trim_eol() assumes that a line always ends either with \n, \r, or \r\n.
This assumption is always valid except for the last line of a document, since it
may miss the trailing newline. LyX does not create such documents, bu they may
result from automatic creation tools, and LyX can read them, so lyx2lyx should
be able to read them as well.
def trim_eol(line):
" Remove end of line char(s)."
+ if line[-1] != '\n':
+ # May happen for the last line of a document
+ return line
if line[-2:-1] == '\r':
return line[:-2]
else: