Message parsing

Zoltern

Hello!

I need to write a program to parse incoming e-mails. The mailbox i need to parse receives only error messages, but i'd need to know what messages and the e-mail used... i cant think of a way to do it, because not every server sends the mail using the standards.. like some use the status code inside the body instead of the head...

Anyone had a problem like that or could point me to a solution (could be another program...)?

Thanks a lot in advance!

Zoltern

Protato

Zoltern,

I don't have any specific code for you, but I have a project outline for something similar.

Basically you're going to have to break it down into two parts. Categorizing the emails, then actually parsing them.

Categorizing you will have to figure out a way to uniquely identify each email that differs in format. You said one has the error code in the header and the other in the body, so those are attributes you can use to distinguish between the two. Or if the reply-to differs for each format that will make it easy.

Next you parse each message where the parse rules are dictated by the category. Read up on all the string and substring functions, they should get you going in the right direction.

Zoltern

Hey, thanks for answering so fast! 🙂

The problem is that some status are in the middle of the body, like this example:

yada yada yada unknown user more yada yada (#5.1.4)

And that makes it dificult to catch..

Im worried about performace too, as its 10k returned e-mails per day.. Of ocurse as the script kicks in, that will drop to a few exceptions only.

Is there any part of the mail that i can look for the status that will be always there? Did i miss something reading the standards or the servers just ignore it?