grepcidr 2.0 - Filter IPv4 and IPv6 addresses matching CIDR patterns
Copyright (C) 2004 - 2014  Jem E. Berkes <jem@berkes.ca>
	http://www.pc-tools.net/unix/grepcidr/
	http://www.berkes.ca/
	
Sponsored in part by the Spamhaus Project, http://www.spamhaus.org/


DESCRIPTION
-----------
grepcidr can be used to filter a list of IP addresses against one or more
Classless Inter-Domain Routing (CIDR) specifications.  As with grep, there
are options to invert matching and load patterns from a file.  grepcidr is
capable of efficiently processing large numbers of IPs and networks.

grepcidr has endless uses in network software, including: mail filtering and
processing, network security, log analysis, and many custom applications.


COMPILING & INSTALLING
----------------------
	make
	make doc
	make install



COMMAND USAGE
-------------

	grepcidr [-V] [-cisvx] PATTERN [FILE...]
	grepcidr [-V] [-cisvx] [-e PATTERN | -f PATFILE] [FILE...]


-V	Show software version
-c	Display count of the matching lines, instead of showing the lines
-i	Inverse match, include lines without an IP, implies -v
-s	Enforce strict alignment of CIDR mask; host portion must be all zero
-v	Invert the sense of matching, output lines with IPs that don't match
-x	Strict matching, only look at start of line
-e	Specify individual IP or CIDR pattern(s) on command-line
-f	Load individual IP or CIDR pattern(s) from file


MAJOR 2.0 CHANGES
-----------------

Version 2.0 adds support for IPv6 addresses.  As with 1.4, version 2.0
finds IP addresses anywhere on the line.  Use -x to search only from the
start of line, like versions 1.3 and earlier.


USAGE NOTES
-----------

PATTERN specified on the command line may contain multiple patterns
separated by whitespace or commas. For long lists of network patterns,
use -f to load a file where each line contains one pattern (can be IPv4
or IPv6).  Blank lines and comments starting with # are ignored.

Each IPv4 pattern, whether on command line or loaded from a file, may be:
a.b.c.d/xy	(CIDR format)
a.b.c.d-e.f.g.h	(IP range)
a.b.c.d		(Single IP)

And similarly for IPv6:
a:b:c::/xyz	(CIDR format)
a:b:c::		(Single IP)

Dotted-decimal IPv4 format, or any legal IPv6 format is supported (see:
man inet_pton).

IP addresses that appear anywhere on the input line will be compared and
matched against the patterns.  To be recognized, an IP (field) must end
with terminating text.  An IPv4 field terminates upon anything other than
alphanumeric or dot.  An IPv6 field terminates upon anything other than
alphanumeric, dot, or colon.  This is to prevent accidental matching of
ambiguous text such as host names containing reverse DNS.

Use the -x option to do a strict parse without searching the whole line,
and grepcidr will only look for the single IP at the start of the line.


EXAMPLES
--------

grepcidr -f ournetworks blacklist > abuse.log
	Find customers (CIDR ranges in file) that appear in blacklist
	
grepcidr 2001:db8::/32 log.1 log.2
	Search for this IPv6 network inside two files

grepcidr 127.0.0.0/8 iplog
	Searches for any localnet IP addresses inside the iplog file

grepcidr "192.168.0.1-192.168.10.13" iplog
	Searches for IPs matching indicated range in the iplog file

script | grepcidr -vf whitelist > blacklist
	Create a blacklist, with whitelisted networks removed (inverse)

grepcidr -f list1 list2
	Cross-reference two lists, outputs IPs common to both lists


EXIT STATUS
-----------

As with grep: the exit status is 0 if matching IPs are found, and 1
if not found.  If an error occurred the exit status is 2. 
	
	
THANKS
------

Sponsored in part by the Spamhaus Project, http://www.spamhaus.org/

Thanks to John Levine <johnl@taugh.com> for sharing his alternative
implementation.  I used a couple ideas from his code, such as portable
128-bit numbers and support for multiple input files. However, John's
version is quite different and represents a significant fork in the project.

Thanks to Ryan Finnie <ryan@finnie.org> for his work on the Debian package.
I've adopted several of his changes including the Makefile, and the DocBook
format which now is the source of the man page.  Thanks to Ryan for writing
the first manual in DocBook format.

Many thanks to Dick Wesseling <ftu@fi.uu.nl> who suggested an improved data
structure format as well as binary search, to improve grepcidr performance.