Return address
From Wikipedia, the free encyclopedia
In both conventional and electronic messaging, a return address is an explicit inclusion of the address of the person sending the message. It provides the recipient (and sometimes authorized intermediaries) with a means to determine how to respond to the sender of the message if needed.
In programming, return address means the position in code where a program shall return after the subroutine in execution terminates.
[edit] Mail
In the postal mail of some countries, the return address is conventionally located in the upper left hand corner of the envelope, card, or label. It may or may not include a sender's name, but should include address or P.O. box, city, national division (e.g. state, province, county, etc.), and postal code. In the United Kingdom, the return address is usually placed on the reverse of the envelope, near the top.
The return address is not required on postal mail. However, lack of a return address prevents the postal service from being able to return the message (or a package's contents) in the case it is undeliverable (such as from damage, lack of postage, or invalid destination.)
[edit] E-mail messages
In e-mail, the return address is the destination for any Delivery Status Notifications and auto-responses. The return address, also called the envelope sender address, is transmitted in the message envelope, separately from the message content (headers and body). In the case of SMTP via the MAIL FROM command.
When a message is delivered to the recipient's mailbox, the return address is typically recorded in the Return-Path header that is prepended to the message at the destination system.
As with physical mail, the return address of email is easily "forged" to indicate someone other than the actual sender of the message. Spammers, phishers, and other nefarious email senders often forge the return address of email resulting in delivery status notifications and other auto-responses to the forged address, see Joe job and e-mail backscatter.
[edit] Programming
Any useful program is usually subdivided in reusable parts called subroutines or functions. They can be called in different parts of the same program. Hence, after executing the subroutine, the program should remember where the execution should return. That's the return address, that's pushed on the call stack.
Early computers and FORTRAN compilers simply reserved a position at or before the first location of the subroutine to store the return address. This method does not require a stack, which were not generally built into computers until the late 1960s, but does not support recursion. A similar technique was used by Lotus 1-2-3, which executed a tree walk to compute calculation order. Since cycles were not allowed, no recursion was necessary, and the return location could be stored in a reserved location within each cell. The advantage is that no large stack was required which could potentially require one location for every cell allocated.
