Leonardo de Moura
6465b16951
fix(library/class_instance_resolution): memory initialization bug in new type class resolution procedure
2015-11-08 14:04:58 -08:00
Leonardo de Moura
95348bc90b
fix(library/class_instance_resolution): uninitialized variables
2015-11-08 14:04:58 -08:00
Leonardo de Moura
20539d698f
fix(tests/lean): tests affected by new type class resolution procedure
2015-11-08 14:04:58 -08:00
Leonardo de Moura
50186e2db3
feat(library/class_instance_resolution): add old type class resolution procedure back to be able to compile HoTT library
2015-11-08 14:04:58 -08:00
Leonardo de Moura
885393f4de
feat(library/type_inference): check types when assigning meta-variables
2015-11-08 14:04:58 -08:00
Leonardo de Moura
4eafcfe945
chore(library/data/fintype/function): fix indentation
2015-11-08 14:04:58 -08:00
Leonardo de Moura
766fdd415a
feat(library/type_inference): postpone "nontrivial" universe unification constraints
2015-11-08 14:04:58 -08:00
Leonardo de Moura
0bf069f016
feat(library/class_instance_resolution): more liberal type class resolution procedure
...
This modification is needed for the group_theory folder
2015-11-08 14:04:58 -08:00
Leonardo de Moura
33006919b3
refactor(library/theories/group_theory/hom): cleanup definition
2015-11-08 14:04:58 -08:00
Leonardo de Moura
21501ccfa4
fix(library/type_inference): temporary hack for solving universe unification problems
...
We need a better solution. It turns out the universe constraints are not
as simple as we expected.
2015-11-08 14:04:58 -08:00
Leonardo de Moura
5f43b9b183
feat(library/class_instance_resolution): recursively invoke type class resolution when parameters are instances
2015-11-08 14:04:58 -08:00
Leonardo de Moura
6acf7afa16
fix(library/type_inference): bug when using apply_beta
2015-11-08 14:04:58 -08:00
Leonardo de Moura
a10aef0792
fix(library/data/tuple): modify definition to make sure we can compile it using new type class resolution procedure
...
The issue are universe level constraints that cannot be solved by the
new procedure.
2015-11-08 14:04:58 -08:00
Leonardo de Moura
8d8e43abfd
fix(library/class_instance_resolution): transitive instances in the new type class resolution procedure
2015-11-08 14:04:58 -08:00
Leonardo de Moura
5468076400
refactor(library/data): cleanup proofs
2015-11-08 14:04:58 -08:00
Leonardo de Moura
bd41b6ea13
refactor(library/data/rat/basic): cleanup proof
2015-11-08 14:04:58 -08:00
Leonardo de Moura
a1d200e1c6
feat(library/class_instance_resolution): add support for nested type
...
class resolution
2015-11-08 14:04:58 -08:00
Leonardo de Moura
ac01e9e352
refactor(library/data/int/basic): cleanup proof
2015-11-08 14:04:58 -08:00
Leonardo de Moura
bceac9ece5
refactor(library/data/nat/sub): cleanup nasty proofs
2015-11-08 14:04:58 -08:00
Leonardo de Moura
c2ed5d3f1f
fix(library/algebra/category/constructions): make proof more robust
2015-11-08 14:04:58 -08:00
Leonardo de Moura
e9b92adf29
feat(library/type_inference,library/class_instance_resolution): add on_is_def_eq_failure "event handler"
2015-11-08 14:04:57 -08:00
Leonardo de Moura
21bd30d51a
fix(library/class_instance_resolution): bug in mk_choice_point
2015-11-08 14:04:57 -08:00
Leonardo de Moura
0446c43ebf
refactor(library/class_instance_resolution): use new generic type_inference module to implement type class resolution
2015-11-08 14:04:57 -08:00
Leonardo de Moura
73543f1279
fix(library/norm_num): use new type-class resolution procedure at norm_num
2015-11-08 14:04:57 -08:00
Leonardo de Moura
f371182a6c
feat(library/type_inference): generic and cheap type inference, unification, whnf
2015-11-08 14:04:57 -08:00
Leonardo de Moura
a7655b7d43
feat(library/class_instance_resolution): reset internal indices
2015-11-08 14:04:57 -08:00
Leonardo de Moura
1f01e5480d
fix(library/class_instance_resolution): remove reset_cache_and_ctx used for debugging purposes
2015-11-08 14:04:57 -08:00
Leonardo de Moura
2edccf007b
fix(library/class_instance_resolution): make sure that terms synthesized by type class resolution have override the ones synthesized by solving unification constraints.
2015-11-08 14:04:57 -08:00
Leonardo de Moura
b71a68c606
fix(frontends/lean/elaborator): visit_prenum was creating unnecessary universe metavariable
...
This was creating problems for the new type class resolution procedure
since visit_prenum was also not creating any constraint that enforced
the universe level of A to be equal to the superfluous universe level.
2015-11-08 14:04:57 -08:00
Leonardo de Moura
65d7c05737
fix(library/algebra/complete_lattice): avoid looping instances
2015-11-08 14:04:57 -08:00
Leonardo de Moura
4787cf179e
fix(library/class_instance_resolution): skip (instance) meta-variables that have been assigned by solving unification constraints
2015-11-08 14:04:57 -08:00
Leonardo de Moura
eb2236f036
feat(library/class_instance_resolution): bug in mk_choice
2015-11-08 14:04:57 -08:00
Leonardo de Moura
9348701a5b
fix(library/logic/examples/instances_test): we do not support the set_option class.conservative false anymore
2015-11-08 14:04:57 -08:00
Leonardo de Moura
97407eb178
fix(library/class_instance_resolution): add hack for mk_subsingleton_instance API
...
The comment in the source code explains why the hack is needed
2015-11-08 14:04:57 -08:00
Leonardo de Moura
1b92a8333e
fix(library/class_instance_resolution): better is_def_eq for universe levels at new type class resolution procedure
2015-11-08 14:04:57 -08:00
Leonardo de Moura
6b4a891adb
fix(library/class_instance_resolution): bug when creating auxiliary meta-variables at try_instance
2015-11-08 14:04:57 -08:00
Leonardo de Moura
919d55b799
fix(library/class_instance_resolution): position information in the type class trace
2015-11-08 14:04:57 -08:00
Leonardo de Moura
088b0fb795
fix(library/class_instance_resolution): relax has_expr_metavar test at mk_choice
2015-11-08 14:04:57 -08:00
Leonardo de Moura
5f90ff0f07
fix(library/class_instance_resolution): incorrect assertion
2015-11-08 14:04:57 -08:00
Leonardo de Moura
443aca6280
chore(library/class_instance_resolution,library/init_module): fix style
2015-11-08 14:04:57 -08:00
Leonardo de Moura
abcfe0d805
feat(library/class_instance_resolution): add support for attribute [multiple-instances] in the new type class resolution procedure
2015-11-08 14:04:57 -08:00
Leonardo de Moura
2c177d595c
fix(library/class_instance_resolution): bug in cache validation
2015-11-08 14:04:57 -08:00
Leonardo de Moura
343ede3fbe
fix(library/class_instance_resolution): trace option for new type class resolution procedure
2015-11-08 14:04:57 -08:00
Leonardo de Moura
98943f7832
fix(library/class_instance_resolution): initialization bug and ignore universe metavariables not instantiated by unifier
2015-11-08 14:04:57 -08:00
Leonardo de Moura
6a36bffe4b
fix(library/class_instance_resolution): bugs in new type class resolution procedure
2015-11-08 14:04:57 -08:00
Leonardo de Moura
bf17440f31
feat(library/class_instance_resolution): throw exception is maximum depth is reached
2015-11-08 14:04:57 -08:00
Leonardo de Moura
f5819fab60
feat(library/class_instance_resolution): new type class resolution procedure
2015-11-08 14:04:57 -08:00
Leonardo de Moura
c69bbd4eb7
chore(library/norm_num,library/tactic/norm_num_tactic): fix style
2015-11-08 14:04:57 -08:00
Leonardo de Moura
bda0b038b0
fix(library/norm_num): include 'constants.h' instead of 'constants.cpp'
...
By including 'constants.cpp' we get compilation errors in some
platforms (e.g., 'g_one has been defined in multiple files...')
The module constants(.h/cpp) does not expose global variables such as g_one.
We have to use the procedures get_one_name() instead.
2015-11-08 14:04:56 -08:00
Rob Lewis
ce1cbcc205
feat(library/norm_num): give better error message when norm_num fails
2015-11-08 14:04:56 -08:00
