v3.0: Mostly rewritten, improved and optimised
To reduce the number of requests to the server, the script downloads the data the first time you run it
- code split in several files
- it is possible to limit the results to N authors with the option
-a N, e.g. N=1 for single-author hits or N=10 for hits with 10 or less authors - usage of list and dictionary comprehension, f-strings, warnings, try/except
- data are downloaded in pages with 50 hits and no longer by year; it is faster and more efficient
- JSON keys that can make the local file very large are not saved
- added JSON keys
publication_or_earliest_dateandage_of_publication - data selection rewritten
- add summary of number of document types with the option
-c all - author-level metrics and breakdown of papers are printed only if data contain more than one (published or citeable) hit