Why is function `bracket` called bracket in Haskell-CodePudding

Why is the bracket function called bracket?

I assume it has to do with the type signature syntax; brackets are used in the type signature to denote a function. Lets see the type signature of function map:

map :: (a -> b) -> [a] -> [b]

The first parameter of map is a function.

Now, let's have a look at bracket on Hackage.

bracket:: IO a -> (a -> IO b) -> (a -> IO c)  -> IO c

Bracket: https://hackage.haskell.org/package/base-4.16.0.0/docs/Control-Exception.html#v:bracket

Okay I get it, the parameters of bracket are functions with IO actions.

Examples on Hackage use bracket like this:

bracket
  (openFile "filename" ReadMode)
  (hClose)
  (\fileHandle -> do { ... })

After contemplating for a while I considered the name "bracket" not favorable.

I would say "performResourceSafely" would be a better fit.

So I ask again: Why is the bracket function called bracket?

CodePudding user response：

bracket v. To bound on both sides, to surround, as enclosing with brackets
— wiktionary

The thing that bracket does is it wraps some operation that uses a resource with the actions that allocate and deallocate that resource (with some care taken to ensure that deallocation happens even when exceptions are thrown). So the (de)allocation actions surround – or bracket – the main operation.

The file example you mention is especially suggestive. Like "open bracket" for [ or ( and "close bracket" for ] or ), the bracketing actions are "open file" and "close file" (well, "close handle"). This parallel is fairly common; e.g. with databases, one opens and closes a connection, with network stuff, one opens and closes a session, with subprocesses, one opens and closes a program, etc.