In Haskell, will calling length on a Lazy ByteString force the entire string into memory?

Question

I am reading a large data stream using lazy bytestrings, and want to know if at least X more bytes is available while parsing it. That is, I want to know if the bytestring is at least X bytes long.

Will calling length on it result in the entire stream getting loaded, hence defeating the purpose of using the lazy bytestring?

If yes, then the followup would be: How to tell if it has at least X bytes without loading the entire stream?

EDIT: Originally I asked in the context of reading files but understand that there are better ways to determine filesize. Te ultimate solution I need however should not depend on the lazy bytestring source.

Answer 1

Yes.

length . take x.

Answer 2

Is there a reason you're not using hFileSize :: Handle -> IO Integer for getting the length of the file?

Answer 3

EDIT: sorry. I guess I was thinking bytestrings were lists. There is no genericLength for bytestrings.

length is strict because the type it returns Int is strict. You can use genericLength from Data.List and import a library that defines lazy Peano numbers and gives you a Num instance for them, e.g the numbers library:

That would let you express your function in the way you would like, but ephemient's answer is functionally the same, and doesn't require importing a new library.

I just did a blog post on the subject here, if that sounds like an approach you might be interested in:

http://coder.bsimmons.name/blog/2010/03/lazy-arithmetic-in-haskell/

Answer 4

It will have to iterate the entire string, but if you don't keep a reference to the entire lazy byte-string anywhere else, I believe it should be able to free the head of the string as it progresses towards its tail.