...

diff --git a/helm/software/matita/contribs/dama/dama/models/q_rebase.ma b/helm/software/matita/contribs/dama/dama/models/q_rebase.ma
index 93e0d19e73d47c5dcde8289963e817efc62afe3b..81ea33e68efb7eee29fb3df77c715894f6905ad4 100644 (file)
--- a/helm/software/matita/contribs/dama/dama/models/q_rebase.ma
+++ b/helm/software/matita/contribs/dama/dama/models/q_rebase.ma
@@ -18,16 +18,31 @@ alias symbol "lt" = "Q less than".
 alias symbol "Q" = "Rationals".
 axiom q_unlimited: ∀x:ℚ.∃y:ratio.x<Qpos y.                
 axiom q_halving: ∀x,y:ℚ.∃z:ℚ.x<z ∧ z<y.
+alias symbol "not" = "logical not".
+axiom q_not_OQ_lt_Qneg: ∀r. ¬ (OQ < Qneg r).
 lemma same_values_unit_OQ: 
-  ∀b1,b2,h1,l. b2 < b1 → sorted q2_lt (〈b1,h1〉::l) →
+  ∀b1,b2,h1,l. OQ < b2 → b2 < b1 → sorted q2_lt (〈b1,h1〉::l) →
     sorted q2_lt  [〈b2,〈OQ,OQ〉〉] → 
     same_values_simpl (〈b1,h1〉::l) [〈b2,〈OQ,OQ〉〉]  → h1 = 〈OQ,OQ〉.
-intros 4; cases l;
+intros 5 (b1 b2 h1 l POS); cases l;
 [1: intros; cases (q_unlimited b1); cut (b2 < Qpos w); [2:apply (q_lt_trans ??? H H4);] 
     lapply (H3 H1 ? H2 ? w H4 Hcut) as K; simplify; [1,2: autobatch]
     rewrite > (value_unit 〈b1,h1〉) in K;
     rewrite > (value_unit 〈b2,〈OQ,OQ〉〉) in K; assumption;
-|2: intros; (* MANCA che le basi sono positive, 
+|2: intros; unfold in H3; lapply depth=0 (H3 H1 ? H2 ?) as K; [1,2:simplify; autobatch]
+    clear H3; cases (q_halving b1 (b1 + \fst p)) (w Pw); cases w in Pw; intros;
+    [cases (q_lt_le_incompat ?? POS); apply q_lt_to_le; cases H3;
+     apply (q_lt_trans ???? H4); assumption;
+    |3: elim H3; lapply (q_lt_trans ??? H H4); lapply (q_lt_trans ??? POS Hletin);
+        cases (q_not_OQ_lt_Qneg ? Hletin1);
+    | cases H3; lapply (K r);
+      [2: simplify; assumption
+      |3: simplify; apply (q_lt_trans ???? H4); assumption;
+      |rewrite > (value_head b1,h1 .. q) in Hletin;
+        
+
+
+ (* MANCA che le basi sono positive, 
                poi con halving prendi tra b1 e \fst p e hai h1=OQ,OQ*)