Dumps from hd are meant to go to a printer, so form feed characters are inserted to cause page ejects at the end of the page. Pages are formatted as follows: At the top of the page is the name of the file being dumped and the column headings. The column headings label the offset of the byte in that column from the address given at the beginning of the line containing the byte.
After the column headings, the dump of the file is presented. Each line contains 16 bytes of data from the file. The first item on the line is the address within the file of the first byte of the line. If the address is preceded by a "+", then it is the beginning of a 512 byte block.
The next item on the line is the hexadecimal representation of the bytes in the file. For machines with the so-called "little-endian" architecture (eg. VAXes, IBM PCs), the bytes are arranged right to left, that is, the byte whose address is given at the beginning of the line is at the rightmost part of the hex dump. For machines with "big-endian" architectures (eg. Sun Workstations), the bytes are arranged left to right.
Following the hex dump is the ASCII dump. It is always arranged left to right. Characters which are not printable (eg. control characters) are represented by a ".".
When the end of file is reached, hd starts a new page.
This manual page uses the term ASCII when it really shouldn’t. Hd will compile and run properly on an EBCDIC machine, since <ctype.h> and isprint() are used.
~src/nemo/util hd.c hd.1 Update History xx-xxx-xx Version 1.0, created (Kenneth J. Montgomery) 28-Jan-87 Version 2.0, various mods (Donald L. Nash) 01-Feb-87 Version 2.0a, nextfile() bug, hexdump left to right (PJT)
Table of Contents