missing kids status code handling

app_urls/fetcher/src/db_utils.py

@@ -6,6 +6,7 @@ from django.utils import timezone
 from datetime import timedelta
 from .fetch_utils_url_processor import process_url, get_with_protocol
 import re
+import requests
 import os
 import traceback
 from .logger import get_logger
@@ -80,24 +81,56 @@ class DB_Handler():
         except Exception as e:
             logger.warning("Exception inserting raw URLs: {}\n{}".format(e, traceback.format_exc()))
 
+    def _set_status(self, obj_url, status):
+        # Update status if setting a new value
+        if (obj_url.status != status):
+            obj_url.status = status
+            obj_url.save()
+
+    def _set_duplicate_and_insert_canonical(self, obj_url, url_canonical):
+        # Update status
+        self._set_status(obj_url, Urls.STATUS_ENUM.DUPLICATE)
+        # Get or create URL with canonical form
+        obj_url_canonical, created = Urls.objects.get_or_create(url=url_canonical)
+        # Get the source-search IDs associated to obj_url.id
+        list_url_source_search = UrlsSourceSearch.objects.filter(id_url=obj_url)
+        for obj_url_source_search in list_url_source_search:
+            # Associate same sources to url_canonical (it might already exist)
+            UrlsSourceSearch.objects.get_or_create(id_url=obj_url_canonical, id_source=obj_url_source_search.id_source, id_search=obj_url_source_search.id_search)
+        # URLs duplciate association
+        UrlsDuplicate.objects.get_or_create(id_url_canonical=obj_url_canonical, id_url_duplicated=obj_url)
 
     def _process_single_url(self, obj_url, status_pattern_match, raise_exception_on_error, paywall_bypass=False):
-        
-        def set_status(obj_url, status):
-            # Update status if setting a new value
-            if (obj_url.status != status):
-                obj_url.status = status
-                obj_url.save()
+        ##########################################################################
+        # URL pattern: missingkids.org/poster OR missingkids.org/new-poster
+        if ("missingkids.org" in obj_url.url) and ("poster" in obj_url.url):
+            # Request
+            r = requests.get(obj_url.url, allow_redirects=True)
+
+            if (r.url != obj_url.url):
+                # Canonical
+                url_canonical = r.url
+                # Set duplicate, and insert new canonical form
+                self._set_duplicate_and_insert_canonical(obj_url, url_canonical)
+            elif (r.status_code == 200):
+                self._set_status(self, obj_url, Urls.STATUS_ENUM.VALID)
+            elif (r.status_code == 404):
+                self._set_status(self, obj_url, Urls.STATUS_ENUM.INVALID)
+            else:
+                logger.info("Unknown request status: {} for missing kids request: {}".format(r.status_code, obj_url.url))
+            
+            return
+        ##########################################################################
 
         # Found a pattern match -> Override status
         if (status_pattern_match is not None):
             logger.debug("Pattern match, status '{}' for input URL: {}".format(status_pattern_match, obj_url.url))
             # Update status
-            set_status(obj_url, status_pattern_match)
+            self._set_status(obj_url, status_pattern_match)
             ##### Filter URL? -> Invalid (don't extract content)
             if (status_pattern_match == "invalid"):
                 return
-        
+
         try:
             # Extract URL content
             dict_url_data = process_url(obj_url.url, paywall_bypass)
@@ -111,19 +144,9 @@ class DB_Handler():
                 dict_url_data = None
         
         ##### Canonical URL different? -> Duplicate
-        if (dict_url_data is not None) and (dict_url_data.get("url_canonical") is not None) and (dict_url_data.get("url") != dict_url_data.get("url_canonical")):
-            # Update status
-            set_status(obj_url, Urls.STATUS_ENUM.DUPLICATE)
-            # Get or create URL with canonical form
-            obj_url_canonical, created = Urls.objects.get_or_create(url=dict_url_data.get("url_canonical"))
-            # Get the source-search IDs associated to obj_url.id
-            list_url_source_search = UrlsSourceSearch.objects.filter(id_url=obj_url)
-            for obj_url_source_search in list_url_source_search:
-                # Associate same sources to url_canonical (it might already exist)
-                UrlsSourceSearch.objects.get_or_create(id_url=obj_url_canonical, id_source=obj_url_source_search.id_source, id_search=obj_url_source_search.id_search)
-            # URLs duplciate association
-            UrlsDuplicate.objects.get_or_create(id_url_canonical=obj_url_canonical, id_url_duplicated=obj_url)
-
+        if (dict_url_data is not None) and (dict_url_data.get("url_canonical") is not None) and (dict_url_data.get("url") != dict_url_data.get("url_canonical")):            
+            # URL as duplicate, insert canonical URL
+            self._set_duplicate_and_insert_canonical(obj_url, dict_url_data.get("url_canonical"))
             # Next URL
             return
 
@@ -132,20 +155,20 @@ class DB_Handler():
             # (dict_url_data is None) or (Exception while processing URL) ? -> Error status
             if (dict_url_data is None):
                 # Update status
-                set_status(obj_url, Urls.STATUS_ENUM.ERROR)
+                self._set_status(obj_url, Urls.STATUS_ENUM.ERROR)
                 # Next URL
                 return
             
             # Invalid? e.g. binary data
             if (dict_url_data.get("override_status") == "invalid"):
                 # Update status
-                set_status(obj_url, Urls.STATUS_ENUM.INVALID)
+                self._set_status(obj_url, Urls.STATUS_ENUM.INVALID)
                 # Next URL
                 return
 
             ##### Valid URL
             # Update status
-            set_status(obj_url, Urls.STATUS_ENUM.VALID)
+            self._set_status(obj_url, Urls.STATUS_ENUM.VALID)
 
         try:
             if (dict_url_data is not None):
@@ -260,14 +283,12 @@ class DB_Handler():
             # Per URL
             for obj_url in missingkids_urls:
                 try:
-                    # Process URL. If no exception -> Valid
-                    self._process_single_url(obj_url, status_pattern_match=None, raise_exception_on_error=True)
+                    # Process URL
+                    self._process_single_url(obj_url, status_pattern_match=None, raise_exception_on_error=None)
                 except Exception as e:
-                    # Raised exception -> Invalid (404 error)
-                    obj_url.status = Urls.STATUS_ENUM.INVALID
-                    obj_url.save()
+                    logger.warning("Unknown error processing missing kids poster for URL: {}\n{}".format(obj_url.url, str(e)))
             
-            logger.info("Verified status of #{} missingkids.org/poster URLs".format(len(missingkids_urls)))
+            logger.info("Verified status of #{} missingkids.org/poster / missingkids.org/new-poster URLs".format(len(missingkids_urls)))
         except Exception as e:
             logger.warning("Exception processing MissingKids URLs: {}\n{}".format(e, traceback.format_exc()))
     

app_urls/init_data.json

@@ -30,6 +30,5 @@
         [".*zerohedge\\.com\\/(economics|political|markets|)\\/.*", "valid", 50],
         [".*breitbart\\.com\\/(economy|entertainment|border|crime|clips)\\/.*", "valid", 50],
         [".*foxnews\\.com\\/(lifestyle|opinion|sports|world)\\/.*", "valid", 50]
-        [".*missingkids\\.org\\/poster\\/.*", "valid", 50]
     ]
 }