- advances on csx ...
--- /dev/null
+(**************************************************************************)
+(* ___ *)
+(* ||M|| *)
+(* ||A|| A project by Andrea Asperti *)
+(* ||T|| *)
+(* ||I|| Developers: *)
+(* ||T|| The HELM team. *)
+(* ||A|| http://helm.cs.unibo.it *)
+(* \ / *)
+(* \ / This file is distributed under the terms of the *)
+(* v GNU General Public License Version 2 *)
+(* *)
+(**************************************************************************)
+
+include "basic_2/rt_transition/cpx_lfdeq.ma".
+include "basic_2/rt_computation/csx_csx.ma".
+
+(* STRONGLY NORMALIZING TERMS FOR UNCOUNTED PARALLEL RT-TRANSITION **********)
+
+(* Properties with degree-based equivalence for local environments **********)
+
+(* Basic_2A1: uses: csx_lleq_conf *)
+lemma csx_lfdeq_conf: ∀h,o,G,L1,T. ⦃G, L1⦄ ⊢ ⬈*[h, o] 𝐒⦃T⦄ →
+ ∀L2. L1 ≡[h, o, T] L2 → ⦃G, L2⦄ ⊢ ⬈*[h, o] 𝐒⦃T⦄.
+#h #o #G #L1 #T #H
+@(csx_ind … H) -T #T1 #_ #IH #L2 #HL12
+@csx_intro #T2 #HT12 #HnT12
+elim (lfdeq_cpx_trans … HL12 … HT12) -HT12
+/5 width=4 by cpx_lfdeq_conf_sn, csx_tdeq_trans, tdeq_trans/
+qed-.
+
+(* Basic_2A1: uses: csx_lleq_conf *)
+lemma csx_lfdeq_trans: ∀h,o,L1,L2,T. L1 ≡[h, o, T] L2 →
+ ∀G. ⦃G, L2⦄ ⊢ ⬈*[h, o] 𝐒⦃T⦄ → ⦃G, L1⦄ ⊢ ⬈*[h, o] 𝐒⦃T⦄.
+/3 width=3 by csx_lfdeq_conf, lfdeq_sym/ qed-.
--- /dev/null
+(**************************************************************************)
+(* ___ *)
+(* ||M|| *)
+(* ||A|| A project by Andrea Asperti *)
+(* ||T|| *)
+(* ||I|| Developers: *)
+(* ||T|| The HELM team. *)
+(* ||A|| http://helm.cs.unibo.it *)
+(* \ / *)
+(* \ / This file is distributed under the terms of the *)
+(* v GNU General Public License Version 2 *)
+(* *)
+(**************************************************************************)
+
+include "basic_2/rt_computation/csx_lfpx.ma".
+include "basic_2/rt_computation/lfpxs_fqup.ma".
+
+(* STRONGLY NORMALIZING TERMS FOR UNCOUNTED PARALLEL RT-TRANSITION **********)
+
+(* Properties with uncounted parallel rt-computation on referred entries ****)
+
+(* Basic_2A1: uses: csx_lpxs_conf *)
+lemma csx_lfpxs_conf: ∀h,o,G,L1,L2,T. ⦃G, L1⦄ ⊢ ⬈*[h, T] L2 →
+ ⦃G, L1⦄ ⊢ ⬈*[h, o] 𝐒⦃T⦄ → ⦃G, L2⦄ ⊢ ⬈*[h, o] 𝐒⦃T⦄.
+#h #o #G #L1 #L2 #T #H @(lfpxs_ind_sn … H) -L2 /3 by lfpxs_step_dx, csx_lfpx_conf/
+qed-.
+++ /dev/null
-(**************************************************************************)
-(* ___ *)
-(* ||M|| *)
-(* ||A|| A project by Andrea Asperti *)
-(* ||T|| *)
-(* ||I|| Developers: *)
-(* ||T|| The HELM team. *)
-(* ||A|| http://helm.cs.unibo.it *)
-(* \ / *)
-(* \ / This file is distributed under the terms of the *)
-(* v GNU General Public License Version 2 *)
-(* *)
-(**************************************************************************)
-
-include "basic_2/reduction/cpx_lleq.ma".
-include "basic_2/computation/csx.ma".
-
-(* CONTEXT-SENSITIVE EXTENDED STRONGLY NORMALIZING TERMS ********************)
-
-(* Properties on lazy equivalence for local environments ********************)
-
-lemma csx_lleq_conf: ∀h,o,G,L1,T. ⦃G, L1⦄ ⊢ ⬊*[h, o] T →
- ∀L2. L1 ≡[T, 0] L2 → ⦃G, L2⦄ ⊢ ⬊*[h, o] T.
-#h #o #G #L1 #T #H @(csx_ind … H) -T
-/4 width=6 by csx_intro, cpx_lleq_conf_dx, lleq_cpx_trans/
-qed-.
-
-lemma csx_lleq_trans: ∀h,o,G,L1,L2,T.
- L1 ≡[T, 0] L2 → ⦃G, L2⦄ ⊢ ⬊*[h, o] T → ⦃G, L1⦄ ⊢ ⬊*[h, o] T.
-/3 width=3 by csx_lleq_conf, lleq_sym/ qed-.
+++ /dev/null
-(**************************************************************************)
-(* ___ *)
-(* ||M|| *)
-(* ||A|| A project by Andrea Asperti *)
-(* ||T|| *)
-(* ||I|| Developers: *)
-(* ||T|| The HELM team. *)
-(* ||A|| http://helm.cs.unibo.it *)
-(* \ / *)
-(* \ / This file is distributed under the terms of the *)
-(* v GNU General Public License Version 2 *)
-(* *)
-(**************************************************************************)
-
-include "basic_2/computation/csx_lpx.ma".
-include "basic_2/computation/lpxs.ma".
-
-(* CONTEXT-SENSITIVE EXTENDED STRONGLY NORMALIZING TERMS ********************)
-
-(* Properties on sn extended parallel computation for local environments ****)
-
-lemma csx_lpxs_conf: ∀h,o,G,L1,L2. ⦃G, L1⦄ ⊢ ➡*[h, o] L2 →
- ∀T. ⦃G, L1⦄ ⊢ ⬊*[h, o] T → ⦃G, L2⦄ ⊢ ⬊*[h, o] T.
-#h #o #G #L1 #L2 #H @(lpxs_ind … H) -L2 /3 by lpxs_strap1, csx_lpx_conf/
-qed-.
G ⊢ ⬈*[h, o, V] 𝐒⦃L⦄.
#h #o #I #G #L #V #T #H @(lfsx_ind … H) -L
#L1 #_ #IHL1 @lfsx_intro
-#L2 #H #HnL12 elim (lfpx_pair_sn_split … o I … T H) -H
+#L2 #H #HnL12 elim (lfpx_pair_sn_split … H o I T) -H
/6 width=3 by lfsx_lfdeq_trans, lfdeq_trans, lfdeq_fwd_pair_sn/
qed-.
G ⊢ ⬈*[h, o, T] 𝐒⦃L⦄.
#h #o #I #G #L #V #T #H @(lfsx_ind … H) -L
#L1 #_ #IHL1 @lfsx_intro
-#L2 #H #HnL12 elim (lfpx_flat_dx_split … o I … V … H) -H
+#L2 #H #HnL12 elim (lfpx_flat_dx_split … H o I V) -H
/6 width=3 by lfsx_lfdeq_trans, lfdeq_trans, lfdeq_fwd_flat_dx/
qed-.
+(* Basic_2A1: uses: lsx_fwd_bind_dx *)
+(* Note: the exclusion binder (ⓧ) makes this more elegant and much simpler *)
+lemma lfsx_fwd_bind_dx: ∀h,o,p,I,G,L,V,T. G ⊢ ⬈*[h, o, ⓑ{p,I}V.T] 𝐒⦃L⦄ →
+ G ⊢ ⬈*[h, o, T] 𝐒⦃L.ⓧ⦄.
+#h #o #p #I #G #L #V #T #H @(lfsx_ind … H) -L
+#L1 #_ #IH @lfsx_intro
+#L2 #H #HT elim (lfpx_bind_dx_split_void … H o p I V) -H
+/6 width=5 by lfsx_lfdeq_trans, lfdeq_trans, lfdeq_fwd_bind_dx_void/
+qed-.
+
(* Advanced inversion lemmas ************************************************)
(* Basic_2A1: uses: lsx_inv_flat *)
lemma lfsx_inv_flat: ∀h,o,I,G,L,V,T. G ⊢ ⬈*[h, o, ⓕ{I}V.T] 𝐒⦃L⦄ →
G ⊢ ⬈*[h, o, V] 𝐒⦃L⦄ ∧ G ⊢ ⬈*[h, o, T] 𝐒⦃L⦄.
/3 width=3 by lfsx_fwd_pair_sn, lfsx_fwd_flat_dx, conj/ qed-.
+
+(* Basic_2A1: uses: lsx_inv_bind *)
+lemma lfsx_inv_bind: ∀h,o,p,I,G,L,V,T. G ⊢ ⬈*[h, o, ⓑ{p,I}V.T] 𝐒⦃L⦄ →
+ G ⊢ ⬈*[h, o, V] 𝐒⦃L⦄ ∧ G ⊢ ⬈*[h, o, T] 𝐒⦃L.ⓧ⦄.
+/3 width=4 by lfsx_fwd_pair_sn, lfsx_fwd_bind_dx, conj/ qed-.
cpxs.ma cpxs_tdeq.ma cpxs_theq.ma cpxs_theq_vector.ma cpxs_drops.ma cpxs_fqus.ma cpxs_lsubr.ma cpxs_lfdeq.ma cpxs_aaa.ma cpxs_lfpx.ma cpxs_cnx.ma cpxs_cpxs.ma
lfpxs.ma lfpxs_length.ma lfpxs_drops.ma lfpxs_fqup.ma lfpxs_lfdeq.ma lfpxs_aaa.ma lfpxs_cpxs.ma lfpxs_lfpxs.ma
-csx.ma csx_simple.ma csx_simple_theq.ma csx_drops.ma csx_lsubr.ma csx_aaa.ma csx_gcp.ma csx_gcr.ma csx_lfpx.ma csx_cnx.ma csx_cpxs.ma csx_csx.ma
+csx.ma csx_simple.ma csx_simple_theq.ma csx_drops.ma csx_lsubr.ma csx_lfdeq.ma csx_aaa.ma csx_gcp.ma csx_gcr.ma csx_lfpx.ma csx_cnx.ma csx_cpxs.ma csx_lfpxs.ma csx_csx.ma
csx_vector.ma csx_cnx_vector.ma csx_csx_vector.ma
lfsx.ma lfsx_fqup.ma lfsx_lfpxs.ma lfsx_lfsx.ma
(* Properties with degree-based equivalence for local environments **********)
-lemma lfpx_pair_sn_split: ∀h,o,I,G,L1,L2,V,T. ⦃G, L1⦄ ⊢ ⬈[h, V] L2 →
+lemma lfpx_pair_sn_split: ∀h,G,L1,L2,V. ⦃G, L1⦄ ⊢ ⬈[h, V] L2 → ∀o,I,T.
∃∃L. ⦃G, L1⦄ ⊢ ⬈[h, ②{I}V.T] L & L ≡[h, o, V] L2.
/3 width=5 by lfpx_frees_conf, lfxs_pair_sn_split/ qed-.
-lemma lfpx_flat_dx_split: ∀h,o,I,G,L1,L2,V,T. ⦃G, L1⦄ ⊢ ⬈[h, T] L2 →
+lemma lfpx_flat_dx_split: ∀h,G,L1,L2,T. ⦃G, L1⦄ ⊢ ⬈[h, T] L2 → ∀o,I,V.
∃∃L. ⦃G, L1⦄ ⊢ ⬈[h, ⓕ{I}V.T] L & L ≡[h, o, T] L2.
/3 width=5 by lfpx_frees_conf, lfxs_flat_dx_split/ qed-.
-lemma lfpx_bind_dx_split: ∀h,o,p,I,G,L1,L2,V1,T. ⦃G, L1.ⓑ{I}V1⦄ ⊢ ⬈[h, T] L2 →
+lemma lfpx_bind_dx_split: ∀h,I,G,L1,L2,V1,T. ⦃G, L1.ⓑ{I}V1⦄ ⊢ ⬈[h, T] L2 → ∀o,p.
∃∃L,V. ⦃G, L1⦄ ⊢ ⬈[h, ⓑ{p,I}V1.T] L & L.ⓑ{I}V ≡[h, o, T] L2 & ⦃G, L1⦄ ⊢ V1 ⬈[h] V.
/3 width=5 by lfpx_frees_conf, lfxs_bind_dx_split/ qed-.
+lemma lfpx_bind_dx_split_void: ∀h,G,K1,L2,T. ⦃G, K1.ⓧ⦄ ⊢ ⬈[h, T] L2 → ∀o,p,I,V.
+ ∃∃K2. ⦃G, K1⦄ ⊢ ⬈[h, ⓑ{p,I}V.T] K2 & K2.ⓧ ≡[h, o, T] L2.
+/3 width=5 by lfpx_frees_conf, lfxs_bind_dx_split_void/ qed-.
+
lemma cpx_tdeq_conf_lexs: ∀h,o,G. R_confluent2_lfxs … (cpx h G) (cdeq h o) (cpx h G) (cdeq h o).
#h #o #G #L0 #T0 #T1 #H @(cpx_ind … H) -G -L0 -T0 -T1 /2 width=3 by ex2_intro/
[ #G #L0 #s0 #X0 #H0 #L1 #HL01 #L2 #HL02
/4 width=7 by sle_lexs_trans, ex3_2_intro, ex2_intro/
qed-.
+lemma lfxs_bind_dx_split_void: ∀R1,R2. (∀L. reflexive … (R1 L)) → (∀L. reflexive … (R2 L)) →
+ lexs_frees_confluent … (cext2 R1) cfull →
+ ∀L1,L2,T. L1.ⓧ ⪤*[R1, T] L2 → ∀p,I,V.
+ ∃∃L. L1 ⪤*[R1, ⓑ{p,I}V.T] L & L.ⓧ ⪤*[R2, T] L2.
+#R1 #R2 #HR1 #HR2 #HR #L1 #L2 #T * #f #Hf #HL12 #p #I #V
+elim (frees_total L1 (ⓑ{p,I}V.T)) #g #Hg
+elim (frees_inv_bind_void … Hg) #y1 #y2 #_ #H #Hy
+lapply(frees_mono … H … Hf) -H #H2
+lapply (tl_eq_repl … H2) -H2 #H2
+lapply (sor_eq_repl_back2 … Hy … H2) -y2 #Hy
+lapply (sor_inv_sle_dx … Hy) -y1 #Hfg
+lapply (sle_inv_tl_sn … Hfg) -Hfg #Hfg
+elim (lexs_sle_split (cext2 R1) (cext2 R2) … HL12 … Hfg) -HL12 /2 width=1 by ext2_refl/ #Y #H #HL2
+lapply (sle_lexs_trans … H … Hfg) // #H0
+elim (lexs_inv_next1 … H) -H #Z #L #HL1 #H
+elim (ext2_inv_unit_sn … H) -H #H destruct
+elim (HR … Hf … H0) -HR -Hf -H0
+/4 width=7 by sle_lexs_trans, ex2_intro/ (* note: 2 ex2_intro *)
+qed-.
+
(* Main properties **********************************************************)
(* Basic_2A1: uses: llpx_sn_bind llpx_sn_bind_O *)
[ { "strongly normalizing rt-computation" * } {
[ "lcosx ( ? ⊢ ~⬊*[?,?,?] ? )" "lcosx_cpx" * ]
[ "lsx_drop" + "lsx_lpx" + "lsx_lpxs" + "llsx_csx" * ]
- [ "csx_lleq" + "csx_lpxs" + "csx_fpbs" * ]
+ [ "csx_fpbs" * ]
}
]
[ { "parallel qrst-computation" * } {
[ { "uncounted context-sensitive rt-computation" * } {
[ "lfsx ( ? ⊢ ⬈*[?,?,?] 𝐒⦃?⦄ )" "lfsx_fqup" + "lfsx_lfpxs" + "lfsx_lfsx" * ]
[ "csx_vector ( ⦃?,?⦄ ⊢ ⬈*[?,?] 𝐒⦃?⦄ )" "csx_cnx_vector" + "csx_csx_vector" * ]
- [ "csx ( ⦃?,?⦄ ⊢ ⬈*[?,?] 𝐒⦃?⦄ )" "csx_simple" + "csx_simple_theq" + "csx_drops" + "csx_lsubr" + "csx_aaa" + "csx_gcp" + "csx_gcr" + "csx_lfpx" + "csx_cnx" + "csx_cpxs" + "csx_csx" * ]
+ [ "csx ( ⦃?,?⦄ ⊢ ⬈*[?,?] 𝐒⦃?⦄ )" "csx_simple" + "csx_simple_theq" + "csx_drops" + "csx_lsubr" + "csx_lfdeq" + "csx_aaa" + "csx_gcp" + "csx_gcr" + "csx_lfpx" + "csx_cnx" + "csx_cpxs" + "csx_lfpxs" + "csx_csx" * ]
[ "lfpxs ( ⦃?,?⦄ ⊢ ⬈*[?,?] ? )" "lfpxs_length" + "lfpxs_drops" + "lfpxs_fqup" + "lfpxs_lfdeq" + "lfpxs_aaa" + "lfpxs_cpxs" + "lfpxs_lfpxs" * ]
[ "cpxs ( ⦃?,?⦄ ⊢ ? ⬈*[?] ? )" "cpxs_tdeq" + "cpxs_theq" + "cpxs_theq_vector" + "cpxs_drops" + "cpxs_fqus" + "cpxs_lsubr" + "cpxs_lfdeq" + "cpxs_aaa" + "cpxs_lfpx" + "cpxs_cnx" + "cpxs_cpxs" * ]
}
projects / helm.git / commitdiff