Fixing proxy checker (#2696)

changedetectionio/blueprint/check_proxies/__init__.py

@@ -1,4 +1,7 @@
+import importlib
 from concurrent.futures import ThreadPoolExecutor
+
+from changedetectionio.processors.text_json_diff.processor import FilterNotFoundInResponse
 from changedetectionio.store import ChangeDetectionStore
 
 from functools import wraps
@@ -30,7 +33,6 @@ def construct_blueprint(datastore: ChangeDetectionStore):
     def long_task(uuid, preferred_proxy):
         import time
         from changedetectionio.content_fetchers import exceptions as content_fetcher_exceptions
-        from changedetectionio.processors.text_json_diff import text_json_diff
         from changedetectionio.safe_jinja import render as jinja_render
 
         status = {'status': '', 'length': 0, 'text': ''}
@@ -38,8 +40,12 @@ def construct_blueprint(datastore: ChangeDetectionStore):
         contents = ''
         now = time.time()
         try:
-            update_handler = text_json_diff.perform_site_check(datastore=datastore, watch_uuid=uuid)
-            update_handler.call_browser()
+            processor_module = importlib.import_module("changedetectionio.processors.text_json_diff.processor")
+            update_handler = processor_module.perform_site_check(datastore=datastore,
+                                                                 watch_uuid=uuid
+                                                                 )
+
+            update_handler.call_browser(preferred_proxy_id=preferred_proxy)
         # title, size is len contents not len xfer
         except content_fetcher_exceptions.Non200ErrorCodeReceived as e:
             if e.status_code == 404:
@@ -48,7 +54,7 @@ def construct_blueprint(datastore: ChangeDetectionStore):
                 status.update({'status': 'ERROR', 'length': len(contents), 'text': f"{e.status_code} - Access denied"})
             else:
                 status.update({'status': 'ERROR', 'length': len(contents), 'text': f"Status code: {e.status_code}"})
-        except text_json_diff.FilterNotFoundInResponse:
+        except FilterNotFoundInResponse:
             status.update({'status': 'OK', 'length': len(contents), 'text': f"OK but CSS/xPath filter not found (page changed layout?)"})
         except content_fetcher_exceptions.EmptyReply as e:
             if e.status_code == 403 or e.status_code == 401:

changedetectionio/processors/__init__.py

@@ -18,6 +18,7 @@ class difference_detection_processor():
     screenshot = None
     watch = None
     xpath_data = None
+    preferred_proxy = None
 
     def __init__(self, *args, datastore, watch_uuid, **kwargs):
         super().__init__(*args, **kwargs)
@@ -26,7 +27,8 @@ class difference_detection_processor():
         # Generic fetcher that should be extended (requests, playwright etc)
         self.fetcher = Fetcher()
 
-    def call_browser(self):
+    def call_browser(self, preferred_proxy_id=None):
+
         from requests.structures import CaseInsensitiveDict
 
         # Protect against file:// access
@@ -42,7 +44,7 @@ class difference_detection_processor():
         prefer_fetch_backend = self.watch.get('fetch_backend', 'system')
 
         # Proxy ID "key"
-        preferred_proxy_id = self.datastore.get_preferred_proxy_for_watch(uuid=self.watch.get('uuid'))
+        preferred_proxy_id = preferred_proxy_id if preferred_proxy_id else self.datastore.get_preferred_proxy_for_watch(uuid=self.watch.get('uuid'))
 
         # Pluggable content self.fetcher
         if not prefer_fetch_backend or prefer_fetch_backend == 'system':

changedetectionio/static/js/recheck-proxy.js

@@ -1,14 +1,14 @@
 $(function () {
     /* add container before each proxy location to show status */
-
-    var option_li = $('.fetch-backend-proxy li').filter(function() {
-        return $("input",this)[0].value.length >0;
-    });
-
-    //var option_li = $('.fetch-backend-proxy li');
     var isActive = false;
-    $(option_li).prepend('<div class="proxy-status"></div>');
-    $(option_li).append('<div class="proxy-timing"></div><div class="proxy-check-details"></div>');
+
+    function setup_html_widget() {
+        var option_li = $('.fetch-backend-proxy li').filter(function () {
+            return $("input", this)[0].value.length > 0;
+        });
+        $(option_li).prepend('<div class="proxy-status"></div>');
+        $(option_li).append('<div class="proxy-timing"></div><div class="proxy-check-details"></div>');
+    }
 
     function set_proxy_check_status(proxy_key, state) {
         // select input by value name
@@ -59,8 +59,14 @@ $(function () {
     }
 
     $('#check-all-proxies').click(function (e) {
+
         e.preventDefault()
-        $('body').addClass('proxy-check-active');
+
+        if (!$('body').hasClass('proxy-check-active')) {
+            setup_html_widget();
+            $('body').addClass('proxy-check-active');
+        }
+
         $('.proxy-check-details').html('');
         $('.proxy-status').html('<span class="spinner"></span>').fadeIn();
         $('.proxy-timing').html('');

changedetectionio/static/styles/scss/parts/_extra_proxies.scss

@@ -25,15 +25,19 @@ ul#requests-extra_proxies {
 
 body.proxy-check-active {
   #request {
+    // Padding set by flex layout
+    /*
     .proxy-status {
       width: 2em;
     }
+    */
 
     .proxy-check-details {
       font-size: 80%;
       color: #555;
       display: block;
-      padding-left: 4em;
+      padding-left: 2em;
+      max-width: 500px;
     }
 
     .proxy-timing {

changedetectionio/static/styles/styles.css

@@ -119,19 +119,22 @@ ul#requests-extra_proxies {
   #request label[for=proxy] {
     display: inline-block; }
 
-body.proxy-check-active #request .proxy-status {
-  width: 2em; }
-
-body.proxy-check-active #request .proxy-check-details {
-  font-size: 80%;
-  color: #555;
-  display: block;
-  padding-left: 4em; }
-
-body.proxy-check-active #request .proxy-timing {
-  font-size: 80%;
-  padding-left: 1rem;
-  color: var(--color-link); }
+body.proxy-check-active #request {
+  /*
+    .proxy-status {
+      width: 2em;
+    }
+    */ }
+  body.proxy-check-active #request .proxy-check-details {
+    font-size: 80%;
+    color: #555;
+    display: block;
+    padding-left: 2em;
+    max-width: 500px; }
+  body.proxy-check-active #request .proxy-timing {
+    font-size: 80%;
+    padding-left: 1rem;
+    color: var(--color-link); }
 
 #recommended-proxy {
   display: grid;

changedetectionio/tests/test_preview_endpoints.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python3
+
+import time
+from flask import url_for
+from .util import set_original_response, set_modified_response, live_server_setup, wait_for_all_checks
+
+
+# `subtractive_selectors` should still work in `source:` type requests
+def test_fetch_pdf(client, live_server, measure_memory_usage):
+    import shutil
+    shutil.copy("tests/test.pdf", "test-datastore/endpoint-test.pdf")
+
+    live_server_setup(live_server)
+    test_url = url_for('test_pdf_endpoint', _external=True)
+    # Add our URL to the import page
+    res = client.post(
+        url_for("import_page"),
+        data={"urls": test_url},
+        follow_redirects=True
+    )
+
+    assert b"1 Imported" in res.data
+
+    wait_for_all_checks(client)
+
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+
+    # PDF header should not be there (it was converted to text)
+    assert b'PDF' not in res.data[:10]
+    assert b'hello world' in res.data
+
+    # So we know if the file changes in other ways
+    import hashlib
+    original_md5 = hashlib.md5(open("test-datastore/endpoint-test.pdf", 'rb').read()).hexdigest().upper()
+    # We should have one
+    assert len(original_md5) > 0
+    # And it's going to be in the document
+    assert b'Document checksum - ' + bytes(str(original_md5).encode('utf-8')) in res.data
+
+    shutil.copy("tests/test2.pdf", "test-datastore/endpoint-test.pdf")
+    changed_md5 = hashlib.md5(open("test-datastore/endpoint-test.pdf", 'rb').read()).hexdigest().upper()
+    res = client.get(url_for("form_watch_checknow"), follow_redirects=True)
+    assert b'1 watches queued for rechecking.' in res.data
+
+    wait_for_all_checks(client)
+
+    # Now something should be ready, indicated by having a 'unviewed' class
+    res = client.get(url_for("index"))
+    assert b'unviewed' in res.data
+
+    # The original checksum should be not be here anymore (cdio adds it to the bottom of the text)
+
+    res = client.get(
+        url_for("preview_page", uuid="first"),
+        follow_redirects=True
+    )
+
+    assert original_md5.encode('utf-8') not in res.data
+    assert changed_md5.encode('utf-8') in res.data
+
+    res = client.get(
+        url_for("diff_history_page", uuid="first"),
+        follow_redirects=True
+    )
+
+    assert original_md5.encode('utf-8') in res.data
+    assert changed_md5.encode('utf-8') in res.data
+
+    assert b'here is a change' in res.data