From 8e207ba4380ee97c549c26c74e700000b881c2d9 Mon Sep 17 00:00:00 2001
From: dgtlmoon <dgtlmoon@gmail.com>
Date: Fri, 1 Dec 2023 18:38:49 +0100
Subject: [PATCH] API - Ability to add/import bulk list of watches as a
 line-feed separated list (#2021)

---
 changedetectionio/api/api_v1.py     | 55 +++++++++++++++++++++++++++++
 changedetectionio/flask_app.py      |  4 +++
 changedetectionio/store.py          |  5 +--
 changedetectionio/tests/test_api.py | 22 ++++++++++++
 4 files changed, 84 insertions(+), 2 deletions(-)

diff --git a/changedetectionio/api/api_v1.py b/changedetectionio/api/api_v1.py
index 052bc94e..689af011 100644
--- a/changedetectionio/api/api_v1.py
+++ b/changedetectionio/api/api_v1.py
@@ -296,6 +296,61 @@ class CreateWatch(Resource):
 
         return list, 200
 
+class Import(Resource):
+    def __init__(self, **kwargs):
+        # datastore is a black box dependency
+        self.datastore = kwargs['datastore']
+
+    @auth.check_token
+    def post(self):
+        """
+        @api {post} /api/v1/import - Import a list of watched URLs
+        @apiDescription Accepts a line-feed separated list of URLs to import, additionally with ?tag_uuids=(tag  id), ?tag=(name), ?proxy={key}, ?dedupe=true (default true) one URL per line.
+        @apiExample {curl} Example usage:
+            curl http://localhost:5000/api/v1/import --data-binary @list-of-sites.txt -H"x-api-key:8a111a21bc2f8f1dd9b9353bbd46049a"
+        @apiName Import
+        @apiGroup Watch
+        @apiSuccess (200) {List} OK List of watch UUIDs added
+        @apiSuccess (500) {String} ERR Some other error
+        """
+
+        extras = {}
+
+        if request.args.get('proxy'):
+            plist = self.datastore.proxy_list
+            if not request.args.get('proxy') in plist:
+                return "Invalid proxy choice, currently supported proxies are '{}'".format(', '.join(plist)), 400
+            else:
+                extras['proxy'] = request.args.get('proxy')
+
+        dedupe = strtobool(request.args.get('dedupe', 'true'))
+
+        tags = request.args.get('tag')
+        tag_uuids = request.args.get('tag_uuids')
+
+        if tag_uuids:
+            tag_uuids = tag_uuids.split(',')
+
+        urls = request.get_data().decode('utf8').splitlines()
+        added = []
+        allow_simplehost = not strtobool(os.getenv('BLOCK_SIMPLEHOSTS', 'False'))
+        for url in urls:
+            url = url.strip()
+            if not len(url):
+                continue
+
+            # If hosts that only contain alphanumerics are allowed ("localhost" for example)
+            if not validators.url(url, simple_host=allow_simplehost):
+                return f"Invalid or unsupported URL - {url}", 400
+
+            if dedupe and self.datastore.url_exists(url):
+                continue
+
+            new_uuid = self.datastore.add_watch(url=url, extras=extras, tag=tags, tag_uuids=tag_uuids)
+            added.append(new_uuid)
+
+        return added
+
 class SystemInfo(Resource):
     def __init__(self, **kwargs):
         # datastore is a black box dependency
diff --git a/changedetectionio/flask_app.py b/changedetectionio/flask_app.py
index cd36f9b5..9345eb9a 100644
--- a/changedetectionio/flask_app.py
+++ b/changedetectionio/flask_app.py
@@ -239,6 +239,10 @@ def changedetection_app(config=None, datastore_o=None):
     watch_api.add_resource(api_v1.SystemInfo, '/api/v1/systeminfo',
                            resource_class_kwargs={'datastore': datastore, 'update_q': update_q})
 
+    watch_api.add_resource(api_v1.Import,
+                           '/api/v1/import',
+                           resource_class_kwargs={'datastore': datastore})
+
     # Setup cors headers to allow all domains
     # https://flask-cors.readthedocs.io/en/latest/
     #    CORS(app)
diff --git a/changedetectionio/store.py b/changedetectionio/store.py
index 9522d582..be2546e4 100644
--- a/changedetectionio/store.py
+++ b/changedetectionio/store.py
@@ -234,7 +234,7 @@ class ChangeDetectionStore:
 
         # Probably their should be dict...
         for watch in self.data['watching'].values():
-            if watch['url'] == url:
+            if watch['url'].lower() == url.lower():
                 return True
 
         return False
@@ -333,7 +333,8 @@ class ChangeDetectionStore:
 
         # Or if UUIDs given directly
         if tag_uuids:
-            apply_extras['tags'] = list(set(apply_extras['tags'] + tag_uuids))
+            for t in tag_uuids:
+                apply_extras['tags'] = list(set(apply_extras['tags'] + [t.strip()]))
 
         # Make any uuids unique
         if apply_extras.get('tags'):
diff --git a/changedetectionio/tests/test_api.py b/changedetectionio/tests/test_api.py
index 8f0eb949..d83ababa 100644
--- a/changedetectionio/tests/test_api.py
+++ b/changedetectionio/tests/test_api.py
@@ -357,3 +357,25 @@ def test_api_watch_PUT_update(client, live_server):
     # Cleanup everything
     res = client.get(url_for("form_delete", uuid="all"), follow_redirects=True)
     assert b'Deleted' in res.data
+
+
+def test_api_import(client, live_server):
+    api_key = extract_api_key_from_UI(client)
+
+    res = client.post(
+        url_for("import") + "?tag=import-test",
+        data='https://website1.com\r\nhttps://website2.com',
+        headers={'x-api-key': api_key},
+        follow_redirects=True
+    )
+
+    assert res.status_code == 200
+    assert len(res.json) == 2
+    res = client.get(url_for("index"))
+    assert b"https://website1.com" in res.data
+    assert b"https://website2.com" in res.data
+
+    # Should see the new tag in the tag/groups list
+    res = client.get(url_for('tags.tags_overview_page'))
+    assert b'import-test' in res.data
+    
\ No newline at end of file