add NOTICE file

NOTICE

@@ -0,0 +1,4 @@
+WordPress Site Extractor
+Copyright 2022-2024 The University of Sheffield
+
+Portions of this code are derived from WPJsonScraper, which is available under the MIT license. For details, see src/extractor/dl/LICENSE.

README.md

@@ -212,4 +212,4 @@ The columns of each type are subset and exported as a JSON file each.
 
 This software is made available under the terms of the [Apache License version 2.0](LICENSE).
 
-A portion of this software (contained within the `extractor.dl` module) is adapted from [WPJsonScraper](https://github.com/MickaelWalter/wp-json-scraper) by Mickael Walter, made available under the terms of the [MIT license](src/extractor/dl/LICENSE.txt).
+Portions of this software are derived from other works, see [the `NOTICE` file](NOTICE) for further information.

pyproject.toml

@@ -6,6 +6,8 @@ authors=["Freddy Heppell <[email protected]>"]
 packages=[
     { include = "extractor", from = "src"}
 ]
+repository="https://github.com/GateNLP/wordpress-site-extractor"
+license="Apache-2.0"
 
 [tool.poetry.scripts]
 wpextract = "extractor.cli.cli:main"

src/extractor/dl/LICENSE.txt → src/extractor/dl/LICENSE

@@ -1,6 +1,3 @@
-This module is adapted from WPJsonScraper (https://github.com/MickaelWalter/wp-json-scraper)
-The original license of this code is as follows:
-
 Copyright (c) 2018-2020 Mickaël "Kilawyn" Walter
 
 Permission is hereby granted, free of charge, to any person obtaining a copy