Text extraction from PDF to annotation not working

[funnym0nk3y]

JabRef version:
JabRef 5.0-beta.438--2020-02-20--5fa1dcf
Windows 10 10.0 amd64
Java 13.0.2

Hi there,

when using JabRef I found a bug in the annotations function. I marked several passages in my pdf-file and they were shown in the annotations tab accordingly. But unfortunately the text extracted from those annotations did not match the marked text. I tried copying the text directly from Adobe Reader DC via right click > copy text and pasting it to notepad, which works fine. There are characters missing within the whole marked area. Also there are passages included which are not marked.

Example:
While these results are excellent from an academic perspective, scaling up the overallcell device area is crucial for achieving practical utility for hybrid perovskite based thin-film solar cells.In this paper, we present a comprehensive study on the use of temperature-controlled doctor bladingtechnique for the growth of large island, crystalline perovskite thin-films. Specifically, we elucidate thephysical conditions such as substrate temperature, solution volume, and blade speed under ambientconditions that control the growth of large area perovskite thin-films with desired island size, thickness,uniformity and crystallinity. Using these doctor-bladed thin-films we fabricated devices of ∼1 cm2areain air that yielded an average efficiency of 7.32% with negligible hysteresis in the current-voltage scans.Further improvements in

In JabRef displayed:
ng up the overall rea is crucial for achieving practical utility for hybrid perovskite based thin-film solar cells.
r, we present a comprehensive study on the use of temperature-controlled doctor blading r the growth of large island

When exported via Acrobat Reader:
Specifically, we elucidate the
physical conditions such as substrate temperature, solution volume, and blade speed under ambient
conditions that control the growth of large area perovskite thin-films with desired island size, thickness,
uniformity and crystallinity

Regards,
funnym0nk3y

[Siedlerchr]

Could you maybe attach or give us a link to the pdf? Is this the only pdf where you noticed that or do experience that with others as well?

[funnym0nk3y]

I experienced this behavior with several PDFs.
Unfortunately I can't attach the PDF here because of copyright protection. But you can find the files here:
https://pubs.acs.org/doi/10.1021/la803646e
https://www.sciencedirect.com/science/article/pii/S2352940716300038?via%3Dihub
If you don't have access through your institution I'll try to find some with OpenAccess.

Besides that I noticed that there are difficulties with some characters like µ,° and sub-/superscript or stuff like 4,23 x 10^2. But that is just an annoyance.

[smihael]

I can reproduce this with the following test.pdf file in a test.bib library.

I am using:
JabRef 5.0--2020-03-06--2e6f433
Linux 5.3.0-42-generic amd64
Java 13.0.2

PDF was generated using LibreOffice and annotations were added in Okular.

[github-actions]

This issue will be closed in 7 days due to inactivity 💤 Please provide the requested information if the problem persists.

[funnym0nk3y]

@smihael provided an example pdf, please update status