Home/ Questions/Q 3879334
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-19T22:42:52+00:00

First of all, I’ll admit this is a school project, so I need more

  • 0

First of all, I’ll admit this is a school project, so I need more of just direction, rather than actual code. Also, I cannot use the java.net.HTTPURLConnection and java.net.URL classes to help out…

Ok… I’m creating a HTTP server in Java, but having trouble reading HTTP messages. I need to allow persistent connections (when using HTTP/1.1 and no Connection: close), which is already working. But because of the persistence, there might be multiple requests coming down the pipeline.

I know I can use the Content-Length header to determine how long the message body will be, in bytes. After that, the next message will come down the pipe.

My questions…

  • Should I read in the message byte by byte or char by char, or can I read in line by line? Going line by line would be fine until hitting the message body…
  • Should the message body be saved in byte format, or is String ok (I know HTML is fine, but will images, etc. break if going through a String?)
  • Is there a scanner that will let me read line by line, and then when I hit the message body, call a getBytes( contentLength ), and it gives me the body?

Thanks again!

Edit: I do not have to support chunked or compressed data.

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    I suggest reading into byte-arrays. And don’t read line wise, but read in chunks of bytes (you can try to use content-length as your chunk size. Beware that read(byte[]) returns the number of bytes actually read and you still may have to call it several times).

    Also beware of chunked encoding, although I admit that I do not know whether servers must support this since I always just cared for coding clients.

    • 0