Firstorder: record with defined field aren't conjonctions (fix #2629)

 The let-in in the type of constructor was perturbating the
 Hipattern.dependent check, leading firstorder to consider too much
 inductives as dependent-free conjonctions, ending with lift errors
 (UNBOUND_REL_-2, ouch).

 This patch is probably overly cautious : if the variable of the
 let-in is used, we consider the constructor to be dependent.
 If someday somebody feels it necessary, he/she could try to reduce
 the let-in's instead...

git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@16339 85f007b7-540e-0410-9357-904b9bb8a0f7

1 files changed, 22 insertions, 0 deletions

diff --git a/test-suite/bugs/closed/shouldsucceed/2629.v b/test-suite/bugs/closed/shouldsucceed/2629.v
new file mode 100644
index 000000000..759cd3dd2
--- /dev/null
+++ b/test-suite/bugs/closed/shouldsucceed/2629.v
@@ -0,0 +1,22 @@
+Class Join (t: Type) : Type := mkJoin {join: t -> t -> t -> Prop}.
+
+Class sepalg (t: Type) {JOIN: Join t} : Type :=
+  SepAlg   {
+   join_eq: forall {x y z z'}, join x y z -> join x y z' -> z = z';
+   join_assoc: forall {a b c d e}, join a b d -> join d c e ->
+                    {f : t & join b c f /\ join a f e};
+   join_com: forall {a b c}, join a b c -> join b a c;
+   join_canc: forall {a1 a2 b c}, join a1 b c -> join a2 b c -> a1=a2;
+
+   unit_for : t -> t -> Prop := fun e a => join e a a;
+   join_ex_units: forall a, {e : t & unit_for e a}
+}.
+
+Definition joins {A} `{Join A} (a b : A) : Prop :=
+ exists c, join a b c.
+
+Lemma join_joins {A} `{sepalg A}: forall {a b c},
+ join a b c -> joins a b.
+Proof.
+ firstorder.
+Qed.