Thanks @JudithCo for pointing this out
Description
"55 @ 77777" is matched as a full number in the snippet below.
How to reproduce the bug
import edsnlp
nlp = edsnlp.blank('eds')
nlp.add_pipe('eds.normalizer')
nlp.add_pipe('eds.measurements')
nlp("Il mesure 55 @ 77777 ")
causes an issue when the measurement module tries to eval "55 @ 77777" (when it should be a simple number or fraction)
However the issue doesn't occur when eds.normalizer is not added