I get iconv errors quite regularly. I have written a python script
(filter) to remove all UTF characters. This is sufficient for my
purposes since I hardly ever use UTF, and I can always use diff to see
if there are any UTF characters I want to preserve. I find it much
faster than manually bisecting the document.
Shall I put the following up on the Tools page of Wiki?
#!/usr/bin/env python
import sys
r=sys.stdin.read(1)
while r > '':
if ord(r) < 128:
sys.stdout.write(r)
r=sys.stdin.read(1)
In the long run I don't think users should be expected to spend time
manually fixing iconv errors. At worst LyX could pop up a dialog box:
"LyX cannot handle the following characters ... [Remove] [Replace]
[Cancel]".
--
John C. McCabe-Dansted
PhD Student
University of Western Australia
