Errors in unicode conversion

[rbawden]

I have come across some problems in the conversion to unicode. I have been using the following code to parse a raw bibtex string (hal_bibtex) using BibTexParser, and although some of the more common escaped characters are correctly converted, others seem to contain some errors:

from bibtexparser.bparser import BibTexParser
from bibtexparser.customization import convert_to_unicode

parser = BibTexParser()
parser.customization = convert_to_unicode
info = bibtexparser.loads(hal_bibtex, parser=parser)

For example, I tried it out on the following bibtex file : https://hal.inria.fr/hal-01682188v1/bibtex, which contains a large number of accented characters. The result contains at least the following errors:

Bibtex:

... Katar{'i}na ... Gr{=u}z{=i}tis ... Jel{'i}nek ... Ljube{\v s}i{'c} ... Mart{'i}nez Alonso ... Ne{\v s}pore-B{=e}rzkalne ... Samard{\v z}i{'c} ... Saul{=i}te, ...

Once converted to unicode:

... Katar\ńa ... and Grūz\̄tis ... and Jel'ék ... and Ljubeši'cŃikola ... Mart'in ́Alonso, ... Nešpore-B=r̄zkalne ... Samardži'c ... Saul=iē, ...

Expected output:

... Katarína ... and Grūzītis ... and Jelínek ... and Ljubešić Nikola ... Martínez Alonso, ... Nešpore‐bērzkalne ... Samardžić ... Saulīte, ...

Thanks in advance for your help!

[MiWeiss]

Duplicate of #272. There's also a workaround provided there.