Box ha a number of benefits over null -- type system enforced null
checks, stringing computations that can fail together in a safe
fashion, error handling, and so on.
Usually you only need to handle the case where there is a value and
map, foreach or for comprehensions are good for that. If you want to
branch and handle both has-a-value (Full) and doesn't (Empty or
Failure) try using pattern matching:
S.cookie("foobar") match {
case Full(cookie) => // do something with cookie
case _ => // do something without cookie
}
You can also combine pattern matching, comprehensions, and so on for
more complicated things (e.g. operate on multiple boxes, etc.)
HTH,
-Ross
On Nov 16, 2009, at 5:08 AM, DMB <combust...@gmail.com> wrote:
>
> Except if there's no cookie, in almost all cases I'll likely want to
> branch and do something else (write out a cookie, for instance), and
> the non-evaluation of stuff doesn't help me much with that.
>
> And there is checking, and memory allocation / deallocation hidden in
> the for comprehension and Box anyway. It's neither expensive, nor
> verbose, so I don't see null checking as a big deal.
>
> On Nov 16, 1:45 am, Sergey Andreev <andser...@gmail.com> wrote:
>> To avoid null checking. If there is no cookie, you will get Empty
>> otherwise
>> Full(cookie). Moreover with for-comprehensions you don't need any
>> checks at
>> all. If there is no cookie then the doSomethingWithValue won't be
>> evaluated.
>>
>> On Mon, Nov 16, 2009 at 12:34 PM, DMB <combust...@gmail.com> wrote:
>>
>>> I guess that could work, but why go to such lengths where there are
>>> much more straightforward solutions available? What do for
>>> comprehensions buy you in this case? I mean, 99% of the time, when I
>>> want to check for a cookie, I don't need the cookie itself or any of
>>> its properties. I need its value, or null if there's no cookie. Why
>>> not do something a-la RoR:
>>
>>> S.cookieValue("cookieName")
>>
>>> or a-la ASP.NET:
>>
>>> val c = S.cookie("cookieName")
>>> if(c != null) {
>>> val v = c.value
>>> }
>>
>>> Or, indeed, both?
>>
>>> On Nov 16, 1:22 am, Sergey Andreev <andser...@gmail.com> wrote:
>>>> Hi,
>>
>>>> For-comprehensions could help you out:
>>
>>>> for{
>>>> cookie <- S.findCookie(cookieName)
>>>> value <- cookie} doSomethingWithValue
>>
>>>> Regards
>>
>>>> On Mon, Nov 16, 2009 at 12:07 PM, DMB <combust...@gmail.com> wrote:
>>
>>>>> When I call findCookie it returns a Box. Then, the value on the
>>>>> cookie
>>>>> itself is also a box. Hence a ruby one-liner turns into something
>>>>> like:
>>
>>>>> val cookie = S.findCookie(cookieName)
>>>>> if(cookie.isDefined) {
>>>>> val cookieVal = cookie.open_!.value.openOr(null)
>>>>> // Do something with the cookie value
>>>>> }
>>
>>>>> This is very ugly, so I'm guessing I'm doing something wrong,
>>>>> but try
>>>>> as I might, I could not find any examples that would look even
>>>>> vaguely
>>>>> "right" to me.
>>
>>>>> Why can't findCookie return a simple, unboxed HTTPCookie object or
>>>>> null if cookie is not found?
>>>>> Why does the value inside a cookie need to also be Box'ed?
>>
>>>>> For the sake of comparison, here's how you do the same thing in
>>>>> RoR:
>>>>> v = cookies["cookieName"]
>>>>> // Do something with the cookie
>>
>>>>> or ASP.NET:
>>>>> var c = Request.Cookies["CookieName"]
>>>>> if(c != null) {
>>>>> var v = c.Value
>>>>> // Do something with the cookie
>>>>> }
>>
>>>>> I fail to see why Lift should be more complicated.
>>
>>>>> This is with Lift 1.1 M7
> >
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Lift" group.
To post to this group, send email to liftweb@googlegroups.com
To unsubscribe from this group, send email to
liftweb+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/liftweb?hl=en
-~----------~----~----~----~------~----~------~--~---
