No TLDRs when using Data Partner's API

[alexmaehon]

Regular use of 'sch.get_paper' has TLDR, while access through Data Partner's API does not.
What's wrong?

[danielnsilva]

Maybe this was caused by Partner's default URL, which calls the API deprecating endpoints.

semanticscholar/semanticscholar/SemanticScholar.py

Line 15 in a27da4d

DEFAULT_PARTNER_API_URL = 'https://partner.semanticscholar.org/v1'

Try setting a custom URL.

from semanticscholar import SemanticScholar
s2_api_key = '40-CharacterPrivateKeyProvidedToPartners'
s2_api_url = 'https://partner.semanticscholar.org/graph/v1'
sch = SemanticScholar(api_key=s2_api_key, api_url=s2_api_url)

Please let me know if this works for you.

[alexmaehon]

It works, thank you!

[xinyi6666]

@danielnsilva I encountered the following error message when I try
results = sch.search_paper('Computing Machinery and Intelligence')
without custom url. It went away after I added the custom url. Probably worthy adding the custom url part in the README?

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
/tmp/ipykernel_74557/4191216135.py in <module>
----> 1 results = sch.search_paper('Computing Machinery and Intelligence')

~/.local/lib/python3.8/site-packages/semanticscholar/SemanticScholar.py in search_paper(self, query, year, fields_of_study, fields, limit)
    141             query += f'&fields_of_study={fields_of_study}'
    142 
--> 143         results = PaginatedResults(
    144                 self._requester,
    145                 Paper,

~/.local/lib/python3.8/site-packages/semanticscholar/PaginatedResults.py in __init__(self, requester, data_type, url, query, fields, limit, headers)
     36         self._items = []
     37 
---> 38         self.__get_next_page()
     39 
     40     @property

~/.local/lib/python3.8/site-packages/semanticscholar/PaginatedResults.py in __get_next_page(self)
    101             )
    102 
--> 103         self._data = results['data']
    104         self._total = results['total']
    105         self._offset = results['offset']

[danielnsilva]

Yes, but changing the default URL in the code is the best solution. I will do that shortly.