Zefram wrote:
Ed Murphy wrote:
Admitted (and corrected in the database). The system eats
backslashes unless I remember to double them up during entry.
Well fix it then. There's no excuse for the code discriminating against
certain (ASCII printable) characters. I have a dim view of tools that
silently corrupt the historical record by default.
I suspect this one is sufficiently embedded that I couldn't reliably
fix it without risk of more significant errors. Backslashes rarely
arise in this context anyway; a whopping six times (818a, 1267, 1285,
1324, 1736, 1881) in 7+ years, though I suppose there might be a few
more undetected errors as well.
