some get_paper results are empty

[piotrm0]

Here are two examples:

print(sch.get_paper("c77cd64f26228442ffff9219bfd870c83c8747c0"))
print(sch.get_paper("d7701e78e0bfc92b03a89582e80cfb751ac03f26"))

results in

{}
{}

But those are real papers according to the API:

curl https://api.semanticscholar.org/graph/v1/paper/c77cd64f26228442ffff9219bfd870c83c8747c0
curl https://api.semanticscholar.org/graph/v1/paper/d7701e78e0bfc92b03a89582e80cfb751ac03f26

results in:

{"paperId": "c77cd64f26228442ffff9219bfd870c83c8747c0", "title": "Abstract interpretation: a unified lattice model for static analysis of programs by construction or approximation of fixpoints"}
{"paperId": "d7701e78e0bfc92b03a89582e80cfb751ac03f26", "title": "Explaining Explanations: An Overview of Interpretability of Machine Learning"}

My version is "0.3.2".

[danielnsilva]

I've noticed last few weeks that requesting particular fields produces an internal server error for some papers. In your examples, I believe the tldr field is the cause.

By default, the library tries to fetch all fields. So, you can work around it by restricting the list of returned fields.

from semanticscholar import SemanticScholar                                       
sch = SemanticScholar()
print(sch.get_paper("c77cd64f26228442ffff9219bfd870c83c8747c0", fields=['year']))

curl https://api.semanticscholar.org/graph/v1/paper/c77cd64f26228442ffff9219bfd870c83c8747c0?fields=year

Output:

{'paperId': 'c77cd64f26228442ffff9219bfd870c83c8747c0', 'year': 1977}

This will cause error:

curl https://api.semanticscholar.org/graph/v1/paper/c77cd64f26228442ffff9219bfd870c83c8747c0?fields=tldr

Ouput:

{"message": "Internal Server Error"}

In addition, future versions of the package will raise exceptions for certain conditions like that instead of returning empty objects.

[piotrm0]

Upthumb.