What Are "Continuation Lines" in a Web Response?

I’m looking at RT 5.0.3 source code. In the “rt” command line tool source there’s a mention of
“continuation lines”. I’m using a proxy so that I can see what RT sends and receives.
Running a simple command, such as “rt show 1” results in a multipart form with 4 sections.
So far so good.

I sometimes see a continuation line in the value for “Attachments:”, e.g.:

1: untitled (39b)

I have the following questions:

  1. Why is it necessary to have a continuation line? Maybe it’s because the line might
    be tricky to parse since there would be two ":"s characters if it were all on one line.

  2. What’s the maximum number of continuation lines? The source code alludes to there
    being more than one, but I’m not sure.

  3. Which keys might have continuations lines? So far, I’ve only seen them on “Attachments:”
    lines but I haven’t looked very hard.

Jon Forrest

In the context of web responses, “continuation lines” refer to a concept in the HTTP protocol specification, specifically in the header section of an HTTP response.

When a server sends an HTTP response, the response headers provide important information about the response, such as the content type, cache control directives, or cookie information. Each header typically consists of a single line with a field name and its corresponding value.