Removing 'bad'chrs from a file

[rikona]

I just installed UB for a friend, and tried to move a previous
Thunderbird email file to a new TB in her new UB. TB indexed it, but
many emails are missing after a certain date. I thought I'd look at
the file with Kate, but Kate says there are 'bad chrs' in the file.
Perhaps that is the problem with TB indexing.

It's a big file, about 1 gig. I also found out that trying to work
with that size file in Kate is essentially impossible. What may be
swaps take many minutes every time I try to do anything to the file.

Tail shows me that all the emails are there, and I know the last one
indexed by TB. One possibility might be to split the file into 3
chunks - one including up to the one before the last 'good' email,
another containing perhaps the next 500 lines or so where the problem
might be, and the third with the rest of the file. That mid-section
might be small enough to be reasonable to view/fix with an editor.

Or, is there a way to just fix/remove the 'bad chrs' from a file? All
[reasonable] {gotta watch out on this list :-) } ideas welcome.

She'd like to get all her emails back... :-))

Thanks,

rikona