Fetching - Always record `server` software reply headers (will be used in the future) (#2143)

changedetectionio/content_fetcher.py

@@ -51,6 +51,7 @@ class BrowserStepsStepException(Exception):
         return
 
 
+# @todo - make base Exception class that announces via logger()
 class PageUnloadable(Exception):
     def __init__(self, status_code, url, message, screenshot=False):
         # Set this so we can use it in other parts of the app
@@ -389,10 +390,24 @@ class base_html_playwright(Fetcher):
             raise PageUnloadable(url=url, status_code=None, message=f"Timed out connecting to browserless, retrying..")
         else:
             # 200 Here means that the communication to browserless worked only, not the page state
-            if response.status_code == 200:
+            try:
+                x = response.json()
+            except Exception as e:
+                raise PageUnloadable(url=url, message="Error reading JSON response from browserless")
+
+            try:
+                self.status_code = response.status_code
+            except Exception as e:
+                raise PageUnloadable(url=url, message="Error reading status_code code response from browserless")
+
+            self.headers = x.get('headers')
+
+            if self.status_code != 200 and not ignore_status_codes:
+                raise Non200ErrorCodeReceived(url=url, status_code=self.status_code, page_html=x.get('content',''))
+
+            if self.status_code == 200:
                 import base64
 
-                x = response.json()
                 if not x.get('screenshot'):
                     # https://github.com/puppeteer/puppeteer/blob/v1.0.0/docs/troubleshooting.md#tips
                     # https://github.com/puppeteer/puppeteer/issues/1834
@@ -403,16 +418,10 @@ class base_html_playwright(Fetcher):
                 if not x.get('content', '').strip():
                     raise EmptyReply(url=url, status_code=None)
 
-                if x.get('status_code', 200) != 200 and not ignore_status_codes:
-                    raise Non200ErrorCodeReceived(url=url, status_code=x.get('status_code', 200), page_html=x['content'])
-
                 self.content = x.get('content')
-                self.headers = x.get('headers')
                 self.instock_data = x.get('instock_data')
                 self.screenshot = base64.b64decode(x.get('screenshot'))
-                self.status_code = x.get('status_code')
                 self.xpath_data = x.get('xpath_data')
-
             else:
                 # Some other error from browserless
                 raise PageUnloadable(url=url, status_code=None, message=response.content.decode('utf-8'))
@@ -742,6 +751,8 @@ class html_requests(Fetcher):
                 if encoding:
                     r.encoding = encoding
 
+        self.headers = r.headers
+
         if not r.content or not len(r.content):
             raise EmptyReply(url=url, status_code=r.status_code)
 
@@ -758,7 +769,7 @@ class html_requests(Fetcher):
         else:
             self.content = r.text
 
-        self.headers = r.headers
+
         self.raw_content = r.content
 
 

changedetectionio/model/Watch.py

@@ -56,6 +56,7 @@ base_config = {
     'previous_md5': False,
     'previous_md5_before_filters': False,  # Used for skipping changedetection entirely
     'proxy': None,  # Preferred proxy connection
+    'remote_server_reply': None, # From 'server' reply header
     'subtractive_selectors': [],
     'tag': '', # Old system of text name for a tag, to be removed
     'tags': [], # list of UUIDs to App.Tags

changedetectionio/store.py

@@ -255,6 +255,7 @@ class ChangeDetectionStore:
                 'last_viewed': 0,
                 'previous_md5': False,
                 'previous_md5_before_filters': False,
+                'remote_server_reply': None,
                 'track_ldjson_price_data': None,
             })
 

changedetectionio/tests/test_request.py

@@ -10,7 +10,7 @@ def test_setup(live_server):
 # Hard to just add more live server URLs when one test is already running (I think)
 # So we add our test here (was in a different file)
 def test_headers_in_request(client, live_server):
-    #live_server_setup(live_server)
+    #ve_server_setup(live_server)
     # Add our URL to the import page
     test_url = url_for('test_headers', _external=True)
     if os.getenv('PLAYWRIGHT_DRIVER_URL'):
@@ -70,16 +70,17 @@ def test_headers_in_request(client, live_server):
 
     wait_for_all_checks(client)
 
-    # Re #137 -  Examine the JSON index file, it should have only one set of headers entered
+    # Re #137 -  It should have only one set of headers entered
     watches_with_headers = 0
-    with open('test-datastore/url-watches.json') as f:
-        app_struct = json.load(f)
-        for uuid in app_struct['watching']:
-            if (len(app_struct['watching'][uuid]['headers'])):
+    for k, watch in client.application.config.get('DATASTORE').data.get('watching').items():
+            if (len(watch['headers'])):
                 watches_with_headers += 1
-
-    # Should be only one with headers set
     assert watches_with_headers == 1
+
+    # 'server' http header was automatically recorded
+    for k, watch in client.application.config.get('DATASTORE').data.get('watching').items():
+        assert 'custom' in watch.get('remote_server_reply') # added in util.py
+
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
 

changedetectionio/tests/util.py

@@ -180,7 +180,11 @@ def live_server_setup(live_server):
         for header in request.headers:
             output.append("{}:{}".format(str(header[0]), str(header[1])))
 
-        return "\n".join(output)
+        content = "\n".join(output)
+
+        resp = make_response(content, 200)
+        resp.headers['server'] = 'custom'
+        return resp
 
     # Just return the body in the request
     @live_server.app.route('/test-body', methods=['POST', 'GET'])

changedetectionio/update_worker.py

@@ -491,6 +491,16 @@ class update_worker(threading.Thread):
                     if self.datastore.data['watching'].get(uuid):
                         # Always record that we atleast tried
                         count = self.datastore.data['watching'][uuid].get('check_count', 0) + 1
+
+                        # Record the 'server' header reply, can be used for actions in the future like cloudflare/akamai workarounds
+                        try:
+                            server_header = update_handler.fetcher.headers.get('server', '').strip().lower()[:255]
+                            self.datastore.update_watch(uuid=uuid,
+                                                        update_obj={'remote_server_reply': server_header}
+                                                        )
+                        except Exception as e:
+                            pass
+
                         self.datastore.update_watch(uuid=uuid, update_obj={'fetch_time': round(time.time() - now, 3),
                                                                            'last_checked': round(time.time()),
                                                                            'check_count': count

docker-compose.yml

@@ -94,7 +94,8 @@ services:
 #
 
      # Used for fetching pages via Playwright+Chrome where you need Javascript support.
-     # Note: works well but is deprecated, does not fetch full page screenshots (doesnt work with Visual Selector) and other issues
+     # Note: Works well but is deprecated, does not fetch full page screenshots (doesnt work with Visual Selector)
+     #       Does not report status codes (200, 404, 403) and other issues
      # More information about the advantages of playwright/browserless https://www.browserless.io/blog/2023/12/13/migrating-selenium-to-playwright/
 #    browser-chrome:
 #        hostname: browser-chrome