parse_file replaces double-spaces in filenames

[geritwagner]

I noticed that parse_file() does not seem to handle double spaces in filenames correctly. I am attaching a simple example for a failing test case:

from pybtex.database.input import bibtex
import io

bibtexstr = """
@test{id1,
 file = {fp  .pdf},
 }
"""

ret = bibtex.Parser().parse_file(io.StringIO(bibtexstr))
assert ret.entries["id1"].fields["file"] == "fp  .pdf"