Chapter of book returns the DOI of the book instead of the chapter

[AllergicMushroom]

I have a pdf of https://link.springer.com/chapter/10.1007/978-3-642-21311-3_5, note the _5 in the DOI.

By using pdf2doi [file] -v, it returns the DOI 10.1007/978-3-642-21311-3, which is the book https://link.springer.com/book/10.1007/978-3-642-21311-3 containing the article.

It also says that it was found during method #1. However, I don't see the DOI in the metadata of the PDF.

Is there something misleading pdf2doi and can I make it return the correct DOI, or is this a bug on pdf2doi side ?

[MicheleCotrufo]

I am not sure that chapters are correctly supported by the current implementation of pdf2doi. Do you know if the notation "DOIBOOK_CHAPTERNUMBER" is a standard one?

It also says that it was found during method #1. However, I don't see the DOI in the metadata of the PDF.

This can happen if you run multiple times pdf2doi on the same file.
The first time that you run pdf2doi on a certain file and a DOI is found, that DOI is stored into the pdf metadata. This can be avoided by adding the command -nostore to the pdf2doi invokation). Then if you run pdf2doi a second time on the same file, it will find the DOI into the metadata. If you run pdf2doi filename.pdf -id '' on the file, it will deleted any DOI previously stored by pdf2doi in the file metadata. The next run of pdf2doi on the same file should then reveal the actual method used.

[AllergicMushroom]

I do not know. However, I do know that Zotero manages to find chapters very accurately, so maybe that could be a first direction.

[jmorris335]

No one should be compared to Zotero. There's some serious wizardry going on behind the scenes there.