Extended hexdump for Go

This package implements extended hexdump designed for ease of recognition of patterns in binary data, while being a superset of classic hexdump -C.

Extended hexdump makes the structure of data stand out:

00000000  01 00 00 00 01 00 0c 00  02 00 00 00              |☺▪▪▪☺▪♀▪☻▪▪▪|

00000000  01 00 00 00 00 00 24 00  01 00 00 00 03 00 00 00  |☺▪▪▪▪▪$▪☺▪▪▪♥▪▪▪|
00000010  10 00 00 00 75 6e 6b 6e  6f 77 6e 20 72 65 71 75  |►▪▪▪unknown requ|
00000020  65 73 74 00                                       |est▪|

The format makes it obvious at a quick glance that the data is little-endian, uses 4-byte and 2-byte integers and NUL-terminated strings.

Compare to the classic hexdump -C:

00000000  01 00 00 00 01 00 0c 00  02 00 00 00              |............|

00000000  01 00 00 00 00 00 24 00  01 00 00 00 03 00 00 00  |......$.........|
00000010  10 00 00 00 75 6e 6b 6e  6f 77 6e 20 72 65 71 75  |....unknown requ|
00000020  65 73 74 00                                       |est.|

Usage

Print any block of data to stdout:

xd.Print([]byte("hello world"), 0)

Or format to string:

xd.Sprint([]byte("hello world"), 0)

Second argument is the offset in the printed stream, to properly print addresses and to align printed data:

xd.Sprint([]byte("first chunk"), 0)
xd.Sprint([]byte("second chunk"), 11)
xd.Sprint([]byte("third chunk"), 23)
...

Command-line tool

There is a trivial command-line tool:

go run github.com/dottedmag/xd/xd@master < your-binary

Credits

Extended hexdump is invented by Ange Albertini in Corkami and implemented in braille dump by Justine Tunney.

This implementation differs only in the character used for NUL byte, as ▁ and ░ used by Justine and Ange are hard to count quickly: they are block characters, so consecutive NUL bytes in data become a continuous pattern. ▪ is perfectly suited for NUL.

Legal

© Mikhail Gusarov dottedmag@dottedmag.net.

Licensed under Apache 2.0 license.