David T-G wrote: > > Hi, all -- Hello,
> I have some large files (1G-2G) that I want to split into 600M chunks > for writing to CD; unfortunately, I don't have enough spare space to hold > two copies (or I'd just use split). > > It seems that it should be straightforward to open the file, seek > forward (chunksize*(chunknum-1)) bytes, and then spit out (chunksize) > bytes to a new filehandle (and, for when I'm really tight on space, > that filehandle could be stdout and I could pipeline straight through > cdrecord, assuming I could then figure out how to read the CDs to append > the pieces back together). The reassembly side should be as simple as > "cat /mnt/cdrom/chunk >> /path/to/growing_big_file" for every chunk I > have. > > Is it that easy, or am I way off base somehow? I plan to start by > playing with 100-byte files :-) but haven't played with seek or trying to > read a fixed amount of data. Perhaps you want the dd program. man dd John -- use Perl; program fulfillment -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
