Enhancement: add DOI to metadata/importer? about plato

[baskerville]

I could add a retriever based on the DOI.

But the problem is the extractor: the papers don't mention their DOI.

[taosxx]

According to the documentation it seems that you can extract the ISBN as long as:
1.) the document does have an OCR text layer that is not scrambled,
2.) the ISBN is given and listed on the first ten pages.

Thus, I assumed that you don't have to extract the information from e.g. an .opf file in an epub or the actual metadata of a pdf file but you're searching the first 10 pages (of a pdf?) for the first mention of "isbn" and an associated number that looks like a valid ISBN number. If that's the case, a lot of (but of course not all) downloaded pdfs from journals can be searched and include a DOI either on the first, second or last page.

BTW, is the use of "retreive" as function name instead of retrieve intentional?

[baskerville]

Thus, I assumed that you don't have to extract the information from e.g. an .opf file in an epub or the actual metadata of a pdf file but you're searching the first 10 pages (of a pdf?) for the first mention of "isbn" and an associated number that looks like a valid ISBN number.

That's correct. There's also `-M` to extract the metadata from the document's metadata: this is mostly useful for ePUBs.

If that's the case, a lot of (but of course not all) downloaded pdfs from journals can be searched and include a DOI either on the first, second or last page.

I've looked at twenty articles from my library and I didn't find any DOI.

BTW, is the use of "retreive" as function name instead of retrieve intentional?

That's a typo, thanks for catching it.