Fix syntax

.github/workflows/test-container-build.yml

@@ -1,8 +1,7 @@
 name: ChangeDetection.io Container Build Test
 
 # Triggers the workflow on push or pull request events
-on:
-  push:
+on: [push, pull_request]
     paths:
       - requirements.txt
       - Dockerfile

MANIFEST.in

@@ -2,6 +2,7 @@ recursive-include changedetectionio/api *
 recursive-include changedetectionio/templates *
 recursive-include changedetectionio/static *
 recursive-include changedetectionio/model *
+recursive-include changedetectionio/tests *
 include changedetection.py
 global-exclude *.pyc
 global-exclude node_modules

README.md

@@ -161,50 +161,14 @@ This will re-parse the JSON and apply formatting to the text, making it super ea
 
 ### JSONPath or jq?
 
-For more complex parsing, filtering, and modifying of JSON data, jq is recommended due to the built-in operators and functions. Refer to the [documentation](https://stedolan.github.io/jq/manual/) for more information on jq.
+For more complex parsing, filtering, and modifying of JSON data, jq is recommended due to the built-in operators and functions. Refer to the [documentation](https://stedolan.github.io/jq/manual/) for more specifc information on jq.
 
-Notes:
-- `jq` must be added manually separately from the installation of changedetection.io (simply run `pip3 install jq`)
-- `jq` is not available on Windows or must be manually compiled (No "wheel" package available on pypi)
+One big advantage of `jq` is that you can use logic in your JSON filter, such as filters to only show items that have a value greater than/less than etc.
 
-- The example below adds the price in dollars to each item in the JSON data, and then filters to only show items that are greater than 10.
+See the wiki https://github.com/dgtlmoon/changedetection.io/wiki/JSON-Selector-Filter-help for more information and examples
 
-#### Sample input data from API
-```
-{
-    "items": [
-        {
-           "name": "Product A",
-           "priceInCents": 2500
-        },
-        {
-           "name": "Product B",
-           "priceInCents": 500
-        },
-        {
-           "name": "Product C",
-           "priceInCents": 2000
-        }
-    ]
-}
-```
+Note: `jq` library must be added separately (`pip3 install jq`)
 
-#### Sample jq
-`jq:.items[] | . + { "priceInDollars": (.priceInCents / 100) } | select(.priceInDollars > 10)`
-
-#### Sample output data
-```
-{
-  "name": "Product A",
-  "priceInCents": 2500,
-  "priceInDollars": 25
-}
-{
-  "name": "Product C",
-  "priceInCents": 2000,
-  "priceInDollars": 20
-}
-```
 
 ### Parse JSON embedded in HTML!
 

changedetectionio/__init__.py

@@ -33,7 +33,7 @@ from flask_wtf import CSRFProtect
 from changedetectionio import html_tools
 from changedetectionio.api import api_v1
 
-__version__ = '0.39.20.2'
+__version__ = '0.39.20.4'
 
 datastore = None
 
@@ -816,8 +816,10 @@ def changedetection_app(config=None, datastore_o=None):
 
         newest_file = history[dates[-1]]
 
+        # Read as binary and force decode as UTF-8
+        # Windows may fail decode in python if we just use 'r' mode (chardet decode exception)
         try:
-            with open(newest_file, 'r') as f:
+            with open(newest_file, 'r', encoding='utf-8', errors='ignore') as f:
                 newest_version_file_contents = f.read()
         except Exception as e:
             newest_version_file_contents = "Unable to read {}.\n".format(newest_file)
@@ -830,7 +832,7 @@ def changedetection_app(config=None, datastore_o=None):
             previous_file = history[dates[-2]]
 
         try:
-            with open(previous_file, 'r') as f:
+            with open(previous_file, 'r', encoding='utf-8', errors='ignore') as f:
                 previous_version_file_contents = f.read()
         except Exception as e:
             previous_version_file_contents = "Unable to read {}.\n".format(previous_file)
@@ -907,7 +909,7 @@ def changedetection_app(config=None, datastore_o=None):
         timestamp = list(watch.history.keys())[-1]
         filename = watch.history[timestamp]
         try:
-            with open(filename, 'r') as f:
+            with open(filename, 'r', encoding='utf-8', errors='ignore') as f:
                 tmp = f.readlines()
 
                 # Get what needs to be highlighted

changedetectionio/model/Watch.py

@@ -151,28 +151,30 @@ class model(dict):
         import uuid
         import logging
 
-        output_path = "{}/{}".format(self.__datastore_path, self['uuid'])
+        output_path = os.path.join(self.__datastore_path, self['uuid'])
 
         self.ensure_data_dir_exists()
+        snapshot_fname = os.path.join(output_path, str(uuid.uuid4()))
 
-        snapshot_fname = "{}/{}.stripped.txt".format(output_path, uuid.uuid4())
         logging.debug("Saving history text {}".format(snapshot_fname))
 
+        # in /diff/ and /preview/ we are going to assume for now that it's UTF-8 when reading
+        # most sites are utf-8 and some are even broken utf-8
         with open(snapshot_fname, 'wb') as f:
             f.write(contents)
             f.close()
 
         # Append to index
         # @todo check last char was \n
-        index_fname = "{}/history.txt".format(output_path)
+        index_fname = os.path.join(output_path, "history.txt")
         with open(index_fname, 'a') as f:
             f.write("{},{}\n".format(timestamp, snapshot_fname))
             f.close()
 
         self.__newest_history_key = timestamp
-        self.__history_n+=1
+        self.__history_n += 1
 
-        #@todo bump static cache of the last timestamp so we dont need to examine the file to set a proper ''viewed'' status
+        # @todo bump static cache of the last timestamp so we dont need to examine the file to set a proper ''viewed'' status
         return snapshot_fname
 
     @property