTimeMachinePlugin(HttpProxyBasePlugin):
     def __init__(self, *args, **kwargs):
         super().__init__(*args, **kwargs)
+        (self.ontoFormat, self.ontoVersion, self.only_ontologies,
+         self.https_intercept, self.inspect_redirects, self.forward_headers,
+         self.subject_binary_search_threshold) = parse_arguments()
+
 
     def before_upstream_connection(self, request: HttpParser):
         logger.info('Before upstream connection hook')
         logger.info(f'Request method: {request.method} - Request host: {request.host} - Request path: {request.path} - Request headers: {request.headers}')
 
-        scheme = 'https' if request.method == b'CONNECT' else 'http'
-        if scheme == 'https':
-            return request
+        if request.method == b'CONNECT':
+            logger.info(f'HTTPS interception mode: {self.https_intercept}')
+            # Only intercept if interception is enabled
+            if self.https_intercept in ['all', 'archivo']:
+                return request
+            else:
+                return None
+
 
         ontology_request = check_if_archivo_ontology_requested(request)
+        # If only ontology mode, return None in all other cases
+        if self.only_ontologies and not ontology_request:
+            logger.warning('Request denied: not an ontology request and only ontologies mode is enabled')
+            self.queue_response(mock_response_403)
+            return None
+
         if ontology_request:
             logger.debug('The request is for an ontology')
-            response = proxy_logic(request)
+            response = proxy_logic(request, self.ontoFormat, self.ontoVersion)
             self.queue_response(response)
             return None
         return request
@@ -47,16 +60,15 @@ def handle_client_request(self, request: HttpParser):
         logger.info(f'Request method: {request.method} - Request host: {request.host} - Request path: {request.path} - Request headers: {request.headers}')
 
         logger.debug(request.method)
-        scheme = 'https' if request.method == b'CONNECT' else 'http'
-        if scheme == 'https':
+        if request.method == b'CONNECT':
             return request
 
         ontology_request = check_if_archivo_ontology_requested(request)
         if not ontology_request:
             logger.info('The requested IRI is not part of DBpedia Archivo')
             return request   
 
-        response = proxy_logic(request)
+        response = proxy_logic(request, self.ontoFormat, self.ontoVersion)
         self.queue_response(response)
 
         return None
@@ -80,6 +92,7 @@ def queue_response(self, response):
 
 
 if __name__ == '__main__':
+
     sys.argv += [
         '--ca-key-file', 'ca-key.pem',
         '--ca-cert-file', 'ca-cert.pem',

ontologytimemachine/utils/mock_responses.py

@@ -0,0 +1,38 @@
+import requests
+
+
+def mock_response_200():
+    mock_response = requests.Response()
+    mock_response.status_code = 200
+    mock_response.url = 'https://example.com/success'
+    mock_response.headers['Content-Type'] = 'text/html'
+    mock_response._content = b'<html><body><h1>To be implemented</h1></body></html>'
+    return mock_response
+
+
+def mock_response_403():
+    mock_response = requests.Response()
+    mock_response.status_code = 403
+    mock_response.url = 'https://example.com/forbidden'
+    mock_response.headers['Content-Type'] = 'text/html'
+    mock_response._content = b'<html><body><h1>403 Forbidden</h1></body></html>'
+    return mock_response
+
+
+
+def mock_response_404():
+    mock_response = requests.Response()
+    mock_response.status_code = 404
+    mock_response.url = 'https://example.com/resource-not-found'
+    mock_response.headers['Content-Type'] = 'text/html'
+    mock_response._content = b'<html><body><h1>404 Not Found</h1></body></html>'
+    return mock_response
+
+
+def mock_response_500():
+    mock_response = requests.Response()
+    mock_response.status_code = 500
+    mock_response.url = 'https://example.com/internal-server-error'
+    mock_response.headers['Content-Type'] = 'text/html'
+    mock_response._content = b'<html><body><h1>500 Internal Server Error</h1></body></html>'
+    return mock_response