German text recognized incorrectly

[rettetdemdativ]

https://boersengefluester.de/wp-content/uploads/assets/annuals/2019/578560.pdf

Configuration:

• PDF.js version: 2.7.570

Steps to reproduce the problem:

1. Load the linked PDF
2. On the second page, highlight the word "Mitarbeiterinnen" by double-clicking it

What is the expected behavior?
The word should be recognized as a single word and highlighted as such.

What went wrong?
Instead, it's recognized as 3 words: "Mit", "arbei", "terinnen"

This happens for many words in this PDF.

Is there anything that can be changed in how I use pdfjs-dist to make it recognize this as a single word? Just looking at the PDF, I can't tell what would cause this behavior.

[calixteman]

In the pdf, we've this code to display Mitarberterinnen:

[(M)5 (i)5 (t)]TJ
/Span<</ActualText<FEFF00AD>>> BDC 
1.516 0 Td
( )Tj
EMC 
T*
(arbei)Tj
/Span<</ActualText<FEFF00AD>>> BDC 
( )Tj
EMC 
2.306 0 Td
[(terinnen und Mitarbeiter arbeiten in \374ber 100 L\344ndern engagiert im\
 Dienste)5 ( )]TJ

and there are 2 extra spaces (the ( )Tj).
Since there are no Td after the Tj then the spaces are behind the char following them.
Anyway those spaces are in some marked content sections (BDC/EMC) and I didn't find anything in specs on how to deal with that case.