From FreeBSD:

    Add new option '-p pattern' for splitting files based on matching
    lines in the file with a regular expression. Useful for e.g.
    'cvs diff' output.  Also compile cleanly with -Wall and use
    defines from <sysexits.h>.

1 files changed, 16 insertions, 1 deletions

diff --git a/usr.bin/split/split.1 b/usr.bin/split/split.1
index 0e59fb11795..c2d0022974c 100644
--- a/usr.bin/split/split.1
+++ b/usr.bin/split/split.1
@@ -1,4 +1,4 @@
-.\"	$OpenBSD: split.1,v 1.2 1996/06/26 05:39:28 deraadt Exp $
+.\"	$OpenBSD: split.1,v 1.3 1999/02/04 03:53:48 millert Exp $
 .\"	$NetBSD: split.1,v 1.5 1994/12/21 08:20:35 jtc Exp $
 .\"
 .\" Copyright (c) 1990, 1991, 1993, 1994
@@ -44,6 +44,7 @@
 .Nm split
 .Op Fl b Ar byte_count[k|m]
 .Op Fl l Ar line_count
+.Op Fl p Ar pattern
 .Op Ar file Op Ar name
 .Sh DESCRIPTION
 The
@@ -73,6 +74,16 @@ megabyte pieces.
 Create smaller files
 .Ar n
 lines in length.
+.It Fl p Ar pattern
+The file is split whenever an input line matches
+.Ar pattern ,
+which is interpreted as an extended regular expression.
+The matching line will be the first line of the next output file.
+This option is incompatible with the
+.Fl b
+and
+.Fl l
+options.
 .El
 .Pp
 If additional arguments are specified, the first is used as the name
@@ -95,6 +106,10 @@ For historical reasons, if you specify
 can only create 676 separate
 files.
 The default naming convention allows 2028 separate files.
+.Pp
+The maximum line length for matching patterns is 65536.
+.Sh SEE ALSO
+.Xr re_format 7 .
 .Sh HISTORY
 A
 .Nm split