Forked from #27

I believe that this package populates the BibTeX title field directly from the input title field. However, if the input subtitle field is present, then the BibTeX title field should instead join the input title and subtitle fields with a colon.

Example: https://doi.org/10.1145/3448016.3452841 ("Vector Quotient Filters: Overcoming the Time/Space Trade-Off in Filter Design")

@jowens for the ProceedingsArticle https://doi.org/10.1145/3448016.3452841 i get this bibtex output:

@inproceedings{https://doi.org/10.1145/3448016.3452841,
    author = {Pandey, Prashant and Conway, Alex and Durie, Joe and Bender, Michael A. and Farach-Colton, Martin and Johnson, Rob},
    booktitle = {Proceedings of the 2021 International Conference on Management of Data},
    copyright = {https://www.acm.org/publications/policies/copyright_policy#Background},
    doi = {10.1145/3448016.3452841},
    month = jun,
    publisher = {Association for Computing Machinery (ACM)},
    title = {Vector Quotient Filters},
    url = {https://dl.acm.org/doi/10.1145/3448016.3452841},
    urldate = {2021-06-09},
    year = {2021}
}

It missed the series information and adds a copyright field, but is it otherwise what you expect?

#40 (comment)

It appears here that your generated BibTeX uses an en-dash for page ranges. I believe it's slightly preferable to use ASCII for everything; not all tools do the right thing (sadly) with non-ASCII. Could you do -- instead?

BibTeX citekeys are currently something like

@inproceedings{https://doi.org/x.y/z.w,

I don't think there's any value to the https://doi.org/ and suggest that should be removed. Shorter is better when having to cite things.

(There's obviously a zillion different citekey styles, and you should just sidestep those; using the DOI is sensible and easy and unique.)

Hello,

We are using commonmeta-py in inveniosoftware/invenio-rdm-records and started running into problems with v0.9.x v0.8.7.

The problem is related to the fact that quartodoc (a documentation dependency), is listed as a main dependency:

The proper way to define such dependencies according to the Poetry documentation is:

[tool.poetry.group.docs]
optional = true

[tool.poetry.group.docs.dependencies]
# Docs deps here...

Do you think you could release a new version with these dependencies out of the main list?

Thanks,
Pablo

Hi @mfenner 👋, very interesting project!

Just a quick heads up that the GitHub landing page still links to the talbot docs:

This is a matter of taste, but when I typeset my own @inproceedings, I follow many bib providers and encode the abbreviation for the conference in the BibTeX series field. This is the abbreviated name of the conference and is much handier to web-search than searching on an entire conference title.

e.g.: ACM: my paper doi:10.1145/3225058.3225122 was published at "ICPP 2018". In the input JSON this is available as event/acronym. I would personally add the field series = ICPP 2018.

I don't see this as a particularly consistent thing across publishers but I would like to see it in the generated BibTeX if possible.

NIH is the big medical funder in the USA, and its CV-making tool appears to be becoming standard across USA funding agencies.

https://www.ncbi.nlm.nih.gov/books/NBK53595/#mybibliography.Adding_Citations_from_a_F
https://support.nlm.nih.gov/knowledgebase/article/KA-05275/en-us

They used to allow import from RIS but are moving to the "MEDLINE/PubMed" format for bibliographies.

https://www.nlm.nih.gov/bsd/mms/medlineelements.html

This format appears to be ... not super common among the various bib formats out there. Nonetheless, here we are.

Any chance of supporting this within commonmeta-py (outputting this format is more important, at least for me, than inputting it)?

The REST API returns an updated date that switches back and forth four hours, presumably because the local server time is set differently (even though the time is returned as GMT + 00:00). If helpful, I can provide examples.

cc @afandian

Would you be willing to take suggestions for bibtex writing? I note

https://community.crossref.org/t/crossref-bibtex-export-improving-it/3306/12

and my very sincere desire to fetch high-quality bibtex from crossref or other sources.

As an example, I note that months in the examples are enclosed in braces (e.g., month = {feb}), whereas bibtex has built-in abbreviations that are preferred (e.g., month=feb). Happy to burst these out into separate issues if you'll take suggestions. Thanks!

The XML API returns publisher names via the "crm-item.publisher-name" endpoint and that name matches with the member database (at https://api.crossref.org/members). In contrast the JSON REST API sometimes returns different publisher names, for example for Springer Nature.

cc @afandian