I've just discovered that something that I thought was working is broken in an odd way and I hope maybe someone here can give me a hint as to what is going on. The short version is that I am trying to process a file that I exported from Quark including the Xpress tags (618k) and am doing a lot of substitution stuff in order to prepare the file for use in a hash. There are lots of strange gotchas in this endeavor, but the one I had not counted on was MacPerl failing to read (or at least write) most of the file.It happens to be the A section of a dictionary and even if I just run: local $/; open(FILE1, "a.tag"); $my_file = <FILE1>; print($my_file); close(FILE1); The only portion that prints is from atxxxx instead of aaxxxx as expected. With another file that is tagged with html (423k), the results are even crazier as it begins by printing from lines 8070-8190 (eof) then starts again with lines 7706-8190 (eof). If I use normal line by line processing instead, this behavior is avoided, but it takes forever to run. I am using MacPerl 5.2.0r4 on an 8100/100, OS8.5.1 and did not see any changes when I reinstalled MacPerl or raised the memory allocation to 5000k from the default with vm on. I have not tried to work with these files under Perl in mkLinux or Solaris yet. So, is this a memory thing or what? If it is memory, is there a formula based on file size that I might use to estimate how much to raise MacPerl's usual allocation? If it's something else, will moving my work to mkLinux get around it? Thanks & Merry Xmas. Richard Gordon Gordon Consulting & Design Voice: 770-565-8267 Fax: 770-971-6887 ***** Want to unsubscribe from this list? ***** Send mail with body "unsubscribe" to mac-perl-request@iis.ee.ethz.ch