Aazhar
Aazhar
Hello we end up with two authors associated to the same ORCID when processing this document : https://mdpi-res.com/d_attachment/foods/foods-11-00274/article_deploy/foods-11-00274-v2.pdf in this case, the DOI was extracted so I think we should...
Hello When using consolidation and the doi is not present in the document, we proceed with fuzzy matching, the problem is that while this should be very helpful in most...
Using some editors formats, extracting ORCID using PDF annotations is not reliable in this case if consolidated , then prefer the crossref metadata for the authors ORCIDS.
https://github.com/kermitt2/grobid-quantities/tree/master/resources/dataset/original/pdf/problematic_files
This is the text from the PDF  and this is the result from the text : > the 2010Hawaii Ironman Triathlon consisting of 3.8 km...
 And the text `appears to be maintained until approximately 35 40 years -of age, followed by modest decreases until 50 years of age`,, you can...
Some requests coming coming about having the possibility to output characters along with their respective attributs (width, height, fonts..)
Content mine regroups a list of some known problematic fonts and maps character to correct unicode (e.g : l -> λ)
This is a suggestion from the user @dlaurie linebreaks except when they would be significant (pdftohtml -xml did that), elision of unnecessary attributes, i.e. rotation=0, angle=0.