I think that lseek(1) as an external program and allowing it to also seek character boundaries based on the current locale is a good idea (in addition to the byte boundaries) and also useful for numerous multibyte locales including UTF-8.
Ienup I. Szczesniak wrote at 05/15/06 07:21: >> > Roland Mainz wrote: >> > ... snip... >> > - Currently there is only support for seeking n-bytes. But what about >> > characters in multibyte locales (CC:'ing Ienup Sung >> <Ienup.Sung at Sun.COM> >> > and i18n-discuss at opensolaris.org for that issue) ? >> ... snip ... >> I see no advantage for adding a built-in for this. > > > Maybe an external program for this may make the way into the POSIX > standard soon - I have forwarded Roland's proposal to some external > people and they seem to be interested. A built in may then the logical > follow up. > The built in can also deal with the support for seeking for holes and > data in a file earlier than syntax extensions to the shell itself. > > Irek
