CAL Hangman on AppEngine part 2
0 views
Skip to first unread message
Tom Davies
May 12, 2009, 10:31:37 PM5/12/09
to CAL Language Discussion
A description of a CAL module for talking to AppEngine's data store:
http://www.kablambda.org/blog/2009/05/12/cal-hangman-on-gae-part-2-the-datastore/
http://www.kablambda.org/blog/2009/05/12/cal-hangman-on-gae-part-2-the-datastore/
Luke
May 14, 2009, 12:34:10 AM5/14/09
to CAL Language Discussion
Excellent. Thanks again.
I have some extra monad utilities, and one of these days I might tidy
them up and share them - along with my CAL Arrows modules perhaps. :-)
Even without supporting syntactic sugar in CAL, monads are still
occasionally a nice way to go - not always for concerns about side
effects, but often just because they provide a nicer encapsulation.
Having some of the common monads supported (including IO, ST and
State) also helps with doing more direct ports of some Haskell stuff -
on the occasions that you want to toy with Haskell examples.
On May 12, 7:31 pm, Tom Davies <tgdav...@gmail.com> wrote:
> A description of a CAL module for talking to AppEngine's data store:http://www.kablambda.org/blog/2009/05/12/cal-hangman-on-gae-part-2-th...
I have some extra monad utilities, and one of these days I might tidy
them up and share them - along with my CAL Arrows modules perhaps. :-)
Even without supporting syntactic sugar in CAL, monads are still
occasionally a nice way to go - not always for concerns about side
effects, but often just because they provide a nicer encapsulation.
Having some of the common monads supported (including IO, ST and
State) also helps with doing more direct ports of some Haskell stuff -
on the occasions that you want to toy with Haskell examples.
On May 12, 7:31 pm, Tom Davies <tgdav...@gmail.com> wrote:
> A description of a CAL module for talking to AppEngine's data store:http://www.kablambda.org/blog/2009/05/12/cal-hangman-on-gae-part-2-th...
Tom Davies
May 14, 2009, 12:53:51 AM5/14/09
to CAL Language Discussion
I think I'm going to refactor my HTTP request processing combinators
as a monad, even though there is no side effect issue -- ending up
with a function which returns ((a,b),c) is not as nice as being able
to bind return values as you go.
I don't mind the lack of 'do' much -- my biggest gripe is that
anonymousBind is so verbose :-)
as a monad, even though there is no side effect issue -- ending up
with a function which returns ((a,b),c) is not as nice as being able
to bind return values as you go.
I don't mind the lack of 'do' much -- my biggest gripe is that
anonymousBind is so verbose :-)
Reply all
Reply to author
Forward
0 new messages