OS: Monterey 12.1

PHP version:

$ php --version
PHP 8.1.1 (cli) (built: Dec 17 2021 22:21:23) (NTS)
Copyright (c) The PHP Group
Zend Engine v4.1.1, Copyright (c) Zend Technologies
    with Zend OPcache v8.1.1, Copyright (c), by Zend Technologies

$ which php
/opt/homebrew/bin/php

Installed with: brew install php

Clone:

git clone https://github.com/maxgalli/bibcompare.git
cd bitcompare

Start local server:

php -S localhost:8888

Search in browser:

http://localhost:8888/gd.php

No idea of how the API of InspireHEP used to be, but some instructions for the modern API can be found here.

The main feature that broke the behavior of this package probably consists in the fact that the elements returned when parsing a URL are now organized in pages with a size which is 25 (?) by default and can't be higher than 1000. Inside bib_item_fetch_inspirehep_container, the constructor have been changed and now creates an array containing URLs with different pages; we then loop over this array to get the items and perform exactly the same operations that were performed before. Methods and attributes that are not needed anymore were removed.

Perorming a match with a query like dissertori still takes around ~3 minutes, probably this can be taken down by applying some more metadata filtering in the URL itself.

To avoid running over all Guenther's publications on Inspire (which are a lot and take some time to be fetched) it is convenient to comment out the while loop inside bib_item_fetch_inspirehep_container.php, in order to fetch only the first page of results.

Tests are located in the directory.