model help

[Jason Clark]

how do i or can i access a model from the console
say i make
src/model/member.erl
-module(member, [Id, Username, Email, Name, PasswordHash, Key, Status]).
-define(SECRET_STRING, "Not telling secrets!").
-compile(export_all).

%% @doc Returns the session identifier.
session_identifier() ->
    mochihex:to_hex(erlang:md5(?SECRET_STRING ++ Id)).
  

then in the console run
member:new().
** exception error: undefined function member:new/0

member:session_identifier().
** exception error: undefined function member:session_identifier/0
 any ideas
im just following the an evening guide but swtched the ward_boss

[chan sisowath]

hi,

Attrs = [{username, "MyName"},{email "M...@email.com"}],

M = boss_record:new(member, Attrs).

M:username().
M:email().

M1 = M:set(name, "MyName").

M1:name().

M1:save().

--
You received this message because you are subscribed to the Google Groups "ChicagoBoss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chicagoboss...@googlegroups.com.
Visit this group at http://groups.google.com/group/chicagoboss.
To view this discussion on the web visit https://groups.google.com/d/msgid/chicagoboss/9c60f373-bcf1-4308-b7fc-b7e7d1c36e8e%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Igor Clark]

Hey Jason, mihawk gave a great example there. Not sure if you're new to CB and/or Erlang, but I also found this really confusing when I started looking at CB.

If you're at all like me it might help to know that the way the models work is using "parameterized modules", which (if I understand them correctly) are a way to pass state into a variable when you create it through parameters on a special "new" function in the module.

They're kind of controversial in Erlang world as they bring an "OO-like" flavour that many dislike (and argue strongly against), and in fact were removed from the runtime a couple of versions ago, so there's not even much in the way of documentation that comes up in a search (http://myotherpants.com/2009/04/parameterized-modules-in-erlang/ gives a brief overview, and there are a few notes about them on Stack Overflow).

"P-mods" are pretty core to CB's models though, as they allow "construction" of "objects" (really just structured variables with tuples in them under the hood) in a Rails-ish way, and so they're still supported in CB by using a "parse transform", which basically converts code using them to non-parameterized "normal" Erlang code.

So the "new" "method" you're trying to call there is really a module:function call which requires the extra "constructor"-like arguments mihawk showed in the example in order to "initialize" the variable as a parameterized module. Once that's done and you have the p-mod variable in play, you can call the module's functions like "accessors" to get hold of the state variables being looked after by the p-mod.

Like I said I found this really confusing to start because I was trying to get the whole Erlang stateless module:function thing in my head and this seemed to be contrary to that approach; what helped me was to grasp and remember that what one's really doing with the models (and also Requests and other CB constructed that have parameters in their module definitions) is using some syntactic sugar to help you deal with some state stuff in a more familiar way, if you're coming from e.g. Rails or Django or some other environment with an ORM.

Hope this makes sense, and hope others will put me straight if I have anything wrong!

Cheers

Igor

[Jason Clark]

thanks that makes it a lot clearer

[Jesse Gumm]

If you're curious about the internals of Pmods and their
implementation (given that that seems to be where this conversation
has gone), I did a talk a few months ago on "Erlang Gotchas" that
featured Tuple modules and the now-deprecated Pmods that covers how
they work within Erlang: It starts with slide number 7:
http://slides.sigma-star.com/view/Erlang%20Gotchas#/7

Maybe you'll find it helpful along your path of understanding some of
Erlang's deep, dark secrets.

-Jesse

> email to chicagoboss...@googlegroups.com.

> Visit this group at http://groups.google.com/group/chicagoboss.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/chicagoboss/ee5f5f5a-9675-47ad-a752-ae828c7c2ea4%40googlegroups.com.

>
> For more options, visit https://groups.google.com/d/optout.

--
Jesse Gumm
Owner, Sigma Star Systems
414.940.4866 || sigma-star.com || @jessegumm

[Nicolas Michel]

thank you very much !

very helpful :) (and fun)

Speaking about P-mod, and being clearly off-topic, it is a good idea to continue to use P-mod, or to start a new project with P-mod ?

Nicolas -

To view this discussion on the web visit https://groups.google.com/d/msgid/chicagoboss/CAPTXyXe3ybyf3_XciLn0w79z8YZHHowxG4uAKYKn7c3ornMUvQ%40mail.gmail.com.

[Jesse Gumm]

Personally, I would shy away from pmods for a new project, instead preferring the tuple module construct with record definitions if the "pmod style function calls" were necessary, this gives you the option of allowing tuple module style calls, while also allowing the more canonical m:f(a) type calls.

And it also let's you take advantage of dialyzer, which cannot be used to properly analyze tuple module calls.

If, however, your new project is using CB or boss_db, then you'll be using pmods anyway.

-Jesse

--
Jesse Gumm
Owner, Sigma Star Systems
414.940.4866 || sigma-star.com || @jessegumm

To view this discussion on the web visit https://groups.google.com/d/msgid/chicagoboss/CAAQ9oZo2b8MaCtx8VbNfO67iE5EvQ3FYS8Seq9hjqs5CY0yJHQ%40mail.gmail.com.