1 | ! The Full Polymorphic Package |
---|
2 | ! The module in this file is, to the best of our knowledge, |
---|
3 | ! the property of Lawrence Berkeley National Laboratory |
---|
4 | ! Its distribution and commercial usage may therefore be governed by the laws of the |
---|
5 | ! United States of America |
---|
6 | |
---|
7 | module lielib_yang_berz |
---|
8 | use dabnew |
---|
9 | ! use precision_constants |
---|
10 | implicit none |
---|
11 | public |
---|
12 | ! private |
---|
13 | PUBLIC XGAM,XGBM !,FILTRES |
---|
14 | PUBLIC LIEPEEK,INITPERT,HYPER,MAPFLOL |
---|
15 | PUBLIC ETALL1,TAKE,ETALL,DAPEK0,ETINI,DACLRD,DACOPD,DIFD |
---|
16 | PUBLIC INTD,ETCCT,TRXFLO,TRX,FACFLOD,EXPFLO,DALIND,ETINV |
---|
17 | PUBLIC INPUTRES,MAPNORMF,DHDJFLO,GETTURA !,SETIDPRIDPRSET, |
---|
18 | PUBLIC FLOFACG,FLOFAC,DACMUD,CTORFLO,RTOCFLO,CTOR,RTOC,ETPIN |
---|
19 | PUBLIC LIEINIT,PERTPEEK,FLOWPARA,COMCFU |
---|
20 | PUBLIC DAPOK0,FACFLO,EXPFLOD,gofix |
---|
21 | public getcct,GETINV,gtrx,eig6 |
---|
22 | private DFILT,DLIE,FILT,REXT,respoke |
---|
23 | private etallnom,simil |
---|
24 | private dapokzer,davar0,taked,daread,daprid,daflo,daflod,fexpo,etcom,etpoi |
---|
25 | private exp1d,expnd2,liefact,mapnorm,orderflo,nuanaflo,h2pluflo,rotflo,rotiflo |
---|
26 | private ctord,rtocd,resvec,reelflo,midbflo,mulnd2,movearou,movemul,cpart |
---|
27 | private ctoi,itoc,etrtc,etctr,etcjg,etdiv,sympl3,ety,etyt,ety2 !,flip |
---|
28 | integer,public,parameter::ndim=4,nreso=100 |
---|
29 | integer,public::no,nv,nd,nd2,ndpt |
---|
30 | integer, private :: ndc,ndc2,ndt,iref,itu,iflow,jtune,nres !,idpr |
---|
31 | integer, private,dimension(ndim)::nplane,idsta,ista |
---|
32 | real(dp), private,dimension(0:20)::xintex |
---|
33 | real(dp), private,dimension(ndim)::dsta,sta,angle,rad,ps,rads |
---|
34 | real(dp), private,dimension(ndim,nreso)::mx |
---|
35 | !real(dp),private::epsplane |
---|
36 | !real(dp),private,dimension(ndim)::xplane |
---|
37 | !integer,public,parameter::ndim2=2*ndim,ntt=40 |
---|
38 | integer,private,parameter::ndim2=2*ndim,ntt=lnv ! joahn 2008 |
---|
39 | ! integer,private,parameter::ndim2=2*ndim,ntt=100 |
---|
40 | character(120), private :: line |
---|
41 | logical :: frankheader=.true. |
---|
42 | logical :: new_ndpt = .true. |
---|
43 | integer,private :: nt_pos,npt_pos |
---|
44 | logical :: perform_flip = .true. |
---|
45 | integer time_plane |
---|
46 | real(dp), private :: stmem(ndim) |
---|
47 | logical(lp) :: courant_snyder=.true. |
---|
48 | logical(lp) :: check_krein=.true. |
---|
49 | real(dp) :: size_krein=1.e-10_dp |
---|
50 | |
---|
51 | real(dp),dimension(ndim2)::rr_eigen,ri_eigen |
---|
52 | contains |
---|
53 | |
---|
54 | |
---|
55 | |
---|
56 | |
---|
57 | subroutine lieinit(no1,nv1,nd1,ndpt1,time_pos,da_init) !,nis |
---|
58 | implicit none |
---|
59 | !! Lieinit initializes AD Package and Lielib |
---|
60 | integer i,nd1,ndc1,ndpt1,no1,nv1 !,nis |
---|
61 | real(dp),dimension(ndim)::ang,ra,st |
---|
62 | integer, optional :: time_pos |
---|
63 | integer ipause,mypauses |
---|
64 | logical, optional :: da_init |
---|
65 | logical dai |
---|
66 | dai=.true. |
---|
67 | if(present(da_init)) dai=da_init |
---|
68 | |
---|
69 | if(present(time_pos)) then |
---|
70 | time_plane=time_pos |
---|
71 | else |
---|
72 | time_plane=0 |
---|
73 | if(ndpt1==0.and.(nd1==3.or.nd1==4) ) time_plane=3 |
---|
74 | endif |
---|
75 | |
---|
76 | do i=1,ndim |
---|
77 | nplane(i)=2*i-1 |
---|
78 | ang(i)=0.0_dp |
---|
79 | ra(i)=0.0_dp |
---|
80 | st(i)=1.0_dp |
---|
81 | enddo |
---|
82 | no=no1 |
---|
83 | nv=nv1 |
---|
84 | nd=nd1 |
---|
85 | nd2=2*nd1 |
---|
86 | ! do i=1,100 |
---|
87 | ! is(i)=0 |
---|
88 | ! enddo |
---|
89 | if(dai) call daini(no,nv,0) |
---|
90 | ! if(nis.gt.0)call etallnom(is,nis,'$$IS ') |
---|
91 | |
---|
92 | if(ndpt1.eq.0) then |
---|
93 | ndpt=0 |
---|
94 | ndt=0 |
---|
95 | ndc1=0 |
---|
96 | else |
---|
97 | if(.not.new_ndpt) then |
---|
98 | ndpt=ndpt1 |
---|
99 | ndc1=1 |
---|
100 | if(ndpt.eq.nd2) then |
---|
101 | ndt=nd2-1 |
---|
102 | else |
---|
103 | ndt=nd2 |
---|
104 | if(ndpt.ne.nd2-1) then |
---|
105 | line=' LETHAL ERROR IN LIEINIT' |
---|
106 | write(6,*) line |
---|
107 | stop |
---|
108 | endif |
---|
109 | endif |
---|
110 | else |
---|
111 | if(mod(ndpt1,2)==0) then |
---|
112 | ndpt=nd2 |
---|
113 | ndt=nd2-1 |
---|
114 | else |
---|
115 | ndpt=nd2-1 |
---|
116 | ndt=nd2 |
---|
117 | endif |
---|
118 | npt_pos=ndpt1 |
---|
119 | ndc1=1 |
---|
120 | if(mod(npt_pos,2)==0) then |
---|
121 | nt_pos=npt_pos-1 |
---|
122 | else |
---|
123 | nt_pos=npt_pos+1 |
---|
124 | endif |
---|
125 | if(nt_pos==nd2.or.nt_pos==nd2-1) then |
---|
126 | perform_flip=.false. |
---|
127 | else |
---|
128 | perform_flip=.true. |
---|
129 | endif |
---|
130 | if(npt_pos<3.or.npt_pos>nd2) then |
---|
131 | line=' LETHAL ERROR IN LIEINIT' |
---|
132 | write(6,*) line |
---|
133 | stop |
---|
134 | endif |
---|
135 | endif |
---|
136 | endif |
---|
137 | ndc=ndc1 |
---|
138 | ndc2=2*ndc1 |
---|
139 | iref=0 |
---|
140 | call initpert(st,ang,ra) |
---|
141 | ! iref=iref1 |
---|
142 | ! if(iref1.eq.0) then |
---|
143 | itu=0 |
---|
144 | ! else |
---|
145 | ! itu=1 |
---|
146 | ! endif |
---|
147 | ! if(iref1.eq.0) |
---|
148 | iref=-1 |
---|
149 | |
---|
150 | if(lielib_print(1)==1) then |
---|
151 | write(6,'(a17,4(1x,i4))') ' no,nv,nd,ndpt = ',no1,nv1,nd1,ndpt1 |
---|
152 | endif |
---|
153 | |
---|
154 | do i=0,20 |
---|
155 | xintex(i)=0.0_dp |
---|
156 | enddo |
---|
157 | xintex(0) =1.0_dp |
---|
158 | xintex(1) =0.5_dp |
---|
159 | xintex(2) =1.0_dp/12.0_dp |
---|
160 | xintex(4) =-1.0_dp/720e0_dp |
---|
161 | xintex(6) =1.0_dp/30240e0_dp |
---|
162 | xintex(8) =-1.0_dp/1209600e0_dp |
---|
163 | xintex(10)=1.0_dp/21772800e0_dp |
---|
164 | |
---|
165 | |
---|
166 | return |
---|
167 | end subroutine lieinit |
---|
168 | subroutine flowpara(ifl,jtu) |
---|
169 | implicit none |
---|
170 | integer ifl,jtu |
---|
171 | |
---|
172 | iflow=ifl |
---|
173 | jtune=jtu |
---|
174 | return |
---|
175 | end subroutine flowpara |
---|
176 | subroutine pertpeek(st,ang,ra) |
---|
177 | implicit none |
---|
178 | integer i |
---|
179 | real(dp),dimension(ndim)::ang,ra,st |
---|
180 | |
---|
181 | do i=1,nd |
---|
182 | st(i)=sta(i) |
---|
183 | ang(i)=angle(i) |
---|
184 | ra(i)=rad(i) |
---|
185 | enddo |
---|
186 | return |
---|
187 | end subroutine pertpeek |
---|
188 | subroutine inputres(mx1,nres1) |
---|
189 | implicit none |
---|
190 | integer i,j,nres1 |
---|
191 | integer,dimension(ndim,nreso)::mx1 |
---|
192 | |
---|
193 | nres=nres1 |
---|
194 | do i=1,nreso |
---|
195 | do j=1,ndim |
---|
196 | mx(j,i)=0 |
---|
197 | enddo |
---|
198 | enddo |
---|
199 | |
---|
200 | do i=1,nres |
---|
201 | do j=1,ndim |
---|
202 | mx(j,i)=mx1(j,i) |
---|
203 | enddo |
---|
204 | enddo |
---|
205 | return |
---|
206 | end subroutine inputres |
---|
207 | subroutine respoke(mres,nre,ire) |
---|
208 | implicit none |
---|
209 | integer i,ire,j,nre |
---|
210 | integer,dimension(ndim,nreso)::mres |
---|
211 | real(dp),dimension(ndim)::ang,ra,st |
---|
212 | |
---|
213 | iref=ire |
---|
214 | nres=nre |
---|
215 | do j=1,nreso |
---|
216 | do i=1,nd |
---|
217 | mx(i,j)=mres(i,j) |
---|
218 | enddo |
---|
219 | enddo |
---|
220 | call initpert(st,ang,ra) |
---|
221 | return |
---|
222 | end subroutine respoke |
---|
223 | subroutine liepeek(iia,icoast) |
---|
224 | implicit none |
---|
225 | integer,dimension(:)::iia,icoast |
---|
226 | |
---|
227 | iia(1)=no |
---|
228 | iia(2)=nv |
---|
229 | iia(3)=nd |
---|
230 | iia(4)=nd2 |
---|
231 | icoast(1)=ndc |
---|
232 | icoast(2)=ndc2 |
---|
233 | icoast(3)=ndt |
---|
234 | icoast(4)=ndpt |
---|
235 | |
---|
236 | return |
---|
237 | end subroutine liepeek |
---|
238 | |
---|
239 | |
---|
240 | subroutine etallnom(x,n) |
---|
241 | implicit none |
---|
242 | ! CREATES A AD-VARIABLE WHICH CAN BE DESTROYED BY DADAL |
---|
243 | ! allocates vector of n polynomials and give it the name NOM=A10 |
---|
244 | integer i,n,nd2 |
---|
245 | integer,dimension(:)::x |
---|
246 | integer,dimension(4)::i1,i2 |
---|
247 | ! character(10) nom |
---|
248 | |
---|
249 | do i=1,iabs(n) |
---|
250 | x(i)=0 |
---|
251 | call daall0(x(i)) |
---|
252 | enddo |
---|
253 | ! call daallno(x,iabs(n),nom) |
---|
254 | if(n.lt.0) then |
---|
255 | call liepeek(i1,i2) |
---|
256 | nd2=i1(4) |
---|
257 | do i=nd2+1,-n |
---|
258 | call davar(x(i),0.0_dp,i) |
---|
259 | enddo |
---|
260 | endif |
---|
261 | return |
---|
262 | end subroutine etallnom |
---|
263 | subroutine etall(x,n) |
---|
264 | implicit none |
---|
265 | ! allocates vector of n polynomials |
---|
266 | integer i,n,nd2 |
---|
267 | integer,dimension(:)::x |
---|
268 | integer,dimension(4)::i1,i2 |
---|
269 | do i=1,iabs(n) |
---|
270 | x(i)=0 |
---|
271 | enddo |
---|
272 | |
---|
273 | if(.not.frankheader) then |
---|
274 | do i=1,iabs(n) |
---|
275 | call daall0(x(i)) |
---|
276 | enddo |
---|
277 | else |
---|
278 | do i=1,iabs(n) |
---|
279 | call daall1(x(i),'etall ',no,nv) |
---|
280 | enddo |
---|
281 | endif |
---|
282 | if(n.lt.0) then |
---|
283 | call liepeek(i1,i2) |
---|
284 | nd2=i1(4) |
---|
285 | do i=nd2+1,-n |
---|
286 | call davar(x(i),0.0_dp,i) |
---|
287 | enddo |
---|
288 | endif |
---|
289 | return |
---|
290 | end subroutine etall |
---|
291 | subroutine etall1(x) |
---|
292 | implicit none |
---|
293 | integer x |
---|
294 | |
---|
295 | x=0 |
---|
296 | if(.not.frankheader) then |
---|
297 | call daall0(x) |
---|
298 | else |
---|
299 | call daall1(x,'etall ',no,nv) |
---|
300 | endif |
---|
301 | return |
---|
302 | end subroutine etall1 |
---|
303 | |
---|
304 | subroutine etcct(x,y,z) |
---|
305 | implicit none |
---|
306 | ! Z=XoY |
---|
307 | integer i,nt |
---|
308 | integer,dimension(ntt)::ie,iv |
---|
309 | integer,dimension(:)::x,y,z |
---|
310 | if(.not.c_%stable_da) return |
---|
311 | |
---|
312 | nt=nv-nd2 |
---|
313 | if(nt.gt.0) then |
---|
314 | call etallnom(ie,nt) !,'IE ') |
---|
315 | do i=nd2+1,nv |
---|
316 | call davar(ie(i-nd2),0.0_dp,i) |
---|
317 | enddo |
---|
318 | do i=nd2+1,nv |
---|
319 | iv(i)=ie(i-nd2) |
---|
320 | enddo |
---|
321 | endif |
---|
322 | do i=1,nd2 |
---|
323 | iv(i)=y(i) |
---|
324 | enddo |
---|
325 | call dacct(x,nd2,iv,nv,z,nd2) |
---|
326 | if(nt.gt.0) then |
---|
327 | call dadal(ie,nt) |
---|
328 | endif |
---|
329 | return |
---|
330 | end subroutine etcct |
---|
331 | |
---|
332 | subroutine getcct(x,y,z,n) |
---|
333 | implicit none |
---|
334 | ! Z=XoY |
---|
335 | integer i,nt,n |
---|
336 | integer,dimension(ntt)::ie,iv |
---|
337 | integer,dimension(:)::x,y,z |
---|
338 | if(.not.c_%stable_da) return |
---|
339 | |
---|
340 | nt=nv-n |
---|
341 | if(nt.gt.0) then |
---|
342 | call etallnom(ie,nt) !,'IE ') |
---|
343 | do i=n+1,nv |
---|
344 | call davar(ie(i-n),0.0_dp,i) |
---|
345 | enddo |
---|
346 | do i=n+1,nv |
---|
347 | iv(i)=ie(i-n) |
---|
348 | enddo |
---|
349 | endif |
---|
350 | do i=1,n |
---|
351 | iv(i)=y(i) |
---|
352 | enddo |
---|
353 | call dacct(x,n,iv,nv,z,n) |
---|
354 | if(nt.gt.0) then |
---|
355 | call dadal(ie,nt) |
---|
356 | endif |
---|
357 | return |
---|
358 | end subroutine getcct |
---|
359 | |
---|
360 | |
---|
361 | subroutine trx(h,rh,y) |
---|
362 | implicit none |
---|
363 | ! :RH: = Y :H: Y^-1 = :HoY: |
---|
364 | integer i,nt,h,rh |
---|
365 | integer,dimension(ntt)::ie,iv |
---|
366 | integer,dimension(1)::h1,rh1 |
---|
367 | integer,dimension(:)::y |
---|
368 | if(.not.c_%stable_da) return |
---|
369 | |
---|
370 | nt=nv-nd2 |
---|
371 | if(nt.gt.0) then |
---|
372 | call etallnom(ie,nt) !,'IE ') |
---|
373 | do i=nd2+1,nv |
---|
374 | call davar(ie(i-nd2),0.0_dp,i) |
---|
375 | enddo |
---|
376 | do i=nd2+1,nv |
---|
377 | iv(i)=ie(i-nd2) |
---|
378 | enddo |
---|
379 | endif |
---|
380 | do i=1,nd2 |
---|
381 | iv(i)=y(i) |
---|
382 | enddo |
---|
383 | h1(1)=h |
---|
384 | rh1(1)=rh |
---|
385 | call dacct(h1,1,iv,nv,rh1,1) |
---|
386 | if(nt.gt.0) then |
---|
387 | call dadal(ie,nt) |
---|
388 | endif |
---|
389 | return |
---|
390 | end subroutine trx |
---|
391 | |
---|
392 | subroutine gtrx(h,rh,y,n) |
---|
393 | implicit none |
---|
394 | ! :RH: = Y :H: Y^-1 = :HoY: |
---|
395 | integer i,nt,h,rh,n |
---|
396 | integer,dimension(ntt)::ie,iv |
---|
397 | integer,dimension(1)::h1,rh1 |
---|
398 | integer,dimension(:)::y |
---|
399 | if(.not.c_%stable_da) return |
---|
400 | |
---|
401 | nt=nv-n |
---|
402 | if(nt.gt.0) then |
---|
403 | call etallnom(ie,nt) !,'IE ') |
---|
404 | do i=n+1,nv |
---|
405 | call davar(ie(i-n),0.0_dp,i) |
---|
406 | enddo |
---|
407 | do i=n+1,nv |
---|
408 | iv(i)=ie(i-n) |
---|
409 | enddo |
---|
410 | endif |
---|
411 | do i=1,n |
---|
412 | iv(i)=y(i) |
---|
413 | enddo |
---|
414 | h1(1)=h |
---|
415 | rh1(1)=rh |
---|
416 | call dacct(h1,1,iv,nv,rh1,1) |
---|
417 | if(nt.gt.0) then |
---|
418 | call dadal(ie,nt) |
---|
419 | endif |
---|
420 | return |
---|
421 | end subroutine gtrx |
---|
422 | |
---|
423 | subroutine trxflo(h,rh,y) |
---|
424 | implicit none |
---|
425 | ! *RH* = Y *H* Y^-1 CHANGE OF A VECTOR FLOW OPERATOR |
---|
426 | integer j,k,b1,b2 |
---|
427 | integer,dimension(:)::h,rh,y |
---|
428 | integer,dimension(ndim2)::yi,ht |
---|
429 | if(.not.c_%stable_da) return |
---|
430 | |
---|
431 | call etallnom(yi,nd2) ! ,'YI ') |
---|
432 | call etallnom(ht,nd2) ! ,'HT ') |
---|
433 | call etall1(b1 ) |
---|
434 | call etall1(b2 ) |
---|
435 | |
---|
436 | call etinv(y,yi) |
---|
437 | !----- HT= H o Y |
---|
438 | call etcct(h,y,ht) |
---|
439 | !---- |
---|
440 | call daclrd(rh) |
---|
441 | do j=1,nd2 |
---|
442 | do k=1,nd2 |
---|
443 | call dader(k,yi(j),b1) |
---|
444 | call trx(b1,b2,y) |
---|
445 | call damul(b2,ht(k),b1) |
---|
446 | call daadd(b1,rh(j),b2) |
---|
447 | call dacop(b2,rh(j)) |
---|
448 | enddo |
---|
449 | enddo |
---|
450 | |
---|
451 | call dadal1(b2) |
---|
452 | call dadal1(b1) |
---|
453 | call dadal(ht,nd2) |
---|
454 | call dadal(yi,nd2) |
---|
455 | return |
---|
456 | end subroutine trxflo |
---|
457 | |
---|
458 | subroutine simil(a,x,ai,y) |
---|
459 | implicit none |
---|
460 | ! Y= AoXoAI |
---|
461 | integer,dimension(:)::x,y,a,ai |
---|
462 | integer,dimension(ndim2)::w,v |
---|
463 | if(.not.c_%stable_da) return |
---|
464 | |
---|
465 | call etallnom(w,nd2) ! ,'W ') |
---|
466 | call etallnom(v,nd2) ! ,'V ') |
---|
467 | |
---|
468 | call etcct(a,x,w) |
---|
469 | call etcct(w,ai,v) |
---|
470 | |
---|
471 | call dacopd(v,y) |
---|
472 | |
---|
473 | call dadal(v,nd2) |
---|
474 | call dadal(w,nd2) |
---|
475 | return |
---|
476 | end subroutine simil |
---|
477 | |
---|
478 | subroutine etini(x) |
---|
479 | implicit none |
---|
480 | ! X=IDENTITY |
---|
481 | integer i |
---|
482 | integer,dimension(:)::x |
---|
483 | if(.not.c_%stable_da) return |
---|
484 | |
---|
485 | do i=1,nd2 |
---|
486 | call davar(x(i),0.0_dp,i) |
---|
487 | enddo |
---|
488 | return |
---|
489 | end subroutine etini |
---|
490 | |
---|
491 | subroutine etinv(x,y) |
---|
492 | implicit none |
---|
493 | ! Y=X^-1 |
---|
494 | integer i,nt |
---|
495 | integer,dimension(ntt)::ie1,ie2,iv1,iv2 |
---|
496 | integer,dimension(:)::x,y |
---|
497 | if(.not.c_%stable_da) return |
---|
498 | nt=nv-nd2 |
---|
499 | if(nt.gt.0) then |
---|
500 | do i=1,nt |
---|
501 | ie1(i)=0 |
---|
502 | ie2(i)=0 |
---|
503 | enddo |
---|
504 | call etallnom(ie1,nt) !,'IE1 ') |
---|
505 | call etallnom(ie2,nt) !,'IE2 ') |
---|
506 | do i=nd2+1,nv |
---|
507 | call davar(ie1(i-nd2),0.0_dp,i) |
---|
508 | enddo |
---|
509 | do i=nd2+1,nv |
---|
510 | iv1(i)=ie1(i-nd2) |
---|
511 | iv2(i)=ie2(i-nd2) |
---|
512 | enddo |
---|
513 | endif |
---|
514 | do i=1,nd2 |
---|
515 | iv1(i)=x(i) |
---|
516 | iv2(i)=y(i) |
---|
517 | enddo |
---|
518 | |
---|
519 | call dainv(iv1,nv,iv2,nv) |
---|
520 | if(nt.gt.0) then |
---|
521 | call dadal(ie2,nt) |
---|
522 | call dadal(ie1,nt) |
---|
523 | endif |
---|
524 | return |
---|
525 | end subroutine etinv |
---|
526 | |
---|
527 | subroutine etpin(x,y,jj) |
---|
528 | implicit none |
---|
529 | |
---|
530 | integer i,nt |
---|
531 | integer,dimension(ntt)::ie1,ie2,iv1,iv2 |
---|
532 | integer,dimension(:)::x,y,jj |
---|
533 | if(.not.c_%stable_da) return |
---|
534 | |
---|
535 | nt=nv-nd2 |
---|
536 | if(nt.gt.0) then |
---|
537 | do i=1,nt |
---|
538 | ie1(i)=0 |
---|
539 | ie2(i)=0 |
---|
540 | enddo |
---|
541 | call etallnom(ie1,nt) !,'IE1 ') |
---|
542 | call etallnom(ie2,nt) !,'IE2 ') |
---|
543 | do i=nd2+1,nv |
---|
544 | call davar(ie1(i-nd2),0.0_dp,i) |
---|
545 | enddo |
---|
546 | do i=nd2+1,nv |
---|
547 | iv1(i)=ie1(i-nd2) |
---|
548 | iv2(i)=ie2(i-nd2) |
---|
549 | enddo |
---|
550 | endif |
---|
551 | do i=1,nd2 |
---|
552 | iv1(i)=x(i) |
---|
553 | iv2(i)=y(i) |
---|
554 | enddo |
---|
555 | |
---|
556 | call dapin(iv1,nv,iv2,nv,jj) |
---|
557 | if(nt.gt.0) then |
---|
558 | call dadal(ie2,nt) |
---|
559 | call dadal(ie1,nt) |
---|
560 | endif |
---|
561 | return |
---|
562 | end subroutine etpin |
---|
563 | |
---|
564 | subroutine getinv(x,y,n) |
---|
565 | implicit none |
---|
566 | ! Y=X^-1 |
---|
567 | integer i,nt,n |
---|
568 | integer,dimension(ntt)::ie1,ie2,iv1,iv2 |
---|
569 | integer,dimension(:)::x,y |
---|
570 | if(.not.c_%stable_da) return |
---|
571 | |
---|
572 | |
---|
573 | nt=nv-n |
---|
574 | if(nt.gt.0) then |
---|
575 | do i=1,nt |
---|
576 | ie1(i)=0 |
---|
577 | ie2(i)=0 |
---|
578 | enddo |
---|
579 | call etallnom(ie1,nt) !,'IE1 ') |
---|
580 | call etallnom(ie2,nt) !,'IE2 ') |
---|
581 | do i=n+1,nv |
---|
582 | call davar(ie1(i-n),0.0_dp,i) |
---|
583 | enddo |
---|
584 | do i=n+1,nv |
---|
585 | iv1(i)=ie1(i-n) |
---|
586 | iv2(i)=ie2(i-n) |
---|
587 | enddo |
---|
588 | endif |
---|
589 | do i=1,n |
---|
590 | iv1(i)=x(i) |
---|
591 | iv2(i)=y(i) |
---|
592 | enddo |
---|
593 | |
---|
594 | call dainv(iv1,nv,iv2,nv) |
---|
595 | if(nt.gt.0) then |
---|
596 | call dadal(ie2,nt) |
---|
597 | call dadal(ie1,nt) |
---|
598 | endif |
---|
599 | return |
---|
600 | end subroutine getinv |
---|
601 | |
---|
602 | subroutine dapek0(v,x,jj) |
---|
603 | implicit none |
---|
604 | |
---|
605 | integer i,jj |
---|
606 | integer,dimension(ntt)::jd |
---|
607 | integer,dimension(:)::v |
---|
608 | real(dp),dimension(:)::x |
---|
609 | if(.not.c_%stable_da) return |
---|
610 | |
---|
611 | do i=1,ntt |
---|
612 | jd(i)=0 |
---|
613 | enddo |
---|
614 | do i=1,jj |
---|
615 | call dapek(v(i),jd,x(i)) |
---|
616 | enddo |
---|
617 | return |
---|
618 | end subroutine dapek0 |
---|
619 | |
---|
620 | subroutine dapok0(v,x,jj) |
---|
621 | implicit none |
---|
622 | integer i,jj |
---|
623 | integer,dimension(ntt)::jd |
---|
624 | integer,dimension(:)::v |
---|
625 | real(dp),dimension(:)::x |
---|
626 | if(.not.c_%stable_da) return |
---|
627 | |
---|
628 | do i=1,ntt |
---|
629 | jd(i)=0 |
---|
630 | enddo |
---|
631 | do i=1,jj |
---|
632 | call dapok(v(i),jd,x(i)) |
---|
633 | enddo |
---|
634 | return |
---|
635 | end subroutine dapok0 |
---|
636 | |
---|
637 | subroutine dapokzer(v,jj) |
---|
638 | implicit none |
---|
639 | integer i,jj |
---|
640 | integer,dimension(:)::v |
---|
641 | integer,dimension(ntt)::jd |
---|
642 | if(.not.c_%stable_da) return |
---|
643 | |
---|
644 | do i=1,ntt |
---|
645 | jd(i)=0 |
---|
646 | enddo |
---|
647 | do i=1,jj |
---|
648 | call dapok(v(i),jd,0.0_dp) |
---|
649 | enddo |
---|
650 | return |
---|
651 | end subroutine dapokzer |
---|
652 | |
---|
653 | subroutine davar0(v,x,jj) |
---|
654 | implicit none |
---|
655 | integer i,jj |
---|
656 | integer,dimension(:)::v |
---|
657 | real(dp),dimension(:)::x |
---|
658 | if(.not.c_%stable_da) return |
---|
659 | |
---|
660 | do i=1,jj |
---|
661 | call davar(v(i),x(i),i) |
---|
662 | enddo |
---|
663 | return |
---|
664 | end subroutine davar0 |
---|
665 | |
---|
666 | subroutine comcfu(b,f1,f2,c) |
---|
667 | implicit none |
---|
668 | ! Complex dacfu |
---|
669 | integer,dimension(2)::b,c |
---|
670 | integer,dimension(4)::t |
---|
671 | real(dp),external::f1,f2 |
---|
672 | logical doflip |
---|
673 | if(.not.c_%stable_da) return |
---|
674 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
675 | perform_flip=.false. |
---|
676 | call flip_i(b(1),b(1),1) |
---|
677 | call flip_i(b(2),b(2),1) |
---|
678 | doflip=.true. |
---|
679 | else |
---|
680 | doflip=.false. |
---|
681 | endif |
---|
682 | |
---|
683 | |
---|
684 | call etall(t,4) |
---|
685 | |
---|
686 | call dacfu(b(1),f1,t(1)) |
---|
687 | call dacfu(b(1),f2,t(2)) |
---|
688 | call dacfu(b(2),f1,t(3)) |
---|
689 | call dacfu(b(2),f2,t(4)) |
---|
690 | |
---|
691 | call dasub(t(1),t(4),c(1)) |
---|
692 | call daadd(t(2),t(3),c(2)) |
---|
693 | call dadal(t,4) |
---|
694 | if(doflip) then |
---|
695 | call flip_i(b(1),b(1),-1) |
---|
696 | call flip_i(b(2),b(2),-1) |
---|
697 | if(c(1)/=b(1).and.c(1)/=b(2)) call flip_i(c(1),c(1),-1) |
---|
698 | if(c(2)/=b(1).and.c(2)/=b(2).and.c(2)/=c(1)) call flip_i(c(2),c(2),-1) |
---|
699 | perform_flip=.true. |
---|
700 | endif |
---|
701 | return |
---|
702 | end subroutine comcfu |
---|
703 | |
---|
704 | subroutine take(h,m,ht) |
---|
705 | implicit none |
---|
706 | ! HT= H_M (TAKES M^th DEGREE PIECE ALL VARIABLES INCLUDED) |
---|
707 | integer i,m,h,ht,b1,b2,b3 |
---|
708 | integer,dimension(ntt)::j |
---|
709 | real(dp) r |
---|
710 | if(.not.c_%stable_da) return |
---|
711 | |
---|
712 | call etall1(b1) |
---|
713 | call etall1(b2) |
---|
714 | call etall1(b3) |
---|
715 | |
---|
716 | if(no.ge.2) then |
---|
717 | if(m.eq.0) then |
---|
718 | do i=1,ntt |
---|
719 | j(i)=0 |
---|
720 | enddo |
---|
721 | call dapek(h,j,r) |
---|
722 | call dacon(ht,r) |
---|
723 | else |
---|
724 | ! call danot(m) |
---|
725 | ! call dacop(h,b1) |
---|
726 | call datrunc(h,m+1,b1) |
---|
727 | ! call danot(m-1) |
---|
728 | ! call dacop(b1,b2) |
---|
729 | call datrunc(b1,m,b2) |
---|
730 | ! call danot(no) |
---|
731 | call dasub(b1,b2,b3) |
---|
732 | call dacop(b3,ht) |
---|
733 | endif |
---|
734 | else |
---|
735 | do i=1,ntt |
---|
736 | j(i)=0 |
---|
737 | enddo |
---|
738 | if(m.eq.0) then |
---|
739 | call dapek(h,j,r) |
---|
740 | call dacon(ht,r) |
---|
741 | elseif(m.eq.1) then |
---|
742 | do i=1,nv |
---|
743 | j(i)=1 |
---|
744 | call dapek(h,j,r) |
---|
745 | call dapok(b3,j,r) |
---|
746 | j(i)=0 |
---|
747 | enddo |
---|
748 | call dacop(b3,ht) |
---|
749 | else |
---|
750 | call daclr(ht) |
---|
751 | endif |
---|
752 | endif |
---|
753 | |
---|
754 | call dadal1(b3) |
---|
755 | call dadal1(b2) |
---|
756 | call dadal1(b1) |
---|
757 | return |
---|
758 | end subroutine take |
---|
759 | |
---|
760 | subroutine taked(h,m,ht) |
---|
761 | implicit none |
---|
762 | ! \VEC{HT}= \VEC{H_M} (TAKES M^th DEGREE PIECE ALL VARIABLES INCLUDED) |
---|
763 | integer i,m,b1,b2 |
---|
764 | integer,dimension(:)::h,ht |
---|
765 | integer,dimension(ntt)::j |
---|
766 | integer,dimension(ndim2)::x |
---|
767 | if(.not.c_%stable_da) return |
---|
768 | |
---|
769 | call etall1(b1) |
---|
770 | call etall1(b2) |
---|
771 | call etallnom(x,nd2) ! ,'X ') |
---|
772 | |
---|
773 | |
---|
774 | do i=1,ntt |
---|
775 | j(i)=0 |
---|
776 | enddo |
---|
777 | |
---|
778 | do i=1,nd2 |
---|
779 | call take(h(i),m,ht(i)) |
---|
780 | enddo |
---|
781 | call dadal(x,nd2) |
---|
782 | call dadal1(b2) |
---|
783 | call dadal1(b1) |
---|
784 | return |
---|
785 | end subroutine taked |
---|
786 | |
---|
787 | subroutine daclrd(h) |
---|
788 | implicit none |
---|
789 | ! clear a map : a vector of nd2 polynomials |
---|
790 | integer i |
---|
791 | integer,dimension(:)::h |
---|
792 | if(.not.c_%stable_da) return |
---|
793 | |
---|
794 | do i=1,nd2 |
---|
795 | call daclr(h(i)) |
---|
796 | enddo |
---|
797 | return |
---|
798 | end subroutine daclrd |
---|
799 | |
---|
800 | subroutine dacopd(h,ht) |
---|
801 | implicit none |
---|
802 | ! H goes into HT (nd2 array) |
---|
803 | integer i |
---|
804 | integer,dimension(:)::h,ht |
---|
805 | if(.not.c_%stable_da) return |
---|
806 | |
---|
807 | do i=1,nd2 |
---|
808 | call dacop(h(i),ht(i)) |
---|
809 | enddo |
---|
810 | return |
---|
811 | end subroutine dacopd |
---|
812 | |
---|
813 | subroutine datruncd(h,io,ht) |
---|
814 | implicit none |
---|
815 | ! H goes into HT (nd2 array) |
---|
816 | integer i |
---|
817 | integer,dimension(:)::h,ht |
---|
818 | integer io |
---|
819 | if(.not.c_%stable_da) return |
---|
820 | |
---|
821 | do i=1,nd2 |
---|
822 | call datrunc(h(i),io,ht(i)) |
---|
823 | enddo |
---|
824 | return |
---|
825 | end subroutine datruncd |
---|
826 | |
---|
827 | subroutine dacmud(h,sca,ht) |
---|
828 | implicit none |
---|
829 | integer i |
---|
830 | integer,dimension(:)::h,ht |
---|
831 | real(dp) sca |
---|
832 | if(.not.c_%stable_da) return |
---|
833 | |
---|
834 | do i=1,nd2 |
---|
835 | call dacmu(h(i),sca,ht(i)) |
---|
836 | enddo |
---|
837 | return |
---|
838 | end subroutine dacmud |
---|
839 | subroutine dalind(h,rh,ht,rt,hr) |
---|
840 | implicit none |
---|
841 | integer i |
---|
842 | integer,dimension(:)::h,ht,hr |
---|
843 | integer,dimension(ndim2)::b |
---|
844 | real(dp) rh,rt |
---|
845 | if(.not.c_%stable_da) return |
---|
846 | |
---|
847 | call etallnom(b,nd2) ! ,'B ') |
---|
848 | |
---|
849 | do i=1,nd2 |
---|
850 | call dalin(h(i),rh,ht(i),rt,b(i)) |
---|
851 | enddo |
---|
852 | call dacopd(b,hr) |
---|
853 | call dadal(b,nd2) |
---|
854 | return |
---|
855 | end subroutine dalind |
---|
856 | subroutine daread(h,nd1,mfile,xipo) |
---|
857 | implicit none |
---|
858 | ! read a map |
---|
859 | integer i,mfile,nd1 |
---|
860 | integer,dimension(:)::h |
---|
861 | integer,dimension(ntt)::j |
---|
862 | real(dp) rx,xipo |
---|
863 | if(.not.c_%stable_da) return |
---|
864 | |
---|
865 | do i=1,ntt |
---|
866 | j(i)=0 |
---|
867 | enddo |
---|
868 | do i=1,nd1 |
---|
869 | call darea(h(i),mfile) |
---|
870 | call dapek(h(i),j,rx) |
---|
871 | rx=rx*xipo |
---|
872 | call dapok(h(i),j,rx) |
---|
873 | enddo |
---|
874 | return |
---|
875 | end subroutine daread |
---|
876 | subroutine daprid(h,n1,n2,mfile) |
---|
877 | implicit none |
---|
878 | ! print a map |
---|
879 | integer i,mfile,n1,n2 |
---|
880 | integer,dimension(:)::h |
---|
881 | if(.not.c_%stable_da) return |
---|
882 | |
---|
883 | if(mfile.le.0) return |
---|
884 | do i=n1,n2 |
---|
885 | call dapri(h(i),mfile) |
---|
886 | enddo |
---|
887 | return |
---|
888 | end subroutine daprid |
---|
889 | |
---|
890 | subroutine daflo(h,x,y) |
---|
891 | implicit none |
---|
892 | ! LIE EXPONENT ROUTINES WITH FLOW OPERATORS |
---|
893 | ! |
---|
894 | ! \VEC{H}.GRAD X =Y |
---|
895 | integer i,x,y,b1,b2,b3 |
---|
896 | integer,dimension(:)::h |
---|
897 | if(.not.c_%stable_da) return |
---|
898 | |
---|
899 | call etall1(b1) |
---|
900 | call etall1(b2) |
---|
901 | call etall1(b3) |
---|
902 | |
---|
903 | call daclr(b1) |
---|
904 | call daclr(b2) |
---|
905 | do i=1,nd2 |
---|
906 | call dader(i,x,b2) |
---|
907 | call damul(b2,h(i),b3) |
---|
908 | call daadd(b3,b1,b2) |
---|
909 | call dacop(b2,b1) |
---|
910 | enddo |
---|
911 | call dacop(b1,y) |
---|
912 | call dadal1(b3) |
---|
913 | call dadal1(b2) |
---|
914 | call dadal1(b1) |
---|
915 | return |
---|
916 | end subroutine daflo |
---|
917 | subroutine daflod(h,x,y) |
---|
918 | implicit none |
---|
919 | integer i |
---|
920 | integer,dimension(:)::h,x,y |
---|
921 | integer,dimension(ndim2)::b1,b2 |
---|
922 | if(.not.c_%stable_da) return |
---|
923 | |
---|
924 | call etall(b1,nd2) |
---|
925 | call etall(b2,nd2) |
---|
926 | |
---|
927 | call dacopd(h,b1) |
---|
928 | call dacopd(x,b2) |
---|
929 | |
---|
930 | do i=1,nd2 |
---|
931 | call daflo(b1,b2(i),y(i)) |
---|
932 | enddo |
---|
933 | |
---|
934 | call dadal(b1,nd2) |
---|
935 | call dadal(b2,nd2) |
---|
936 | return |
---|
937 | end subroutine daflod |
---|
938 | |
---|
939 | subroutine intd(v,h,sca) |
---|
940 | implicit none |
---|
941 | ! IF SCA=-one |
---|
942 | ! \VEC{V}.GRAD = J GRAD H . GRAD = :H: |
---|
943 | ! |
---|
944 | ! IF SCA=one |
---|
945 | ! \VEC{V}.GRAD = GRAD H . GRAD |
---|
946 | integer i,h,b1,b2,b3,b4 |
---|
947 | integer,dimension(:)::v |
---|
948 | integer,dimension(ndim2)::x |
---|
949 | real(dp) sca |
---|
950 | if(.not.c_%stable_da) return |
---|
951 | |
---|
952 | call etall1(b1) |
---|
953 | call etall1(b2) |
---|
954 | call etall1(b3) |
---|
955 | call etall1(b4) |
---|
956 | call etallnom(x,nd2) ! ,'X ') |
---|
957 | |
---|
958 | call daclr(b4) |
---|
959 | call daclr(h) |
---|
960 | call etini(x) |
---|
961 | do i=1,nd |
---|
962 | call dacfu(v(2*i-1),dlie,b3) |
---|
963 | call dacfu(v(2*i),dlie,b1) |
---|
964 | call damul(b1,x(2*i-1),b2) |
---|
965 | call damul(b3,x(2*i),b1) |
---|
966 | call dalin(b2,1.0_dp,b1,sca,b3) |
---|
967 | call daadd(b3,b4,b2) |
---|
968 | call dacop(b2,b4) |
---|
969 | enddo |
---|
970 | call dacop(b4,h) |
---|
971 | call dadal(x,nd2) |
---|
972 | call dadal1(b4) |
---|
973 | call dadal1(b3) |
---|
974 | call dadal1(b2) |
---|
975 | call dadal1(b1) |
---|
976 | return |
---|
977 | end subroutine intd |
---|
978 | |
---|
979 | subroutine difd(h1,v,sca) |
---|
980 | implicit none |
---|
981 | ! INVERSE OF INTD ROUTINE |
---|
982 | integer i,h1,b1,h |
---|
983 | integer,dimension(:)::v |
---|
984 | real(dp) sca |
---|
985 | if(.not.c_%stable_da) return |
---|
986 | |
---|
987 | call etall1(b1) |
---|
988 | call etall1(h) |
---|
989 | call dacop(h1,h) |
---|
990 | do i=1,nd |
---|
991 | call dader(2*i-1,h,v(2*i)) |
---|
992 | call dader(2*i,h,b1) |
---|
993 | call dacmu(b1,sca,v(2*i-1)) |
---|
994 | enddo |
---|
995 | call dadal1(h) |
---|
996 | call dadal1(b1) |
---|
997 | return |
---|
998 | end subroutine difd |
---|
999 | |
---|
1000 | subroutine expflo(h,x,y,eps,nrmax) |
---|
1001 | implicit none |
---|
1002 | ! DOES EXP( \VEC{H} ) X = Y |
---|
1003 | logical(lp) more |
---|
1004 | integer i,nrmax,x,y,b1,b2,b3,b4 |
---|
1005 | integer,dimension(:)::h |
---|
1006 | real(dp) coe,eps,r,rbefore |
---|
1007 | if(.not.c_%stable_da) return |
---|
1008 | |
---|
1009 | call etall1(b1) |
---|
1010 | call etall1(b2) |
---|
1011 | call etall1(b3) |
---|
1012 | call etall1(b4) |
---|
1013 | |
---|
1014 | call dacop(x,b4) |
---|
1015 | call dacop(x,b1) |
---|
1016 | more=.true. |
---|
1017 | rbefore=1e30_dp |
---|
1018 | do i=1,nrmax |
---|
1019 | coe=1.0_dp/REAL(i,kind=DP) |
---|
1020 | call dacmu(b1,coe,b2) |
---|
1021 | call daflo(h,b2,b1) |
---|
1022 | call daadd(b4,b1,b3) |
---|
1023 | call daabs(b1,r) |
---|
1024 | if(more) then |
---|
1025 | if(r.gt.eps) then |
---|
1026 | rbefore=r |
---|
1027 | goto 100 |
---|
1028 | else |
---|
1029 | rbefore=r |
---|
1030 | more=.false. |
---|
1031 | endif |
---|
1032 | else |
---|
1033 | if(r.ge.rbefore) then |
---|
1034 | call dacop(b3,y) |
---|
1035 | call dadal1(b4) |
---|
1036 | call dadal1(b3) |
---|
1037 | call dadal1(b2) |
---|
1038 | call dadal1(b1) |
---|
1039 | return |
---|
1040 | endif |
---|
1041 | rbefore=r |
---|
1042 | endif |
---|
1043 | 100 continue |
---|
1044 | call dacop(b3,b4) |
---|
1045 | enddo |
---|
1046 | if(lielib_print(2)==1) then |
---|
1047 | write(6,'(a6,1x,G21.14,1x,a25)') ' NORM ',eps,' NEVER REACHED IN EXPFLO ' |
---|
1048 | endif |
---|
1049 | call dacop(b3,y) |
---|
1050 | call dadal1(b4) |
---|
1051 | call dadal1(b3) |
---|
1052 | call dadal1(b2) |
---|
1053 | call dadal1(b1) |
---|
1054 | return |
---|
1055 | end subroutine expflo |
---|
1056 | |
---|
1057 | subroutine expflod(h,x,w,eps,nrmax) |
---|
1058 | implicit none |
---|
1059 | ! DOES EXP( \VEC{H} ) \VEC{X} = \VEC{Y} |
---|
1060 | integer j,nrmax,b0 |
---|
1061 | integer,dimension(:)::x,w,h |
---|
1062 | integer,dimension(ndim2)::v |
---|
1063 | real(dp) eps |
---|
1064 | if(.not.c_%stable_da) return |
---|
1065 | |
---|
1066 | call etall1(b0 ) |
---|
1067 | call etallnom(v,nd2) ! ,'V ') |
---|
1068 | |
---|
1069 | call dacopd(x,v) |
---|
1070 | do j=1,nd2 |
---|
1071 | call expflo(h,v(j),b0,eps,nrmax) |
---|
1072 | call dacop(b0,v(j)) |
---|
1073 | enddo |
---|
1074 | call dacopd(v,w) |
---|
1075 | call dadal(v,nd2) |
---|
1076 | call dadal1(b0) |
---|
1077 | return |
---|
1078 | end subroutine expflod |
---|
1079 | subroutine facflo(h,x,w,nrmin,nrmax,sca,ifac) |
---|
1080 | implicit none |
---|
1081 | ! IFAC=1 |
---|
1082 | ! DOES EXP(SCA \VEC{H}_MRMIN ) ... EXP(SCA \VEC{H}_NRMAX ) X= Y |
---|
1083 | ! IFAC=-1 |
---|
1084 | ! DOES EXP(SCA \VEC{H}_NRMAX ) ... EXP(SCA \VEC{H}_MRMIN ) X= Y |
---|
1085 | integer i,ifac,nmax,nrmax,nrmin,x,w,v |
---|
1086 | integer,dimension(:)::h |
---|
1087 | integer,dimension(ndim2)::bm,b0 |
---|
1088 | real(dp) eps,sca |
---|
1089 | if(.not.c_%stable_da) return |
---|
1090 | |
---|
1091 | call etallnom(bm,nd2) ! ,'BM ') |
---|
1092 | call etallnom(b0,nd2) ! ,'B0 ') |
---|
1093 | call etall1(v) |
---|
1094 | |
---|
1095 | call dacop(x,v) |
---|
1096 | |
---|
1097 | ! eps=-one |
---|
1098 | ! call daeps(eps) |
---|
1099 | eps=epsflo |
---|
1100 | nmax=100 |
---|
1101 | ! |
---|
1102 | ! IFAC =1 ---> V = EXP(:SCA*H(NRMAX):)...EXP(:SCA*H(NRMIN):)X |
---|
1103 | if(ifac.eq.1) then |
---|
1104 | do i=nrmax,nrmin,-1 |
---|
1105 | call taked(h,i,b0) |
---|
1106 | call dacmud(b0,sca,bm) |
---|
1107 | |
---|
1108 | call expflo(bm,v,b0(1),eps,nmax) |
---|
1109 | call dacop(b0(1),v) |
---|
1110 | enddo |
---|
1111 | else |
---|
1112 | ! IFAC =-1 ---> V = EXP(:SCA*H(NRMIN):)...EXP(:SCA*H(NRMAX):)X |
---|
1113 | do i=nrmin,nrmax |
---|
1114 | call taked(h,i,b0) |
---|
1115 | call dacmud(b0,sca,bm) |
---|
1116 | |
---|
1117 | call expflo(bm,v,b0(1),eps,nmax) |
---|
1118 | call dacop(b0(1),v) |
---|
1119 | enddo |
---|
1120 | endif |
---|
1121 | call dacop(v,w) |
---|
1122 | call dadal1(v) |
---|
1123 | call dadal(b0,nd2) |
---|
1124 | call dadal(bm,nd2) |
---|
1125 | return |
---|
1126 | end subroutine facflo |
---|
1127 | subroutine facflod(h,x,w,nrmin,nrmax,sca,ifac) |
---|
1128 | implicit none |
---|
1129 | ! IFAC=1 |
---|
1130 | ! DOES EXP(SCA \VEC{H}_MRMIN ) ... EXP(SCA \VEC{H}_NRMAX ) \VEC{X}= \VEC{Y} |
---|
1131 | ! IFAC=-1 |
---|
1132 | ! DOES EXP(SCA \VEC{H}_NRMAX ) ... EXP(SCA \VEC{H}_MRMIN ) \VEC{X}= \VEC{Y} |
---|
1133 | integer i,ifac,nrmax,nrmin |
---|
1134 | integer,dimension(:)::x,w,h |
---|
1135 | real(dp) sca |
---|
1136 | if(.not.c_%stable_da) return |
---|
1137 | |
---|
1138 | do i=1,nd2 |
---|
1139 | call facflo(h,x(i),w(i),nrmin,nrmax,sca,ifac) |
---|
1140 | enddo |
---|
1141 | |
---|
1142 | return |
---|
1143 | end subroutine facflod |
---|
1144 | subroutine fexpo(h,x,w,nrmin,nrmax,sca,ifac) |
---|
1145 | implicit none |
---|
1146 | ! WRAPPED ROUTINES FOR THE OPERATOR \VEC{H}=:H: |
---|
1147 | ! WRAPPING FACFLOD |
---|
1148 | integer ifac,nrma,nrmax,nrmi,nrmin,h |
---|
1149 | integer,dimension(:)::x,w |
---|
1150 | integer,dimension(ndim2)::v |
---|
1151 | real(dp) sca |
---|
1152 | if(.not.c_%stable_da) return |
---|
1153 | |
---|
1154 | nrmi=nrmin-1 |
---|
1155 | nrma=nrmax-1 |
---|
1156 | call etall(v,nd2) |
---|
1157 | call difd(h,v,-1.0_dp) |
---|
1158 | call facflod(v,x,w,nrmi,nrma,sca,ifac) |
---|
1159 | |
---|
1160 | call dadal(v,nd2) |
---|
1161 | |
---|
1162 | return |
---|
1163 | end subroutine fexpo |
---|
1164 | subroutine etcom(x,y,h) |
---|
1165 | implicit none |
---|
1166 | ! ETCOM TAKES THE BRACKET OF TWO VECTOR FIELDS. |
---|
1167 | integer i,j,t1,t2 |
---|
1168 | integer,dimension(:)::h,x,y |
---|
1169 | integer,dimension(ndim2)::t3 |
---|
1170 | if(.not.c_%stable_da) return |
---|
1171 | |
---|
1172 | call etall1(t1) |
---|
1173 | call etall1(t2) |
---|
1174 | call etall(t3,nd2) |
---|
1175 | |
---|
1176 | do j=1,nd2 |
---|
1177 | do i=1,nd2 |
---|
1178 | |
---|
1179 | call dader(i,x(j),t1) |
---|
1180 | call dader(i,y(j),t2) |
---|
1181 | call damul(x(i),t2,t2) |
---|
1182 | call damul(y(i),t1,t1) |
---|
1183 | call dalin(t2,1.0_dp,t1,-1.0_dp,t1) |
---|
1184 | call daadd(t1,t3(j),t3(j)) |
---|
1185 | |
---|
1186 | enddo |
---|
1187 | enddo |
---|
1188 | |
---|
1189 | call dacopd(t3,h) |
---|
1190 | |
---|
1191 | call dadal1(t1) |
---|
1192 | call dadal1(t2) |
---|
1193 | call dadal(t3,nd2) |
---|
1194 | return |
---|
1195 | end subroutine etcom |
---|
1196 | subroutine etpoi(x,y,h) |
---|
1197 | implicit none |
---|
1198 | ! ETPOI TAKES THE POISSON BRACKET OF TWO FUNCTIONS |
---|
1199 | integer i,h,x,y,t1,t2,t3 |
---|
1200 | if(.not.c_%stable_da) return |
---|
1201 | |
---|
1202 | call etall1(t1) |
---|
1203 | call etall1(t2) |
---|
1204 | call etall1(t3) |
---|
1205 | |
---|
1206 | do i=1,nd |
---|
1207 | |
---|
1208 | call dader(2*i-1,x,t1) |
---|
1209 | call dader(2*i,y,t2) |
---|
1210 | call damul(t1,t2,t1) |
---|
1211 | |
---|
1212 | call dalin(t1,1.0_dp,t3,1.0_dp,t3) |
---|
1213 | call dader(2*i-1,y,t1) |
---|
1214 | call dader(2*i,x,t2) |
---|
1215 | call damul(t1,t2,t1) |
---|
1216 | |
---|
1217 | call dalin(t1,-1.0_dp,t3,1.0_dp,t3) |
---|
1218 | |
---|
1219 | enddo |
---|
1220 | |
---|
1221 | call dacop(t3,h) |
---|
1222 | |
---|
1223 | call dadal1(t1) |
---|
1224 | call dadal1(t2) |
---|
1225 | call dadal1(t3) |
---|
1226 | return |
---|
1227 | end subroutine etpoi |
---|
1228 | subroutine exp1d(h,x,y,eps,non) |
---|
1229 | implicit none |
---|
1230 | ! WRAPPING EXPFLO |
---|
1231 | integer non,h,x,y |
---|
1232 | integer,dimension(ndim2)::v |
---|
1233 | real(dp) eps |
---|
1234 | if(.not.c_%stable_da) return |
---|
1235 | |
---|
1236 | call etall(v,nd2) |
---|
1237 | call difd(h,v,-1.0_dp) |
---|
1238 | call expflo(v,x,y,eps,non) |
---|
1239 | |
---|
1240 | call dadal(v,nd2) |
---|
1241 | |
---|
1242 | return |
---|
1243 | end subroutine exp1d |
---|
1244 | subroutine expnd2(h,x,w,eps,nrmax) |
---|
1245 | implicit none |
---|
1246 | ! WRAPPING EXPFLOD USING EXP1D |
---|
1247 | integer j,nrmax,b0,h |
---|
1248 | integer,dimension(:)::x,w |
---|
1249 | integer,dimension(ndim2)::v |
---|
1250 | real(dp) eps |
---|
1251 | if(.not.c_%stable_da) return |
---|
1252 | |
---|
1253 | call etall1(b0) |
---|
1254 | call etallnom(v,nd2) ! ,'V ') |
---|
1255 | |
---|
1256 | call dacopd(x,v) |
---|
1257 | do j=1,nd2 |
---|
1258 | call exp1d(h,v(j),b0,eps,nrmax) |
---|
1259 | call dacop(b0,v(j)) |
---|
1260 | enddo |
---|
1261 | call dacopd(v,w) |
---|
1262 | call dadal(v,nd2) |
---|
1263 | call dadal1(b0) |
---|
1264 | return |
---|
1265 | end subroutine expnd2 |
---|
1266 | |
---|
1267 | subroutine flofacg(xy,h,epsone) |
---|
1268 | implicit none |
---|
1269 | ! GENERAL ONE EXPONENT FACTORIZATION |
---|
1270 | logical(lp) more |
---|
1271 | integer i,k,kk,nrmax |
---|
1272 | integer,dimension(:)::xy,h |
---|
1273 | integer,dimension(ndim2)::x,v,w,t,z |
---|
1274 | integer,dimension(ntt)::jj |
---|
1275 | real(dp) eps,epsone,r,xn,xnbefore,xnorm,xnorm1,xx |
---|
1276 | if(.not.c_%stable_da) return |
---|
1277 | |
---|
1278 | jj(1)=1 |
---|
1279 | ! |
---|
1280 | call etallnom(v,nd2) ! ,'V ') |
---|
1281 | call etallnom(w,nd2) ! ,'W ') |
---|
1282 | call etallnom(t,nd2) ! ,'T ') |
---|
1283 | call etallnom(x,nd2) ! ,'Z ') |
---|
1284 | call etallnom(z,nd2) ! ,'Z ') |
---|
1285 | |
---|
1286 | call etini(v) |
---|
1287 | call daclrd(w) |
---|
1288 | xnorm1=0.0_dp |
---|
1289 | do i=1,nd2 |
---|
1290 | call daabs(xy(i),r) |
---|
1291 | xnorm1=xnorm1+r |
---|
1292 | enddo |
---|
1293 | xnbefore=1e36_dp |
---|
1294 | more=.false. |
---|
1295 | eps=1e-5_dp |
---|
1296 | nrmax=1000 |
---|
1297 | xn=1e4_dp |
---|
1298 | |
---|
1299 | if(epsone>0.0_dp) then !epsone>zero |
---|
1300 | do k=1,nrmax |
---|
1301 | call dacmud(h,-1.0_dp,t) |
---|
1302 | call expflod(t,xy,x,eps,nrmax) |
---|
1303 | call dalind(x,1.0_dp,v,-1.0_dp,t) |
---|
1304 | ! write(20,*) "$$$$$$$$$$$$$$",k,"$$$$$$$$$$$$$$$$$$$$" |
---|
1305 | ! call daprid(t,1,1,20) |
---|
1306 | if(xn.lt.epsone) then |
---|
1307 | if(lielib_print(3)==1) then |
---|
1308 | w_p=0 |
---|
1309 | w_p%nc=1 |
---|
1310 | write(w_p%c(1),'(a14,g21.14)') " xn quadratic ",xn |
---|
1311 | w_p%fc='(1((1X,A72)))' |
---|
1312 | ! CALL !WRITE_a |
---|
1313 | endif |
---|
1314 | call daflod(t,t,w) |
---|
1315 | call dalind(t,1.0_dp,w,-0.5_dp,t) |
---|
1316 | call dacopd(t,z) |
---|
1317 | call dacopd(t,w) |
---|
1318 | ! second order in W |
---|
1319 | call etcom(h,w,x) |
---|
1320 | call etcom(x,w,x) |
---|
1321 | ! END OF order in W |
---|
1322 | |
---|
1323 | do kk=1,3 !10 |
---|
1324 | call etcom(h,w,w) |
---|
1325 | call dalind(z,1.0_dp,w,xintex(kk),z) |
---|
1326 | enddo |
---|
1327 | call dacopd(z,t) |
---|
1328 | xx=1.0_dp/12.0_dp |
---|
1329 | call dalind(x,xx,h,1.0_dp,h) |
---|
1330 | endif |
---|
1331 | |
---|
1332 | call dalind(t,1.0_dp,h,1.0_dp,h) |
---|
1333 | xnorm=0.0_dp |
---|
1334 | do i=1,nd2 |
---|
1335 | call daabs(t(i),r) |
---|
1336 | xnorm=xnorm+r |
---|
1337 | enddo |
---|
1338 | xn=xnorm/xnorm1 |
---|
1339 | if(xn.ge.epsone.and.(lielib_print(3)==1)) then |
---|
1340 | w_p=0 |
---|
1341 | w_p%nc=1 |
---|
1342 | write(w_p%c(1),'(a11,g21.14)') " xn linear ",xn |
---|
1343 | w_p%fc='(1((1X,A72)))' |
---|
1344 | !CALL !WRITE_a |
---|
1345 | endif |
---|
1346 | if(xn.lt.eps.or.more) then |
---|
1347 | more=.true. |
---|
1348 | if(xn.ge.xnbefore) goto 1000 |
---|
1349 | xnbefore=xn |
---|
1350 | endif |
---|
1351 | enddo |
---|
1352 | 1000 continue |
---|
1353 | else !epsone>zero |
---|
1354 | do k=1,nint(abs(epsone))-1 |
---|
1355 | call dacmud(h,-1.0_dp,t) |
---|
1356 | call expflod(t,xy,x,eps,nrmax) |
---|
1357 | call dalind(x,1.0_dp,v,-1.0_dp,t) |
---|
1358 | ! write(20,*) "$$$$$$$$$$$$$$",k,"$$$$$$$$$$$$$$$$$$$$" |
---|
1359 | ! call daprid(t,1,1,20) |
---|
1360 | |
---|
1361 | call dalind(t,1.0_dp,h,1.0_dp,h) |
---|
1362 | enddo |
---|
1363 | endif |
---|
1364 | if(lielib_print(3)==1) WRITE(6,*) " K ", K,epsone |
---|
1365 | if(lielib_print(3)==1) then |
---|
1366 | w_p=0 |
---|
1367 | w_p%nc=1 |
---|
1368 | write(w_p%c(1),'(a11,i4)') " iteration " , k |
---|
1369 | w_p%fc='(1((1X,A72)))' |
---|
1370 | endif |
---|
1371 | ! if(lielib_print(3)==1) CALL WRITE_a |
---|
1372 | call dadal(x,nd2) |
---|
1373 | call dadal(w,nd2) |
---|
1374 | call dadal(v,nd2) |
---|
1375 | call dadal(t,nd2) |
---|
1376 | call dadal(z,nd2) |
---|
1377 | return |
---|
1378 | end subroutine flofacg |
---|
1379 | |
---|
1380 | subroutine flofac(xy,x,h) |
---|
1381 | implicit none |
---|
1382 | ! GENERAL DRAGT-FINN FACTORIZATION |
---|
1383 | integer k |
---|
1384 | integer,dimension(:)::xy,x,h |
---|
1385 | integer,dimension(ndim2)::v,w |
---|
1386 | if(.not.c_%stable_da) return |
---|
1387 | |
---|
1388 | call etallnom(v,nd2) ! ,'V ') |
---|
1389 | call etallnom(w,nd2) ! ,'W ') |
---|
1390 | |
---|
1391 | call dacopd(xy,x) |
---|
1392 | ! call dacopd(x,v) |
---|
1393 | call datruncd(x,2,v) |
---|
1394 | call daclrd(w) |
---|
1395 | ! call danot(1) |
---|
1396 | call etinv(v,w) |
---|
1397 | ! call danot(no) |
---|
1398 | call etcct(x,w,v) |
---|
1399 | ! call danot(1) |
---|
1400 | ! call dacopd(xy,x) |
---|
1401 | ! call datruncd(xy,1,x) ! lethal error |
---|
1402 | call datruncd(xy,2,x) |
---|
1403 | |
---|
1404 | |
---|
1405 | ! call danot(no) |
---|
1406 | call dacopd(v,w) |
---|
1407 | call daclrd(h) |
---|
1408 | do k=2,no |
---|
1409 | call taked(w,k,v) |
---|
1410 | call dalind(v,1.0_dp,h,1.0_dp,h) |
---|
1411 | call facflod(h,w,v,k,k,-1.0_dp,-1) |
---|
1412 | call dacopd(v,w) |
---|
1413 | enddo |
---|
1414 | call dadal(w,nd2) |
---|
1415 | call dadal(v,nd2) |
---|
1416 | return |
---|
1417 | end subroutine flofac |
---|
1418 | subroutine liefact(xy,x,h) |
---|
1419 | implicit none |
---|
1420 | ! SYMPLECTIC DRAGT-FINN FACTORIZATION WRAPPING FLOFAC |
---|
1421 | integer h |
---|
1422 | integer,dimension(:)::xy,x |
---|
1423 | integer,dimension(ndim2)::v |
---|
1424 | if(.not.c_%stable_da) return |
---|
1425 | |
---|
1426 | call etall(v,nd2) |
---|
1427 | |
---|
1428 | call flofac(xy,x,v) |
---|
1429 | call intd(v,h,-1.0_dp) |
---|
1430 | ! |
---|
1431 | call dadal(v,nd2) |
---|
1432 | |
---|
1433 | return |
---|
1434 | end subroutine liefact |
---|
1435 | subroutine mapnorm(x,ft,a2,a1,xy,h,nord) |
---|
1436 | implicit none |
---|
1437 | !--NORMALIZATION ROUTINES OF LIELIB |
---|
1438 | !- WRAPPING MAPNORMF |
---|
1439 | integer isi,nord,ft,h |
---|
1440 | integer,dimension(:)::x,a1,a2,xy |
---|
1441 | integer,dimension(ndim2)::hf,ftf |
---|
1442 | if(.not.c_%stable_da) return |
---|
1443 | |
---|
1444 | call etall(ftf,nd2) |
---|
1445 | call etall(hf,nd2) |
---|
1446 | isi=0 |
---|
1447 | call mapnormf(x,ftf,a2,a1,xy,hf,nord,isi) |
---|
1448 | call intd(hf,h,-1.0_dp) |
---|
1449 | call intd(ftf,ft,-1.0_dp) |
---|
1450 | call dadal(ftf,nd2) |
---|
1451 | call dadal(hf,nd2) |
---|
1452 | |
---|
1453 | return |
---|
1454 | end subroutine mapnorm |
---|
1455 | |
---|
1456 | subroutine gettura(psq,radsq) |
---|
1457 | implicit none |
---|
1458 | integer ik,j1,j2 |
---|
1459 | real(dp),dimension(ndim)::psq,radsq,st |
---|
1460 | if(.not.c_%stable_da) return |
---|
1461 | |
---|
1462 | if(new_ndpt) then |
---|
1463 | if(ndpt/=0) then |
---|
1464 | if(mod(ndpt,2)==0) then |
---|
1465 | j1=ndpt/2 |
---|
1466 | j2=npt_pos/2 |
---|
1467 | else |
---|
1468 | j1=(ndpt+1)/2 |
---|
1469 | j2=(npt_pos+1)/2 |
---|
1470 | endif |
---|
1471 | endif |
---|
1472 | do ik=1,nd |
---|
1473 | psq(ik)=ps(ik) |
---|
1474 | radsq(ik)=rads(ik) |
---|
1475 | st(ik)=stmem(ik) |
---|
1476 | enddo |
---|
1477 | if(ndpt/=0) then |
---|
1478 | psq(j2) = ps(j1) |
---|
1479 | psq(j1) = ps(j2) |
---|
1480 | radsq(j2) = rads(j1) |
---|
1481 | radsq(j1) = rads(j2) |
---|
1482 | st(j2) = stmem(j1) |
---|
1483 | st(j1) = stmem(j2) |
---|
1484 | endif |
---|
1485 | do ik=1,nd |
---|
1486 | if(ik/=time_plane) then |
---|
1487 | if(st(ik)+1e-3_dp.gt.1.0_dp.and.psq(ik).lt.0.0_dp) psq(ik)=psq(ik)+1.0_dp |
---|
1488 | endif |
---|
1489 | enddo |
---|
1490 | if(time_plane>0) then |
---|
1491 | if(st(time_plane)+1e-3_dp.gt.1.0_dp.and.psq(time_plane).lt.-0.5_dp) psq(time_plane)=psq(time_plane)+1.0_dp |
---|
1492 | endif |
---|
1493 | else |
---|
1494 | do ik=1,nd |
---|
1495 | psq(ik)=ps(ik) |
---|
1496 | radsq(ik)=rads(ik) |
---|
1497 | enddo |
---|
1498 | endif |
---|
1499 | return |
---|
1500 | end subroutine gettura |
---|
1501 | |
---|
1502 | subroutine setidpr(nplan) |
---|
1503 | implicit none |
---|
1504 | integer ik |
---|
1505 | integer,dimension(ndim)::nplan |
---|
1506 | if(.not.c_%stable_da) return |
---|
1507 | |
---|
1508 | do ik=1,nd |
---|
1509 | nplane(ik)=nplan(ik) |
---|
1510 | enddo |
---|
1511 | return |
---|
1512 | end subroutine setidpr |
---|
1513 | ! subroutine idprset(idprint) |
---|
1514 | ! implicit none |
---|
1515 | ! integer idprint |
---|
1516 | ! if(.not.c_%stable_da) return! |
---|
1517 | ! |
---|
1518 | ! idpr=idprint |
---|
1519 | ! |
---|
1520 | ! return |
---|
1521 | ! end subroutine idprset |
---|
1522 | ! CALL MAPNORMF(junk%V%i,s2%a%nonlinear%v%i,s2%a%linear%v%i,& |
---|
1523 | ! & s2%A1%v%i,s2%normal%linear%v%i,s2%normal%nonlinear%v%i,s2%NORD,s2%jtune) |
---|
1524 | subroutine mapnormf(x,ft,a2,a1,xy,h,nord,isi) |
---|
1525 | implicit none |
---|
1526 | integer ij,isi,nord |
---|
1527 | integer,dimension(ndim2)::a1i,a2i |
---|
1528 | integer,dimension(:)::x,a1,a2,ft,xy,h |
---|
1529 | real(dp),dimension(ndim)::angle,rad,st,p |
---|
1530 | logical doflip |
---|
1531 | if(.not.c_%stable_da) return |
---|
1532 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
1533 | perform_flip=.false. |
---|
1534 | call flip(x,x) |
---|
1535 | doflip=.true. |
---|
1536 | else |
---|
1537 | doflip=.false. |
---|
1538 | endif |
---|
1539 | |
---|
1540 | call etallnom(a1i,nd2) ! ,'A1I ') |
---|
1541 | call etallnom(a2i,nd2) ! ,'A2I ') |
---|
1542 | ! frank/etienne |
---|
1543 | do itu=1,ndim |
---|
1544 | angle(itu)=0.0_dp |
---|
1545 | p(itu)=0.0_dp |
---|
1546 | st(itu)=0.0_dp |
---|
1547 | rad(itu)=0.0_dp |
---|
1548 | ps(itu)=0.0_dp |
---|
1549 | rads(itu)=0.0_dp |
---|
1550 | enddo |
---|
1551 | jtune=isi |
---|
1552 | call dacopd(x,xy) |
---|
1553 | ! go to fix point in the parameters + pt to order nord>=1 |
---|
1554 | if(nv>nd2.or.ndc==1) then |
---|
1555 | call gofix(xy,a1,a1i,nord) |
---|
1556 | call simil(a1i,xy,a1,xy) |
---|
1557 | else ! this "if" was added to remove crashes when y-=plane is nearly identity |
---|
1558 | call etini(a1) ! in stochastic kick calculations |
---|
1559 | call etini(a1i) ! 2002.10.20 |
---|
1560 | endif |
---|
1561 | ! linear part |
---|
1562 | call midbflo(xy,a2,a2i,angle,rad,st) |
---|
1563 | do ij=1,nd-ndc |
---|
1564 | p(ij)=angle(ij)*(st(ij)*(twopii-1.0_dp)+1.0_dp) |
---|
1565 | enddo |
---|
1566 | stmem=st |
---|
1567 | if(ndc.eq.1) p(nd)=angle(nd) |
---|
1568 | if(lielib_print(4)==1) then |
---|
1569 | w_p=1 |
---|
1570 | w_p%nc=1 |
---|
1571 | w_p%nr=2 |
---|
1572 | w_p%c(1)='tune ' |
---|
1573 | do ij=1,nd |
---|
1574 | w_p%r(ij)=p(ij) |
---|
1575 | enddo |
---|
1576 | w_p%fc='((1X,A8))' |
---|
1577 | w_p%fr='(3(1x,g21.14))' |
---|
1578 | !CALL !WRITE_a |
---|
1579 | w_p=1 |
---|
1580 | w_p%nc=1 |
---|
1581 | w_p%nr=2 |
---|
1582 | w_p%c(1)='damping ' |
---|
1583 | do ij=1,nd |
---|
1584 | w_p%r(ij)=rad(ij) |
---|
1585 | enddo |
---|
1586 | w_p%fc='((1X,A8))' |
---|
1587 | w_p%fr='(3(1x,g21.14))' |
---|
1588 | !CALL !WRITE_a |
---|
1589 | endif |
---|
1590 | do ij=1,nd ! -ndc Frank |
---|
1591 | ps(ij)=p(ij) |
---|
1592 | rads(ij)=rad(ij) |
---|
1593 | enddo |
---|
1594 | call initpert(st,angle,rad) |
---|
1595 | call simil(a2i,xy,a2,xy) |
---|
1596 | call dacopd(xy,a2i) |
---|
1597 | call orderflo(h,ft,xy,angle,rad) |
---|
1598 | do ij=1,nd-ndc |
---|
1599 | p(ij)=angle(ij) |
---|
1600 | ! if(angle(ij).gt.pi.and.st(ij).gt.zero) then !.and.itu.eq.1)then |
---|
1601 | ! p(ij)=angle(ij)-twopi |
---|
1602 | ! w_p=0 |
---|
1603 | ! w_p%nc=1 |
---|
1604 | ! w_p%fc='((1X,A72))' |
---|
1605 | ! write(w_p%c(1),'(i4,a27,g21.14)') ij,' TH TUNE MODIFIED IN H2 TO ',p(ij)*twopii |
---|
1606 | ! CALL WRITE_a |
---|
1607 | ! endif |
---|
1608 | enddo |
---|
1609 | call h2pluflo(h,p,rad) |
---|
1610 | ! CALL TAKED(A2I,1,XY) |
---|
1611 | call taked(a2i,1,a1i) |
---|
1612 | call etcct(xy,a1i,xy) |
---|
1613 | |
---|
1614 | if(doflip) then |
---|
1615 | call flip(x,x) |
---|
1616 | call flip(a2,a2) |
---|
1617 | call flip(a1,a1) |
---|
1618 | call flip(xy,xy) |
---|
1619 | call flipflo(ft,ft,-1) |
---|
1620 | call flipflo(h,h,-1) |
---|
1621 | perform_flip=.true. |
---|
1622 | endif |
---|
1623 | call dadal(a2i,nd2) |
---|
1624 | call dadal(a1i,nd2) |
---|
1625 | return |
---|
1626 | end subroutine mapnormf |
---|
1627 | |
---|
1628 | subroutine get_flip_info(nt_pos1,npt_pos1) |
---|
1629 | implicit none |
---|
1630 | integer nt_pos1,npt_pos1 |
---|
1631 | nt_pos1=nt_pos |
---|
1632 | npt_pos1=npt_pos |
---|
1633 | end subroutine get_flip_info |
---|
1634 | |
---|
1635 | subroutine flip(xy,xyf) |
---|
1636 | implicit none |
---|
1637 | integer i,nord |
---|
1638 | integer,dimension(:):: xy,xyf |
---|
1639 | integer,dimension(ndim2)::x,xi |
---|
1640 | if(.not.c_%stable_da) return |
---|
1641 | |
---|
1642 | if(nt_pos>=nd2-1) return |
---|
1643 | |
---|
1644 | |
---|
1645 | call etallnom(x,nd2) |
---|
1646 | call etallnom(xi,nd2) |
---|
1647 | |
---|
1648 | call etini(x) |
---|
1649 | |
---|
1650 | call davar(x(npt_pos),0.0_dp,ndpt) |
---|
1651 | call davar(x(nt_pos),0.0_dp,ndt) |
---|
1652 | call davar(x(ndpt),0.0_dp,npt_pos) |
---|
1653 | call davar(x(ndt),0.0_dp,nt_pos) |
---|
1654 | call etinv(x,xi) |
---|
1655 | |
---|
1656 | call simil(x,xy,xi,xyf) |
---|
1657 | |
---|
1658 | ! call davar(x(ndpt),zero,ndpt) |
---|
1659 | |
---|
1660 | |
---|
1661 | call dadal(xi,nd2) |
---|
1662 | call dadal(x,nd2) |
---|
1663 | |
---|
1664 | end subroutine flip |
---|
1665 | |
---|
1666 | subroutine flip_real_array(xy,xyf,i) |
---|
1667 | implicit none |
---|
1668 | integer i |
---|
1669 | real(dp),dimension(:):: xy,xyf |
---|
1670 | real(dp),dimension(ndim)::x |
---|
1671 | if(.not.c_%stable_da) return |
---|
1672 | |
---|
1673 | if(nt_pos>=nd2-1) return |
---|
1674 | x=0.0_dp |
---|
1675 | x(1:nd) = xy(1:nd) |
---|
1676 | xyf(1:nd) = xy(1:nd) |
---|
1677 | if(mod(ndpt,2)==0) then |
---|
1678 | if(i==1) then |
---|
1679 | xyf(ndpt/2)=x(npt_pos/2) |
---|
1680 | xyf(npt_pos/2)=x(ndpt/2) |
---|
1681 | else |
---|
1682 | xyf(npt_pos/2)=x(ndpt/2) |
---|
1683 | xyf(ndpt/2)=x(npt_pos/2) |
---|
1684 | endif |
---|
1685 | else |
---|
1686 | if(i==1) then |
---|
1687 | xyf((ndpt+1)/2)=x((npt_pos+1)/2) |
---|
1688 | xyf((npt_pos+1)/2)=x((ndpt+1)/2) |
---|
1689 | else |
---|
1690 | xyf((npt_pos+1)/2)=x((ndpt+1)/2) |
---|
1691 | xyf((ndpt+1)/2)=x((npt_pos+1)/2) |
---|
1692 | endif |
---|
1693 | endif |
---|
1694 | |
---|
1695 | |
---|
1696 | end subroutine flip_real_array |
---|
1697 | |
---|
1698 | subroutine flip_resonance(xy,xyf,i) |
---|
1699 | implicit none |
---|
1700 | integer i,NRES,j |
---|
1701 | integer,dimension(:,:):: xy,xyf |
---|
1702 | integer,dimension(NDIM,NRESO) ::x |
---|
1703 | if(.not.c_%stable_da) return |
---|
1704 | |
---|
1705 | if(nt_pos>=nd2-1) return |
---|
1706 | x=0 |
---|
1707 | x = xy |
---|
1708 | xyf = xy |
---|
1709 | if(mod(ndpt,2)==0) then |
---|
1710 | do j=1,nreso |
---|
1711 | if(i==1) then |
---|
1712 | xyf(ndpt/2,j)=x(npt_pos/2,j) |
---|
1713 | xyf(npt_pos/2,j)=x(ndpt/2,j) |
---|
1714 | else |
---|
1715 | xyf(npt_pos/2,j)=x(ndpt/2,j) |
---|
1716 | xyf(ndpt/2,j)=x(npt_pos/2,j) |
---|
1717 | endif |
---|
1718 | enddo |
---|
1719 | else |
---|
1720 | do j=1,nreso |
---|
1721 | if(i==1) then |
---|
1722 | xyf((ndpt+1)/2,j)=x((npt_pos+1)/2,j) |
---|
1723 | xyf((npt_pos+1)/2,j)=x((ndpt+1)/2,j) |
---|
1724 | else |
---|
1725 | xyf((npt_pos+1)/2,j)=x((ndpt+1)/2,j) |
---|
1726 | xyf((ndpt+1)/2,j)=x((npt_pos+1)/2,j) |
---|
1727 | endif |
---|
1728 | enddo |
---|
1729 | endif |
---|
1730 | |
---|
1731 | |
---|
1732 | end subroutine flip_resonance |
---|
1733 | |
---|
1734 | subroutine flipflo(xy,xyf,i) |
---|
1735 | implicit none |
---|
1736 | integer i |
---|
1737 | integer,dimension(:):: xy,xyf |
---|
1738 | integer,dimension(ndim2)::x,xi |
---|
1739 | if(.not.c_%stable_da) return |
---|
1740 | |
---|
1741 | if(nt_pos>=nd2-1) return |
---|
1742 | |
---|
1743 | |
---|
1744 | call etallnom(x,nd2) |
---|
1745 | call etallnom(xi,nd2) |
---|
1746 | |
---|
1747 | call etini(x) |
---|
1748 | |
---|
1749 | call davar(x(npt_pos),0.0_dp,ndpt) |
---|
1750 | call davar(x(nt_pos),0.0_dp,ndt) |
---|
1751 | call davar(x(ndpt),0.0_dp,npt_pos) |
---|
1752 | call davar(x(ndt),0.0_dp,nt_pos) |
---|
1753 | call etinv(x,xi) |
---|
1754 | |
---|
1755 | if(i==1) then |
---|
1756 | call trxflo(xy,xyf,xi) |
---|
1757 | else |
---|
1758 | call trxflo(xy,xyf,x) |
---|
1759 | endif |
---|
1760 | |
---|
1761 | |
---|
1762 | call dadal(xi,nd2) |
---|
1763 | call dadal(x,nd2) |
---|
1764 | |
---|
1765 | end subroutine flipflo |
---|
1766 | |
---|
1767 | subroutine flip_i(xy,xyf,i) |
---|
1768 | implicit none |
---|
1769 | integer i,nord |
---|
1770 | integer xy,xyf |
---|
1771 | integer,dimension(ndim2)::x,xi |
---|
1772 | if(.not.c_%stable_da) return |
---|
1773 | |
---|
1774 | if(nt_pos>=nd2-1) return |
---|
1775 | |
---|
1776 | |
---|
1777 | call etallnom(x,nd2) |
---|
1778 | call etallnom(xi,nd2) |
---|
1779 | |
---|
1780 | call etini(x) |
---|
1781 | |
---|
1782 | call davar(x(npt_pos),0.0_dp,ndpt) |
---|
1783 | call davar(x(nt_pos),0.0_dp,ndt) |
---|
1784 | call davar(x(ndpt),0.0_dp,npt_pos) |
---|
1785 | call davar(x(ndt),0.0_dp,nt_pos) |
---|
1786 | call etinv(x,xi) |
---|
1787 | |
---|
1788 | if(i==1) then |
---|
1789 | call trx(xy,xyf,xi) |
---|
1790 | else |
---|
1791 | call trx(xy,xyf,x) |
---|
1792 | endif |
---|
1793 | |
---|
1794 | |
---|
1795 | call dadal(xi,nd2) |
---|
1796 | call dadal(x,nd2) |
---|
1797 | |
---|
1798 | end subroutine flip_i |
---|
1799 | |
---|
1800 | |
---|
1801 | subroutine gofix(xy,a1,a1i,nord) |
---|
1802 | implicit none |
---|
1803 | ! GETTING TO THE FIXED POINT AND CHANGING TIME APPROPRIATELY IN THE |
---|
1804 | ! COASTING BEAM CASE |
---|
1805 | !**************************************************************** |
---|
1806 | ! X = A1 XY A1I WHERE X IS TO THE FIXED POINT TO ORDER NORD |
---|
1807 | ! for ndpt not zero, works in all cases. (coasting beam: eigenvalue |
---|
1808 | !1 in Jordan form) |
---|
1809 | !**************************************************************** |
---|
1810 | integer i,nord |
---|
1811 | integer,dimension(:)::xy,a1,a1i |
---|
1812 | integer,dimension(ndim2)::x,w,v,rel |
---|
1813 | real(dp) xic |
---|
1814 | logical doflip |
---|
1815 | if(.not.c_%stable_da) return |
---|
1816 | |
---|
1817 | |
---|
1818 | |
---|
1819 | call etallnom(x,nd2) ! , 'X ') |
---|
1820 | call etallnom(w,nd2) ! , 'W ') |
---|
1821 | call etallnom(v,nd2) ! , 'V ') |
---|
1822 | call etallnom(rel,nd2) ! ,'REL ') |
---|
1823 | |
---|
1824 | |
---|
1825 | ! COMPUTATION OF A1 AND A1I USING DAINV |
---|
1826 | |
---|
1827 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
1828 | perform_flip=.false. |
---|
1829 | call flip(xy,xy) |
---|
1830 | doflip=.true. |
---|
1831 | else |
---|
1832 | doflip=.false. |
---|
1833 | endif |
---|
1834 | |
---|
1835 | call etini(rel) |
---|
1836 | |
---|
1837 | ! call danot(nord) |
---|
1838 | |
---|
1839 | call etini(v) |
---|
1840 | |
---|
1841 | do i=1,nd2-ndc2 |
---|
1842 | ! call dacop(xy(i),x(i)) |
---|
1843 | call datrunc(xy(i),nord+1,x(i)) |
---|
1844 | call dalin(x(i),1.0_dp,rel(i),-1.0_dp,v(i)) |
---|
1845 | enddo |
---|
1846 | call etinv(v,w) |
---|
1847 | call datruncd(w,nord+1,w) |
---|
1848 | call daclrd(x) |
---|
1849 | if(ndc.eq.1) then |
---|
1850 | call davar(x(ndpt),0.0_dp,ndpt) |
---|
1851 | endif |
---|
1852 | call etcct(w,x,v) |
---|
1853 | if(ndc.eq.1) then |
---|
1854 | call daclr(v(nd2)) |
---|
1855 | call daclr(v(nd2-ndc)) |
---|
1856 | endif |
---|
1857 | call dalind(rel,1.0_dp,v,1.0_dp,a1) |
---|
1858 | call dalind(rel,1.0_dp,v,-1.0_dp,a1i) |
---|
1859 | |
---|
1860 | if(ndpt.ne.0) then |
---|
1861 | |
---|
1862 | ! CORRECTIONS |
---|
1863 | call daclrd(w) |
---|
1864 | call daclrd(v) |
---|
1865 | call daclrd(x) |
---|
1866 | |
---|
1867 | do i=1,nd2-ndc2 |
---|
1868 | call dalin(a1(i),1.0_dp,rel(i),-1.0_dp,w(i)) |
---|
1869 | enddo |
---|
1870 | |
---|
1871 | ! COMPUTE Deta/Ddelta |
---|
1872 | call dacopd(w,a1) |
---|
1873 | |
---|
1874 | do i=1,nd2-ndc2 |
---|
1875 | call dader(ndpt,w(i),w(i)) |
---|
1876 | enddo |
---|
1877 | ! COMPUTE J*Deta/dDELTA |
---|
1878 | |
---|
1879 | do i=1,nd-ndc |
---|
1880 | call dacmu(w(2*i),1.0_dp,v(2*i-1) ) |
---|
1881 | call dacmu(w(2*i-1),-1.0_dp,v(2*i) ) |
---|
1882 | enddo |
---|
1883 | |
---|
1884 | xic=(-1)**(ndt) |
---|
1885 | |
---|
1886 | do i=1,nd2-ndc2 |
---|
1887 | call damul(v(i),rel(i),x(1)) |
---|
1888 | call daadd(x(1),w(ndt),w(ndt)) |
---|
1889 | call dacop(a1(i),w(i)) |
---|
1890 | enddo |
---|
1891 | call dacmu(w(ndt),xic,w(ndt)) |
---|
1892 | |
---|
1893 | call expflod(w,rel,a1,1e-7_dp,10000) |
---|
1894 | ! END OF CORRECTIONS |
---|
1895 | |
---|
1896 | call datruncd(a1,nord+1,a1) |
---|
1897 | call etinv(a1,a1i) |
---|
1898 | call datruncd(a1i,nord+1,a1i) |
---|
1899 | |
---|
1900 | endif |
---|
1901 | if(doflip) then |
---|
1902 | call flip(xy,xy) |
---|
1903 | call flip(a1,a1) |
---|
1904 | call flip(a1i,a1i) |
---|
1905 | perform_flip=.true. |
---|
1906 | endif |
---|
1907 | |
---|
1908 | |
---|
1909 | ! call danot(no) |
---|
1910 | |
---|
1911 | call dadal(rel,nd2) |
---|
1912 | call dadal(v,nd2) |
---|
1913 | call dadal(w,nd2) |
---|
1914 | call dadal(x,nd2) |
---|
1915 | return |
---|
1916 | end subroutine gofix |
---|
1917 | |
---|
1918 | |
---|
1919 | subroutine orderflo(h,ft,x,ang,ra) |
---|
1920 | implicit none |
---|
1921 | !- NONLINEAR NORMALIZATION PIECE OF MAPNORMF |
---|
1922 | integer k |
---|
1923 | integer,dimension(ndim2)::w,v,rel,roi,b1,b5,b6,b9 |
---|
1924 | integer,dimension(:)::x,h,ft |
---|
1925 | real(dp),dimension(ndim)::ang,ra |
---|
1926 | if(.not.c_%stable_da) return |
---|
1927 | |
---|
1928 | call etallnom(w,nd2) ! ,'W ') |
---|
1929 | call etallnom(v,nd2) ! ,'V ') |
---|
1930 | call etallnom(rel,nd2) ! ,'REL ') |
---|
1931 | call etallnom(roi,nd2) ! ,'ROI ') |
---|
1932 | call etallnom(b1,nd2) ! ,'B1 ') |
---|
1933 | call etallnom(b5,nd2) ! ,'B5 ') |
---|
1934 | call etallnom(b6,nd2) ! ,'B6 ') |
---|
1935 | call etallnom(b9,nd2) ! ,'B9 ') |
---|
1936 | call rotiflo(roi,ang,ra) |
---|
1937 | call etini(rel) |
---|
1938 | call daclrd(h) |
---|
1939 | call daclrd(ft) |
---|
1940 | call etcct(x,roi,x) |
---|
1941 | do k=2,no |
---|
1942 | ! IF K>2 V = H(K)^-1 X(K) |
---|
1943 | call facflod(h,x,v,2,k-1,-1.0_dp,-1) |
---|
1944 | ! EXTRACTING K TH DEGREE OF V ----> W |
---|
1945 | call taked(v,k,w) |
---|
1946 | ! write(16,*) "$$$$$$$$ K $$$$$$$$$$", k |
---|
1947 | ! W = EXP(B5) + ... |
---|
1948 | call dacopd(w,b5) |
---|
1949 | ! CALL INTD(W,B5,-one) |
---|
1950 | ! B5 ON EXIT IS THE NEW CONTRIBUTION TO H |
---|
1951 | ! B6 IS THE NEW CONTRIBUTION TO FT |
---|
1952 | call nuanaflo(b5,b6) |
---|
1953 | call dalind(b5,1.0_dp,h,1.0_dp,b1) |
---|
1954 | call dacopd(b1,h) |
---|
1955 | ! EXP(B9) = EXP( : ROTI B6 :) |
---|
1956 | call trxflo(b6,b9,roi) |
---|
1957 | |
---|
1958 | ! V = EXP(-B6) REL |
---|
1959 | call facflod(b6,rel,v,k,k,-1.0_dp,1) |
---|
1960 | ! W = V o X |
---|
1961 | call etcct(v,x,w) |
---|
1962 | if(lielib_print(5)==1) then |
---|
1963 | w_p=0 |
---|
1964 | w_p%nc=1 |
---|
1965 | w_p%fc='(1((1X,A72),/))' |
---|
1966 | write(w_p%c(1),'(a13,i4)') ' ORDERFLO K= ', k |
---|
1967 | !CALL !WRITE_a |
---|
1968 | endif |
---|
1969 | ! X = EXP(B9) W |
---|
1970 | call facflod(b9,w,x,k,k,1.0_dp,1) |
---|
1971 | ! B6 IS THE NEW CONTRIBUTION TO FT |
---|
1972 | call dalind(b6,1.0_dp,ft,1.0_dp,b1) |
---|
1973 | call dacopd(b1,ft) |
---|
1974 | enddo |
---|
1975 | call dadal(b9,nd2) |
---|
1976 | call dadal(b6,nd2) |
---|
1977 | call dadal(b5,nd2) |
---|
1978 | call dadal(b1,nd2) |
---|
1979 | call dadal(roi,nd2) |
---|
1980 | call dadal(rel,nd2) |
---|
1981 | call dadal(v,nd2) |
---|
1982 | call dadal(w,nd2) |
---|
1983 | return |
---|
1984 | end subroutine orderflo |
---|
1985 | subroutine nuanaflo(h,ft) |
---|
1986 | implicit none |
---|
1987 | ! RESONANCE DENOMINATOR OPERATOR (1-R^-1)^-1 |
---|
1988 | integer i |
---|
1989 | integer,dimension(:)::h,ft |
---|
1990 | integer,dimension(ndim2)::br,bi,c,ci |
---|
1991 | integer,dimension(2)::t1,t2 |
---|
1992 | if(.not.c_%stable_da) return |
---|
1993 | |
---|
1994 | call etall(br,nd2) |
---|
1995 | call etall(bi,nd2) |
---|
1996 | call etall(c,nd2) |
---|
1997 | call etall(ci,nd2) |
---|
1998 | |
---|
1999 | call ctorflo(h,br,bi) |
---|
2000 | |
---|
2001 | ! FILTERING RESONANCES AND TUNE SHIFTS |
---|
2002 | ! ASSUMING REALITY I.E. B(2*I-1)=CMPCJG(B(2*I)) |
---|
2003 | |
---|
2004 | do i=1,nd2 |
---|
2005 | iflow=i |
---|
2006 | call dacfu(br(i),filt,c(i)) |
---|
2007 | call dacfu(bi(i),filt,ci(i)) |
---|
2008 | enddo |
---|
2009 | call rtocflo(c,ci,h) |
---|
2010 | |
---|
2011 | do i=1,nd2 |
---|
2012 | |
---|
2013 | iflow=i |
---|
2014 | call dacfu(br(i),dfilt,br(i)) |
---|
2015 | call dacfu(bi(i),dfilt,bi(i)) |
---|
2016 | enddo |
---|
2017 | ! NOW WE MUST REORDER C AND CI TO SEPARATE THE REAL AND IMAGINARY PART |
---|
2018 | ! THIS IS NOT NECESSARY WITH :H: OPERATORS |
---|
2019 | |
---|
2020 | do i=1,nd2 |
---|
2021 | t1(1)=br(i) |
---|
2022 | t1(2)=bi(i) |
---|
2023 | t2(1)=c(i) |
---|
2024 | t2(2)=ci(i) |
---|
2025 | iflow=i |
---|
2026 | call comcfu(t1,xgam,xgbm,t2) |
---|
2027 | enddo |
---|
2028 | |
---|
2029 | call rtocflo(c,ci,ft) |
---|
2030 | |
---|
2031 | call dadal(br,nd2) |
---|
2032 | call dadal(bi,nd2) |
---|
2033 | call dadal(c,nd2) |
---|
2034 | call dadal(ci,nd2) |
---|
2035 | |
---|
2036 | return |
---|
2037 | end subroutine nuanaflo |
---|
2038 | |
---|
2039 | real(dp) function xgam(j) |
---|
2040 | implicit none |
---|
2041 | ! XGAM AND XGBM ARE THE EIGENVALUES OF THE OPERATOR NEWANAFLO |
---|
2042 | integer i,ic,ij,ik |
---|
2043 | ! INTEGER J(NTT),JJ(NDIM),JP(NDIM) |
---|
2044 | integer,dimension(:)::j |
---|
2045 | integer,dimension(ndim)::jj,jp |
---|
2046 | real(dp) ad,ans,as,ex,exh |
---|
2047 | if(.not.c_%stable_da) return |
---|
2048 | |
---|
2049 | xgam=0.0_dp |
---|
2050 | ad=0.0_dp |
---|
2051 | as=0.0_dp |
---|
2052 | ic=0 |
---|
2053 | do i=1,nd-ndc |
---|
2054 | ik=2*i-1 |
---|
2055 | ij=2*i |
---|
2056 | jp(i)=j(ik)+j(ij) |
---|
2057 | jj(i)=j(ik)-j(ij) |
---|
2058 | if(ik.eq.iflow.or.ij.eq.iflow) then |
---|
2059 | jj(i)=jj(i)+(-1)**iflow |
---|
2060 | jp(i)=jp(i)-1 |
---|
2061 | endif |
---|
2062 | ic=ic+iabs(jj(i)) |
---|
2063 | enddo |
---|
2064 | |
---|
2065 | do i=1,nd-ndc |
---|
2066 | ad=dsta(i)*REAL(jj(i),kind=DP)*angle(i)-REAL(jp(i),kind=DP)*rad(i)+ad |
---|
2067 | as=sta(i)*REAL(jj(i),kind=DP)*angle(i)+as |
---|
2068 | enddo |
---|
2069 | |
---|
2070 | exh=EXP(ad/2.0_dp) |
---|
2071 | ex=exh**2 |
---|
2072 | ans=4.0_dp*ex*(SINH(ad/2.0_dp)**2+SIN(as/2.0_dp)**2) |
---|
2073 | if(ans.eq.0.0_dp) then |
---|
2074 | print*,"NormalForm makes no sense!" |
---|
2075 | print*,"no,nv,nd,nd2",no,nv,nd,nd2 |
---|
2076 | print*,"ndc,ndc2,ndt,ndpt",ndc,ndc2,ndt,ndpt |
---|
2077 | stop |
---|
2078 | endif |
---|
2079 | xgam=2.0_dp*(-exh*SINH(ad/2.0_dp)+ex*SIN(as/2.0_dp)**2)/ans |
---|
2080 | |
---|
2081 | return |
---|
2082 | end function xgam |
---|
2083 | real(dp) function xgbm(j) |
---|
2084 | implicit none |
---|
2085 | integer i,ic,ij,ik |
---|
2086 | real(dp) ad,ans,as,ex,exh |
---|
2087 | ! INTEGER J(NTT),JJ(NDIM),JP(NDIM) |
---|
2088 | integer,dimension(:)::j |
---|
2089 | integer,dimension(ndim)::jj,jp |
---|
2090 | if(.not.c_%stable_da) return |
---|
2091 | |
---|
2092 | xgbm=0.0_dp |
---|
2093 | ad=0.0_dp |
---|
2094 | as=0.0_dp |
---|
2095 | ic=0 |
---|
2096 | do i=1,nd-ndc |
---|
2097 | ik=2*i-1 |
---|
2098 | ij=2*i |
---|
2099 | jp(i)=j(ik)+j(ij) |
---|
2100 | jj(i)=j(ik)-j(ij) |
---|
2101 | if(ik.eq.iflow.or.ij.eq.iflow) then |
---|
2102 | jj(i)=jj(i)+(-1)**iflow |
---|
2103 | jp(i)=jp(i)-1 |
---|
2104 | endif |
---|
2105 | ic=ic+iabs(jj(i)) |
---|
2106 | enddo |
---|
2107 | |
---|
2108 | do i=1,nd-ndc |
---|
2109 | ad=dsta(i)*REAL(jj(i),kind=DP)*angle(i)-REAL(jp(i),kind=DP)*rad(i)+ad |
---|
2110 | as=sta(i)*REAL(jj(i),kind=DP)*angle(i)+as |
---|
2111 | enddo |
---|
2112 | |
---|
2113 | exh=EXP(ad/2.0_dp) |
---|
2114 | ex=exh**2 |
---|
2115 | ans=4.0_dp*ex*(SINH(ad/2.0_dp)**2+SIN(as/2.0_dp)**2) |
---|
2116 | if(ans.eq.0.0_dp) then |
---|
2117 | print*,"NormalForm makes no sense!" |
---|
2118 | print*,"no,nv,nd,nd2",no,nv,nd,nd2 |
---|
2119 | print*,"ndc,ndc2,ndt,ndpt",ndc,ndc2,ndt,ndpt |
---|
2120 | stop |
---|
2121 | endif |
---|
2122 | xgbm=SIN(as)*ex/ans |
---|
2123 | |
---|
2124 | return |
---|
2125 | end function xgbm |
---|
2126 | real(dp) function filt(j) |
---|
2127 | implicit none |
---|
2128 | ! PROJECTION FUNCTIONS ON THE KERNEL ANMD RANGE OF (1-R^-1) |
---|
2129 | !- THE KERNEL OF (1-R^-1) |
---|
2130 | integer i,ic,ic1,ic2,ij,ik,ji |
---|
2131 | ! INTEGER J(NTT),JJ(NDIM) |
---|
2132 | integer,dimension(:)::j |
---|
2133 | integer,dimension(ndim)::jj |
---|
2134 | if(.not.c_%stable_da) return |
---|
2135 | |
---|
2136 | filt=1.0_dp |
---|
2137 | |
---|
2138 | ic=0 |
---|
2139 | do i=1,nd-ndc |
---|
2140 | ik=2*i-1 |
---|
2141 | ij=2*i |
---|
2142 | jj(i)=j(ik)-j(ij) |
---|
2143 | if(ik.eq.iflow.or.ij.eq.iflow) then |
---|
2144 | jj(i)=jj(i)+(-1)**iflow |
---|
2145 | endif |
---|
2146 | ic=ic+iabs(jj(i)) |
---|
2147 | enddo |
---|
2148 | |
---|
2149 | if(ic.eq.0.and.jtune.eq.0) return |
---|
2150 | |
---|
2151 | do i=1,nres |
---|
2152 | ic1=1 |
---|
2153 | ic2=1 |
---|
2154 | do ji=1,nd-ndc |
---|
2155 | if(mx(ji,i).ne.jj(ji)) ic1=0 |
---|
2156 | if(mx(ji,i).ne.-jj(ji)) ic2=0 |
---|
2157 | if(ic1.eq.0.and.ic2.eq.0) goto 3 |
---|
2158 | enddo |
---|
2159 | return |
---|
2160 | 3 continue |
---|
2161 | enddo |
---|
2162 | |
---|
2163 | filt=0.0_dp |
---|
2164 | return |
---|
2165 | end function filt |
---|
2166 | |
---|
2167 | real(dp) function dfilt(j) |
---|
2168 | implicit none |
---|
2169 | !- THE RANGE OF (1-R^-1)^1 |
---|
2170 | !- CALLS FILT AND EXCHANGES 1 INTO 0 AND 0 INTO 1. |
---|
2171 | ! INTEGER J(NTT) |
---|
2172 | integer,dimension(:)::j |
---|
2173 | real(dp) fil |
---|
2174 | if(.not.c_%stable_da) return |
---|
2175 | |
---|
2176 | fil=filt(j) |
---|
2177 | if(fil.gt.0.5_dp) then |
---|
2178 | dfilt=0.0_dp |
---|
2179 | else |
---|
2180 | dfilt=1.0_dp |
---|
2181 | endif |
---|
2182 | return |
---|
2183 | end function dfilt |
---|
2184 | |
---|
2185 | subroutine dhdjflo(h,t) |
---|
2186 | implicit none |
---|
2187 | ! CONVENIENT TUNE SHIFT FINDED FOR SYMPLECTIC CASE (NU,DL)(H)=T |
---|
2188 | integer i,bb1,bb2,j1,j2 |
---|
2189 | integer,dimension(:)::h,t |
---|
2190 | integer,dimension(ndim2)::b1,b2,temp |
---|
2191 | logical doflip |
---|
2192 | if(.not.c_%stable_da) return |
---|
2193 | |
---|
2194 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
2195 | perform_flip=.false. |
---|
2196 | call flipflo(h,h,1) |
---|
2197 | doflip=.true. |
---|
2198 | else |
---|
2199 | doflip=.false. |
---|
2200 | endif |
---|
2201 | |
---|
2202 | |
---|
2203 | call etall(b1,nd2) |
---|
2204 | call etall(b2,nd2) |
---|
2205 | call etall1(bb1) |
---|
2206 | call etall1(bb2) |
---|
2207 | |
---|
2208 | call ctorflo(h,b1,b2) |
---|
2209 | |
---|
2210 | do i=1,nd-ndc |
---|
2211 | call datra(2*i,b2(2*i),bb1) |
---|
2212 | call dacmu(bb1,twopii,t(i+nd)) |
---|
2213 | call dacop(t(i+nd),bb1) |
---|
2214 | call daclr(bb2) |
---|
2215 | call rtoc(bb1,bb2,bb1) |
---|
2216 | call dacop(bb1,t(i)) |
---|
2217 | enddo |
---|
2218 | |
---|
2219 | if(ndpt.ne.0) then |
---|
2220 | call dacop(h(ndt),t(nd)) |
---|
2221 | call dacop(b1(ndt),t(nd2)) |
---|
2222 | endif |
---|
2223 | |
---|
2224 | if(doflip) then |
---|
2225 | call flipflo(h,h,-1) |
---|
2226 | call etall(temp,nd2) |
---|
2227 | do i=1,nd2 |
---|
2228 | call flip_i(t(i),temp(i),-1) |
---|
2229 | enddo |
---|
2230 | |
---|
2231 | if(mod(ndpt,2)==0) then |
---|
2232 | j1=ndpt/2 |
---|
2233 | j2=npt_pos/2 |
---|
2234 | else |
---|
2235 | j1=(ndpt+1)/2 |
---|
2236 | j2=(npt_pos+1)/2 |
---|
2237 | endif |
---|
2238 | |
---|
2239 | do i=1,nd2 |
---|
2240 | call dacop(temp(i),t(i)) |
---|
2241 | enddo |
---|
2242 | |
---|
2243 | call dacop(temp(j1),t(j2)) |
---|
2244 | call dacop(temp(j1+nd),t(j2+nd)) |
---|
2245 | call dacop(temp(j2),t(j1)) |
---|
2246 | call dacop(temp(j2+nd),t(j1+nd)) |
---|
2247 | |
---|
2248 | call dadal(temp,nd2) |
---|
2249 | perform_flip=.true. |
---|
2250 | endif |
---|
2251 | |
---|
2252 | call dadal1(bb2) |
---|
2253 | call dadal1(bb1) |
---|
2254 | call dadal(b2,nd2) |
---|
2255 | call dadal(b1,nd2) |
---|
2256 | return |
---|
2257 | end subroutine dhdjflo |
---|
2258 | |
---|
2259 | |
---|
2260 | |
---|
2261 | subroutine h2pluflo(h,ang,ra) |
---|
2262 | implicit none |
---|
2263 | ! POKES IN \VEC{H} ANGLES AND DAMPING COEFFFICIENTS |
---|
2264 | ! |
---|
2265 | integer i |
---|
2266 | integer,dimension(ntt)::j |
---|
2267 | integer,dimension(:)::h |
---|
2268 | real(dp) r1,r2 |
---|
2269 | real(dp),dimension(ndim)::ang,ra,st |
---|
2270 | if(.not.c_%stable_da) return |
---|
2271 | |
---|
2272 | do i=1,nd |
---|
2273 | st(i)=2.0_dp*sta(i)-1.0_dp |
---|
2274 | enddo |
---|
2275 | |
---|
2276 | do i=1,ntt |
---|
2277 | j(i)=0 |
---|
2278 | enddo |
---|
2279 | |
---|
2280 | do i=1,nd-ndc |
---|
2281 | j(2*i-1)=1 |
---|
2282 | r1=-ang(i) |
---|
2283 | !----- |
---|
2284 | call dapok(h(2*i),j,r1) |
---|
2285 | |
---|
2286 | r2=ra(i) |
---|
2287 | call dapok(h(2*i-1),j,r2) |
---|
2288 | j(2*i-1)=0 |
---|
2289 | |
---|
2290 | j(2*i)=1 |
---|
2291 | r1=ang(i)*st(i) |
---|
2292 | call dapok(h(2*i-1),j,r1) |
---|
2293 | call dapok(h(2*i),j,r2) |
---|
2294 | j(2*i)=0 |
---|
2295 | |
---|
2296 | enddo |
---|
2297 | |
---|
2298 | if(ndpt.eq.nd2-1) then |
---|
2299 | j(ndpt)=1 |
---|
2300 | call dapok(h(ndt),j,ang(nd)) |
---|
2301 | elseif(ndpt.eq.nd2) then |
---|
2302 | j(ndpt)=1 |
---|
2303 | call dapok(h(ndt),j,-ang(nd)) |
---|
2304 | endif |
---|
2305 | return |
---|
2306 | end subroutine h2pluflo |
---|
2307 | |
---|
2308 | subroutine rotflo(ro,ang,ra) |
---|
2309 | implicit none |
---|
2310 | ! CREATES R AND R^-1 USING THE EXISTING ANGLES AND DAMPING |
---|
2311 | ! COULD BE REPLACED BY A CALL H2PLUFLO FOLLOWED BY EXPFLOD |
---|
2312 | ! CREATES R |
---|
2313 | integer i |
---|
2314 | integer,dimension(ntt)::j |
---|
2315 | integer,dimension(:)::ro |
---|
2316 | real(dp) ch,sh,sim,xx |
---|
2317 | real(dp),dimension(ndim)::co,si,ang,ra |
---|
2318 | if(.not.c_%stable_da) return |
---|
2319 | |
---|
2320 | call daclrd(ro) |
---|
2321 | do i=1,nd-ndc |
---|
2322 | xx=EXP(ra(i)) |
---|
2323 | if(ista(i).eq.0) then |
---|
2324 | call hyper(ang(i),ch,sh) |
---|
2325 | co(i)=ch*xx |
---|
2326 | si(i)=-sh*xx |
---|
2327 | else |
---|
2328 | co(i)=COS(ang(i))*xx |
---|
2329 | si(i)=SIN(ang(i))*xx |
---|
2330 | endif |
---|
2331 | enddo |
---|
2332 | do i=1,nd-ndc |
---|
2333 | if(ista(i).eq.0)then |
---|
2334 | sim=si(i) |
---|
2335 | else |
---|
2336 | sim=-si(i) |
---|
2337 | endif |
---|
2338 | j(2*i-1)=1 |
---|
2339 | call dapok(ro(2*i-1),j,co(i)) |
---|
2340 | call dapok(ro(2*i),j,sim) |
---|
2341 | j(2*i-1)=0 |
---|
2342 | j(2*i)=1 |
---|
2343 | call dapok(ro(2*i),j,co(i)) |
---|
2344 | call dapok(ro(2*i-1),j,si(i)) |
---|
2345 | j(2*i)=0 |
---|
2346 | enddo |
---|
2347 | |
---|
2348 | if(ndc.eq.1) then |
---|
2349 | j(ndt)=1 |
---|
2350 | call dapok(ro(ndt),j,1.0_dp) |
---|
2351 | call dapok(ro(ndpt),j,0.0_dp) |
---|
2352 | j(ndt)=0 |
---|
2353 | j(ndpt)=1 |
---|
2354 | call dapok(ro(ndt),j,ang(nd)) |
---|
2355 | call dapok(ro(ndpt),j,1.0_dp) |
---|
2356 | j(ndpt)=0 |
---|
2357 | endif |
---|
2358 | |
---|
2359 | return |
---|
2360 | end subroutine rotflo |
---|
2361 | subroutine rotiflo(roi,ang,ra) |
---|
2362 | implicit none |
---|
2363 | ! CREATES R^-1 |
---|
2364 | integer i |
---|
2365 | real(dp) ch,sh,sim,simv,xx |
---|
2366 | integer,dimension(ntt)::j |
---|
2367 | integer,dimension(:)::roi |
---|
2368 | real(dp),dimension(ndim)::co,si,ang,ra |
---|
2369 | if(.not.c_%stable_da) return |
---|
2370 | |
---|
2371 | ! do i=1,10 |
---|
2372 | j=0 |
---|
2373 | ! enddo |
---|
2374 | |
---|
2375 | call daclrd(roi) |
---|
2376 | do i=1,nd-ndc |
---|
2377 | xx=EXP(-ra(i)) |
---|
2378 | if(ista(i).eq.0) then |
---|
2379 | call hyper(ang(i),ch,sh) |
---|
2380 | co(i)=ch*xx |
---|
2381 | si(i)=-sh*xx |
---|
2382 | else |
---|
2383 | co(i)=COS(ang(i))*xx |
---|
2384 | si(i)=SIN(ang(i))*xx |
---|
2385 | endif |
---|
2386 | enddo |
---|
2387 | do i=1,nd-ndc |
---|
2388 | if(ista(i).eq.0)then |
---|
2389 | sim=si(i) |
---|
2390 | else |
---|
2391 | sim=-si(i) |
---|
2392 | endif |
---|
2393 | j(2*i-1)=1 |
---|
2394 | call dapok(roi(2*i-1),j,co(i)) |
---|
2395 | simv=-sim |
---|
2396 | call dapok(roi(2*i),j,simv) |
---|
2397 | j(2*i-1)=0 |
---|
2398 | j(2*i)=1 |
---|
2399 | simv=-si(i) |
---|
2400 | call dapok(roi(2*i),j,co(i)) |
---|
2401 | call dapok(roi(2*i-1),j,simv) |
---|
2402 | j(2*i)=0 |
---|
2403 | enddo |
---|
2404 | |
---|
2405 | if(ndc.eq.1) then |
---|
2406 | j(ndt)=1 |
---|
2407 | call dapok(roi(ndt),j,1.0_dp) |
---|
2408 | call dapok(roi(ndpt),j,0.0_dp) |
---|
2409 | j(ndt)=0 |
---|
2410 | j(ndpt)=1 |
---|
2411 | call dapok(roi(ndt),j,-ang(nd)) |
---|
2412 | call dapok(roi(ndpt),j,1.0_dp) |
---|
2413 | j(ndpt)=0 |
---|
2414 | endif |
---|
2415 | |
---|
2416 | return |
---|
2417 | end subroutine rotiflo |
---|
2418 | |
---|
2419 | subroutine hyper(a,ch,sh) |
---|
2420 | implicit none |
---|
2421 | real(dp) a,ch,sh,x,xi |
---|
2422 | if(.not.c_%stable_da) return |
---|
2423 | ! USED IN ROTIFLO AND ROTFLO |
---|
2424 | x=EXP(a) |
---|
2425 | xi=1.0_dp/x |
---|
2426 | ch=(x+xi)/2.0_dp |
---|
2427 | sh=(x-xi)/2.0_dp |
---|
2428 | return |
---|
2429 | end subroutine hyper |
---|
2430 | |
---|
2431 | subroutine ctor(c1,r2,i2) |
---|
2432 | implicit none |
---|
2433 | ! CHANGES OF BASIS |
---|
2434 | ! C1------> R2+I R1 |
---|
2435 | integer c1,r2,i2,b1,b2 |
---|
2436 | integer,dimension(ndim2)::x |
---|
2437 | logical doflip |
---|
2438 | if(.not.c_%stable_da) return |
---|
2439 | |
---|
2440 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
2441 | perform_flip=.false. |
---|
2442 | call flip_i(c1,c1,1) |
---|
2443 | doflip=.true. |
---|
2444 | else |
---|
2445 | doflip=.false. |
---|
2446 | endif |
---|
2447 | |
---|
2448 | call etall1(b1) |
---|
2449 | call etall1(b2) |
---|
2450 | call etallnom(x,nd2) ! ,'X ') |
---|
2451 | |
---|
2452 | |
---|
2453 | call ctoi(c1,b1) |
---|
2454 | call etcjg(x) |
---|
2455 | call trx(b1,b2,x) |
---|
2456 | call dalin(b1,0.5_dp,b2,0.5_dp,r2) |
---|
2457 | call dalin(b1,0.5_dp,b2,-0.5_dp,i2) |
---|
2458 | |
---|
2459 | if(doflip) then |
---|
2460 | call flip_i(c1,c1,-1) |
---|
2461 | if(r2/=c1) call flip_i(r2,r2,-1) |
---|
2462 | if(i2/=c1.and.i2/=r2) call flip_i(i2,i2,-1) |
---|
2463 | perform_flip=.true. |
---|
2464 | endif |
---|
2465 | |
---|
2466 | |
---|
2467 | call dadal(x,nd2) |
---|
2468 | call dadal1(b2) |
---|
2469 | call dadal1(b1) |
---|
2470 | return |
---|
2471 | end subroutine ctor |
---|
2472 | subroutine rtoc(r1,i1,c2) |
---|
2473 | implicit none |
---|
2474 | ! INVERSE OF CTOR |
---|
2475 | integer c2,r1,i1,b1 |
---|
2476 | logical doflip |
---|
2477 | if(.not.c_%stable_da) return |
---|
2478 | |
---|
2479 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
2480 | perform_flip=.false. |
---|
2481 | call flip_i(r1,r1,1) |
---|
2482 | call flip_i(i1,i1,1) |
---|
2483 | doflip=.true. |
---|
2484 | else |
---|
2485 | doflip=.false. |
---|
2486 | endif |
---|
2487 | |
---|
2488 | call etall1(b1) |
---|
2489 | |
---|
2490 | call daadd(r1,i1,b1) |
---|
2491 | call itoc(b1,c2) |
---|
2492 | call dadal1(b1) |
---|
2493 | |
---|
2494 | |
---|
2495 | if(doflip) then |
---|
2496 | call flip_i(r1,r1,-1) |
---|
2497 | if(i1/=r1) call flip_i(i1,i1,-1) |
---|
2498 | if(c2/=r1.and.c2/=i1) call flip_i(c2,c2,-1) |
---|
2499 | perform_flip=.true. |
---|
2500 | endif |
---|
2501 | |
---|
2502 | return |
---|
2503 | end subroutine rtoc |
---|
2504 | subroutine ctorflo(c,dr,di) |
---|
2505 | implicit none |
---|
2506 | ! FLOW CTOR |
---|
2507 | integer,dimension(:)::dr,di,c |
---|
2508 | if(.not.c_%stable_da) return |
---|
2509 | |
---|
2510 | call ctord(c,dr,di) |
---|
2511 | call resvec(dr,di) |
---|
2512 | |
---|
2513 | return |
---|
2514 | end subroutine ctorflo |
---|
2515 | subroutine rtocflo(dr,di,c) |
---|
2516 | implicit none |
---|
2517 | ! FLOW RTOC |
---|
2518 | integer,dimension(:)::dr,di,c |
---|
2519 | integer,dimension(ndim2)::er,ei |
---|
2520 | if(.not.c_%stable_da) return |
---|
2521 | |
---|
2522 | call etall(er,nd2) |
---|
2523 | call etall(ei,nd2) |
---|
2524 | |
---|
2525 | call reelflo(dr,di,er,ei) |
---|
2526 | call rtocd(er,ei,c) |
---|
2527 | |
---|
2528 | call dadal(er,nd2) |
---|
2529 | call dadal(ei,nd2) |
---|
2530 | |
---|
2531 | return |
---|
2532 | end subroutine rtocflo |
---|
2533 | subroutine ctord(c,cr,ci) |
---|
2534 | implicit none |
---|
2535 | ! ROUTINES USED IN THE INTERMEDIATE STEPS OF CTORFLO AND RTOCFLO |
---|
2536 | ! SAME AS CTOR OVER ARRAYS CONTAINING ND2 COMPONENTS |
---|
2537 | ! ROUTINE USEFUL IN INTERMEDIATE FLOW CHANGE OF BASIS |
---|
2538 | integer i |
---|
2539 | integer,dimension(:)::c,ci,cr |
---|
2540 | if(.not.c_%stable_da) return |
---|
2541 | |
---|
2542 | do i=1,nd2 |
---|
2543 | call ctor(c(i),cr(i),ci(i)) |
---|
2544 | enddo |
---|
2545 | return |
---|
2546 | end subroutine ctord |
---|
2547 | subroutine rtocd(cr,ci,c) |
---|
2548 | implicit none |
---|
2549 | ! INVERSE OF CTORD |
---|
2550 | integer i |
---|
2551 | integer,dimension(:)::c,ci,cr |
---|
2552 | if(.not.c_%stable_da) return |
---|
2553 | |
---|
2554 | do i=1,nd2 |
---|
2555 | call rtoc(cr(i),ci(i),c(i)) |
---|
2556 | enddo |
---|
2557 | return |
---|
2558 | end subroutine rtocd |
---|
2559 | subroutine resvec(cr,ci) |
---|
2560 | implicit none |
---|
2561 | ! DOES THE SPINOR PART IN CTORFLO |
---|
2562 | integer i |
---|
2563 | integer,dimension(:)::ci,cr |
---|
2564 | integer,dimension(2)::tr,ti |
---|
2565 | logical doflip |
---|
2566 | if(.not.c_%stable_da) return |
---|
2567 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
2568 | perform_flip=.false. |
---|
2569 | call flipflo(cr,cr,1) |
---|
2570 | call flipflo(ci,ci,1) |
---|
2571 | doflip=.true. |
---|
2572 | else |
---|
2573 | doflip=.false. |
---|
2574 | endif |
---|
2575 | |
---|
2576 | call etall(tr,2) |
---|
2577 | call etall(ti,2) |
---|
2578 | |
---|
2579 | do i=1,nd-ndc |
---|
2580 | if(ista(i).eq.1) then |
---|
2581 | call dasub(cr(2*i-1),ci(2*i),tr(1)) |
---|
2582 | call daadd(ci(2*i-1),cr(2*i),ti(1)) |
---|
2583 | call daadd(cr(2*i-1),ci(2*i),tr(2)) |
---|
2584 | call dasub(ci(2*i-1),cr(2*i),ti(2)) |
---|
2585 | call dacop(tr(1),cr(2*i-1)) |
---|
2586 | call dacop(tr(2),cr(2*i)) |
---|
2587 | call dacop(ti(1),ci(2*i-1)) |
---|
2588 | call dacop(ti(2),ci(2*i)) |
---|
2589 | else |
---|
2590 | call daadd(cr(2*i-1),cr(2*i),tr(1)) |
---|
2591 | call daadd(ci(2*i-1),ci(2*i),ti(1)) |
---|
2592 | call dasub(cr(2*i-1),cr(2*i),tr(2)) |
---|
2593 | call dasub(ci(2*i-1),ci(2*i),ti(2)) |
---|
2594 | call dacop(tr(1),cr(2*i-1)) |
---|
2595 | call dacop(tr(2),cr(2*i)) |
---|
2596 | call dacop(ti(1),ci(2*i-1)) |
---|
2597 | call dacop(ti(2),ci(2*i)) |
---|
2598 | endif |
---|
2599 | enddo |
---|
2600 | |
---|
2601 | ! do i=nd2-ndc2+1,nd2 |
---|
2602 | ! call dacop(cr(i),dr(i)) |
---|
2603 | ! call dacop(ci(i),di(i)) |
---|
2604 | ! enddo |
---|
2605 | |
---|
2606 | if(doflip) then |
---|
2607 | call flipflo(cr,cr,-1) |
---|
2608 | call flipflo(ci,ci,-1) |
---|
2609 | perform_flip=.true. |
---|
2610 | endif |
---|
2611 | |
---|
2612 | call dadal(tr,2) |
---|
2613 | call dadal(ti,2) |
---|
2614 | return |
---|
2615 | end subroutine resvec |
---|
2616 | |
---|
2617 | subroutine reelflo(c,ci,f,fi) |
---|
2618 | implicit none |
---|
2619 | ! DOES THE SPINOR PART IN RTOCFLO |
---|
2620 | integer i |
---|
2621 | integer,dimension(:)::c,ci,f,fi |
---|
2622 | integer,dimension(ndim2)::e,ei |
---|
2623 | logical doflip |
---|
2624 | if(.not.c_%stable_da) return |
---|
2625 | |
---|
2626 | if(perform_flip.and.new_ndpt.and.ndpt/=0) then |
---|
2627 | perform_flip=.false. |
---|
2628 | call flipflo(c,c,1) |
---|
2629 | call flipflo(ci,ci,1) |
---|
2630 | doflip=.true. |
---|
2631 | else |
---|
2632 | doflip=.false. |
---|
2633 | endif |
---|
2634 | |
---|
2635 | call etall(e,nd2) |
---|
2636 | call etall(ei,nd2) |
---|
2637 | |
---|
2638 | do i=1,nd-ndc |
---|
2639 | call dalin(c(2*i-1),0.5_dp,c(2*i),0.5_dp,e(2*i-1)) |
---|
2640 | call dalin(ci(2*i-1),0.5_dp,ci(2*i),0.5_dp,ei(2*i-1)) |
---|
2641 | if(ista(i).eq.1) then |
---|
2642 | call dalin(ci(2*i-1),0.5_dp,ci(2*i),-0.5_dp,e(2*i)) |
---|
2643 | call dalin(c(2*i-1),-0.5_dp,c(2*i),0.5_dp,ei(2*i)) |
---|
2644 | else |
---|
2645 | call dalin(ci(2*i-1),0.5_dp,ci(2*i),-0.5_dp,ei(2*i)) |
---|
2646 | call dalin(c(2*i-1),0.5_dp,c(2*i),-0.5_dp,e(2*i)) |
---|
2647 | endif |
---|
2648 | enddo |
---|
2649 | |
---|
2650 | do i=nd2-ndc2+1,nd2 |
---|
2651 | call dacop(c(i),e(i)) |
---|
2652 | call dacop(ci(i),ei(i)) |
---|
2653 | enddo |
---|
2654 | |
---|
2655 | call dacopd(e,f) |
---|
2656 | call dacopd(ei,fi) |
---|
2657 | |
---|
2658 | call dadal(e,nd2) |
---|
2659 | call dadal(ei,nd2) |
---|
2660 | |
---|
2661 | if(doflip) then |
---|
2662 | call flipflo(c,c,-1) |
---|
2663 | call flipflo(ci,ci,-1) |
---|
2664 | if(c(1)/=f(1).and.ci(1)/=f(1)) then |
---|
2665 | call flipflo(f,f,-1) |
---|
2666 | endif |
---|
2667 | if(c(1)/=fi(1).and.ci(1)/=fi(1)) then |
---|
2668 | call flipflo(fi,fi,-1) |
---|
2669 | endif |
---|
2670 | perform_flip=.true. |
---|
2671 | endif |
---|
2672 | |
---|
2673 | return |
---|
2674 | end subroutine reelflo |
---|
2675 | |
---|
2676 | subroutine midbflo(c,a2,a2i,q,a,st) |
---|
2677 | implicit none |
---|
2678 | ! LINEAR EXACT NORMALIZATION USING EIGENVALUE PACKAGE OF NERI |
---|
2679 | ! |
---|
2680 | integer i,j |
---|
2681 | integer,dimension(ntt)::jx |
---|
2682 | integer,dimension(:)::c,a2,a2i |
---|
2683 | real(dp) ch,r,shm |
---|
2684 | real(dp),dimension(ndim2,ndim2)::cr,sa,sai,cm |
---|
2685 | real(dp),dimension(ndim)::st,q,a |
---|
2686 | if(.not.c_%stable_da) return |
---|
2687 | |
---|
2688 | do i=1,ntt |
---|
2689 | jx(i)=0 |
---|
2690 | enddo |
---|
2691 | |
---|
2692 | ! frank/etienne |
---|
2693 | do i=1,ndim |
---|
2694 | st(i)=0.0_dp |
---|
2695 | q(i)=0.0_dp |
---|
2696 | a(i)=0.0_dp |
---|
2697 | enddo |
---|
2698 | ! frank/etienne |
---|
2699 | do i=1,ndim2 |
---|
2700 | ! frank/etienne |
---|
2701 | do j=1,ndim2 |
---|
2702 | sai(i,j)=0.0_dp |
---|
2703 | sa(i,j)=0.0_dp |
---|
2704 | cm(i,j)=0.0_dp |
---|
2705 | cr(i,j)=0.0_dp |
---|
2706 | enddo |
---|
2707 | enddo |
---|
2708 | |
---|
2709 | do i=1,nd2 |
---|
2710 | do j=1,nd2 |
---|
2711 | jx(j)=1 |
---|
2712 | call dapek(c(i),jx,r) |
---|
2713 | jx(j)=0 |
---|
2714 | cm(i,j)=r |
---|
2715 | enddo |
---|
2716 | enddo |
---|
2717 | |
---|
2718 | call mapflol(sa,sai,cr,cm,st) |
---|
2719 | do i=1,nd-ndc |
---|
2720 | if(st(i)+1e-3_dp.gt.1.0_dp) then |
---|
2721 | a(i)=sqrt(cr(2*i-1,2*i-1)**2+cr(2*i-1,2*i)**2) |
---|
2722 | q(i)=ARCCOS_lielib(cr(2*i-1,2*i-1)/a(i)) |
---|
2723 | a(i)=LOGE_lielib(a(i)) |
---|
2724 | if(cr(2*i-1,2*i).lt.0.0_dp) q(i)=twopi-q(i) |
---|
2725 | else |
---|
2726 | a(i)=sqrt(cr(2*i-1,2*i-1)**2-cr(2*i-1,2*i)**2) |
---|
2727 | ch=cr(2*i-1,2*i-1)/a(i) |
---|
2728 | shm=cr(2*i-1,2*i)/a(i) |
---|
2729 | ! CH=CH+SQRT(CH**2-one) |
---|
2730 | ! q(i)=LOG(CH) |
---|
2731 | q(i)=-LOGE_lielib(ch+shm) ! half integer ???? blows up |
---|
2732 | ! IF(cr(2*i-1,2*i).gt.zero) Q(I)=-Q(I) |
---|
2733 | a(i)=LOGE_lielib(a(i)) |
---|
2734 | endif |
---|
2735 | enddo |
---|
2736 | if(ndc.eq.0) then |
---|
2737 | if(time_plane>0) then |
---|
2738 | if(new_ndpt) then |
---|
2739 | ! do i=1,nd |
---|
2740 | ! write(6,*) i,q(i)/twopi |
---|
2741 | ! if(st(i)+c_1d_3.gt.one.and.q(i).gt.pi) q(i)=q(i)-twopi |
---|
2742 | ! write(6,*) i,q(i)/twopi |
---|
2743 | ! pause 77 |
---|
2744 | ! enddo |
---|
2745 | if(st(time_plane)+1e-3_dp.gt.1.0_dp.and.nd.ge.3.and.q(time_plane).gt.pi) q(time_plane)=q(time_plane)-twopi |
---|
2746 | else |
---|
2747 | if(st(time_plane)+1e-3_dp.gt.1.0_dp.and.nd.ge.3.and.q(time_plane).gt.pi) q(time_plane)=q(time_plane)-twopi |
---|
2748 | endif |
---|
2749 | endif |
---|
2750 | else |
---|
2751 | if(new_ndpt) then |
---|
2752 | ! do i=1,nd-1 |
---|
2753 | ! if(st(i)+c_1d_3.gt.one.and.q(i).gt.pi) q(i)=q(i)-twopi |
---|
2754 | ! enddo |
---|
2755 | endif |
---|
2756 | q(nd)=cr(ndt,ndpt) |
---|
2757 | endif |
---|
2758 | |
---|
2759 | call daclrd(a2) |
---|
2760 | call daclrd(a2i) |
---|
2761 | |
---|
2762 | do i=1,nd2 |
---|
2763 | do j=1,nd2 |
---|
2764 | jx(j)=1 |
---|
2765 | r=sa(i,j) |
---|
2766 | if(r.ne.0.0_dp)call dapok(a2(i),jx,r) |
---|
2767 | jx(j)=1 |
---|
2768 | r=sai(i,j) |
---|
2769 | if(r.ne.0.0_dp)call dapok(a2i(i),jx,r) |
---|
2770 | jx(j)=0 |
---|
2771 | enddo |
---|
2772 | enddo |
---|
2773 | |
---|
2774 | return |
---|
2775 | end subroutine midbflo |
---|
2776 | |
---|
2777 | subroutine mapflol(sa,sai,cr,cm,st) |
---|
2778 | implicit none |
---|
2779 | !---- FROM TRACKING CODE |
---|
2780 | ! --------------------- |
---|
2781 | integer i,ier,iunst,j,l,n1 |
---|
2782 | integer,dimension(ndim)::n |
---|
2783 | real(dp) ap,ax,rd,rd1,xd,xsu |
---|
2784 | real(dp),dimension(ndim2,ndim2)::cr,xj,sa,sai,cm,w,vr,vi,s1 |
---|
2785 | real(dp),dimension(ndim)::x,xx,st |
---|
2786 | real(dp),dimension(ndim2)::rr,ri,p |
---|
2787 | logical hyp |
---|
2788 | if(.not.c_%stable_da) return |
---|
2789 | |
---|
2790 | n1=0 |
---|
2791 | ! frank/etienne |
---|
2792 | do i=1,ndim2 |
---|
2793 | do j=1,ndim2 |
---|
2794 | cr(j,i)=cm(i,j) |
---|
2795 | xj(i,j)=0.0_dp |
---|
2796 | s1(i,j)=0.0_dp |
---|
2797 | enddo |
---|
2798 | enddo |
---|
2799 | |
---|
2800 | ! frank/etienne |
---|
2801 | do i=1,ndim |
---|
2802 | n(i)=0 |
---|
2803 | xj(2*i-1,2*i)=1.0_dp |
---|
2804 | xj(2*i,2*i-1)=-1.0_dp |
---|
2805 | enddo |
---|
2806 | ! frank/etienne |
---|
2807 | do i=1,ndim2 |
---|
2808 | do j=1,ndim2 |
---|
2809 | sai(i,j)=0.0_dp |
---|
2810 | w(i,j)=cm(i,j) |
---|
2811 | enddo |
---|
2812 | enddo |
---|
2813 | if(ndc.eq.1) then |
---|
2814 | s1(nd2-ndc,nd2-ndc)=1.0_dp |
---|
2815 | s1(nd2,nd2)=1.0_dp |
---|
2816 | sai(nd2-ndc,nd2-ndc)=1.0_dp |
---|
2817 | sai(nd2,nd2)=1.0_dp |
---|
2818 | endif |
---|
2819 | call mulnd2(xj,w) |
---|
2820 | call mulnd2(cr,w) |
---|
2821 | if(lielib_print(6)==1) then |
---|
2822 | w_p=0 |
---|
2823 | w_p%nc=1 |
---|
2824 | w_p%fc='(1((1X,A72),/))' |
---|
2825 | w_p%c(1)= 'Check of the symplectic condition on the linear part' |
---|
2826 | !CALL !WRITE_a |
---|
2827 | xsu=0.0_dp |
---|
2828 | do i=1,nd2 |
---|
2829 | w_p=0 |
---|
2830 | w_p%nr=nd2 |
---|
2831 | w_p%fr='(6(2x,g23.16))' |
---|
2832 | do j=1,nd2 |
---|
2833 | w_p%r(j)=w(i,j) |
---|
2834 | enddo |
---|
2835 | !CALL !WRITE_a |
---|
2836 | |
---|
2837 | do j=1,nd2 |
---|
2838 | xsu=xsu+abs(w(i,j)-XJ(I,J)) |
---|
2839 | enddo |
---|
2840 | enddo |
---|
2841 | w_p=0 |
---|
2842 | w_p%nc=1 |
---|
2843 | w_p%fc='((1X,A120))' |
---|
2844 | ! write(w_p%c(1),'(a29,g23.16,a2)') 'Deviation from symplecticity ',c_100*(xsu)/ND2, ' %' |
---|
2845 | write(6,'(a29,g23.16,a2)') 'Deviation from symplecticity ',100.0_dp*(xsu)/ND2, ' %' |
---|
2846 | !CALL !WRITE_a |
---|
2847 | endif |
---|
2848 | call eig6(cr,rr,ri,vr,vi) |
---|
2849 | rr_eigen=0.0_dp |
---|
2850 | ri_eigen=0.0_dp |
---|
2851 | rr_eigen=rr |
---|
2852 | ri_eigen=ri |
---|
2853 | hyp=.false. |
---|
2854 | ! write(6,*) " checking no_hyperbolic_in_normal_form " |
---|
2855 | do i=1,nd2-ndc2 |
---|
2856 | if(ri(i)==0.0_dp) then |
---|
2857 | hyp=.true. |
---|
2858 | c_%stable_da=.false. |
---|
2859 | c_%check_stable=.false. |
---|
2860 | endif |
---|
2861 | enddo |
---|
2862 | |
---|
2863 | |
---|
2864 | if(hyp) then |
---|
2865 | if(no_hyperbolic_in_normal_form) then !no_hyperbolic_in_normal_form |
---|
2866 | write(6,*) " Eigenvalues are " |
---|
2867 | do i=1,nd2-ndc2 |
---|
2868 | write(6,*) i,rr(i),ri(i) |
---|
2869 | enddo |
---|
2870 | write(6,*) " HYPERBOLIC NORMAL FORM DETECTED " |
---|
2871 | write(6,*) " All TPSA/DA/LIE CALCULATIONS INTERRUPTED AT YOUR REQUEST " |
---|
2872 | write(6,*) " PLEASE RESET STABLE FLAGS " |
---|
2873 | else |
---|
2874 | write(6,*) " HYPERBOLIC NORMAL FORM DETECTED " |
---|
2875 | write(6,*) " HOPE YOU KNOW WHAT YOU ARE DOING " |
---|
2876 | endif |
---|
2877 | endif ! no_hyperbolic_in_normal_form |
---|
2878 | |
---|
2879 | ! checking for Krein |
---|
2880 | |
---|
2881 | if(check_krein.and.(.not.hyp)) then |
---|
2882 | |
---|
2883 | if(.not.hyp.and.nd2>2) then |
---|
2884 | xsu=0.0_dp |
---|
2885 | xd=0.0_dp |
---|
2886 | do i=1,4 |
---|
2887 | xsu=log(rr(i)**2+ri(i)**2)+xsu |
---|
2888 | xd=abs(log(rr(i)**2+ri(i)**2))+xd |
---|
2889 | enddo |
---|
2890 | |
---|
2891 | if(xsu<size_krein.and.xd>size_krein) then |
---|
2892 | write(6,*) " A Krein collision seemed to have happened " |
---|
2893 | write(6,*) " All calculations interrupted " |
---|
2894 | do i=1,nd2-ndc |
---|
2895 | write(6,*)"damping ", log(rr(i)**2+ri(i)**2) |
---|
2896 | enddo |
---|
2897 | do i=1,nd2-ndc |
---|
2898 | write(6,*)"tunes ", atan2(ri(i),rr(i))/twopi |
---|
2899 | enddo |
---|
2900 | do i=1,nd2 |
---|
2901 | write(6,*)"eigenvalues ", rr(i),ri(i) |
---|
2902 | enddo |
---|
2903 | c_%stable_da=.false. |
---|
2904 | c_%check_stable=.false. |
---|
2905 | |
---|
2906 | endif |
---|
2907 | |
---|
2908 | endif |
---|
2909 | endif |
---|
2910 | |
---|
2911 | if(lielib_print(7)==-1) then |
---|
2912 | w_p=0 |
---|
2913 | w_p%nc=3 |
---|
2914 | w_p%fc='(2(1X,A120,/),(1X,A120))' |
---|
2915 | w_p%c(2)= ' Index Real Part ArcSin(Imaginary Part)/2/pi' |
---|
2916 | write(6,w_p%fc) w_p%c(2) |
---|
2917 | !CALL !WRITE_a |
---|
2918 | do i=1,nd-ndc |
---|
2919 | rd1=SQRT(rr(2*i-1)**2+ri(2*i-1)**2) |
---|
2920 | rd=SQRT(rr(2*i)**2+ri(2*i)**2) |
---|
2921 | ! write(6,*) "modulus ",rd1,rd |
---|
2922 | w_p=0 |
---|
2923 | w_p%nc=3 |
---|
2924 | w_p%fc='(2(1X,A120,/),(1X,A120))' |
---|
2925 | write(6,'(i4,2(1x,g21.14))') 2*i-1,rr(2*i-1),ASIN(ri(2*i-1)/rd1)*twopii |
---|
2926 | write(6,'(i4,2(1x,g21.14))') 2*i,rr(2*i),ASIN(ri(2*i)/rd)*twopii |
---|
2927 | write(6,'(a8,g21.14)') ' alphas ', LOG(SQRT(rd*rd1)) |
---|
2928 | !CALL !WRITE_a |
---|
2929 | enddo |
---|
2930 | w_p=0 |
---|
2931 | w_p%nc=1 |
---|
2932 | w_p%fc='((1X,A120))' |
---|
2933 | ! write(w_p%c(1),'(a8,i4,a40)') ' select ',nd-ndc,' eigenplanes (odd integers <0 real axis)' |
---|
2934 | write(6,'(a8,i4,a40)') ' select ',nd-ndc,' eigenplanes (odd integers <0 real axis)' |
---|
2935 | !CALL !WRITE_a |
---|
2936 | call read(n,nd-ndc) |
---|
2937 | elseif(lielib_print(8)==-1) then |
---|
2938 | do i=1,nd-ndc |
---|
2939 | n(i)=nplane(i) |
---|
2940 | enddo |
---|
2941 | ! elseif(idpr.eq.-101.or.idpr.eq.-102) then |
---|
2942 | else ! new line |
---|
2943 | do i=1,nd-ndc |
---|
2944 | if(ri(2*i).ne.0.0_dp) then |
---|
2945 | n(i)=2*i-1 |
---|
2946 | else |
---|
2947 | n(i)=-2*i+1 |
---|
2948 | endif |
---|
2949 | enddo |
---|
2950 | ! else |
---|
2951 | ! do i=1,nd-ndc |
---|
2952 | ! n(i)=2*i-1 |
---|
2953 | ! enddo |
---|
2954 | endif |
---|
2955 | iunst=0 |
---|
2956 | do i=1,nd-ndc ! Frank NDC kept |
---|
2957 | if(n(i).lt.0) then |
---|
2958 | n(i)=-n(i) |
---|
2959 | st(i)=0.0_dp |
---|
2960 | iunst=1 |
---|
2961 | else |
---|
2962 | st(i)=1.0_dp |
---|
2963 | endif |
---|
2964 | x(i)=0.0_dp |
---|
2965 | xx(i)=1.0_dp |
---|
2966 | do j=1,nd-ndc |
---|
2967 | x(i)=vr(2*j-1,n(i))*vi(2*j,n(i))-vr(2*j,n(i))*vi(2*j-1,n(i))+x(i) |
---|
2968 | enddo |
---|
2969 | enddo |
---|
2970 | |
---|
2971 | do i=1,nd-ndc |
---|
2972 | if(x(i).lt.0.0_dp) xx(i)=-1.0_dp |
---|
2973 | x(i)=SQRT(abs(x(i))) |
---|
2974 | if(.not.courant_snyder) x(i)=1.0_dp |
---|
2975 | enddo |
---|
2976 | do i=1,nd2-ndc2 |
---|
2977 | do j=1,nd-ndc |
---|
2978 | if(st(j)+1e-3_dp.gt.1.0_dp) then |
---|
2979 | sai(2*j-1,i)=vr(i,n(j))*xx(j)/x(j) |
---|
2980 | sai(2*j,i)=vi(i,n(j))/x(j) |
---|
2981 | else |
---|
2982 | ax=vr(i,n(j))*xx(j)/x(j) |
---|
2983 | ap=vi(i,n(j))/x(j) |
---|
2984 | sai(2*j-1,i)=(ax+ap)/SQRT(2.0_dp) |
---|
2985 | sai(2*j,i)=(ap-ax)/SQRT(2.0_dp) |
---|
2986 | endif |
---|
2987 | enddo |
---|
2988 | enddo |
---|
2989 | ! if(idpr.eq.-101.or.idpr.eq.-102) then |
---|
2990 | if(lielib_print(7)/=-1) call movearou(sai) |
---|
2991 | ! endif |
---|
2992 | ! adjust sa such that sa(1,2)=0 and sa(3,4)=zero (courant-snyder-edwards-teng |
---|
2993 | ! phase advances) |
---|
2994 | if(iunst.ne.1) then |
---|
2995 | do i=1,nd-ndc |
---|
2996 | p(i)=ATAN(-sai(2*i-1,2*i)/sai(2*i,2*i)) |
---|
2997 | s1(2*i-1,2*i-1)=COS(p(i)) |
---|
2998 | s1(2*i,2*i)=COS(p(i)) |
---|
2999 | s1(2*i-1,2*i)=SIN(p(i)) |
---|
3000 | s1(2*i,2*i-1)=-SIN(p(i)) |
---|
3001 | enddo |
---|
3002 | call mulnd2(s1,sai) |
---|
3003 | ! adjust sa to have sa(1,1)>0 and sa(3,3)>0 rotate by pi if necessary. |
---|
3004 | do i=1,nd-ndc |
---|
3005 | xd=1.0_dp |
---|
3006 | if(sai(2*i-1,2*i-1).lt.0.0_dp) xd=-1.0_dp |
---|
3007 | s1(2*i-1,2*i-1)=xd |
---|
3008 | s1(2*i-1,2*i)=0.0_dp |
---|
3009 | s1(2*i,2*i-1)=0.0_dp |
---|
3010 | s1(2*i,2*i)=xd |
---|
3011 | enddo |
---|
3012 | if(courant_snyder) call mulnd2(s1,sai) |
---|
3013 | ! sa is now uniquely and unambigeously determined. |
---|
3014 | endif |
---|
3015 | ! do i=1,nd2 |
---|
3016 | ! do l=1,nd2 |
---|
3017 | ! sa(i,l)=sai(i,l) |
---|
3018 | ! enddo |
---|
3019 | ! enddo |
---|
3020 | call matinv(sai,sa,nd2,ndim2,ier) |
---|
3021 | |
---|
3022 | call mulnd2(sai,cm) |
---|
3023 | do i=1,nd2 |
---|
3024 | do j=1,nd2 |
---|
3025 | cr(i,j)=sa(i,j) |
---|
3026 | enddo |
---|
3027 | enddo |
---|
3028 | |
---|
3029 | call mulnd2(cm,cr) |
---|
3030 | |
---|
3031 | return |
---|
3032 | end subroutine mapflol |
---|
3033 | |
---|
3034 | subroutine mulnd2(rt,r) |
---|
3035 | implicit none |
---|
3036 | integer i,ia,j |
---|
3037 | real(dp),dimension(ndim2,ndim2)::rt,r,rtt |
---|
3038 | if(.not.c_%stable_da) return |
---|
3039 | |
---|
3040 | do i=1,nd2 |
---|
3041 | do j=1,nd2 |
---|
3042 | rtt(i,j)=0.0_dp |
---|
3043 | enddo |
---|
3044 | enddo |
---|
3045 | do i=1,nd2 |
---|
3046 | do j=1,nd2 |
---|
3047 | do ia=1,nd2 |
---|
3048 | rtt(i,ia)=rt(i,j)*r(j,ia)+rtt(i,ia) |
---|
3049 | enddo |
---|
3050 | enddo |
---|
3051 | enddo |
---|
3052 | |
---|
3053 | do i=1,nd2 |
---|
3054 | do j=1,nd2 |
---|
3055 | r(i,j)=rtt(i,j) |
---|
3056 | enddo |
---|
3057 | enddo |
---|
3058 | return |
---|
3059 | end subroutine mulnd2 |
---|
3060 | |
---|
3061 | subroutine movearou(rt) |
---|
3062 | implicit none |
---|
3063 | ! integer ipause, mypause |
---|
3064 | integer i,ic,j |
---|
3065 | real(dp) xr,xrold |
---|
3066 | real(dp),dimension(ndim2,ndim2)::rt,rto,s |
---|
3067 | real(dp),dimension(ndim2,ndim2):: xt,yt,zt,xy,xz,yz |
---|
3068 | real(dp),dimension(ndim2,ndim2):: xyz,xzy,xyt,yxt,yzt,zyt,xzt,zxt |
---|
3069 | real(dp),dimension(ndim2,ndim2):: xyzt,xytz,xzyt,xzty,xtzy,xtyz |
---|
3070 | |
---|
3071 | if(.not.c_%stable_da) return |
---|
3072 | |
---|
3073 | do i=1,nd2 |
---|
3074 | do j=1,nd2 |
---|
3075 | s(i,j)=0.0_dp |
---|
3076 | s(i,i)=1.0_dp |
---|
3077 | enddo |
---|
3078 | enddo |
---|
3079 | xt=0.0_dp;yt=0.0_dp;zt=0.0_dp;xy=0.0_dp;xz=0.0_dp;yz=0.0_dp; |
---|
3080 | xyzt=0.0_dp;xytz=0.0_dp;xzyt=0.0_dp;xzty=0.0_dp;xtzy=0.0_dp;xtyz=0.0_dp; |
---|
3081 | xyz=0.0_dp;xzy=0.0_dp;xyt=0.0_dp;yxt=0.0_dp;yzt=0.0_dp;zyt=0.0_dp;xzt=0.0_dp;zxt=0.0_dp; |
---|
3082 | |
---|
3083 | do i=0,1 |
---|
3084 | |
---|
3085 | xy(1+i,3+i)=1.0_dp |
---|
3086 | xy(3+i,1+i)=1.0_dp |
---|
3087 | xy(5+i,5+i)=1.0_dp |
---|
3088 | xy(7+i,7+i)=1.0_dp |
---|
3089 | |
---|
3090 | xz(1+i,5+i)=1.0_dp |
---|
3091 | xz(5+i,1+i)=1.0_dp |
---|
3092 | xz(3+i,3+i)=1.0_dp |
---|
3093 | xz(7+i,7+i)=1.0_dp |
---|
3094 | |
---|
3095 | xt(1+i,7+i)=1.0_dp |
---|
3096 | xt(7+i,1+i)=1.0_dp |
---|
3097 | xt(3+i,3+i)=1.0_dp |
---|
3098 | xt(5+i,5+i)=1.0_dp |
---|
3099 | |
---|
3100 | yz(3+i,5+i)=1.0_dp |
---|
3101 | yz(5+i,3+i)=1.0_dp |
---|
3102 | yz(1+i,1+i)=1.0_dp |
---|
3103 | yz(7+i,7+i)=1.0_dp |
---|
3104 | |
---|
3105 | yt(3+i,7+i)=1.0_dp |
---|
3106 | yt(7+i,3+i)=1.0_dp |
---|
3107 | yt(1+i,1+i)=1.0_dp |
---|
3108 | yt(5+i,5+i)=1.0_dp |
---|
3109 | |
---|
3110 | zt(5+i,7+i)=1.0_dp |
---|
3111 | zt(7+i,5+i)=1.0_dp |
---|
3112 | zt(1+i,1+i)=1.0_dp |
---|
3113 | zt(3+i,3+i)=1.0_dp |
---|
3114 | |
---|
3115 | xyz(1+i,3+i)=1.0_dp |
---|
3116 | xyz(3+i,5+i)=1.0_dp |
---|
3117 | xyz(5+i,1+i)=1.0_dp |
---|
3118 | xyz(7+i,7+i)=1.0_dp |
---|
3119 | |
---|
3120 | xyz(1+i,3+i)=1.0_dp |
---|
3121 | xyz(3+i,5+i)=1.0_dp |
---|
3122 | xyz(5+i,1+i)=1.0_dp |
---|
3123 | xyz(7+i,7+i)=1.0_dp |
---|
3124 | |
---|
3125 | xzy(1+i,5+i)=1.0_dp |
---|
3126 | xzy(5+i,3+i)=1.0_dp |
---|
3127 | xzy(3+i,1+i)=1.0_dp |
---|
3128 | xzy(7+i,7+i)=1.0_dp |
---|
3129 | |
---|
3130 | xyt(1+i,3+i)=1.0_dp |
---|
3131 | xyt(3+i,7+i)=1.0_dp |
---|
3132 | xyt(7+i,1+i)=1.0_dp |
---|
3133 | xyt(5+i,5+i)=1.0_dp |
---|
3134 | |
---|
3135 | yxt(3+i,1+i)=1.0_dp |
---|
3136 | yxt(1+i,7+i)=1.0_dp |
---|
3137 | yxt(7+i,3+i)=1.0_dp |
---|
3138 | yxt(5+i,5+i)=1.0_dp |
---|
3139 | |
---|
3140 | yzt(3+i,5+i)=1.0_dp |
---|
3141 | yzt(5+i,7+i)=1.0_dp |
---|
3142 | yzt(7+i,3+i)=1.0_dp |
---|
3143 | yzt(1+i,1+i)=1.0_dp |
---|
3144 | |
---|
3145 | zyt(5+i,3+i)=1.0_dp |
---|
3146 | zyt(3+i,7+i)=1.0_dp |
---|
3147 | zyt(7+i,5+i)=1.0_dp |
---|
3148 | zyt(1+i,1+i)=1.0_dp |
---|
3149 | |
---|
3150 | xzt(1+i,5+i)=1.0_dp |
---|
3151 | xzt(5+i,7+i)=1.0_dp |
---|
3152 | xzt(7+i,1+i)=1.0_dp |
---|
3153 | xzt(3+i,3+i)=1.0_dp |
---|
3154 | |
---|
3155 | zxt(5+i,1+i)=1.0_dp |
---|
3156 | zxt(1+i,7+i)=1.0_dp |
---|
3157 | zxt(7+i,5+i)=1.0_dp |
---|
3158 | zxt(3+i,3+i)=1.0_dp |
---|
3159 | |
---|
3160 | xyzt(1+i,3+i)=1.0_dp |
---|
3161 | xyzt(3+i,5+i)=1.0_dp |
---|
3162 | xyzt(5+i,7+i)=1.0_dp |
---|
3163 | xyzt(7+i,1+i)=1.0_dp |
---|
3164 | |
---|
3165 | xytz(1+i,3+i)=1.0_dp |
---|
3166 | xytz(3+i,7+i)=1.0_dp |
---|
3167 | xytz(7+i,5+i)=1.0_dp |
---|
3168 | xytz(5+i,1+i)=1.0_dp |
---|
3169 | |
---|
3170 | xzyt(1+i,5+i)=1.0_dp |
---|
3171 | xzyt(5+i,3+i)=1.0_dp |
---|
3172 | xzyt(3+i,7+i)=1.0_dp |
---|
3173 | xzyt(7+i,1+i)=1.0_dp |
---|
3174 | |
---|
3175 | xzty(1+i,5+i)=1.0_dp |
---|
3176 | xzty(5+i,7+i)=1.0_dp |
---|
3177 | xzty(7+i,3+i)=1.0_dp |
---|
3178 | xzty(3+i,1+i)=1.0_dp |
---|
3179 | |
---|
3180 | xtzy(1+i,7+i)=1.0_dp |
---|
3181 | xtzy(7+i,5+i)=1.0_dp |
---|
3182 | xtzy(5+i,3+i)=1.0_dp |
---|
3183 | xtzy(3+i,1+i)=1.0_dp |
---|
3184 | |
---|
3185 | xtyz(1+i,7+i)=1.0_dp |
---|
3186 | xtyz(7+i,3+i)=1.0_dp |
---|
3187 | xtyz(3+i,5+i)=1.0_dp |
---|
3188 | xtyz(5+i,1+i)=1.0_dp |
---|
3189 | enddo |
---|
3190 | |
---|
3191 | !do i=1,8 |
---|
3192 | !write(6,100) (rt(i,j),j=1,8) |
---|
3193 | !enddo |
---|
3194 | !write(6,*) " " |
---|
3195 | !100 FORMAT(8(1x, E12.6)) |
---|
3196 | |
---|
3197 | ic=0 |
---|
3198 | xrold=1e9_dp |
---|
3199 | call movemul(rt,s,rto,xr) |
---|
3200 | |
---|
3201 | if(xr.lt.xrold) then |
---|
3202 | xrold=xr |
---|
3203 | endif |
---|
3204 | |
---|
3205 | if(nd>=2) then |
---|
3206 | call movemul(rt,xy,rto,xr) |
---|
3207 | if(xr.lt.xrold) then |
---|
3208 | xrold=xr |
---|
3209 | ic=1 |
---|
3210 | endif |
---|
3211 | endif |
---|
3212 | |
---|
3213 | if(nd>=3) then |
---|
3214 | call movemul(rt,xz,rto,xr) |
---|
3215 | if(xr.lt.xrold) then |
---|
3216 | xrold=xr |
---|
3217 | ic=2 |
---|
3218 | endif |
---|
3219 | call movemul(rt,yz,rto,xr) |
---|
3220 | if(xr.lt.xrold) then |
---|
3221 | xrold=xr |
---|
3222 | ic=3 |
---|
3223 | endif |
---|
3224 | call movemul(rt,xyz,rto,xr) |
---|
3225 | if(xr.lt.xrold) then |
---|
3226 | xrold=xr |
---|
3227 | ic=4 |
---|
3228 | endif |
---|
3229 | call movemul(rt,xzy,rto,xr) |
---|
3230 | if(xr.lt.xrold) then |
---|
3231 | xrold=xr |
---|
3232 | ic=5 |
---|
3233 | endif |
---|
3234 | endif |
---|
3235 | |
---|
3236 | if(nd.eq.4) then |
---|
3237 | call movemul(rt,xt,rto,xr) |
---|
3238 | if(xr.lt.xrold) then |
---|
3239 | xrold=xr |
---|
3240 | ic=6 |
---|
3241 | endif |
---|
3242 | call movemul(rt,yt,rto,xr) |
---|
3243 | if(xr.lt.xrold) then |
---|
3244 | xrold=xr |
---|
3245 | ic=7 |
---|
3246 | endif |
---|
3247 | call movemul(rt,zt,rto,xr) |
---|
3248 | if(xr.lt.xrold) then |
---|
3249 | xrold=xr |
---|
3250 | ic=8 |
---|
3251 | endif |
---|
3252 | |
---|
3253 | call movemul(rt,xyt,rto,xr) |
---|
3254 | if(xr.lt.xrold) then |
---|
3255 | xrold=xr |
---|
3256 | ic=9 |
---|
3257 | endif |
---|
3258 | call movemul(rt,yxt,rto,xr) |
---|
3259 | if(xr.lt.xrold) then |
---|
3260 | xrold=xr |
---|
3261 | ic=10 |
---|
3262 | endif |
---|
3263 | call movemul(rt,yzt,rto,xr) |
---|
3264 | if(xr.lt.xrold) then |
---|
3265 | xrold=xr |
---|
3266 | ic=11 |
---|
3267 | |
---|
3268 | endif |
---|
3269 | call movemul(rt,zyt,rto,xr) |
---|
3270 | if(xr.lt.xrold) then |
---|
3271 | xrold=xr |
---|
3272 | ic=12 |
---|
3273 | endif |
---|
3274 | call movemul(rt,xzt,rto,xr) |
---|
3275 | if(xr.lt.xrold) then |
---|
3276 | xrold=xr |
---|
3277 | ic=13 |
---|
3278 | endif |
---|
3279 | call movemul(rt,zxt,rto,xr) |
---|
3280 | if(xr.lt.xrold) then |
---|
3281 | xrold=xr |
---|
3282 | ic=14 |
---|
3283 | endif |
---|
3284 | call movemul(rt,xyzt,rto,xr) |
---|
3285 | if(xr.lt.xrold) then |
---|
3286 | xrold=xr |
---|
3287 | ic=15 |
---|
3288 | endif |
---|
3289 | call movemul(rt,xytz,rto,xr) |
---|
3290 | if(xr.lt.xrold) then |
---|
3291 | xrold=xr |
---|
3292 | ic=16 |
---|
3293 | endif |
---|
3294 | call movemul(rt,xzyt,rto,xr) |
---|
3295 | if(xr.lt.xrold) then |
---|
3296 | xrold=xr |
---|
3297 | ic=17 |
---|
3298 | endif |
---|
3299 | call movemul(rt,xzty,rto,xr) |
---|
3300 | if(xr.lt.xrold) then |
---|
3301 | xrold=xr |
---|
3302 | ic=18 |
---|
3303 | endif |
---|
3304 | call movemul(rt,xtzy,rto,xr) |
---|
3305 | if(xr.lt.xrold) then |
---|
3306 | xrold=xr |
---|
3307 | ic=19 |
---|
3308 | endif |
---|
3309 | call movemul(rt,xtyz,rto,xr) |
---|
3310 | if(xr.lt.xrold) then |
---|
3311 | xrold=xr |
---|
3312 | ic=20 |
---|
3313 | endif |
---|
3314 | endif |
---|
3315 | |
---|
3316 | |
---|
3317 | w_p=0 |
---|
3318 | i=0 |
---|
3319 | if(ic.eq.0) then |
---|
3320 | call movemul(rt,s,rto,xr) |
---|
3321 | i=i+1 |
---|
3322 | w_p%c(i)= " no exchanged" |
---|
3323 | elseif(ic.eq.1) then |
---|
3324 | call movemul(rt,xy,rto,xr) |
---|
3325 | i=i+1 |
---|
3326 | w_p%c(i)= " x-y exchanged" |
---|
3327 | elseif(ic.eq.2) then |
---|
3328 | call movemul(rt,xz,rto,xr) |
---|
3329 | i=i+1 |
---|
3330 | w_p%c(i)= " x-z exchanged" |
---|
3331 | elseif(ic.eq.3) then |
---|
3332 | call movemul(rt,yz,rto,xr) |
---|
3333 | i=i+1 |
---|
3334 | w_p%c(i)= " y-z exchanged" |
---|
3335 | elseif(ic.eq.4) then |
---|
3336 | call movemul(rt,xyz,rto,xr) |
---|
3337 | i=i+1 |
---|
3338 | w_p%c(i)= " x-y-z permuted" |
---|
3339 | elseif(ic.eq.5) then |
---|
3340 | call movemul(rt,xzy,rto,xr) |
---|
3341 | i=i+1 |
---|
3342 | w_p%c(i)= " x-z-y permuted" |
---|
3343 | elseif(ic.eq.6) then |
---|
3344 | call movemul(rt,xt,rto,xr) |
---|
3345 | elseif(ic.eq.7) then |
---|
3346 | call movemul(rt,yt,rto,xr) |
---|
3347 | elseif(ic.eq.8) then |
---|
3348 | call movemul(rt,zt,rto,xr) |
---|
3349 | elseif(ic.eq.9) then |
---|
3350 | call movemul(rt,xyt,rto,xr) |
---|
3351 | elseif(ic.eq.10) then |
---|
3352 | call movemul(rt,yxt,rto,xr) |
---|
3353 | elseif(ic.eq.11) then |
---|
3354 | call movemul(rt,yzt,rto,xr) |
---|
3355 | elseif(ic.eq.12) then |
---|
3356 | call movemul(rt,zyt,rto,xr) |
---|
3357 | elseif(ic.eq.13) then |
---|
3358 | call movemul(rt,xzt,rto,xr) |
---|
3359 | elseif(ic.eq.14) then |
---|
3360 | call movemul(rt,zxt,rto,xr) |
---|
3361 | elseif(ic.eq.15) then |
---|
3362 | call movemul(rt,xyzt,rto,xr) |
---|
3363 | elseif(ic.eq.16) then |
---|
3364 | call movemul(rt,xytz,rto,xr) |
---|
3365 | elseif(ic.eq.17) then |
---|
3366 | call movemul(rt,xzyt,rto,xr) |
---|
3367 | elseif(ic.eq.18) then |
---|
3368 | call movemul(rt,xzty,rto,xr) |
---|
3369 | elseif(ic.eq.19) then |
---|
3370 | call movemul(rt,xtzy,rto,xr) |
---|
3371 | elseif(ic.eq.20) then |
---|
3372 | call movemul(rt,xtyz,rto,xr) |
---|
3373 | endif |
---|
3374 | |
---|
3375 | |
---|
3376 | |
---|
3377 | |
---|
3378 | |
---|
3379 | do i=1,nd2 |
---|
3380 | do j=1,nd2 |
---|
3381 | rt(i,j)=rto(i,j) |
---|
3382 | enddo |
---|
3383 | enddo |
---|
3384 | |
---|
3385 | return |
---|
3386 | end subroutine movearou |
---|
3387 | |
---|
3388 | subroutine movemul(rt,xy,rto,xr) |
---|
3389 | implicit none |
---|
3390 | integer i,j,k |
---|
3391 | real(dp) xr |
---|
3392 | real(dp),dimension(ndim2,ndim2)::rt,xy,rto |
---|
3393 | if(.not.c_%stable_da) return |
---|
3394 | |
---|
3395 | do i=1,nd2 |
---|
3396 | do j=1,nd2 |
---|
3397 | rto(i,j)=0.0_dp |
---|
3398 | enddo |
---|
3399 | enddo |
---|
3400 | |
---|
3401 | do i=1,nd2 |
---|
3402 | do j=1,nd2 |
---|
3403 | do k=1,nd2 |
---|
3404 | rto(i,k)=xy(i,j)*rt(j,k)+rto(i,k) |
---|
3405 | enddo |
---|
3406 | enddo |
---|
3407 | enddo |
---|
3408 | |
---|
3409 | xr=0.0_dp |
---|
3410 | do i=1,nd2 |
---|
3411 | do j=1,nd2 |
---|
3412 | xr=xr+abs(rto(i,j)) |
---|
3413 | enddo |
---|
3414 | enddo |
---|
3415 | do i=1,nd |
---|
3416 | xr=xr-abs(rto(2*i-1,2*i-1)) |
---|
3417 | xr=xr-abs(rto(2*i-1,2*i)) |
---|
3418 | xr=xr-abs(rto(2*i,2*i)) |
---|
3419 | xr=xr-abs(rto(2*i,2*i-1)) |
---|
3420 | enddo |
---|
3421 | return |
---|
3422 | end subroutine movemul |
---|
3423 | |
---|
3424 | subroutine initpert(st,ang,ra) |
---|
3425 | implicit none |
---|
3426 | ! X-RATED |
---|
3427 | !- SETS UP ALL THE !COMMON BLOCKS RELEVANT TO NORMAL FORM AND THE BASIS |
---|
3428 | !- CHANGES INSIDE MAPNORMF |
---|
3429 | integer i,nn,ipause,mypauses |
---|
3430 | real(dp),dimension(ndim)::ang,ra,st |
---|
3431 | if(.not.c_%stable_da) return |
---|
3432 | |
---|
3433 | if(iref.gt.0) then |
---|
3434 | w_p=0 |
---|
3435 | w_p%nc=1 |
---|
3436 | w_p%fc='((1X,A120))' |
---|
3437 | write(w_p%c(1),'(a19,i4)') " resonance in file ",iref |
---|
3438 | ! call ! WRITE_I |
---|
3439 | read(iref,*) nres |
---|
3440 | if(nres.ge.nreso) then |
---|
3441 | line= ' NRESO IN LIELIB TOO SMALL ' |
---|
3442 | ipause=mypauses(-999,line) |
---|
3443 | endif |
---|
3444 | elseif(iref.eq.0) then |
---|
3445 | nres=0 |
---|
3446 | endif |
---|
3447 | if(nres.ne.0.and.global_verbose) then |
---|
3448 | w_p=0 |
---|
3449 | w_p%nc=1 |
---|
3450 | w_p%fc='((1X,A120))' |
---|
3451 | w_p%c(1) =' warning resonances left in the map' |
---|
3452 | ! call ! WRITE_I |
---|
3453 | endif |
---|
3454 | if(iref.gt.0) then |
---|
3455 | do i=1,nres |
---|
3456 | read(iref,*) (mx(nn,i),nn=1,nd-ndc) |
---|
3457 | enddo |
---|
3458 | endif |
---|
3459 | do i=nres+1,nreso |
---|
3460 | do nn=1,ndim |
---|
3461 | mx(nn,i)=0 |
---|
3462 | enddo |
---|
3463 | enddo |
---|
3464 | ! frank/Etienne |
---|
3465 | do i=1,ndim |
---|
3466 | angle(i)=0.0_dp |
---|
3467 | rad(i)=0.0_dp |
---|
3468 | sta(i)=0.0_dp |
---|
3469 | dsta(i)=1.0_dp-sta(i) |
---|
3470 | ista(i)=0 |
---|
3471 | idsta(i)=0 |
---|
3472 | enddo |
---|
3473 | do i=1,nd ! Frank -ndc |
---|
3474 | angle(i)=ang(i) |
---|
3475 | rad(i)=ra(i) |
---|
3476 | sta(i)=st(i) |
---|
3477 | dsta(i)=1.0_dp-sta(i) |
---|
3478 | enddo |
---|
3479 | do i=1,nd |
---|
3480 | ista(i)=int(sta(i)+1e-2_dp) |
---|
3481 | idsta(i)=int(dsta(i)+1e-2_dp) |
---|
3482 | enddo |
---|
3483 | return |
---|
3484 | end subroutine initpert |
---|
3485 | |
---|
3486 | real(dp) function dlie(j) |
---|
3487 | implicit none |
---|
3488 | integer i |
---|
3489 | ! INTEGER J(NTT) |
---|
3490 | integer,dimension(:)::j |
---|
3491 | if(.not.c_%stable_da) return |
---|
3492 | |
---|
3493 | dlie=0.0_dp |
---|
3494 | do i=1,nd |
---|
3495 | dlie=REAL(j(2*i-1)+j(2*i),kind=DP)+dlie |
---|
3496 | enddo |
---|
3497 | dlie=dlie+1.0_dp |
---|
3498 | dlie=1.0_dp/dlie |
---|
3499 | return |
---|
3500 | end function dlie |
---|
3501 | |
---|
3502 | real(dp) function rext(j) ! no flip needed |
---|
3503 | implicit none |
---|
3504 | integer i,lie,mo |
---|
3505 | integer,dimension(:)::j |
---|
3506 | if(.not.c_%stable_da) return |
---|
3507 | |
---|
3508 | lie=0 |
---|
3509 | do i=1,nd-ndc |
---|
3510 | lie=ista(i)*j(2*i)+lie |
---|
3511 | enddo |
---|
3512 | mo=mod(lie,4)+1 |
---|
3513 | !frs |
---|
3514 | select case(mo) |
---|
3515 | case(1,4) |
---|
3516 | rext = 1.0_dp |
---|
3517 | case(2,3) |
---|
3518 | rext = -1.0_dp |
---|
3519 | end select |
---|
3520 | return |
---|
3521 | !frs goto(11,12,13,14),mo |
---|
3522 | ! 11 rext = one |
---|
3523 | ! return |
---|
3524 | ! 12 rext = -one |
---|
3525 | ! return |
---|
3526 | ! 13 rext = -one |
---|
3527 | ! return |
---|
3528 | ! 14 rext = one |
---|
3529 | ! return |
---|
3530 | !frs |
---|
3531 | end function rext |
---|
3532 | subroutine cpart(h,ch) ! no flip needed |
---|
3533 | implicit none |
---|
3534 | integer h,ch |
---|
3535 | if(.not.c_%stable_da) return |
---|
3536 | |
---|
3537 | call dacfu(h,rext,ch) |
---|
3538 | return |
---|
3539 | end subroutine cpart |
---|
3540 | |
---|
3541 | subroutine ctoi(f1,f2) ! no flip needed |
---|
3542 | implicit none |
---|
3543 | integer f1,f2,b1 |
---|
3544 | integer,dimension(ndim2)::x |
---|
3545 | if(.not.c_%stable_da) return |
---|
3546 | |
---|
3547 | call etall1(b1) |
---|
3548 | call etallnom(x,nd2) ! ,'X ') |
---|
3549 | |
---|
3550 | call cpart(f1,b1) |
---|
3551 | call etctr(x) |
---|
3552 | call trx(b1,f2,x) |
---|
3553 | call dadal(x,nd2) |
---|
3554 | call dadal1(b1) |
---|
3555 | return |
---|
3556 | end subroutine ctoi |
---|
3557 | |
---|
3558 | subroutine itoc(f1,f2) ! no flip needed |
---|
3559 | implicit none |
---|
3560 | integer f1,f2,b1 |
---|
3561 | integer,dimension(ndim2)::x |
---|
3562 | if(.not.c_%stable_da) return |
---|
3563 | |
---|
3564 | call etall1(b1) |
---|
3565 | call etallnom(x,nd2) ! ,'X ') |
---|
3566 | |
---|
3567 | call etrtc(x) |
---|
3568 | call trx(f1,b1,x) |
---|
3569 | call cpart(b1,f2) |
---|
3570 | call dadal(x,nd2) |
---|
3571 | call dadal1(b1) |
---|
3572 | return |
---|
3573 | end subroutine itoc |
---|
3574 | |
---|
3575 | subroutine etrtc(x) ! no flip needed |
---|
3576 | implicit none |
---|
3577 | integer i |
---|
3578 | integer,dimension(:)::x |
---|
3579 | integer,dimension(ndim2)::rel |
---|
3580 | if(.not.c_%stable_da) return |
---|
3581 | |
---|
3582 | call etallnom(rel,nd2) ! ,'REL ') |
---|
3583 | |
---|
3584 | call etini(rel) |
---|
3585 | call etini(x) |
---|
3586 | do i=1,nd-ndc |
---|
3587 | call daadd(rel(2*i-1),rel(2*i),x(2*i-1)) |
---|
3588 | call dasub(rel(2*i-1),rel(2*i),x(2*i)) |
---|
3589 | enddo |
---|
3590 | call dadal(rel,nd2) |
---|
3591 | return |
---|
3592 | end subroutine etrtc |
---|
3593 | |
---|
3594 | subroutine etctr(x) ! no flip needed |
---|
3595 | implicit none |
---|
3596 | integer i |
---|
3597 | integer,dimension(:)::x |
---|
3598 | integer,dimension(ndim2)::rel |
---|
3599 | if(.not.c_%stable_da) return |
---|
3600 | |
---|
3601 | call etallnom(rel,nd2) ! ,'REL ') |
---|
3602 | |
---|
3603 | call etini(rel) |
---|
3604 | call etini(x) |
---|
3605 | do i=1,nd-ndc |
---|
3606 | call dalin(rel(2*i-1),0.5_dp,rel(2*i),0.5_dp,x(2*i-1)) |
---|
3607 | call dalin(rel(2*i-1),0.5_dp,rel(2*i),-0.5_dp,x(2*i)) |
---|
3608 | enddo |
---|
3609 | call dadal(rel,nd2) |
---|
3610 | return |
---|
3611 | end subroutine etctr |
---|
3612 | |
---|
3613 | subroutine etcjg(x) ! no flip needed |
---|
3614 | implicit none |
---|
3615 | integer i |
---|
3616 | integer,dimension(:)::x |
---|
3617 | integer,dimension(ndim2)::rel |
---|
3618 | if(.not.c_%stable_da) return |
---|
3619 | |
---|
3620 | call etallnom(rel,nd2) ! ,'REL ') |
---|
3621 | |
---|
3622 | call etini(rel) |
---|
3623 | call etini(x) |
---|
3624 | do i=1,nd-ndc |
---|
3625 | if(ista(i).eq.1) then |
---|
3626 | call dacop(rel(2*i-1),x(2*i)) |
---|
3627 | call dacop(rel(2*i),x(2*i-1)) |
---|
3628 | else |
---|
3629 | call dacop(rel(2*i-1),x(2*i-1)) |
---|
3630 | call dacop(rel(2*i),x(2*i)) |
---|
3631 | endif |
---|
3632 | enddo |
---|
3633 | call dadal(rel,nd2) |
---|
3634 | return |
---|
3635 | end subroutine etcjg |
---|
3636 | |
---|
3637 | ! Neri's Routine below |
---|
3638 | |
---|
3639 | subroutine eig6(fm,reval,aieval,revec,aievec) |
---|
3640 | implicit none |
---|
3641 | !************************************************************************** |
---|
3642 | |
---|
3643 | ! Diagonalization routines of NERI |
---|
3644 | |
---|
3645 | !ccccccccccccccccc |
---|
3646 | ! |
---|
3647 | ! this routine finds the eigenvalues and eigenvectors |
---|
3648 | ! of the full matrix fm. |
---|
3649 | ! the eigenvectors are normalized so that the real and |
---|
3650 | ! imaginary part of vectors 1, 3, and 5 have +1 antisymmetric |
---|
3651 | ! product: |
---|
3652 | ! revec1 J aivec1 = 1 ; revec3 J aivec3 = 1 ; |
---|
3653 | ! revec5 J aivec5 = one |
---|
3654 | ! the eigenvectors 2 ,4, and 6 have the opposite normalization. |
---|
3655 | ! written by F. Neri, Feb 26 1986. |
---|
3656 | ! |
---|
3657 | integer jet,nn,i,i1,ilo,ihi,mdim,info |
---|
3658 | real(dp),dimension(ndim2)::reval,aieval,ort |
---|
3659 | real(dp),dimension(ndim2,ndim2)::revec,aievec,fm,aa,vv |
---|
3660 | INTEGER IPAUSE,MYPAUSES |
---|
3661 | if(.not.c_%stable_da) return |
---|
3662 | |
---|
3663 | ! copy matrix to temporary storage (the matrix aa is destroyed) |
---|
3664 | do i=1,nd2-ndc2 |
---|
3665 | do i1=1,nd2-ndc2 |
---|
3666 | aa(i1,i) = fm(i1,i) |
---|
3667 | enddo |
---|
3668 | enddo |
---|
3669 | ilo = 1 |
---|
3670 | ihi = nd2-ndc2 |
---|
3671 | mdim = ndim2 |
---|
3672 | nn = nd2-ndc2 |
---|
3673 | ! compute eigenvalues and eigenvectors using double |
---|
3674 | ! precision Eispack routines: |
---|
3675 | call ety(mdim,nn,ilo,ihi,aa,ort) |
---|
3676 | call etyt(mdim,nn,ilo,ihi,aa,ort,vv) |
---|
3677 | call ety2(mdim,nn,ilo,ihi,aa,reval,aieval,vv,info) |
---|
3678 | if ( info .ne. 0 ) then |
---|
3679 | LINE= ' ERROR IN EIG6' |
---|
3680 | IPAUSE=MYPAUSES(0,LINE) |
---|
3681 | endif |
---|
3682 | ! call neigv(vv,pbkt) |
---|
3683 | do i=1,nd-ndc |
---|
3684 | do jet=1,nd2-ndc2 |
---|
3685 | revec(jet,2*i-1)=vv(jet,2*i-1) |
---|
3686 | revec(jet,2*i)=vv(jet,2*i-1) |
---|
3687 | aievec(jet,2*i-1)=vv(jet,2*i) |
---|
3688 | aievec(jet,2*i)=-vv(jet,2*i) |
---|
3689 | enddo |
---|
3690 | enddo |
---|
3691 | do i=1,nd2-ndc2 |
---|
3692 | if(abs(reval(i)**2+aieval(i)**2 -1.0_dp).gt.1e-10_dp) then |
---|
3693 | w_p=0 |
---|
3694 | w_p%nc=1 |
---|
3695 | w_p%fc='((1X,A120))' |
---|
3696 | w_p%c(1) =' EIG6: Eigenvalues off the unit circle!' |
---|
3697 | if(lielib_print(4)==1) then |
---|
3698 | !CALL !WRITE_a |
---|
3699 | write(6,*) sqrt(reval(i)**2+aieval(i)**2) |
---|
3700 | endif |
---|
3701 | endif |
---|
3702 | enddo |
---|
3703 | return |
---|
3704 | end subroutine eig6 |
---|
3705 | |
---|
3706 | subroutine ety(nm,n,low,igh,a,ort) |
---|
3707 | implicit none |
---|
3708 | ! |
---|
3709 | ! this subroutine is a translation of the algol procedure orthes, |
---|
3710 | ! num. math. 12, 349-368(1968) by martin and wilkinson. |
---|
3711 | ! handbook for auto. comp., vol.ii-linear algebra, 339-358(1971). |
---|
3712 | ! |
---|
3713 | ! given a real general matrix, this subroutine |
---|
3714 | ! reduces a submatrix situated in rows and columns |
---|
3715 | ! low through igh to upper hessenberg form by |
---|
3716 | ! orthogonal similarity transformations. |
---|
3717 | ! |
---|
3718 | ! on input- |
---|
3719 | ! |
---|
3720 | ! nm must be set to the row dimension of two-dimensional |
---|
3721 | ! array parameters as declared in the calling program |
---|
3722 | ! dimension statement, |
---|
3723 | ! |
---|
3724 | ! n is the order of the matrix, |
---|
3725 | ! |
---|
3726 | ! low and igh are integers determined by the balancing |
---|
3727 | ! subroutine balanc. if balanc has not been used, |
---|
3728 | ! set low=1, igh=n, |
---|
3729 | ! |
---|
3730 | ! a contains the input matrix. |
---|
3731 | ! |
---|
3732 | ! on output- |
---|
3733 | ! |
---|
3734 | ! a contains the hessenberg matrix. information about |
---|
3735 | ! the orthogonal transformations used in the reduction |
---|
3736 | ! is stored in the remaining triangle under the |
---|
3737 | ! hessenberg matrix, |
---|
3738 | ! |
---|
3739 | ! ort contains further information about the transformations. |
---|
3740 | ! only elements low through igh are used. |
---|
3741 | ! |
---|
3742 | ! fortran routine by b. s. garbow |
---|
3743 | ! modified by filippo neri. |
---|
3744 | ! |
---|
3745 | ! |
---|
3746 | integer i,j,m,n,ii,jj,la,mp,nm,igh,kp1,low |
---|
3747 | real(dp),dimension(nm,n)::a |
---|
3748 | real(dp),dimension(igh)::ort |
---|
3749 | real(dp) f,g,h,scale |
---|
3750 | if(.not.c_%stable_da) return |
---|
3751 | |
---|
3752 | la = igh - 1 |
---|
3753 | kp1 = low + 1 |
---|
3754 | if (la .lt. kp1) go to 200 |
---|
3755 | ! |
---|
3756 | do m = kp1, la |
---|
3757 | h = 0.0_dp |
---|
3758 | ort(m) = 0.0_dp |
---|
3759 | scale = 0.0_dp |
---|
3760 | ! ********** scale column (algol tol then not needed) ********** |
---|
3761 | do i = m, igh |
---|
3762 | scale = scale + abs(a(i,m-1)) |
---|
3763 | enddo |
---|
3764 | ! |
---|
3765 | if (scale .eq. 0.0_dp) go to 180 |
---|
3766 | mp = m + igh |
---|
3767 | ! ********** for i=igh step -1 until m do -- ********** |
---|
3768 | do ii = m, igh |
---|
3769 | i = mp - ii |
---|
3770 | ort(i) = a(i,m-1) / scale |
---|
3771 | h = h + ort(i) * ort(i) |
---|
3772 | enddo |
---|
3773 | ! |
---|
3774 | g = -sign(SQRT(h),ort(m)) |
---|
3775 | h = h - ort(m) * g |
---|
3776 | ort(m) = ort(m) - g |
---|
3777 | ! ********** form (i-(u*ut)/h) * a ********** |
---|
3778 | do j = m, n |
---|
3779 | f = 0.0_dp |
---|
3780 | ! ********** for i=igh step -1 until m do -- ********** |
---|
3781 | do ii = m, igh |
---|
3782 | i = mp - ii |
---|
3783 | f = f + ort(i) * a(i,j) |
---|
3784 | enddo |
---|
3785 | ! |
---|
3786 | f = f / h |
---|
3787 | ! |
---|
3788 | do i = m, igh |
---|
3789 | a(i,j) = a(i,j) - f * ort(i) |
---|
3790 | enddo |
---|
3791 | ! |
---|
3792 | enddo |
---|
3793 | ! ********** form (i-(u*ut)/h)*a*(i-(u*ut)/h) ********** |
---|
3794 | do i = 1, igh |
---|
3795 | f = 0.0_dp |
---|
3796 | ! ********** for j=igh step -1 until m do -- ********** |
---|
3797 | do jj = m, igh |
---|
3798 | j = mp - jj |
---|
3799 | f = f + ort(j) * a(i,j) |
---|
3800 | enddo |
---|
3801 | ! |
---|
3802 | f = f / h |
---|
3803 | ! |
---|
3804 | do j = m, igh |
---|
3805 | a(i,j) = a(i,j) - f * ort(j) |
---|
3806 | enddo |
---|
3807 | ! |
---|
3808 | enddo |
---|
3809 | ! |
---|
3810 | ort(m) = scale * ort(m) |
---|
3811 | a(m,m-1) = scale * g |
---|
3812 | 180 continue |
---|
3813 | enddo |
---|
3814 | ! |
---|
3815 | 200 return |
---|
3816 | ! ********** last card of ety ********** |
---|
3817 | end subroutine ety |
---|
3818 | subroutine etyt(nm,n,low,igh,a,ort,z) |
---|
3819 | implicit none |
---|
3820 | ! |
---|
3821 | ! this subroutine is a translation of the algol procedure ortrans, |
---|
3822 | ! num. math. 16, 181-204(1970) by peters and wilkinson. |
---|
3823 | ! handbook for auto. comp., vol.ii-linear algebra, 372-395(1971). |
---|
3824 | ! |
---|
3825 | ! this subroutine accumulates the orthogonal similarity |
---|
3826 | ! transformations used in the reduction of a real general |
---|
3827 | ! matrix to upper hessenberg form by ety. |
---|
3828 | ! |
---|
3829 | ! on input- |
---|
3830 | ! |
---|
3831 | ! nm must be set to the row dimension of two-dimensional |
---|
3832 | ! array parameters as declared in the calling program |
---|
3833 | ! dimension statement, |
---|
3834 | ! |
---|
3835 | ! n is the order of the matrix, |
---|
3836 | ! |
---|
3837 | ! low and igh are integers determined by the balancing |
---|
3838 | ! subroutine balanc. if balanc has not been used, |
---|
3839 | ! set low=1, igh=n, |
---|
3840 | ! |
---|
3841 | ! a contains information about the orthogonal trans- |
---|
3842 | ! formations used in the reduction by orthes |
---|
3843 | ! in its strict lower triangle, |
---|
3844 | ! |
---|
3845 | ! ort contains further information about the trans- |
---|
3846 | ! formations used in the reduction by ety. |
---|
3847 | ! only elements low through igh are used. |
---|
3848 | ! |
---|
3849 | ! on output- |
---|
3850 | ! |
---|
3851 | ! z contains the transformation matrix produced in the |
---|
3852 | ! reduction by ety, |
---|
3853 | ! |
---|
3854 | ! ort has been altered. |
---|
3855 | ! |
---|
3856 | ! fortran routine by b. s. garbow. |
---|
3857 | ! modified by f. neri. |
---|
3858 | ! |
---|
3859 | ! |
---|
3860 | integer i,j,n,kl,mm,mp,nm,igh,low,mp1 |
---|
3861 | real(dp) g |
---|
3862 | real(dp),dimension(igh)::ort |
---|
3863 | real(dp),dimension(nm,igh)::a |
---|
3864 | real(dp),dimension(nm,n)::z |
---|
3865 | if(.not.c_%stable_da) return |
---|
3866 | |
---|
3867 | ! ********** initialize z to identity matrix ********** |
---|
3868 | do i = 1, n |
---|
3869 | ! |
---|
3870 | do j = 1, n |
---|
3871 | z(i,j) = 0.0_dp |
---|
3872 | enddo |
---|
3873 | ! |
---|
3874 | z(i,i) = 1.0_dp |
---|
3875 | enddo |
---|
3876 | ! |
---|
3877 | kl = igh - low - 1 |
---|
3878 | if (kl .lt. 1) go to 200 |
---|
3879 | ! ********** for mp=igh-1 step -1 until low+1 do -- ********** |
---|
3880 | do mm = 1, kl |
---|
3881 | mp = igh - mm |
---|
3882 | if (a(mp,mp-1) .eq. 0.0_dp) go to 140 |
---|
3883 | mp1 = mp + 1 |
---|
3884 | ! |
---|
3885 | do i = mp1, igh |
---|
3886 | ort(i) = a(i,mp-1) |
---|
3887 | enddo |
---|
3888 | ! |
---|
3889 | do j = mp, igh |
---|
3890 | g = 0.0_dp |
---|
3891 | ! |
---|
3892 | do i = mp, igh |
---|
3893 | g = g + ort(i) * z(i,j) |
---|
3894 | enddo |
---|
3895 | ! ********** divisor below is negative of h formed in orthes. |
---|
3896 | ! double division avoids possible underflow ********** |
---|
3897 | g = (g / ort(mp)) / a(mp,mp-1) |
---|
3898 | ! |
---|
3899 | do i = mp, igh |
---|
3900 | z(i,j) = z(i,j) + g * ort(i) |
---|
3901 | enddo |
---|
3902 | ! |
---|
3903 | enddo |
---|
3904 | ! |
---|
3905 | 140 continue |
---|
3906 | enddo |
---|
3907 | ! |
---|
3908 | 200 return |
---|
3909 | ! ********** last card of etyt ********** |
---|
3910 | end subroutine etyt |
---|
3911 | subroutine ety2(nm,n,low,igh,h,wr,wi,z,ierr) |
---|
3912 | implicit none |
---|
3913 | ! |
---|
3914 | ! |
---|
3915 | ! |
---|
3916 | ! this subroutine is a translation of the algol procedure hqr2, |
---|
3917 | ! num. math. 16, 181-204(1970) by peters and wilkinson. |
---|
3918 | ! handbook for auto. comp., vol.ii-linear algebra, 372-395(1971). |
---|
3919 | ! |
---|
3920 | ! this subroutine finds the eigenvalues and eigenvectors |
---|
3921 | ! of a real upper hessenberg matrix by the qr method. the |
---|
3922 | ! eigenvectors of a real general matrix can also be found |
---|
3923 | ! if elmhes and eltran or orthes and ortran have |
---|
3924 | ! been used to reduce this general matrix to hessenberg form |
---|
3925 | ! and to accumulate the similarity transformations. |
---|
3926 | ! |
---|
3927 | ! on input- |
---|
3928 | ! |
---|
3929 | ! nm must be set to the row dimension of two-dimensional |
---|
3930 | ! array parameters as declared in the calling program |
---|
3931 | ! dimension statement, |
---|
3932 | ! |
---|
3933 | ! n is the order of the matrix, |
---|
3934 | ! |
---|
3935 | ! low and igh are integers determined by the balancing |
---|
3936 | ! subroutine balanc. if balanc has not been used, |
---|
3937 | ! set low=1, igh=n, |
---|
3938 | ! |
---|
3939 | ! h contains the upper hessenberg matrix, |
---|
3940 | ! |
---|
3941 | ! z contains the transformation matrix produced by eltran |
---|
3942 | ! after the reduction by elmhes, or by ortran after the |
---|
3943 | ! reduction by orthes, if performed. if the eigenvectors |
---|
3944 | ! of the hessenberg matrix are desired, z must contain the |
---|
3945 | ! identity matrix. |
---|
3946 | ! |
---|
3947 | ! on output- |
---|
3948 | ! |
---|
3949 | ! h has been destroyed, |
---|
3950 | ! |
---|
3951 | ! wr and wi contain the real and imaginary parts, |
---|
3952 | ! respectively, of the eigenvalues. the eigenvalues |
---|
3953 | ! are unordered except that complex conjugate pairs |
---|
3954 | ! of values appear consecutively with the eigenvalue |
---|
3955 | ! having the positive imaginary part first. if an |
---|
3956 | ! error exit is made, the eigenvalues should be correct |
---|
3957 | ! for indices ierr+1,...,n, |
---|
3958 | ! |
---|
3959 | ! z contains the real and imaginary parts of the eigenvectors. |
---|
3960 | ! if the i-th eigenvalue is real, the i-th column of z |
---|
3961 | ! contains its eigenvector. if the i-th eigenvalue is complex |
---|
3962 | ! with positive imaginary part, the i-th and (i+1)-th |
---|
3963 | ! columns of z contain the real and imaginary parts of its |
---|
3964 | ! eigenvector. the eigenvectors are unnormalized. if an |
---|
3965 | ! error exit is made, none of the eigenvectors has been found, |
---|
3966 | ! |
---|
3967 | ! ierr is set to |
---|
3968 | ! zero for normal return, |
---|
3969 | ! j if the j-th eigenvalue has not been |
---|
3970 | ! determined after 200 iterations. |
---|
3971 | ! |
---|
3972 | ! arithmetic is real(dp). complex division |
---|
3973 | ! is simulated by routin etdiv. |
---|
3974 | ! |
---|
3975 | ! fortran routine by b. s. garbow. |
---|
3976 | ! modified by f. neri. |
---|
3977 | ! |
---|
3978 | ! |
---|
3979 | logical(lp) notlas |
---|
3980 | integer i,j,k,l,m,n,en,ii,jj,ll,mm,na,nm,nn,igh,its,low,mp2,enm2,ierr |
---|
3981 | real(dp) p,q,r,s,t,w,x,y,ra,sa,vi,vr,zz,norm,z3r,z3i |
---|
3982 | real(dp),dimension(n)::wr,wi |
---|
3983 | real(dp),dimension(nm,n)::h,z |
---|
3984 | if(.not.c_%stable_da) return |
---|
3985 | |
---|
3986 | ! ********** machep is a machine dependent parameter specifying |
---|
3987 | ! the relative precision of floating point arithmetic. |
---|
3988 | ! |
---|
3989 | ! ********** |
---|
3990 | ! machep = r1mach(4) |
---|
3991 | ! |
---|
3992 | ierr = 0 |
---|
3993 | norm = 0.0_dp |
---|
3994 | k = 1 |
---|
3995 | ! ********** store roots isolated by balanc |
---|
3996 | ! and compute matrix norm ********** |
---|
3997 | do i = 1, n |
---|
3998 | ! |
---|
3999 | do j = k, n |
---|
4000 | norm = norm + abs(h(i,j)) |
---|
4001 | enddo |
---|
4002 | ! |
---|
4003 | k = i |
---|
4004 | if (i .ge. low .and. i .le. igh) go to 50 |
---|
4005 | wr(i) = h(i,i) |
---|
4006 | wi(i) = 0.0_dp |
---|
4007 | 50 continue |
---|
4008 | enddo |
---|
4009 | ! |
---|
4010 | en = igh |
---|
4011 | t = 0.0_dp |
---|
4012 | ! ********** search for next eigenvalues ********** |
---|
4013 | 60 if (en .lt. low) go to 340 |
---|
4014 | its = 0 |
---|
4015 | na = en - 1 |
---|
4016 | enm2 = na - 1 |
---|
4017 | ! ********** look for single small sub-diagonal element |
---|
4018 | ! for l=en step -1 until low do -- ********** |
---|
4019 | 70 do ll = low, en |
---|
4020 | l = en + low - ll |
---|
4021 | if (l .eq. low) go to 100 |
---|
4022 | s = abs(h(l-1,l-1)) + abs(h(l,l)) |
---|
4023 | if (s .eq. 0.0_dp) s = norm |
---|
4024 | if (abs(h(l,l-1)) .le. machep * s) go to 100 |
---|
4025 | enddo |
---|
4026 | ! ********** form shift ********** |
---|
4027 | 100 x = h(en,en) |
---|
4028 | if (l .eq. en) go to 270 |
---|
4029 | y = h(na,na) |
---|
4030 | w = h(en,na) * h(na,en) |
---|
4031 | if (l .eq. na) go to 280 |
---|
4032 | if (its .eq. 200) go to 1000 |
---|
4033 | if (its .ne. 10 .and. its .ne. 20) go to 130 |
---|
4034 | ! ********** form exceptional shift ********** |
---|
4035 | t = t + x |
---|
4036 | ! |
---|
4037 | do i = low, en |
---|
4038 | h(i,i) = h(i,i) - x |
---|
4039 | enddo |
---|
4040 | ! |
---|
4041 | s = abs(h(en,na)) + abs(h(na,enm2)) |
---|
4042 | x = 0.75_dp * s |
---|
4043 | y = x |
---|
4044 | w = -0.4375_dp * s * s |
---|
4045 | 130 its = its + 1 |
---|
4046 | ! ********** look for two consecutive small |
---|
4047 | ! sub-diagonal elements. |
---|
4048 | ! for m=en-2 step -1 until l do -- ********** |
---|
4049 | do mm = l, enm2 |
---|
4050 | m = enm2 + l - mm |
---|
4051 | zz = h(m,m) |
---|
4052 | r = x - zz |
---|
4053 | s = y - zz |
---|
4054 | p = (r * s - w) / h(m+1,m) + h(m,m+1) |
---|
4055 | q = h(m+1,m+1) - zz - r - s |
---|
4056 | r = h(m+2,m+1) |
---|
4057 | s = abs(p) + abs(q) + abs(r) |
---|
4058 | p = p / s |
---|
4059 | q = q / s |
---|
4060 | r = r / s |
---|
4061 | if (m .eq. l) go to 150 |
---|
4062 | if (abs(h(m,m-1)) * (abs(q) + abs(r)) .le. machep * abs(p) * (abs(h(m-1,m-1)) + abs(zz) + abs(h(m+1,m+1)))) go to 150 |
---|
4063 | enddo |
---|
4064 | ! |
---|
4065 | 150 mp2 = m + 2 |
---|
4066 | ! |
---|
4067 | do i = mp2, en |
---|
4068 | h(i,i-2) = 0.0_dp |
---|
4069 | if (i .eq. mp2) go to 160 |
---|
4070 | h(i,i-3) = 0.0_dp |
---|
4071 | 160 continue |
---|
4072 | enddo |
---|
4073 | ! ********** double qr step involving rows l to en and |
---|
4074 | ! columns m to en ********** |
---|
4075 | do k = m, na |
---|
4076 | notlas = k .ne. na |
---|
4077 | if (k .eq. m) go to 170 |
---|
4078 | p = h(k,k-1) |
---|
4079 | q = h(k+1,k-1) |
---|
4080 | r = 0.0_dp |
---|
4081 | if (notlas) r = h(k+2,k-1) |
---|
4082 | x = abs(p) + abs(q) + abs(r) |
---|
4083 | if (x .eq. 0.0_dp) go to 260 |
---|
4084 | p = p / x |
---|
4085 | q = q / x |
---|
4086 | r = r / x |
---|
4087 | 170 s = sign(SQRT(p*p+q*q+r*r),p) |
---|
4088 | if (k .eq. m) go to 180 |
---|
4089 | h(k,k-1) = -s * x |
---|
4090 | go to 190 |
---|
4091 | 180 if (l .ne. m) h(k,k-1) = -h(k,k-1) |
---|
4092 | 190 p = p + s |
---|
4093 | x = p / s |
---|
4094 | y = q / s |
---|
4095 | zz = r / s |
---|
4096 | q = q / p |
---|
4097 | r = r / p |
---|
4098 | ! ********** row modification ********** |
---|
4099 | do j = k, n |
---|
4100 | p = h(k,j) + q * h(k+1,j) |
---|
4101 | if (.not. notlas) go to 200 |
---|
4102 | p = p + r * h(k+2,j) |
---|
4103 | h(k+2,j) = h(k+2,j) - p * zz |
---|
4104 | 200 h(k+1,j) = h(k+1,j) - p * y |
---|
4105 | h(k,j) = h(k,j) - p * x |
---|
4106 | enddo |
---|
4107 | ! |
---|
4108 | j = min0(en,k+3) |
---|
4109 | ! ********** column modification ********** |
---|
4110 | do i = 1, j |
---|
4111 | p = x * h(i,k) + y * h(i,k+1) |
---|
4112 | if (.not. notlas) go to 220 |
---|
4113 | p = p + zz * h(i,k+2) |
---|
4114 | h(i,k+2) = h(i,k+2) - p * r |
---|
4115 | 220 h(i,k+1) = h(i,k+1) - p * q |
---|
4116 | h(i,k) = h(i,k) - p |
---|
4117 | enddo |
---|
4118 | ! ********** accumulate transformations ********** |
---|
4119 | do i = low, igh |
---|
4120 | p = x * z(i,k) + y * z(i,k+1) |
---|
4121 | if (.not. notlas) go to 240 |
---|
4122 | p = p + zz * z(i,k+2) |
---|
4123 | z(i,k+2) = z(i,k+2) - p * r |
---|
4124 | 240 z(i,k+1) = z(i,k+1) - p * q |
---|
4125 | z(i,k) = z(i,k) - p |
---|
4126 | enddo |
---|
4127 | ! |
---|
4128 | 260 continue |
---|
4129 | enddo |
---|
4130 | ! |
---|
4131 | go to 70 |
---|
4132 | ! ********** one root found ********** |
---|
4133 | 270 h(en,en) = x + t |
---|
4134 | wr(en) = h(en,en) |
---|
4135 | wi(en) = 0.0_dp |
---|
4136 | en = na |
---|
4137 | go to 60 |
---|
4138 | ! ********** two roots found ********** |
---|
4139 | 280 p = (y - x) / 2.0_dp |
---|
4140 | q = p * p + w |
---|
4141 | zz = SQRT(abs(q)) |
---|
4142 | h(en,en) = x + t |
---|
4143 | x = h(en,en) |
---|
4144 | h(na,na) = y + t |
---|
4145 | if (q .lt. 0.0_dp) go to 320 |
---|
4146 | ! ********** real pair ********** |
---|
4147 | zz = p + sign(zz,p) |
---|
4148 | wr(na) = x + zz |
---|
4149 | wr(en) = wr(na) |
---|
4150 | if (zz .ne. 0.0_dp) wr(en) = x - w / zz |
---|
4151 | wi(na) = 0.0_dp |
---|
4152 | wi(en) = 0.0_dp |
---|
4153 | x = h(en,na) |
---|
4154 | s = abs(x) + abs(zz) |
---|
4155 | p = x / s |
---|
4156 | q = zz / s |
---|
4157 | r = SQRT(p*p+q*q) |
---|
4158 | p = p / r |
---|
4159 | q = q / r |
---|
4160 | ! ********** row modification ********** |
---|
4161 | do j = na, n |
---|
4162 | zz = h(na,j) |
---|
4163 | h(na,j) = q * zz + p * h(en,j) |
---|
4164 | h(en,j) = q * h(en,j) - p * zz |
---|
4165 | enddo |
---|
4166 | ! ********** column modification ********** |
---|
4167 | do i = 1, en |
---|
4168 | zz = h(i,na) |
---|
4169 | h(i,na) = q * zz + p * h(i,en) |
---|
4170 | h(i,en) = q * h(i,en) - p * zz |
---|
4171 | enddo |
---|
4172 | ! ********** accumulate transformations ********** |
---|
4173 | do i = low, igh |
---|
4174 | zz = z(i,na) |
---|
4175 | z(i,na) = q * zz + p * z(i,en) |
---|
4176 | z(i,en) = q * z(i,en) - p * zz |
---|
4177 | enddo |
---|
4178 | ! |
---|
4179 | go to 330 |
---|
4180 | ! ********** complex pair ********** |
---|
4181 | 320 wr(na) = x + p |
---|
4182 | wr(en) = x + p |
---|
4183 | wi(na) = zz |
---|
4184 | wi(en) = -zz |
---|
4185 | 330 en = enm2 |
---|
4186 | go to 60 |
---|
4187 | ! ********** all roots found. backsubstitute to find |
---|
4188 | ! vectors of upper triangular form ********** |
---|
4189 | 340 if (norm .eq. 0.0_dp) go to 1001 |
---|
4190 | ! ********** for en=n step -1 until 1 do -- ********** |
---|
4191 | do nn = 1, n |
---|
4192 | en = n + 1 - nn |
---|
4193 | p = wr(en) |
---|
4194 | q = wi(en) |
---|
4195 | na = en - 1 |
---|
4196 | if (q.lt.0) goto 710 |
---|
4197 | if (q.eq.0) goto 600 |
---|
4198 | if (q.gt.0) goto 800 |
---|
4199 | ! ********** real vector ********** |
---|
4200 | 600 m = en |
---|
4201 | h(en,en) = 1.0_dp |
---|
4202 | if (na .eq. 0) go to 800 |
---|
4203 | ! ********** for i=en-1 step -1 until 1 do -- ********** |
---|
4204 | do ii = 1, na |
---|
4205 | i = en - ii |
---|
4206 | w = h(i,i) - p |
---|
4207 | r = h(i,en) |
---|
4208 | if (m .gt. na) go to 620 |
---|
4209 | ! |
---|
4210 | do j = m, na |
---|
4211 | r = r + h(i,j) * h(j,en) |
---|
4212 | enddo |
---|
4213 | ! |
---|
4214 | 620 if (wi(i) .ge. 0.0_dp) go to 630 |
---|
4215 | zz = w |
---|
4216 | s = r |
---|
4217 | go to 700 |
---|
4218 | 630 m = i |
---|
4219 | if (wi(i) .ne. 0.0_dp) go to 640 |
---|
4220 | t = w |
---|
4221 | if (w .eq. 0.0_dp) t = machep * norm |
---|
4222 | h(i,en) = -r / t |
---|
4223 | go to 700 |
---|
4224 | ! ********** solve real equations ********** |
---|
4225 | 640 x = h(i,i+1) |
---|
4226 | y = h(i+1,i) |
---|
4227 | q = (wr(i) - p) * (wr(i) - p) + wi(i) * wi(i) |
---|
4228 | t = (x * s - zz * r) / q |
---|
4229 | h(i,en) = t |
---|
4230 | if (abs(x) .le. abs(zz)) go to 650 |
---|
4231 | h(i+1,en) = (-r - w * t) / x |
---|
4232 | go to 700 |
---|
4233 | 650 h(i+1,en) = (-s - y * t) / zz |
---|
4234 | 700 continue |
---|
4235 | enddo |
---|
4236 | ! ********** end real vector ********** |
---|
4237 | go to 800 |
---|
4238 | ! ********** complex vector ********** |
---|
4239 | 710 m = na |
---|
4240 | ! ********** last vector component chosen imaginary so that |
---|
4241 | ! eigenvector matrix is triangular ********** |
---|
4242 | if (abs(h(en,na)) .le. abs(h(na,en))) go to 720 |
---|
4243 | h(na,na) = q / h(en,na) |
---|
4244 | h(na,en) = -(h(en,en) - p) / h(en,na) |
---|
4245 | go to 730 |
---|
4246 | ! 720 z3 = cmplx(zero,-h(na,en)) / cmplx(h(na,na)-p,q) |
---|
4247 | ! h(na,na) = real(z3,kind=dp) |
---|
4248 | ! h(na,en) = aimag(z3) |
---|
4249 | 720 call etdiv(z3r,z3i,0.0_dp,-h(na,en),h(na,na)-p,q) |
---|
4250 | h(na,na) = z3r |
---|
4251 | h(na,en) = z3i |
---|
4252 | 730 h(en,na) = 0.0_dp |
---|
4253 | h(en,en) = 1.0_dp |
---|
4254 | enm2 = na - 1 |
---|
4255 | if (enm2 .eq. 0) go to 800 |
---|
4256 | ! ********** for i=en-2 step -1 until 1 do -- ********** |
---|
4257 | do ii = 1, enm2 |
---|
4258 | i = na - ii |
---|
4259 | w = h(i,i) - p |
---|
4260 | ra = 0.0_dp |
---|
4261 | sa = h(i,en) |
---|
4262 | ! |
---|
4263 | do j = m, na |
---|
4264 | ra = ra + h(i,j) * h(j,na) |
---|
4265 | sa = sa + h(i,j) * h(j,en) |
---|
4266 | enddo |
---|
4267 | ! |
---|
4268 | if (wi(i) .ge. 0.0_dp) go to 770 |
---|
4269 | zz = w |
---|
4270 | r = ra |
---|
4271 | s = sa |
---|
4272 | go to 790 |
---|
4273 | 770 m = i |
---|
4274 | if (wi(i) .ne. 0.0_dp) go to 780 |
---|
4275 | ! z3 = cmplx(-ra,-sa) / cmplx(w,q) |
---|
4276 | ! h(i,na) = real(z3,kind=dp) |
---|
4277 | ! h(i,en) = aimag(z3) |
---|
4278 | call etdiv(z3r,z3i,-ra,-sa,w,q) |
---|
4279 | h(i,na) = z3r |
---|
4280 | h(i,en) = z3i |
---|
4281 | go to 790 |
---|
4282 | ! ********** solve complex equations ********** |
---|
4283 | 780 x = h(i,i+1) |
---|
4284 | y = h(i+1,i) |
---|
4285 | vr = (wr(i) - p) * (wr(i) - p) + wi(i) * wi(i) - q * q |
---|
4286 | vi = (wr(i) - p) * 2.0_dp * q |
---|
4287 | if (vr .eq. 0.0_dp .and. vi .eq. 0.0_dp) vr = machep * norm * (abs(w) + abs(q) + abs(x) + abs(y) + abs(zz)) |
---|
4288 | ! z3 = cmplx(x*r-zz*ra+q*sa,x*s-zz*sa-q*ra) / cmplx(vr,vi) |
---|
4289 | ! h(i,na) = real(z3,kind=dp) |
---|
4290 | ! h(i,en) = aimag(z3) |
---|
4291 | call etdiv(z3r,z3i,x*r-zz*ra+q*sa,x*s-zz*sa-q*ra,vr,vi) |
---|
4292 | h(i,na) = z3r |
---|
4293 | h(i,en) = z3i |
---|
4294 | if (abs(x) .le. abs(zz) + abs(q)) go to 785 |
---|
4295 | h(i+1,na) = (-ra - w * h(i,na) + q * h(i,en)) / x |
---|
4296 | h(i+1,en) = (-sa - w * h(i,en) - q * h(i,na)) / x |
---|
4297 | go to 790 |
---|
4298 | ! 785 z3 = cmplx(-r-y*h(i,na),-s-y*h(i,en)) / cmplx(zz,q) |
---|
4299 | ! h(i+1,na) = real(z3,kind=dp) |
---|
4300 | ! h(i+1,en) = aimag(z3) |
---|
4301 | 785 call etdiv(z3r,z3i,-r-y*h(i,na),-s-y*h(i,en),zz,q) |
---|
4302 | h(i+1,na) = z3r |
---|
4303 | h(i+1,en) = z3i |
---|
4304 | 790 continue |
---|
4305 | enddo |
---|
4306 | ! ********** end complex vector ********** |
---|
4307 | 800 continue |
---|
4308 | enddo |
---|
4309 | ! ********** end back substitution. |
---|
4310 | ! vectors of isolated roots ********** |
---|
4311 | do i = 1, n |
---|
4312 | if (i .ge. low .and. i .le. igh) go to 840 |
---|
4313 | ! |
---|
4314 | do j = i, n |
---|
4315 | z(i,j) = h(i,j) |
---|
4316 | enddo |
---|
4317 | ! |
---|
4318 | 840 continue |
---|
4319 | enddo |
---|
4320 | ! ********** multiply by transformation matrix to give |
---|
4321 | ! vectors of original full matrix. |
---|
4322 | ! for j=n step -1 until low do -- ********** |
---|
4323 | do jj = low, n |
---|
4324 | j = n + low - jj |
---|
4325 | m = min0(j,igh) |
---|
4326 | ! |
---|
4327 | do i = low, igh |
---|
4328 | zz = 0.0_dp |
---|
4329 | ! |
---|
4330 | do k = low, m |
---|
4331 | zz = zz + z(i,k) * h(k,j) |
---|
4332 | enddo |
---|
4333 | ! |
---|
4334 | z(i,j) = zz |
---|
4335 | enddo |
---|
4336 | enddo |
---|
4337 | ! |
---|
4338 | go to 1001 |
---|
4339 | ! ********** set error -- no convergence to an |
---|
4340 | ! eigenvalue after 200 iterations ********** |
---|
4341 | 1000 ierr = en |
---|
4342 | 1001 return |
---|
4343 | ! ********** last card of ety2 ********** |
---|
4344 | end subroutine ety2 |
---|
4345 | subroutine etdiv(a,b,c,d,e,f) |
---|
4346 | implicit none |
---|
4347 | ! computes the complex division |
---|
4348 | ! a + ib = (c + id)/(e + if) |
---|
4349 | ! very slow, but tries to be as accurate as |
---|
4350 | ! possible by changing the order of the |
---|
4351 | ! operations, so to avoid under(over)flow |
---|
4352 | ! problems. |
---|
4353 | ! Written by F. Neri Feb. 12 1986 |
---|
4354 | ! |
---|
4355 | integer flip |
---|
4356 | real(dp) a,b,c,d,e,f,s,t,cc,dd,ee,ff,temp |
---|
4357 | if(.not.c_%stable_da) return |
---|
4358 | |
---|
4359 | flip = 0 |
---|
4360 | cc = c |
---|
4361 | dd = d |
---|
4362 | ee = e |
---|
4363 | ff = f |
---|
4364 | if( abs(f).ge.abs(e) ) then |
---|
4365 | ee = f |
---|
4366 | ff = e |
---|
4367 | cc = d |
---|
4368 | dd = c |
---|
4369 | flip = 1 |
---|
4370 | endif |
---|
4371 | s = 1.0_dp/ee |
---|
4372 | t = 1.0_dp/(ee+ ff*(ff*s)) |
---|
4373 | if ( abs(ff) .ge. abs(s) ) then |
---|
4374 | temp = ff |
---|
4375 | ff = s |
---|
4376 | s = temp |
---|
4377 | endif |
---|
4378 | if( abs(dd) .ge. abs(s) ) then |
---|
4379 | a = t*(cc + s*(dd*ff)) |
---|
4380 | else if ( abs(dd) .ge. abs(ff) ) then |
---|
4381 | a = t*(cc + dd*(s*ff)) |
---|
4382 | else |
---|
4383 | a = t*(cc + ff*(s*dd)) |
---|
4384 | endif |
---|
4385 | if ( abs(cc) .ge. abs(s)) then |
---|
4386 | b = t*(dd - s*(cc*ff)) |
---|
4387 | else if ( abs(cc) .ge. abs(ff)) then |
---|
4388 | b = t*(dd - cc*(s*ff)) |
---|
4389 | else |
---|
4390 | b = t*(dd - ff*(s*cc)) |
---|
4391 | endif |
---|
4392 | if (flip.ne.0 ) then |
---|
4393 | b = -b |
---|
4394 | endif |
---|
4395 | return |
---|
4396 | end subroutine etdiv |
---|
4397 | subroutine sympl3(m) |
---|
4398 | implicit none |
---|
4399 | !********************************************************** |
---|
4400 | ! |
---|
4401 | ! SYMPL3 |
---|
4402 | ! |
---|
4403 | ! |
---|
4404 | ! On return ,the matrix m(*,*), supposed to be almost |
---|
4405 | ! symplectic on entry is made exactly symplectic by |
---|
4406 | ! using a non iterative, constructive method. |
---|
4407 | ! |
---|
4408 | !********************************************************** |
---|
4409 | ! |
---|
4410 | ! Written by F. Neri Feb 7 1986 |
---|
4411 | ! |
---|
4412 | integer,parameter::n=3 |
---|
4413 | integer kp,kq,lp,lq,jp,jq,i |
---|
4414 | real(dp) qq,pq,qp,pp |
---|
4415 | real(dp),dimension(2*n,2*n)::m |
---|
4416 | if(.not.c_%stable_da) return |
---|
4417 | |
---|
4418 | ! |
---|
4419 | do kp=2,2*n,2 |
---|
4420 | kq = kp-1 |
---|
4421 | do lp=2,kp-2,2 |
---|
4422 | lq = lp-1 |
---|
4423 | qq = 0.0_dp |
---|
4424 | pq = 0.0_dp |
---|
4425 | qp = 0.0_dp |
---|
4426 | pp = 0.0_dp |
---|
4427 | do jp=2,2*n,2 |
---|
4428 | jq = jp-1 |
---|
4429 | qq = qq + m(lq,jq)*m(kq,jp) - m(lq,jp)*m(kq,jq) |
---|
4430 | pq = pq + m(lp,jq)*m(kq,jp) - m(lp,jp)*m(kq,jq) |
---|
4431 | qp = qp + m(lq,jq)*m(kp,jp) - m(lq,jp)*m(kp,jq) |
---|
4432 | pp = pp + m(lp,jq)*m(kp,jp) - m(lp,jp)*m(kp,jq) |
---|
4433 | enddo |
---|
4434 | |
---|
4435 | do i=1,2*n |
---|
4436 | m(kq,i) = m(kq,i) - qq*m(lp,i) + pq*m(lq,i) |
---|
4437 | m(kp,i) = m(kp,i) - qp*m(lp,i) + pp*m(lq,i) |
---|
4438 | enddo |
---|
4439 | enddo |
---|
4440 | qp = 0.0_dp |
---|
4441 | do jp=2,2*n,2 |
---|
4442 | jq = jp-1 |
---|
4443 | qp = qp + m(kq,jq)*m(kp,jp) - m(kq,jp)*m(kp,jq) |
---|
4444 | enddo |
---|
4445 | do i=1,2*n |
---|
4446 | m(kp,i) = m(kp,i)/qp |
---|
4447 | enddo |
---|
4448 | enddo |
---|
4449 | return |
---|
4450 | end subroutine sympl3 |
---|
4451 | |
---|
4452 | subroutine diagonalise_envelope_a(b,br,a,ai,kick) |
---|
4453 | implicit none |
---|
4454 | |
---|
4455 | |
---|
4456 | integer i,j |
---|
4457 | real(dp) a(6,6),ai(6,6),b(6,6) |
---|
4458 | |
---|
4459 | real(dp) xj(6,6),mj(6,6),xn,jb(6,6),kick(3),br(6,6) |
---|
4460 | |
---|
4461 | |
---|
4462 | |
---|
4463 | xj=0.0_dp |
---|
4464 | |
---|
4465 | do i=1,3 |
---|
4466 | xj(2*i,2*i-1)=-1.0_dp |
---|
4467 | xj(2*i-1,2*i)=1.0_dp |
---|
4468 | enddo |
---|
4469 | |
---|
4470 | |
---|
4471 | jb=matmul(xj,b) |
---|
4472 | xn=30.0_dp*mat_norm(jb) |
---|
4473 | jb=jb/xn |
---|
4474 | |
---|
4475 | ! mj=0.0_dp |
---|
4476 | ! xj=0.0_dp |
---|
4477 | ! do i=1,6 |
---|
4478 | ! xj(i,i)=one |
---|
4479 | ! enddo |
---|
4480 | |
---|
4481 | ! mj=xj+matmul(xj,jb) |
---|
4482 | |
---|
4483 | call mapflol6s(a,ai,br,jb) |
---|
4484 | |
---|
4485 | |
---|
4486 | do i=1,3 |
---|
4487 | kick(i)=sqrt(abs(br(2*i-1,2*i)*xn)) |
---|
4488 | enddo |
---|
4489 | |
---|
4490 | |
---|
4491 | |
---|
4492 | end subroutine diagonalise_envelope_a |
---|
4493 | |
---|
4494 | subroutine mapflol6s(sa,sai,cr,cm) |
---|
4495 | implicit none |
---|
4496 | !---- FROM TRACKING CODE |
---|
4497 | ! --------------------- |
---|
4498 | integer, parameter :: ndimt=3,ndimt2=6 |
---|
4499 | integer i,ier,iunst,j,l,n1,n(ndimt) |
---|
4500 | real(dp) ap,ax,rd,rd1,xd,xsu |
---|
4501 | real(dp),dimension(ndimt2,ndimt2)::cr,xj,sa,sai,cm,w,vr,vi,s1 |
---|
4502 | real(dp),dimension(ndimt)::x,xx,st |
---|
4503 | real(dp),dimension(ndimt2)::rr,ri,p |
---|
4504 | logical hyp |
---|
4505 | if(.not.c_%stable_da) return |
---|
4506 | |
---|
4507 | n1=0 |
---|
4508 | ! frank/etienne |
---|
4509 | do i=1,ndimt2 |
---|
4510 | do j=1,ndimt2 |
---|
4511 | cr(j,i)=cm(i,j) |
---|
4512 | enddo |
---|
4513 | enddo |
---|
4514 | xj=0.0_dp |
---|
4515 | s1=0.0_dp |
---|
4516 | |
---|
4517 | ! frank/etienne |
---|
4518 | do i=1,ndimt |
---|
4519 | n(i)=0 |
---|
4520 | xj(2*i-1,2*i)=1.0_dp |
---|
4521 | xj(2*i,2*i-1)=-1.0_dp |
---|
4522 | enddo |
---|
4523 | ! frank/etienne |
---|
4524 | |
---|
4525 | |
---|
4526 | sai=0.0_dp |
---|
4527 | w=cm |
---|
4528 | |
---|
4529 | w=matmul(xj,w) |
---|
4530 | w=matmul(cr,w) |
---|
4531 | |
---|
4532 | ! call mulnd2(xj,w) |
---|
4533 | ! call mulnd2(cr,w) |
---|
4534 | |
---|
4535 | call eig6s(cr,rr,ri,vr,vi) |
---|
4536 | |
---|
4537 | do i=1,6 |
---|
4538 | write(6,*) rr(i),ri(i) |
---|
4539 | enddo |
---|
4540 | |
---|
4541 | |
---|
4542 | do i=1,ndimt |
---|
4543 | n(i)=2*i-1 |
---|
4544 | st(i)=1.0_dp |
---|
4545 | enddo |
---|
4546 | ! elseif(idpr.eq.-101.or.idpr.eq.-102) then |
---|
4547 | |
---|
4548 | iunst=0 |
---|
4549 | do i=1,ndimt ! Frank NDC kept |
---|
4550 | x(i)=0.0_dp |
---|
4551 | xx(i)=1.0_dp |
---|
4552 | do j=1,ndimt |
---|
4553 | x(i)=vr(2*j-1,n(i))*vi(2*j,n(i))-vr(2*j,n(i))*vi(2*j-1,n(i))+x(i) |
---|
4554 | enddo |
---|
4555 | enddo |
---|
4556 | |
---|
4557 | do i=1,ndimt |
---|
4558 | if(x(i).lt.0.0_dp) xx(i)=-1.0_dp |
---|
4559 | x(i)=SQRT(abs(x(i))) |
---|
4560 | enddo |
---|
4561 | do i=1,ndimt2 |
---|
4562 | do j=1,ndimt |
---|
4563 | sai(2*j-1,i)=vr(i,n(j))*xx(j)/x(j) |
---|
4564 | sai(2*j,i)=vi(i,n(j))/x(j) |
---|
4565 | enddo |
---|
4566 | enddo |
---|
4567 | ! if(idpr.eq.-101.or.idpr.eq.-102) then |
---|
4568 | call movearous(sai) |
---|
4569 | ! endif |
---|
4570 | ! adjust sa such that sa(1,2)=0 and sa(3,4)=zero (courant-snyder-edwards-teng |
---|
4571 | ! phase advances) |
---|
4572 | ! sa=sai |
---|
4573 | |
---|
4574 | |
---|
4575 | call matinv(sai,sa,ndimt2,ndimt2,ier) |
---|
4576 | |
---|
4577 | ! call mulnd2(sai,cm) |
---|
4578 | cm=matmul(sai,cm) |
---|
4579 | cr=sa |
---|
4580 | |
---|
4581 | |
---|
4582 | cr=matmul(cm,cr) |
---|
4583 | |
---|
4584 | ! call mulnd2(cm,cr) |
---|
4585 | |
---|
4586 | return |
---|
4587 | end subroutine mapflol6s |
---|
4588 | |
---|
4589 | subroutine eig6s(fm,reval,aieval,revec,aievec) |
---|
4590 | implicit none |
---|
4591 | !************************************************************************** |
---|
4592 | |
---|
4593 | ! Diagonalization routines of NERI |
---|
4594 | |
---|
4595 | !ccccccccccccccccc |
---|
4596 | ! |
---|
4597 | ! this routine finds the eigenvalues and eigenvectors |
---|
4598 | ! of the full matrix fm. |
---|
4599 | ! the eigenvectors are normalized so that the real and |
---|
4600 | ! imaginary part of vectors 1, 3, and 5 have +1 antisymmetric |
---|
4601 | ! product: |
---|
4602 | ! revec1 J aivec1 = 1 ; revec3 J aivec3 = 1 ; |
---|
4603 | ! revec5 J aivec5 = one |
---|
4604 | ! the eigenvectors 2 ,4, and 6 have the opposite normalization. |
---|
4605 | ! written by F. Neri, Feb 26 1986. |
---|
4606 | ! |
---|
4607 | integer, parameter :: ndimt=3,ndimt2=6 |
---|
4608 | integer jet,nn,i,i1,ilo,ihi,mdim,info |
---|
4609 | real(dp),dimension(ndimt2)::reval,aieval,ort |
---|
4610 | real(dp),dimension(ndimt2,ndimt2)::revec,aievec,fm,aa,vv |
---|
4611 | INTEGER IPAUSE,MYPAUSES |
---|
4612 | if(.not.c_%stable_da) return |
---|
4613 | |
---|
4614 | ! copy matrix to temporary storage (the matrix aa is destroyed) |
---|
4615 | do i=1,ndimt2 |
---|
4616 | do i1=1,ndimt2 |
---|
4617 | aa(i1,i) = fm(i1,i) |
---|
4618 | enddo |
---|
4619 | enddo |
---|
4620 | ilo = 1 |
---|
4621 | ihi = ndimt2 |
---|
4622 | mdim = ndimt2 |
---|
4623 | nn = ndimt2 |
---|
4624 | ! compute eigenvalues and eigenvectors using double |
---|
4625 | ! precision Eispack routines: |
---|
4626 | call ety(mdim,nn,ilo,ihi,aa,ort) |
---|
4627 | call etyt(mdim,nn,ilo,ihi,aa,ort,vv) |
---|
4628 | call ety2(mdim,nn,ilo,ihi,aa,reval,aieval,vv,info) |
---|
4629 | |
---|
4630 | |
---|
4631 | ! call neigv(vv,pbkt) |
---|
4632 | do i=1,ndimt |
---|
4633 | do jet=1,ndimt2 |
---|
4634 | revec(jet,2*i-1)=vv(jet,2*i-1) |
---|
4635 | revec(jet,2*i)=vv(jet,2*i-1) |
---|
4636 | aievec(jet,2*i-1)=vv(jet,2*i) |
---|
4637 | aievec(jet,2*i)=-vv(jet,2*i) |
---|
4638 | enddo |
---|
4639 | enddo |
---|
4640 | do i=1,ndimt2 |
---|
4641 | if(abs(reval(i)**2+aieval(i)**2 -1.0_dp).gt.1e-10_dp) then |
---|
4642 | w_p=0 |
---|
4643 | w_p%nc=1 |
---|
4644 | w_p%fc='((1X,A120))' |
---|
4645 | w_p%c(1) =' EIG6: Eigenvalues off the unit circle!' |
---|
4646 | if(lielib_print(4)==1) then |
---|
4647 | !CALL !WRITE_a |
---|
4648 | write(6,*) sqrt(reval(i)**2+aieval(i)**2) |
---|
4649 | endif |
---|
4650 | endif |
---|
4651 | enddo |
---|
4652 | return |
---|
4653 | end subroutine eig6s |
---|
4654 | |
---|
4655 | subroutine movearous(rt) |
---|
4656 | implicit none |
---|
4657 | ! integer ipause, mypause |
---|
4658 | integer, parameter :: ndimt=3,ndimt2=6 |
---|
4659 | integer i,ic,j |
---|
4660 | real(dp) xr,xrold |
---|
4661 | real(dp),dimension(ndimt2,ndimt2)::rt,rto,s |
---|
4662 | real(dp),dimension(ndimt2,ndimt2):: xt,yt,zt,xy,xz,yz |
---|
4663 | real(dp),dimension(ndimt2,ndimt2):: xyz,xzy,xyt,yxt,yzt,zyt,xzt,zxt |
---|
4664 | real(dp),dimension(ndimt2,ndimt2):: xyzt,xytz,xzyt,xzty,xtzy,xtyz |
---|
4665 | |
---|
4666 | if(.not.c_%stable_da) return |
---|
4667 | |
---|
4668 | do i=1,ndimt2 |
---|
4669 | do j=1,ndimt2 |
---|
4670 | s(i,j)=0.0_dp |
---|
4671 | s(i,i)=1.0_dp |
---|
4672 | enddo |
---|
4673 | enddo |
---|
4674 | xt=0.0_dp;yt=0.0_dp;zt=0.0_dp;xy=0.0_dp;xz=0.0_dp;yz=0.0_dp; |
---|
4675 | xyzt=0.0_dp;xytz=0.0_dp;xzyt=0.0_dp;xzty=0.0_dp;xtzy=0.0_dp;xtyz=0.0_dp; |
---|
4676 | xyz=0.0_dp;xzy=0.0_dp;xyt=0.0_dp;yxt=0.0_dp;yzt=0.0_dp;zyt=0.0_dp;xzt=0.0_dp;zxt=0.0_dp; |
---|
4677 | |
---|
4678 | do i=0,1 |
---|
4679 | |
---|
4680 | xy(1+i,3+i)=1.0_dp |
---|
4681 | xy(3+i,1+i)=1.0_dp |
---|
4682 | xy(5+i,5+i)=1.0_dp |
---|
4683 | ! xy(7+i,7+i)=one |
---|
4684 | |
---|
4685 | xz(1+i,5+i)=1.0_dp |
---|
4686 | xz(5+i,1+i)=1.0_dp |
---|
4687 | xz(3+i,3+i)=1.0_dp |
---|
4688 | ! xz(7+i,7+i)=one |
---|
4689 | |
---|
4690 | ! xt(1+i,7+i)=one |
---|
4691 | ! xt(7+i,1+i)=one |
---|
4692 | xt(3+i,3+i)=1.0_dp |
---|
4693 | xt(5+i,5+i)=1.0_dp |
---|
4694 | |
---|
4695 | yz(3+i,5+i)=1.0_dp |
---|
4696 | yz(5+i,3+i)=1.0_dp |
---|
4697 | yz(1+i,1+i)=1.0_dp |
---|
4698 | ! yz(7+i,7+i)=one |
---|
4699 | |
---|
4700 | ! yt(3+i,7+i)=one |
---|
4701 | ! yt(7+i,3+i)=one |
---|
4702 | yt(1+i,1+i)=1.0_dp |
---|
4703 | yt(5+i,5+i)=1.0_dp |
---|
4704 | |
---|
4705 | ! zt(5+i,7+i)=one |
---|
4706 | ! zt(7+i,5+i)=one |
---|
4707 | zt(1+i,1+i)=1.0_dp |
---|
4708 | zt(3+i,3+i)=1.0_dp |
---|
4709 | |
---|
4710 | xyz(1+i,3+i)=1.0_dp |
---|
4711 | xyz(3+i,5+i)=1.0_dp |
---|
4712 | xyz(5+i,1+i)=1.0_dp |
---|
4713 | ! xyz(7+i,7+i)=one |
---|
4714 | |
---|
4715 | xyz(1+i,3+i)=1.0_dp |
---|
4716 | xyz(3+i,5+i)=1.0_dp |
---|
4717 | xyz(5+i,1+i)=1.0_dp |
---|
4718 | ! xyz(7+i,7+i)=one |
---|
4719 | |
---|
4720 | xzy(1+i,5+i)=1.0_dp |
---|
4721 | xzy(5+i,3+i)=1.0_dp |
---|
4722 | xzy(3+i,1+i)=1.0_dp |
---|
4723 | ! xzy(7+i,7+i)=one |
---|
4724 | |
---|
4725 | xyt(1+i,3+i)=1.0_dp |
---|
4726 | ! xyt(3+i,7+i)=one |
---|
4727 | ! xyt(7+i,1+i)=one |
---|
4728 | xyt(5+i,5+i)=1.0_dp |
---|
4729 | |
---|
4730 | yxt(3+i,1+i)=1.0_dp |
---|
4731 | ! yxt(1+i,7+i)=one |
---|
4732 | ! yxt(7+i,3+i)=one |
---|
4733 | yxt(5+i,5+i)=1.0_dp |
---|
4734 | |
---|
4735 | yzt(3+i,5+i)=1.0_dp |
---|
4736 | ! yzt(5+i,7+i)=one |
---|
4737 | ! yzt(7+i,3+i)=one |
---|
4738 | yzt(1+i,1+i)=1.0_dp |
---|
4739 | |
---|
4740 | zyt(5+i,3+i)=1.0_dp |
---|
4741 | ! zyt(3+i,7+i)=one |
---|
4742 | ! zyt(7+i,5+i)=one |
---|
4743 | zyt(1+i,1+i)=1.0_dp |
---|
4744 | |
---|
4745 | xzt(1+i,5+i)=1.0_dp |
---|
4746 | ! xzt(5+i,7+i)=one |
---|
4747 | ! xzt(7+i,1+i)=one |
---|
4748 | xzt(3+i,3+i)=1.0_dp |
---|
4749 | |
---|
4750 | zxt(5+i,1+i)=1.0_dp |
---|
4751 | ! zxt(1+i,7+i)=one |
---|
4752 | ! zxt(7+i,5+i)=one |
---|
4753 | zxt(3+i,3+i)=1.0_dp |
---|
4754 | |
---|
4755 | xyzt(1+i,3+i)=1.0_dp |
---|
4756 | xyzt(3+i,5+i)=1.0_dp |
---|
4757 | ! xyzt(5+i,7+i)=one |
---|
4758 | ! xyzt(7+i,1+i)=one |
---|
4759 | |
---|
4760 | xytz(1+i,3+i)=1.0_dp |
---|
4761 | ! xytz(3+i,7+i)=one |
---|
4762 | ! xytz(7+i,5+i)=one |
---|
4763 | xytz(5+i,1+i)=1.0_dp |
---|
4764 | |
---|
4765 | xzyt(1+i,5+i)=1.0_dp |
---|
4766 | xzyt(5+i,3+i)=1.0_dp |
---|
4767 | ! xzyt(3+i,7+i)=one |
---|
4768 | ! xzyt(7+i,1+i)=one |
---|
4769 | |
---|
4770 | xzty(1+i,5+i)=1.0_dp |
---|
4771 | ! xzty(5+i,7+i)=one |
---|
4772 | ! xzty(7+i,3+i)=one |
---|
4773 | xzty(3+i,1+i)=1.0_dp |
---|
4774 | |
---|
4775 | ! xtzy(1+i,7+i)=one |
---|
4776 | ! xtzy(7+i,5+i)=one |
---|
4777 | xtzy(5+i,3+i)=1.0_dp |
---|
4778 | xtzy(3+i,1+i)=1.0_dp |
---|
4779 | |
---|
4780 | ! xtyz(1+i,7+i)=one |
---|
4781 | ! xtyz(7+i,3+i)=one |
---|
4782 | xtyz(3+i,5+i)=1.0_dp |
---|
4783 | xtyz(5+i,1+i)=1.0_dp |
---|
4784 | enddo |
---|
4785 | |
---|
4786 | !do i=1,8 |
---|
4787 | !write(6,100) (rt(i,j),j=1,8) |
---|
4788 | !enddo |
---|
4789 | !write(6,*) " " |
---|
4790 | !100 FORMAT(8(1x, E12.6)) |
---|
4791 | |
---|
4792 | ic=0 |
---|
4793 | xrold=1e9_dp |
---|
4794 | call movemuls(rt,s,rto,xr) |
---|
4795 | |
---|
4796 | if(xr.lt.xrold) then |
---|
4797 | xrold=xr |
---|
4798 | endif |
---|
4799 | |
---|
4800 | if(ndimt>=2) then |
---|
4801 | call movemuls(rt,xy,rto,xr) |
---|
4802 | if(xr.lt.xrold) then |
---|
4803 | xrold=xr |
---|
4804 | ic=1 |
---|
4805 | endif |
---|
4806 | endif |
---|
4807 | |
---|
4808 | if(ndimt>=3) then |
---|
4809 | call movemuls(rt,xz,rto,xr) |
---|
4810 | if(xr.lt.xrold) then |
---|
4811 | xrold=xr |
---|
4812 | ic=2 |
---|
4813 | endif |
---|
4814 | call movemuls(rt,yz,rto,xr) |
---|
4815 | if(xr.lt.xrold) then |
---|
4816 | xrold=xr |
---|
4817 | ic=3 |
---|
4818 | endif |
---|
4819 | call movemuls(rt,xyz,rto,xr) |
---|
4820 | if(xr.lt.xrold) then |
---|
4821 | xrold=xr |
---|
4822 | ic=4 |
---|
4823 | endif |
---|
4824 | call movemuls(rt,xzy,rto,xr) |
---|
4825 | if(xr.lt.xrold) then |
---|
4826 | xrold=xr |
---|
4827 | ic=5 |
---|
4828 | endif |
---|
4829 | endif |
---|
4830 | |
---|
4831 | if(ndimt.eq.4) then |
---|
4832 | call movemuls(rt,xt,rto,xr) |
---|
4833 | if(xr.lt.xrold) then |
---|
4834 | xrold=xr |
---|
4835 | ic=6 |
---|
4836 | endif |
---|
4837 | call movemuls(rt,yt,rto,xr) |
---|
4838 | if(xr.lt.xrold) then |
---|
4839 | xrold=xr |
---|
4840 | ic=7 |
---|
4841 | endif |
---|
4842 | call movemuls(rt,zt,rto,xr) |
---|
4843 | if(xr.lt.xrold) then |
---|
4844 | xrold=xr |
---|
4845 | ic=8 |
---|
4846 | endif |
---|
4847 | |
---|
4848 | call movemuls(rt,xyt,rto,xr) |
---|
4849 | if(xr.lt.xrold) then |
---|
4850 | xrold=xr |
---|
4851 | ic=9 |
---|
4852 | endif |
---|
4853 | call movemuls(rt,yxt,rto,xr) |
---|
4854 | if(xr.lt.xrold) then |
---|
4855 | xrold=xr |
---|
4856 | ic=10 |
---|
4857 | endif |
---|
4858 | call movemuls(rt,yzt,rto,xr) |
---|
4859 | if(xr.lt.xrold) then |
---|
4860 | xrold=xr |
---|
4861 | ic=11 |
---|
4862 | |
---|
4863 | endif |
---|
4864 | call movemuls(rt,zyt,rto,xr) |
---|
4865 | if(xr.lt.xrold) then |
---|
4866 | xrold=xr |
---|
4867 | ic=12 |
---|
4868 | endif |
---|
4869 | call movemuls(rt,xzt,rto,xr) |
---|
4870 | if(xr.lt.xrold) then |
---|
4871 | xrold=xr |
---|
4872 | ic=13 |
---|
4873 | endif |
---|
4874 | call movemuls(rt,zxt,rto,xr) |
---|
4875 | if(xr.lt.xrold) then |
---|
4876 | xrold=xr |
---|
4877 | ic=14 |
---|
4878 | endif |
---|
4879 | call movemuls(rt,xyzt,rto,xr) |
---|
4880 | if(xr.lt.xrold) then |
---|
4881 | xrold=xr |
---|
4882 | ic=15 |
---|
4883 | endif |
---|
4884 | call movemuls(rt,xytz,rto,xr) |
---|
4885 | if(xr.lt.xrold) then |
---|
4886 | xrold=xr |
---|
4887 | ic=16 |
---|
4888 | endif |
---|
4889 | call movemuls(rt,xzyt,rto,xr) |
---|
4890 | if(xr.lt.xrold) then |
---|
4891 | xrold=xr |
---|
4892 | ic=17 |
---|
4893 | endif |
---|
4894 | call movemuls(rt,xzty,rto,xr) |
---|
4895 | if(xr.lt.xrold) then |
---|
4896 | xrold=xr |
---|
4897 | ic=18 |
---|
4898 | endif |
---|
4899 | call movemuls(rt,xtzy,rto,xr) |
---|
4900 | if(xr.lt.xrold) then |
---|
4901 | xrold=xr |
---|
4902 | ic=19 |
---|
4903 | endif |
---|
4904 | call movemuls(rt,xtyz,rto,xr) |
---|
4905 | if(xr.lt.xrold) then |
---|
4906 | xrold=xr |
---|
4907 | ic=20 |
---|
4908 | endif |
---|
4909 | endif |
---|
4910 | |
---|
4911 | |
---|
4912 | w_p=0 |
---|
4913 | i=0 |
---|
4914 | if(ic.eq.0) then |
---|
4915 | call movemuls(rt,s,rto,xr) |
---|
4916 | i=i+1 |
---|
4917 | w_p%c(i)= " no exchanged" |
---|
4918 | elseif(ic.eq.1) then |
---|
4919 | call movemuls(rt,xy,rto,xr) |
---|
4920 | i=i+1 |
---|
4921 | w_p%c(i)= " x-y exchanged" |
---|
4922 | elseif(ic.eq.2) then |
---|
4923 | call movemuls(rt,xz,rto,xr) |
---|
4924 | i=i+1 |
---|
4925 | w_p%c(i)= " x-z exchanged" |
---|
4926 | elseif(ic.eq.3) then |
---|
4927 | call movemuls(rt,yz,rto,xr) |
---|
4928 | i=i+1 |
---|
4929 | w_p%c(i)= " y-z exchanged" |
---|
4930 | elseif(ic.eq.4) then |
---|
4931 | call movemuls(rt,xyz,rto,xr) |
---|
4932 | i=i+1 |
---|
4933 | w_p%c(i)= " x-y-z permuted" |
---|
4934 | elseif(ic.eq.5) then |
---|
4935 | call movemuls(rt,xzy,rto,xr) |
---|
4936 | i=i+1 |
---|
4937 | w_p%c(i)= " x-z-y permuted" |
---|
4938 | elseif(ic.eq.6) then |
---|
4939 | call movemuls(rt,xt,rto,xr) |
---|
4940 | elseif(ic.eq.7) then |
---|
4941 | call movemuls(rt,yt,rto,xr) |
---|
4942 | elseif(ic.eq.8) then |
---|
4943 | call movemuls(rt,zt,rto,xr) |
---|
4944 | elseif(ic.eq.9) then |
---|
4945 | call movemuls(rt,xyt,rto,xr) |
---|
4946 | elseif(ic.eq.10) then |
---|
4947 | call movemuls(rt,yxt,rto,xr) |
---|
4948 | elseif(ic.eq.11) then |
---|
4949 | call movemuls(rt,yzt,rto,xr) |
---|
4950 | elseif(ic.eq.12) then |
---|
4951 | call movemuls(rt,zyt,rto,xr) |
---|
4952 | elseif(ic.eq.13) then |
---|
4953 | call movemuls(rt,xzt,rto,xr) |
---|
4954 | elseif(ic.eq.14) then |
---|
4955 | call movemuls(rt,zxt,rto,xr) |
---|
4956 | elseif(ic.eq.15) then |
---|
4957 | call movemuls(rt,xyzt,rto,xr) |
---|
4958 | elseif(ic.eq.16) then |
---|
4959 | call movemuls(rt,xytz,rto,xr) |
---|
4960 | elseif(ic.eq.17) then |
---|
4961 | call movemuls(rt,xzyt,rto,xr) |
---|
4962 | elseif(ic.eq.18) then |
---|
4963 | call movemuls(rt,xzty,rto,xr) |
---|
4964 | elseif(ic.eq.19) then |
---|
4965 | call movemuls(rt,xtzy,rto,xr) |
---|
4966 | elseif(ic.eq.20) then |
---|
4967 | call movemuls(rt,xtyz,rto,xr) |
---|
4968 | endif |
---|
4969 | |
---|
4970 | |
---|
4971 | |
---|
4972 | |
---|
4973 | |
---|
4974 | do i=1,ndimt2 |
---|
4975 | do j=1,ndimt2 |
---|
4976 | rt(i,j)=rto(i,j) |
---|
4977 | enddo |
---|
4978 | enddo |
---|
4979 | |
---|
4980 | return |
---|
4981 | end subroutine movearous |
---|
4982 | |
---|
4983 | subroutine movemuls(rt,xy,rto,xr) |
---|
4984 | implicit none |
---|
4985 | integer, parameter :: ndimt=3,ndimt2=6 |
---|
4986 | integer i,j,k |
---|
4987 | real(dp) xr |
---|
4988 | real(dp),dimension(ndimt2,ndimt2)::rt,xy,rto |
---|
4989 | if(.not.c_%stable_da) return |
---|
4990 | |
---|
4991 | do i=1,ndimt2 |
---|
4992 | do j=1,ndimt2 |
---|
4993 | rto(i,j)=0.0_dp |
---|
4994 | enddo |
---|
4995 | enddo |
---|
4996 | |
---|
4997 | do i=1,ndimt2 |
---|
4998 | do j=1,ndimt2 |
---|
4999 | do k=1,ndimt2 |
---|
5000 | rto(i,k)=xy(i,j)*rt(j,k)+rto(i,k) |
---|
5001 | enddo |
---|
5002 | enddo |
---|
5003 | enddo |
---|
5004 | |
---|
5005 | xr=0.0_dp |
---|
5006 | do i=1,ndimt2 |
---|
5007 | do j=1,ndimt2 |
---|
5008 | xr=xr+abs(rto(i,j)) |
---|
5009 | enddo |
---|
5010 | enddo |
---|
5011 | do i=1,ndimt |
---|
5012 | xr=xr-abs(rto(2*i-1,2*i-1)) |
---|
5013 | xr=xr-abs(rto(2*i-1,2*i)) |
---|
5014 | xr=xr-abs(rto(2*i,2*i)) |
---|
5015 | xr=xr-abs(rto(2*i,2*i-1)) |
---|
5016 | enddo |
---|
5017 | return |
---|
5018 | end subroutine movemuls |
---|
5019 | |
---|
5020 | end module lielib_yang_berz |
---|