Une passe sur l'unification des evars (suite aux commits 10124, 10125, 10145)

- Correction bug des filtres dans define_evar_as_abstraction
- Nettoyage, documentation et réorganisations diverses



git-svn-id: svn+ssh://scm.gforge.inria.fr/svn/coq/trunk@10650 85f007b7-540e-0410-9357-904b9bb8a0f7

1 files changed, 7 insertions, 0 deletions

diff --git a/test-suite/success/unification.v b/test-suite/success/unification.v
index e31d67d8c..9de78b138 100644
--- a/test-suite/success/unification.v
+++ b/test-suite/success/unification.v
@@ -109,3 +109,10 @@ intros.
 apply H0. (* Check that equation ?n[H] = ?n[H] is correctly considered true *)
 reflexivity.
 Qed.
+
+(* An example that failed at some time in early 2008 *)
+(* The "fun t => match ..." used to raise an anomaly instead of an error *)
+
+Ltac f := set (x:=fun t => match t with (f,_) => f 0 end).
+Goal True.
+try f. (* if an error, it is caught *)