close over the coercion sub separately
[gitmo/Moose.git] / TODO
CommitLineData
85910734 1== Todo for 2.0200
2
3=== Revise MetaRole API to reunify class/role metaroles:
4
5 apply_metaroles(
6 for => $meta,
7 roles => {
8 attribute => [...],
9 class => [...],
10 role_attribute => [ ... ],
11 }
12 );
13
14If the $meta is a class, we apply the roles to the class. If it's a role, we
15hold onto them and apply them as part of applying the role to a class.
16
17To make this all work nicely, we'll probably want to track the original role
18where a method was defined, just like we do with attributes currently. We'll
19also need to store method modifiers with their original role, which may mean
20adding some sort of Moose::Meta::Role::MethodModifier class.
21
22For each role-specific thing (methods, attributes, etc.) we should allow a
23role_attribute, role_method, etc. key. The common case will be that the
24metaroles are intended for the consuming class, but we should allow for
25metaroles on the role's metaobjects as well.
26
27=== Remove a bunch of deprecated crap
28
29See Moose::Manual::Delta for a list. Some of this is important to make merging
30Class::MOP into Moose core much easier. Other is just removing complexity
31needed to support ancient APIs.
32
33== Old todo (does anyone look at this?)
34
8b59f8d6 35-------------------------------------------------------------------------------
7af2c1d2 36BUGS
d03bd989 37-------------------------------------------------------------------------------
43d599e5 38
39-------------------------------------------------------------------------------
7af2c1d2 40FEATURES
8b59f8d6 41-------------------------------------------------------------------------------
42
d4967760 43- DDuncan's Str types
44
d03bd989 45subtype 'Str'
46 => as 'Value'
47 => where { Encode::is_utf8( $_[0] ) or $_[0] !~ m/[^0x00-0x7F]/x }
d4967760 48 => optimize_as { defined($_[0]) && !ref($_[0]) };
49
d03bd989 50subtype 'Blob'
51 => as 'Value'
52 => where { !Encode::is_utf8( $_[0] ) }
d4967760 53 => optimize_as { defined($_[0]) && !ref($_[0]) };
54
8b59f8d6 55- type unions
56
d03bd989 57Add support for doing it with Classes which do not have
8b59f8d6 58a type constraint yet created
59
60- type intersections
61
62Mostly just for Roles
63
64- inherited slot specs
65
db1ab48d 66'does' can be added to,.. but not changed
67(need type unions for this)
8b59f8d6 68
8b59f8d6 69- proxy attributes
70
db1ab48d 71a proxied attribute is an attribute
d03bd989 72which looks like an attribute,
73talks like an attribute, smells
74like an attribute,.. but if you
75look behind the curtain,.. its
db1ab48d 76over there.. in that other object
77
78(... probably be a custom metaclass)
8b59f8d6 79
f90e052d 80- local coerce
81
82[13:16] mst stevan: slight problem with coerce
83[13:16] mst I only get to declare it once
84[13:17] mst so if I'm trying to declare it cast-style per-source-class rather than per-target-class
85[13:17] mst I am extremely screwed
86[13:17] stevan yes
d03bd989 87[13:17] stevan they are not class specific
f90e052d 88[13:18] stevan they are attached to the type constraint itself
89[13:18] * stevan ponders anon-coercion-metaobjects
90[13:18] mst yes, that's fine
91[13:19] mst but when I declare a class
92[13:19] mst I want to be able to say "this class coerces to X type via <this>"
93[13:19] stevan yeah something like that
94[13:19] stevan oh,.. hmm
95[13:20] stevan sort of like inflate/deflate?
96[13:20] stevan around the accessors?
97[13:25] * bluefeet has quit (Remote host closed the connection)
98[13:27] mst no
99[13:27] mst nothing like that
100[13:27] mst like a cast
101[13:31] mst stevan: $obj->foo($bar); where 'foo' expects a 'Foo' object
102[13:31] mst stevan: is effectively <Foo>$bar, right?
103[13:32] mst stevan: I want to be able to say in package Bar
104[13:32] mst stevan: coerce_to 'Foo' via { ... };
105[13:32] mst etc.
d03bd989 106[13:53] stevan hmm
107
f3dc2e80 108-----------------------------------------------------------
109-- Type Constraints refactor
110-----------------------------------------------------------
111
112- add support for locally scoped TC
113
114This would borrow from MooseX::TypeLibrary to prefix the TC with the name
115of the package. It would then be accesible from the outside as the fully
116scoped name, but the local attributes would use it first. (this would need support
117in the registry for this).
118
119- look into sugar extensions
120
121Use roles as sugar layer function providers (ala MooseX::AttributeHelpers). This
122would allow custom metaclasses to provide roles to extend the sugar syntax with.
123
124(NOTE: Talk to phaylon a bit more on this)
125
126- allow a switch of some kind to optionally turn TC checking off at runtime
127
128The type checks can get expensive and some people have suggested that allowing
129the checks to be turned off would be helpful for deploying into performance
130intensive systems. Perhaps this can actually be done as an option to make_immutable?
131
132- misc. minor bits
133
134* make the errors for TCs use ->message
135* look into localizing the messages too
136* make ANON TCs be lazy, so they can possibly be subsituted for the real thing later
137* make ANON TCs more introspectable
138* add this ...
139
140#
141# Type Definition
142#
143subtype 'Username',
144 from 'Str',
145 where { (/[a-z][a-z0-9]+/i or fail('Invalid character(s)'))
146 and (length($_) >= 5 or fail('Too short (less than 5 chars)'))
147 }
148on_fail { MyException->throw(value => $_[0], message => $_[1]) };
149
150# fail() will just return false unless the call is made via
151$tc->check_or_fail($value);
152
153* and then something like this:
154
155subtype Foo => as Bar => where { ... } => scoped => -global;
156subtype Foo => as Bar => where { ... } => scoped => -local;
157
158# or
159
160subtype Foo => as Bar => where { ... } => in __PACKAGE__ ;
161
162# or (not sure if it would be possible)
163
164my $Foo = subtype Bar => where { ... };
165
166# ----------
167
168[17:10] <autarch> stevan: it should do it if I pass coerce => 1 as part of the attribute definition
169[17:12] <stevan> autarch: what I am not 100% sure of is how to tell it to deep coerce and when to not
170[17:13] <stevan> cause a basic coerce is from A to B
171[17:13] <autarch> hmm
172[17:13] <stevan> which is valid for collection types too
173[17:13] <stevan> deep coercion is what you are asking for
174[17:13] <autarch> yeah
175[17:13] <stevan> so perhaps we add deep_coerce => 1
176[17:13] <stevan> which will do it
177[17:13] <autarch> that's fine for me
178[17:13] <stevan> k
179
180coerce_deeply => 1 # reads better
181
7af2c1d2 182-------------------------------------------------------------------------------
183INTERNALS
184-------------------------------------------------------------------------------
185
186- rationalize all the get_X methods for classes (and roles)
187
188We have get_attribute, get_attributes_list, get_all_attributes,
189etc. First, we need to make the method names consistent. If something
190returns an attribute vs a name, that needs to be clear from the method
191name. We also need to make sure that local vs. "entire inheritance
192chain" is clear from the name.
193
7af2c1d2 194This is mostly a CMOP change.
195
196- Metaclass constructors
197
198There's a _lot_ of different conventions in here. Some things to consider:
199
200* new vs _new
201* allowing new( 'name', %args ) vs ( name => 'name', %args )
202* Method->wrap vs Method->new
203
7af2c1d2 204- Moose::Meta::TypeConstraint::Parameter{izable,ized}
205
206The relationship between these two classes is very odd. In particular,
207this line in Parameterized is insane:
208
209 foreach my $type (Moose::Util::TypeConstraints::get_all_parameterizable_types()) {
210
211Why does it need to loop through all parameterizable types? Shouldn't
212it know which parameterizable type it "came from"?
213
214- Moose::Util::TypeConstraints vs Moose::Meta::Type{Coercion,Constraint}
215
216The Util module has _way_ too much functionality. It needs to be
217refactored so it's a thin sugar layer on top of the meta API. As it
218stands now, it does things like parse type names (and determine if
219they're valid), manage the registry, and much more.
220
7af2c1d2 221- Anything with a _(meta)?class method
222
223Every method that returns a class name needs to become a rw attribute
224that can be set via the constructor.
d03bd989 225
cac484fa 226- The Moose::Error stuff
227
228This is sort of half-implemented. We still use Carp directly, and the
229internals can't decide how to throw an error (is it
230Moose->throw_error, __PACKAGE__->throw_error, what?).
231
232The internals need to be made consistent before we expose this to the
233rest of the world.
234
8b59f8d6 235-------------------------------------------------------------------------------
236TO PONDER
237-------------------------------------------------------------------------------
238
239- Moose "strict" mode
240
241use Moose 'strict'; This would allow us to have all sort of expensive tests
d03bd989 242which can be turned off in prod.
243
8b59f8d6 244- Moose::Philosophy.pod
245
246To explain Moose from a very high level
247
687e52bb 248- moosedoc
8b59f8d6 249
687e52bb 250We certainly have enough meta-information to make pretty complete POD docs.
d03bd989 251
252
253