/src/INCHI-1-SRC/INCHI_BASE/src/ichitaut.c
Line | Count | Source (jump to first uncovered line) |
1 | | /* |
2 | | * International Chemical Identifier (InChI) |
3 | | * Version 1 |
4 | | * Software version 1.06 |
5 | | * December 15, 2020 |
6 | | * |
7 | | * The InChI library and programs are free software developed under the |
8 | | * auspices of the International Union of Pure and Applied Chemistry (IUPAC). |
9 | | * Originally developed at NIST. |
10 | | * Modifications and additions by IUPAC and the InChI Trust. |
11 | | * Some portions of code were developed/changed by external contributors |
12 | | * (either contractor or volunteer) which are listed in the file |
13 | | * 'External-contributors' included in this distribution. |
14 | | * |
15 | | * IUPAC/InChI-Trust Licence No.1.0 for the |
16 | | * International Chemical Identifier (InChI) |
17 | | * Copyright (C) IUPAC and InChI Trust |
18 | | * |
19 | | * This library is free software; you can redistribute it and/or modify it |
20 | | * under the terms of the IUPAC/InChI Trust InChI Licence No.1.0, |
21 | | * or any later version. |
22 | | * |
23 | | * Please note that this library is distributed WITHOUT ANY WARRANTIES |
24 | | * whatsoever, whether expressed or implied. |
25 | | * See the IUPAC/InChI-Trust InChI Licence No.1.0 for more details. |
26 | | * |
27 | | * You should have received a copy of the IUPAC/InChI Trust InChI |
28 | | * Licence No. 1.0 with this library; if not, please e-mail: |
29 | | * |
30 | | * info@inchi-trust.org |
31 | | * |
32 | | */ |
33 | | |
34 | | |
35 | | #include <stdlib.h> |
36 | | #include <string.h> |
37 | | |
38 | | #include "mode.h" |
39 | | |
40 | | #include "ichitaut.h" |
41 | | #include "ichicomn.h" |
42 | | #include "ichitime.h" |
43 | | #include "ichicant.h" |
44 | | #include "util.h" |
45 | | |
46 | | /* Local prototypes */ |
47 | | |
48 | | int SetTautomericBonds( inp_ATOM *at, int nNumBondPos, T_BONDPOS *BondPos ); |
49 | | int CompRankTautomer( const void* a1, const void* a2, void * ); |
50 | | int RegisterEndPoints( CANON_GLOBALS *pCG, |
51 | | T_GROUP_INFO *t_group_info, |
52 | | /* T_GROUP *t_group, int *pnum_t, int max_num_t,*/ |
53 | | T_ENDPOINT *EndPoint, |
54 | | int nNumEndPoints, |
55 | | inp_ATOM *at, |
56 | | int num_atoms, |
57 | | C_GROUP_INFO *cgi, |
58 | | struct BalancedNetworkStructure *pBNS ); |
59 | | int cmpTGroupNumber( const void *a1, const void *a2 ); |
60 | | int comp_candidates( const void *a1, const void *a2 ); |
61 | | int MoveEndpoint( inp_ATOM *at, |
62 | | S_CANDIDATE *s_candidate, |
63 | | AT_NUMB endpoint, |
64 | | AT_NUMB *nTGroupNewNumbers, |
65 | | AT_NUMB *nTGroupPosition, |
66 | | int nNewTGroupOrd, |
67 | | T_GROUP_INFO *t_group_info ); |
68 | | int FindAccessibleEndPoints( CANON_GLOBALS *pCG, |
69 | | T_ENDPOINT *EndPoint, |
70 | | int *nNumEndPoints, |
71 | | T_BONDPOS *BondPos, |
72 | | int *nNumBondPos, |
73 | | struct BalancedNetworkStructure *pBNS, |
74 | | struct BalancedNetworkData *pBD, |
75 | | inp_ATOM *at, |
76 | | int num_atoms, |
77 | | C_GROUP_INFO *cgi, |
78 | | int taut_mode ); |
79 | | int GetChargeType( inp_ATOM *atom, int iat, S_CHAR *cChargeSubtype ); |
80 | | int GetNeutralRepsIfNeeded( AT_NUMB *pri, |
81 | | AT_NUMB *prj, |
82 | | inp_ATOM *at, |
83 | | int num_atoms, |
84 | | T_ENDPOINT *EndPoint, |
85 | | int nNumEndPoints, |
86 | | C_GROUP_INFO *cgi ); |
87 | | int bCanBeACPoint( inp_ATOM *at, |
88 | | S_CHAR cCharge, |
89 | | S_CHAR cChangeValence, |
90 | | S_CHAR neutral_bonds_valence, |
91 | | S_CHAR neutral_valence, |
92 | | S_CHAR nEndpointValence, S_CHAR *cChargeSubtype ); |
93 | | int CmpCCandidates( const void *a1, const void *a2 ); |
94 | | int RegisterCPoints( C_GROUP *c_group, |
95 | | int *pnum_c, |
96 | | int max_num_c, |
97 | | T_GROUP_INFO *t_group_info, |
98 | | int point1, |
99 | | int point2, |
100 | | int ctype, |
101 | | inp_ATOM *at, |
102 | | int num_atoms ); |
103 | | int GetSaltChargeType( inp_ATOM *at, |
104 | | int at_no, |
105 | | T_GROUP_INFO *t_group_info, |
106 | | int *s_subtype ); |
107 | | int GetOtherSaltChargeType( inp_ATOM *at, |
108 | | int at_no, |
109 | | T_GROUP_INFO *t_group_info, |
110 | | int *s_subtype, |
111 | | int bAccept_O ); |
112 | | int MergeSaltTautGroupsBlind( inp_ATOM *at, |
113 | | int s_type, int num_atoms, |
114 | | S_GROUP_INFO *s_group_info, |
115 | | int nNumCandidates, |
116 | | T_GROUP_INFO *t_group_info, |
117 | | C_GROUP_INFO *c_group_info, |
118 | | struct BalancedNetworkStructure *pBNS ); |
119 | | int ConnectSaltTGroups2SuperTGroup( inp_ATOM *at, |
120 | | int num_atoms, |
121 | | S_GROUP_INFO *s_group_info, |
122 | | int nNumCandidates, |
123 | | T_GROUP_INFO *t_group_info, |
124 | | C_GROUP_INFO *c_group_info, |
125 | | struct BalancedNetworkStructure *pBNS, |
126 | | int *nNewTGroupNumber, int *vertSuperTGroup ); |
127 | | int bDoNotMergeNonTautAtom( inp_ATOM *at, int at_no ); |
128 | | int GetOtherSaltType( inp_ATOM *at, int at_no, int *s_subtype ); |
129 | | |
130 | | /* Bits for GetChargeType */ |
131 | | #define C_SUBTYPE_CHARGED 0 |
132 | | #define C_SUBTYPE_p_DONOR 1 /* new */ |
133 | | #define C_SUBTYPE_p_ACCEPT 2 /* new */ |
134 | 0 | #define C_SUBTYPE_H_ACCEPT 4 |
135 | 0 | #define C_SUBTYPE_H_DONOR 8 |
136 | | #define C_SUBTYPE_NEUTRAL 16 |
137 | | |
138 | | /* Internal stack array size */ |
139 | 4 | #define MAX_STACK_ARRAY_LEN 127 |
140 | | #define MAX_TGROUP_ARRAY_LEN 127 |
141 | | |
142 | | |
143 | | /* |
144 | | is_centerpoint... functions |
145 | | */ |
146 | | |
147 | | |
148 | | /****************************************************************************/ |
149 | | int is_centerpoint_elem( U_CHAR el_number ) |
150 | 112k | { |
151 | 112k | static U_CHAR el_numb[12]; |
152 | 112k | static int len; |
153 | 112k | int len2; |
154 | 112k | int i; |
155 | 112k | if (!len) |
156 | 1 | { |
157 | 1 | len2 = 0; |
158 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "C" ); |
159 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "N" ); |
160 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "P" ); |
161 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "S" ); |
162 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "I" ); |
163 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "As" ); |
164 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "Sb" ); |
165 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "Se" ); |
166 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "Te" ); |
167 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "Cl" ); |
168 | 1 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "Br" ); |
169 | 1 | len = len2; |
170 | 1 | } |
171 | 1.34M | for (i = 0; i < len; i++) |
172 | 1.23M | { |
173 | 1.23M | if (el_numb[i] == el_number) |
174 | 1.63k | { |
175 | 1.63k | return 1; |
176 | 1.63k | } |
177 | 1.23M | } |
178 | | |
179 | 111k | return 0; |
180 | 112k | } |
181 | | |
182 | | |
183 | | #if ( KETO_ENOL_TAUT == 1 ) /* post v.1 feature */ |
184 | | |
185 | | |
186 | | /****************************************************************************/ |
187 | | int is_centerpoint_elem_KET( U_CHAR el_number ) |
188 | 0 | { |
189 | 0 | static U_CHAR el_numb[1]; |
190 | 0 | static int len; |
191 | 0 | int i; |
192 | 0 | if (!el_numb[0] && !len) |
193 | 0 | { |
194 | 0 | el_numb[len++] = (U_CHAR) get_periodic_table_number( "C" ); |
195 | 0 | } |
196 | 0 | for (i = 0; i < len; i++) |
197 | 0 | { |
198 | 0 | if (el_numb[i] == el_number) |
199 | 0 | { |
200 | 0 | return 1; |
201 | 0 | } |
202 | 0 | } |
203 | | |
204 | 0 | return 0; |
205 | 0 | } |
206 | | #endif |
207 | | |
208 | | |
209 | | /****************************************************************************/ |
210 | | int is_centerpoint_elem_strict( U_CHAR el_number ) |
211 | 0 | { |
212 | 0 | static U_CHAR el_numb[6]; |
213 | 0 | static int len; |
214 | 0 | int len2; |
215 | 0 | int i; |
216 | 0 | if (!len) |
217 | 0 | { |
218 | 0 | len2 = 0; |
219 | 0 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "C" ); |
220 | 0 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "N" ); |
221 | 0 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "P" ); |
222 | 0 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "As" ); |
223 | 0 | el_numb[len2++] = (U_CHAR) get_periodic_table_number( "Sb" ); |
224 | 0 | len = len2; |
225 | 0 | } |
226 | 0 | for (i = 0; i < len; i++) |
227 | 0 | { |
228 | 0 | if (el_numb[i] == el_number) |
229 | 0 | { |
230 | 0 | return 1; |
231 | 0 | } |
232 | 0 | } |
233 | 0 | return 0; |
234 | 0 | } |
235 | | |
236 | | |
237 | | /* |
238 | | AddAtom2... |
239 | | */ |
240 | | |
241 | | |
242 | | /****************************************************************************/ |
243 | | int AddAtom2num( AT_RANK num[], inp_ATOM *atom, int at_no, int bSubtract ) |
244 | 8 | { /* bSubtract: 0=> add, 1=>subtract, 2=> fill */ |
245 | 8 | inp_ATOM *at = atom + at_no; |
246 | 8 | int k; |
247 | 8 | int nMobile = ( at->charge == -1 ); |
248 | 8 | if (bSubtract == 1) |
249 | 0 | { |
250 | | /* 1: subtract */ |
251 | 0 | num[1] -= nMobile; |
252 | 0 | nMobile += at->num_H; |
253 | 0 | num[0] -= nMobile; |
254 | 0 | for (k = 0; k < T_NUM_ISOTOPIC; k++) |
255 | 0 | { |
256 | | /* T (3H isotope) first because it has higher weight */ |
257 | 0 | num[T_NUM_NO_ISOTOPIC + k] -= at->num_iso_H[NUM_H_ISOTOPES - k - 1]; |
258 | 0 | } |
259 | 0 | } |
260 | 8 | else |
261 | 8 | { |
262 | 8 | if (bSubtract == 2) |
263 | 8 | { |
264 | | /* fill */ |
265 | 8 | memset( num, 0, ( T_NUM_NO_ISOTOPIC + T_NUM_ISOTOPIC ) * sizeof( num[0] ) ); |
266 | 8 | } |
267 | | /* else (0): add */ |
268 | 8 | num[1] += nMobile; |
269 | 8 | nMobile += at->num_H; |
270 | 8 | num[0] += nMobile; |
271 | 32 | for (k = 0; k < T_NUM_ISOTOPIC; k++) |
272 | 24 | { |
273 | | /* T (3H isotope) first because it has higher weight */ |
274 | 24 | num[T_NUM_NO_ISOTOPIC + k] += at->num_iso_H[NUM_H_ISOTOPES - k - 1]; |
275 | 24 | } |
276 | 8 | } |
277 | 8 | return nMobile; |
278 | 8 | } |
279 | | |
280 | | |
281 | | /****************************************************************************/ |
282 | | void AddAtom2DA( AT_RANK num_DA[], inp_ATOM *atom, int at_no, int bSubtract ) |
283 | 8 | { /* bSubtract: 0=> add, 1=>subtract, 2=> fill */ |
284 | 8 | inp_ATOM *at = atom + at_no; |
285 | 8 | int nDelta, nAcidic_O; |
286 | | |
287 | 8 | if (at->charge < -1 || at->charge == 1 && !at->c_point || at->charge > 1) |
288 | 0 | { |
289 | 0 | return; |
290 | 0 | } |
291 | | |
292 | 8 | nDelta = ( bSubtract == 1 ) ? -1 : 1; |
293 | | |
294 | | /* "Acidic" O, S, Se, Te recognition */ |
295 | 8 | if (at->at_type & ATT_ACIDIC_CO) |
296 | 0 | { |
297 | 0 | nAcidic_O = nDelta; |
298 | 0 | } |
299 | 8 | else |
300 | 8 | { |
301 | 8 | nAcidic_O = 0; |
302 | 8 | } |
303 | | |
304 | 8 | if (bSubtract == 2) |
305 | 8 | { |
306 | | /* 2: fill, otherwise add */ |
307 | 8 | memset( num_DA, 0, TG_NUM_DA * sizeof( num_DA[0] ) ); |
308 | 8 | } |
309 | 8 | if (at->charge <= 0 && at->valence == at->chem_bonds_valence || |
310 | | /* neutral or negative donor */ |
311 | 8 | at->charge > 0 && at->valence + 1 == at->chem_bonds_valence |
312 | | /* positively charged donor */ |
313 | 8 | ) |
314 | 8 | { |
315 | 8 | if (at->charge < 0) |
316 | 4 | { |
317 | 4 | num_DA[TG_Num_dM] += nDelta; |
318 | 4 | num_DA[TG_Num_dO] += nAcidic_O; |
319 | 4 | } |
320 | 4 | else |
321 | 4 | { |
322 | 4 | if (at->num_H) |
323 | 4 | { |
324 | 4 | num_DA[TG_Num_dH] += nDelta; |
325 | 4 | num_DA[TG_Num_dO] += nAcidic_O; |
326 | 4 | } |
327 | 4 | } |
328 | 8 | } |
329 | 0 | else |
330 | 0 | { |
331 | 0 | if (at->charge <= 0 && at->valence + 1 == at->chem_bonds_valence || |
332 | 0 | at->charge > 0 && at->valence + 2 == at->chem_bonds_valence) |
333 | 0 | { |
334 | | /* acceptor */ |
335 | 0 | if (at->charge < 0) |
336 | 0 | { |
337 | 0 | num_DA[TG_Num_aM] += nDelta; |
338 | 0 | } |
339 | 0 | else |
340 | 0 | { |
341 | 0 | if (at->num_H) |
342 | 0 | { |
343 | 0 | num_DA[TG_Num_aH] += nDelta; |
344 | 0 | } |
345 | 0 | else |
346 | 0 | { |
347 | 0 | num_DA[TG_Num_aO] += nAcidic_O; /* acidic O-acceptor has no H or charge */ |
348 | 0 | } |
349 | 0 | } |
350 | 0 | } |
351 | 0 | } |
352 | 8 | return; |
353 | 8 | } |
354 | | |
355 | | |
356 | | /* |
357 | | |
358 | | */ |
359 | | |
360 | | |
361 | | /****************************************************************************/ |
362 | | int AddEndPoint( T_ENDPOINT *pEndPoint, inp_ATOM *at, int iat ) |
363 | 8 | { |
364 | 8 | pEndPoint->nAtomNumber = iat; |
365 | 8 | pEndPoint->nEquNumber = 0; |
366 | 8 | pEndPoint->nGroupNumber = at[iat].endpoint; |
367 | 8 | if (at[iat].endpoint) |
368 | 0 | { |
369 | | /* already an endpoint */ |
370 | 0 | memset( pEndPoint->num, 0, sizeof( pEndPoint->num ) ); |
371 | 0 | } |
372 | 8 | else |
373 | 8 | { |
374 | | /* not an endpoint yet, make it an endpoint */ |
375 | 8 | AddAtom2num( pEndPoint->num, at, iat, 2 ); /* fill */ |
376 | 8 | AddAtom2DA( pEndPoint->num_DA, at, iat, 2 ); |
377 | | /* |
378 | | nMobile = pEndPoint->num[1] = (at[iat].charge == -1); |
379 | | nMobile = pEndPoint->num[0] = at[iat].num_H + nMobile; |
380 | | for ( k = 0; k < T_NUM_ISOTOPIC; k ++ ) { |
381 | | pEndPoint->num[T_NUM_NO_ISOTOPIC+k] = at[iat].num_iso_H[NUM_H_ISOTOPES-k-1]; |
382 | | } |
383 | | */ |
384 | 8 | } |
385 | | |
386 | 8 | return 0; |
387 | 8 | } |
388 | | |
389 | | |
390 | | /****************************************************************************/ |
391 | | int nGetEndpointInfo( inp_ATOM *atom, int iat, ENDPOINT_INFO *eif ) |
392 | 564k | { |
393 | 564k | int nEndpointValence; |
394 | 564k | int nMobile; |
395 | 564k | S_CHAR cChargeSubtype; |
396 | | |
397 | 564k | if (atom[iat].radical && atom[iat].radical != RADICAL_SINGLET) |
398 | 30 | { |
399 | 30 | return 0; /* a radical */ |
400 | 30 | } |
401 | 564k | if (!( nEndpointValence = get_endpoint_valence( atom[iat].el_number ) )) |
402 | 560k | { |
403 | 560k | return 0; /* not an endpoint */ |
404 | 560k | } |
405 | 3.92k | if (nEndpointValence <= atom[iat].valence) |
406 | 20 | { |
407 | 20 | return 0; /* not an endpoint, for example >N(+)< or >N< or >O(+)- or >O- or >N- or -O- */ |
408 | 20 | } |
409 | | |
410 | 3.90k | if (atom[iat].charge == -1 || atom[iat].charge == 0) |
411 | 3.90k | { |
412 | | /* not a positive charge-point */ |
413 | 3.90k | if (nEndpointValence < atom[iat].chem_bonds_valence) |
414 | 0 | { |
415 | 0 | return 0; /* abnormal valence > standard endpoint valence */ |
416 | 0 | } |
417 | 3.90k | nMobile = atom[iat].num_H + ( atom[iat].charge == -1 ); |
418 | 3.90k | if (nMobile + atom[iat].chem_bonds_valence != nEndpointValence) |
419 | 3.83k | { |
420 | 3.83k | return 0; /* non-standard endpoint valence */ |
421 | 3.83k | } |
422 | 68 | switch (atom[iat].chem_bonds_valence - atom[iat].valence) |
423 | 68 | { |
424 | 68 | case 0: |
425 | 68 | eif->cDonor = 1; |
426 | 68 | eif->cAcceptor = 0; |
427 | 68 | break; |
428 | 0 | case 1: |
429 | 0 | eif->cDonor = 0; |
430 | 0 | eif->cAcceptor = 1; |
431 | 0 | break; |
432 | 0 | default: |
433 | 0 | return 0; |
434 | 68 | } |
435 | 68 | eif->cMobile = nMobile; |
436 | 68 | eif->cNeutralBondsValence = nEndpointValence - nMobile; |
437 | 68 | eif->cMoveableCharge = 0; |
438 | 68 | #if ( KETO_ENOL_TAUT == 1 ) |
439 | 68 | eif->cKetoEnolCode = 0; |
440 | 68 | #endif |
441 | 68 | return nEndpointValence; |
442 | 68 | } |
443 | 0 | else |
444 | 0 | { |
445 | 0 | if (atom[iat].c_point && |
446 | 0 | 0 <= GetChargeType( atom, iat, &cChargeSubtype ) && |
447 | 0 | ( (int) cChargeSubtype & ( C_SUBTYPE_H_ACCEPT | C_SUBTYPE_H_DONOR ) )) |
448 | 0 | { |
449 | | /* charge-point */ |
450 | 0 | { |
451 | 0 | if (cChargeSubtype & C_SUBTYPE_H_ACCEPT) |
452 | 0 | { |
453 | 0 | eif->cDonor = 0; |
454 | 0 | eif->cAcceptor = 1; |
455 | 0 | } |
456 | 0 | else |
457 | 0 | { |
458 | 0 | if (cChargeSubtype & C_SUBTYPE_H_DONOR) |
459 | 0 | { |
460 | 0 | eif->cDonor = 1; |
461 | 0 | eif->cAcceptor = 0; |
462 | 0 | } |
463 | 0 | else |
464 | 0 | { |
465 | 0 | return 0; |
466 | 0 | } |
467 | 0 | } |
468 | 0 | } |
469 | 0 | eif->cMobile = atom[iat].num_H; |
470 | 0 | eif->cNeutralBondsValence = nEndpointValence - atom[iat].num_H; |
471 | 0 | eif->cMoveableCharge = atom[iat].charge; |
472 | 0 | #if ( KETO_ENOL_TAUT == 1 ) |
473 | 0 | eif->cKetoEnolCode = 0; |
474 | 0 | #endif |
475 | 0 | return nEndpointValence; |
476 | 0 | } |
477 | 0 | } |
478 | | |
479 | 0 | return 0; |
480 | 3.90k | } |
481 | | |
482 | | |
483 | | /*****************************************************************************/ |
484 | | #if ( KETO_ENOL_TAUT == 1 ) /* post v.1 feature */ |
485 | | |
486 | | |
487 | | /****************************************************************************/ |
488 | | int nGetEndpointInfo_KET( inp_ATOM *atom, |
489 | | int iat, |
490 | | ENDPOINT_INFO *eif ) |
491 | 0 | { |
492 | 0 | int nEndpointValence; |
493 | 0 | int nMobile; |
494 | 0 | S_CHAR cChargeSubtype; |
495 | |
|
496 | 0 | if (atom[iat].radical && atom[iat].radical != RADICAL_SINGLET) |
497 | 0 | { |
498 | 0 | return 0; /* a radical */ |
499 | 0 | } |
500 | 0 | if (!( nEndpointValence = get_endpoint_valence_KET( atom[iat].el_number ) )) |
501 | 0 | { |
502 | 0 | return 0; /* not an endpoint; only O and C can be an endpoint for keto-enol tautomerism */ |
503 | 0 | } |
504 | 0 | if (nEndpointValence <= atom[iat].valence) |
505 | 0 | { |
506 | 0 | return 0; /* not an endpoint, for example >N(+)< or >N< or >O(+)- or >O- or >N- or -O- */ |
507 | 0 | } |
508 | 0 | if (nEndpointValence == 4 && atom[iat].valence < 2) |
509 | 0 | { |
510 | 0 | return 0; /* exclude O==C--CH3 <=> HO--C==CH2 */ |
511 | 0 | } |
512 | 0 | if (nEndpointValence == 2 && atom[iat].valence > 1) |
513 | 0 | { |
514 | 0 | return 0; /* exclude --O--C==CH-- */ |
515 | 0 | } |
516 | | |
517 | 0 | if (atom[iat].charge == -1 || atom[iat].charge == 0) |
518 | 0 | { |
519 | | /* not a positive charge-point */ |
520 | 0 | if (nEndpointValence < atom[iat].chem_bonds_valence) |
521 | 0 | { |
522 | 0 | return 0; /* abnormal valence > standard endpoint valence */ |
523 | 0 | } |
524 | 0 | nMobile = atom[iat].num_H + ( atom[iat].charge == -1 ); |
525 | 0 | if (nMobile + atom[iat].chem_bonds_valence != nEndpointValence) |
526 | 0 | { |
527 | 0 | return 0; /* non-standard endpoint valence */ |
528 | 0 | } |
529 | | |
530 | 0 | switch (atom[iat].chem_bonds_valence - atom[iat].valence) |
531 | 0 | { |
532 | 0 | case 0: |
533 | 0 | eif->cDonor = 1; |
534 | 0 | eif->cAcceptor = 0; |
535 | 0 | break; |
536 | 0 | case 1: |
537 | 0 | eif->cDonor = 0; |
538 | 0 | eif->cAcceptor = 1; |
539 | 0 | break; |
540 | 0 | default: |
541 | 0 | return 0; |
542 | 0 | } |
543 | 0 | eif->cMobile = nMobile; |
544 | 0 | eif->cNeutralBondsValence = nEndpointValence - nMobile; |
545 | 0 | eif->cMoveableCharge = 0; |
546 | 0 | eif->cKetoEnolCode = ( nEndpointValence == 2 ) ? 1 : ( nEndpointValence == 4 ) ? 2 : 0; |
547 | 0 | return nEndpointValence; |
548 | 0 | } |
549 | 0 | else |
550 | 0 | { |
551 | 0 | if (atom[iat].c_point && |
552 | 0 | 0 <= GetChargeType( atom, iat, &cChargeSubtype ) && |
553 | 0 | ( (int) cChargeSubtype & ( C_SUBTYPE_H_ACCEPT | C_SUBTYPE_H_DONOR ) )) |
554 | 0 | { |
555 | | /* charge-point; currently only O for keto-enol tautomerism */ |
556 | 0 | if (cChargeSubtype & C_SUBTYPE_H_ACCEPT) |
557 | 0 | { |
558 | 0 | eif->cDonor = 0; |
559 | 0 | eif->cAcceptor = 1; |
560 | 0 | } |
561 | 0 | else |
562 | 0 | { |
563 | 0 | if (cChargeSubtype & C_SUBTYPE_H_DONOR) |
564 | 0 | { |
565 | 0 | eif->cDonor = 1; |
566 | 0 | eif->cAcceptor = 0; |
567 | 0 | } |
568 | 0 | else |
569 | 0 | { |
570 | 0 | return 0; |
571 | 0 | } |
572 | 0 | } |
573 | 0 | eif->cMobile = atom[iat].num_H; |
574 | 0 | eif->cNeutralBondsValence = nEndpointValence - atom[iat].num_H; |
575 | 0 | eif->cMoveableCharge = atom[iat].charge; |
576 | 0 | eif->cKetoEnolCode = ( nEndpointValence == 2 ) ? 1 : ( nEndpointValence == 4 ) ? 2 : 0; |
577 | 0 | return nEndpointValence; |
578 | 0 | } |
579 | 0 | } |
580 | | |
581 | 0 | return 0; |
582 | 0 | } |
583 | | #endif |
584 | | |
585 | | |
586 | | /**************************************************************************** |
587 | | RegisterEndPoints |
588 | | |
589 | | Returns >0 => new registration happened, |
590 | | =0 => no changes, |
591 | | -1 => program error (debug) |
592 | | ****************************************************************************/ |
593 | | int RegisterEndPoints( CANON_GLOBALS *pCG, |
594 | | T_GROUP_INFO *t_group_info, |
595 | | /* |
596 | | T_GROUP *t_group, |
597 | | int *pnum_t, |
598 | | int max_num_t, |
599 | | */ |
600 | | T_ENDPOINT *EndPoint, |
601 | | int nNumEndPoints, |
602 | | inp_ATOM *at, |
603 | | int num_atoms, |
604 | | C_GROUP_INFO *cgi, |
605 | | struct BalancedNetworkStructure *pBNS ) |
606 | 4 | { |
607 | 4 | T_GROUP *t_group = t_group_info->t_group; |
608 | 4 | int *pnum_t = &t_group_info->num_t_groups; |
609 | 4 | int max_num_t = t_group_info->max_num_t_groups; |
610 | 4 | int nNumZeroEqu, nNumNewTGroups; |
611 | 4 | AT_NUMB group, prev_group, prev_eqnum, nNextGroupNumber, nLeastGroupNumber; |
612 | 4 | int nNumGroups, num_t, difference; |
613 | 4 | int i, j, k, ret; |
614 | 4 | AT_NUMB nNewTgNumberStackArray[MAX_STACK_ARRAY_LEN + 1]; |
615 | 4 | AT_NUMB nGroupNumberStackArray[MAX_STACK_ARRAY_LEN + 1]; |
616 | 4 | AT_NUMB nGroupNewNumberStackArray[MAX_STACK_ARRAY_LEN + 1]; |
617 | 4 | AT_NUMB *nNewTgNumber = nNewTgNumberStackArray; |
618 | 4 | AT_NUMB *nGroupNumber = nGroupNumberStackArray; |
619 | 4 | AT_NUMB *nGroupNewNumber = nGroupNewNumberStackArray; |
620 | | |
621 | 4 | if (nNumEndPoints <= 0) |
622 | 0 | { |
623 | 0 | return 0; /* nothing to do */ |
624 | 0 | } |
625 | 4 | num_t = *pnum_t; |
626 | 4 | difference = 0; |
627 | 4 | nNextGroupNumber = 0; |
628 | 4 | nNumZeroEqu = 0; |
629 | 4 | ret = 0; |
630 | | /* find max group number; increment it to obtain next available group number */ |
631 | 4 | for (i = 0; i < num_t; i++) |
632 | 0 | { |
633 | 0 | if (nNextGroupNumber < t_group[i].nGroupNumber) |
634 | 0 | { |
635 | 0 | nNextGroupNumber = t_group[i].nGroupNumber; |
636 | 0 | } |
637 | 0 | } |
638 | 4 | nNextGroupNumber++; |
639 | | |
640 | | /* find min non-zero group number nLeastGroupNumber; |
641 | | count zero EndPoint[i].nEquNumber |
642 | | if all EndPoint[i].nGroupNumber are equal and non-zero then exit: nothing to do. |
643 | | */ |
644 | 4 | nLeastGroupNumber = nNextGroupNumber; |
645 | 4 | prev_group = EndPoint[0].nGroupNumber; |
646 | 4 | prev_eqnum = EndPoint[0].nEquNumber; |
647 | 12 | for (i = j = k = 0; i < nNumEndPoints; i++) |
648 | 8 | { |
649 | 8 | if (group = EndPoint[i].nGroupNumber) |
650 | 0 | { |
651 | 0 | if (group < nLeastGroupNumber) |
652 | 0 | { |
653 | 0 | nLeastGroupNumber = group; |
654 | 0 | } |
655 | 0 | } |
656 | 8 | j += ( prev_group == EndPoint[i].nGroupNumber ); /* count endpoints that belong to the 1st group */ |
657 | 8 | k += ( prev_eqnum == EndPoint[i].nEquNumber ); /* count endpoints that belongo to a group equivalent to the 1st group */ |
658 | 8 | nNumZeroEqu += !EndPoint[i].nEquNumber; /* count endpoints that have been processed by FindAccessibleEndPoints() */ |
659 | 8 | } |
660 | 4 | if (j == nNumEndPoints && prev_group && k == nNumEndPoints) |
661 | 0 | { |
662 | | /* all endpoints already belong to one t-group; |
663 | | the last comparison is not needed for now |
664 | | because EndPoint[i].nEquNumber cannot make |
665 | | endpont partitioning finer |
666 | | */ |
667 | 0 | return 0; |
668 | 0 | } |
669 | | |
670 | 4 | nNumNewTGroups = 0; |
671 | | |
672 | 4 | if (!nNumZeroEqu) |
673 | 0 | { |
674 | | /* EndPoint[] has been processed by FindAccessibleEndPoints; |
675 | | * equal EndPoint[i].nEquNumber mark endpoints belonging to |
676 | | * the same t-group |
677 | | * Since now the next available t-group number, nNextGroupNumber, |
678 | | * is known,replace fict. IDs assigned by FindAccessibleEndPoints |
679 | | * with correct new t-group numbers. |
680 | | */ |
681 | 0 | for (i = 0; i < nNumEndPoints; i++) |
682 | 0 | { |
683 | 0 | if (( group = EndPoint[i].nEquNumber ) >= nNextGroupNumber) |
684 | 0 | { |
685 | | /* replace fict. IDs assigned by FindAccessibleEndPoints() with new t-group numbers */ |
686 | | /* these fict. IDs have values = (num_atoms+1), (num_atoms+2),...; they may be non-contiguous */ |
687 | 0 | for (j = 0; j < nNumNewTGroups; j++) |
688 | 0 | { |
689 | 0 | if (group == nGroupNewNumber[j]) |
690 | 0 | { |
691 | 0 | break; |
692 | 0 | } |
693 | 0 | } |
694 | 0 | if (j == nNumNewTGroups) |
695 | 0 | { |
696 | | /* found new fict. ID = group */ |
697 | 0 | if (j == MAX_STACK_ARRAY_LEN && nGroupNewNumber == nGroupNewNumberStackArray) |
698 | 0 | { |
699 | | /* stack array overflow; allocate more memory than may be needed */ |
700 | 0 | nGroupNewNumber = (AT_NUMB *) inchi_malloc( nNumEndPoints * sizeof( nGroupNewNumber[0] ) ); |
701 | 0 | if (!nGroupNewNumber) |
702 | 0 | { |
703 | 0 | ret = -1; |
704 | 0 | goto exit_function; |
705 | 0 | } |
706 | 0 | memcpy( nGroupNewNumber, nGroupNewNumberStackArray, nNumNewTGroups * sizeof( nGroupNewNumber[0] ) ); |
707 | 0 | } |
708 | | /* save newly found fict. t-group ID to compare to the next values of EndPoint[].nEquNumber */ |
709 | 0 | nGroupNewNumber[j] = group; |
710 | 0 | nNumNewTGroups++; |
711 | 0 | } |
712 | 0 | EndPoint[i].nEquNumber = nNextGroupNumber + j; |
713 | 0 | } |
714 | 0 | } /* after this point the values just stored in nGroupNewNumber[] will not |
715 | | be used. However, the obtained nNumNewTGroups value will be used */ |
716 | 0 | } |
717 | 4 | else |
718 | 4 | { |
719 | 4 | if (nNumZeroEqu == nNumEndPoints) |
720 | 4 | { |
721 | | /* EndPoint[] has NOT been processed by FindAccessibleEndPoints; |
722 | | all atoms and t-groups to which endpoints belong should |
723 | | be merged into a single t-group */ |
724 | 4 | if (nLeastGroupNumber == nNextGroupNumber) |
725 | 4 | { |
726 | | /* flag to create a new t-group: none of the found |
727 | | endpoints belong to an already known t-group */ |
728 | 4 | nNumNewTGroups = 1; /* otherwise 0 */ |
729 | 4 | } |
730 | | /* All EndPoint[*].nEquNumber are zeroes. All endpoints will |
731 | | * belong to one new or old t-group; its ID is nLeastGroupNumber. |
732 | | * Set EndPoint[i].nEquNumber = nLeastGroupNumber; */ |
733 | 12 | for (i = 0; i < nNumEndPoints; i++) |
734 | 8 | { |
735 | 8 | EndPoint[i].nEquNumber = nLeastGroupNumber; |
736 | 8 | } |
737 | 4 | } |
738 | 0 | else |
739 | 0 | { |
740 | 0 | ret = -1; /* program error: only some of EndPoint[i].nEquNumber are zero */ /* <BRKPT> */ |
741 | 0 | goto exit_function; |
742 | 0 | } |
743 | 4 | } |
744 | | |
745 | 4 | if (nNumNewTGroups) |
746 | 4 | { |
747 | | /* create new nNumNewTGroups t-group(s) */ |
748 | 4 | if (num_t + nNumNewTGroups > max_num_t) |
749 | 0 | { |
750 | 0 | ret = -1; /* found too many t-groups */ /* <BRKPT> */ |
751 | 0 | goto exit_function; |
752 | 0 | } |
753 | | /* initialize new t-group(s) */ |
754 | 4 | memset( t_group + num_t, 0, nNumNewTGroups * sizeof( t_group[0] ) ); |
755 | 8 | for (i = 0; i < nNumNewTGroups; i++) |
756 | 4 | { |
757 | 4 | t_group[num_t + i].nGroupNumber = nNextGroupNumber + i; |
758 | 4 | } |
759 | 4 | } |
760 | | |
761 | | /* At this point: |
762 | | * EndPoint[i].nGroupNumber == 0 => the endpoint atom does not belong to a t-group yet |
763 | | * EndPoint[i].nGroupNumber > 0 => current t-group ID of the endpoint atom |
764 | | * EndPoint[i].nEquNumber --> new ID of a tautomeric group of this endpoint atom |
765 | | * EndPoint[i].nAtomNumber --> number of the endpoint atom |
766 | | */ |
767 | | |
768 | 4 | nNumGroups = 0; /* counts the groups to be renumbered */ |
769 | 12 | for (i = j = 0; i < nNumEndPoints; i++) |
770 | 8 | { |
771 | 8 | if (group = EndPoint[i].nGroupNumber) |
772 | 0 | { |
773 | 0 | if (group == EndPoint[i].nEquNumber) |
774 | 0 | { |
775 | 0 | continue; /* ignore: the endpoint belongs to the same t-group as before */ |
776 | 0 | } |
777 | | /* save information for renumbering of the existing t-groups */ |
778 | 0 | for (j = 0; j < nNumGroups; j++) |
779 | 0 | { |
780 | 0 | if (group == nGroupNumber[j]) |
781 | 0 | { |
782 | 0 | if (EndPoint[i].nEquNumber != nGroupNewNumber[j]) |
783 | 0 | { |
784 | 0 | ret = -1; /* program error */ /* <BRKPT> */ |
785 | 0 | goto exit_function; |
786 | 0 | } |
787 | 0 | break; |
788 | 0 | } |
789 | 0 | } |
790 | 0 | if (j == nNumGroups) |
791 | 0 | { |
792 | | /* discovered a new t-group number; store it together with its nEquNumber */ |
793 | 0 | if (j == MAX_STACK_ARRAY_LEN) |
794 | 0 | { |
795 | 0 | if (nGroupNewNumber == nGroupNewNumberStackArray) |
796 | 0 | { |
797 | 0 | nGroupNewNumber = (AT_NUMB *) inchi_malloc( nNumEndPoints * sizeof( nGroupNewNumber[0] ) ); |
798 | 0 | if (!nGroupNewNumber) |
799 | 0 | { |
800 | 0 | ret = -1; |
801 | 0 | goto exit_function; |
802 | 0 | } |
803 | 0 | memcpy( nGroupNewNumber, nGroupNewNumberStackArray, nNumGroups * sizeof( nGroupNewNumber[0] ) ); |
804 | 0 | } |
805 | 0 | if (nGroupNumber == nGroupNumberStackArray) |
806 | 0 | { |
807 | 0 | nGroupNumber = (AT_NUMB *) inchi_malloc( nNumEndPoints * sizeof( nGroupNumber[0] ) ); |
808 | 0 | if (!nGroupNumber) |
809 | 0 | { |
810 | 0 | ret = -1; |
811 | 0 | goto exit_function; |
812 | 0 | } |
813 | 0 | memcpy( nGroupNumber, nGroupNumberStackArray, nNumGroups * sizeof( nGroupNumber[0] ) ); |
814 | 0 | } |
815 | 0 | } |
816 | | |
817 | 0 | nGroupNumber[j] = group; /* old t-group ID */ |
818 | 0 | nGroupNewNumber[j] = EndPoint[i].nEquNumber; /* new t-group ID */ |
819 | 0 | nNumGroups++; |
820 | 0 | } |
821 | 0 | } |
822 | 8 | else |
823 | 8 | { |
824 | | /* add a new endpoint to the newly created or previously existing t-groups */ |
825 | 8 | group = EndPoint[i].nEquNumber; |
826 | 8 | if (group >= nNextGroupNumber) |
827 | 8 | { |
828 | | /* get index of a new t-group from equ number */ |
829 | 8 | j = num_t + group - nNextGroupNumber; /* newly assigned IDs are contiguous */ |
830 | 8 | } |
831 | 0 | else |
832 | 0 | { |
833 | | /* old t-group */ |
834 | 0 | if (j >= num_t || group != t_group[j].nGroupNumber) |
835 | 0 | { |
836 | | /* search only if j is not a needed group index */ |
837 | 0 | for (j = 0; j < num_t; j++) |
838 | 0 | { |
839 | 0 | if (group == t_group[j].nGroupNumber) |
840 | 0 | { |
841 | 0 | break; |
842 | 0 | } |
843 | 0 | } |
844 | 0 | if (j == num_t) |
845 | 0 | { |
846 | 0 | ret = -1; /* program error: t-group not found */ /* <BRKPT> */ |
847 | 0 | goto exit_function; |
848 | 0 | } |
849 | 0 | } |
850 | 0 | } |
851 | | /* add aton to existing or new t-group */ |
852 | 8 | t_group[j].nNumEndpoints++; |
853 | 48 | for (k = 0; k < (int) ( sizeof( t_group->num ) / sizeof( t_group->num[0] ) ); k++) |
854 | 40 | { |
855 | 40 | t_group[j].num[k] += EndPoint[i].num[k]; |
856 | 40 | } |
857 | 56 | for (k = 0; k < (int) ( sizeof( t_group->num_DA ) / sizeof( t_group->num_DA[0] ) ); k++) |
858 | 48 | { |
859 | 48 | t_group[j].num_DA[k] += EndPoint[i].num_DA[k]; |
860 | 48 | } |
861 | | /* mark endpoint */ |
862 | 8 | at[EndPoint[i].nAtomNumber].endpoint = group; |
863 | 8 | difference++; |
864 | 8 | } |
865 | 8 | } |
866 | | |
867 | 4 | difference += nNumGroups; |
868 | 4 | num_t += nNumNewTGroups; |
869 | 4 | if (!difference) |
870 | 0 | { |
871 | 0 | ret = 0; /* nothing to do. Not necessarily a program error: happens if all EndPoint[i].nGroupNumber==EndPoint[i].nEquNumber */ |
872 | 0 | goto exit_function; |
873 | 0 | } |
874 | | |
875 | 4 | if (nNumGroups) |
876 | 0 | { |
877 | | /* prepare for renumbering: find max t-group number */ |
878 | 0 | for (i = 0, nNextGroupNumber = 0; i < num_t; i++) |
879 | 0 | { |
880 | 0 | if (nNextGroupNumber < t_group[i].nGroupNumber) |
881 | 0 | { |
882 | 0 | nNextGroupNumber = t_group[i].nGroupNumber; |
883 | 0 | } |
884 | 0 | } |
885 | 0 | } |
886 | | /* renumber and merge t-groups */ |
887 | 4 | for (i = 0; i < nNumGroups; i++) |
888 | 0 | { |
889 | 0 | int i1, i2; |
890 | 0 | AT_NUMB group1 = nGroupNumber[i]; |
891 | 0 | AT_NUMB group2 = nGroupNewNumber[i]; |
892 | | /* add group1 to group2, then delete group1. */ |
893 | 0 | for (j = 0, i1 = i2 = -1; j < num_t && ( i1 < 0 || i2 < 0 ); j++) |
894 | 0 | { |
895 | 0 | if (i1 < 0 && group1 == t_group[j].nGroupNumber) |
896 | 0 | { |
897 | 0 | i1 = j; |
898 | 0 | } |
899 | 0 | if (i2 < 0 && group2 == t_group[j].nGroupNumber) |
900 | 0 | { |
901 | 0 | i2 = j; |
902 | 0 | } |
903 | 0 | } |
904 | 0 | if (i1 < 0 || i2 < 0) |
905 | 0 | { |
906 | 0 | ret = -1; /* program error */ /* <BRKPT> */ |
907 | 0 | goto exit_function; |
908 | 0 | } |
909 | | /* add t_group[i1] to t_group[i2] and remove t_group[i1] */ |
910 | 0 | for (k = 0; k < (int) ( sizeof( t_group->num ) / sizeof( t_group->num[0] ) ); k++) |
911 | 0 | { |
912 | 0 | t_group[i2].num[k] += t_group[i1].num[k]; |
913 | 0 | } |
914 | 0 | for (k = 0; k < (int) ( sizeof( t_group->num_DA ) / sizeof( t_group->num_DA[0] ) ); k++) |
915 | 0 | { |
916 | 0 | t_group[i2].num_DA[k] += t_group[i1].num_DA[k]; |
917 | 0 | } |
918 | 0 | t_group[i2].nNumEndpoints += t_group[i1].nNumEndpoints; |
919 | 0 | num_t--; |
920 | 0 | if (num_t > i1) |
921 | 0 | { |
922 | 0 | memmove( t_group + i1, t_group + i1 + 1, ( num_t - i1 ) * sizeof( t_group[0] ) ); |
923 | 0 | } |
924 | 0 | } |
925 | | |
926 | 4 | if (nNumGroups) |
927 | 0 | { |
928 | | /* there are groups to merge */ |
929 | 0 | if (nNextGroupNumber >= MAX_STACK_ARRAY_LEN) |
930 | 0 | { |
931 | 0 | nNewTgNumber = (AT_NUMB *) inchi_malloc( ( nNextGroupNumber + 1 ) * sizeof( *nNewTgNumber ) ); |
932 | 0 | if (!nNewTgNumber) |
933 | 0 | { |
934 | 0 | ret = -1; |
935 | 0 | goto exit_function; /* error: out of RAM */ |
936 | 0 | } |
937 | 0 | } |
938 | 0 | memset( nNewTgNumber, 0, ( nNextGroupNumber + 1 ) * sizeof( *nNewTgNumber ) ); |
939 | 0 | for (i = 0; i < num_t; i++) |
940 | 0 | { |
941 | 0 | nNewTgNumber[t_group[i].nGroupNumber] = i + 1; /* new t-group numbers */ |
942 | 0 | } |
943 | 0 | for (j = 0; j < nNumGroups; j++) |
944 | 0 | { |
945 | 0 | if (!nNewTgNumber[nGroupNumber[j]] && nNewTgNumber[nGroupNewNumber[j]]) |
946 | 0 | { |
947 | 0 | nNewTgNumber[nGroupNumber[j]] = nNewTgNumber[nGroupNewNumber[j]]; |
948 | 0 | } |
949 | 0 | else |
950 | 0 | { |
951 | 0 | ret = -1; /* program error: all new numbers must have been marked */ |
952 | 0 | goto exit_function; |
953 | 0 | } |
954 | 0 | } |
955 | | /* renumber t-groups */ |
956 | 0 | for (i = 0; i < num_t; i++) |
957 | 0 | { |
958 | 0 | t_group[i].nGroupNumber = nNewTgNumber[t_group[i].nGroupNumber]; |
959 | 0 | } |
960 | | #if ( bRELEASE_VERSION != 1 ) |
961 | | /* Check: debug only */ |
962 | | for (i = 1; i < num_t; i++) |
963 | | { |
964 | | if (1 != t_group[i].nGroupNumber - t_group[i - 1].nGroupNumber) |
965 | | { |
966 | | ret = -1; /* debug */ |
967 | | goto exit_function; |
968 | | } |
969 | | } |
970 | | #endif |
971 | | /* renumber endpoints */ |
972 | 0 | for (i = 0; i < num_atoms; i++) |
973 | 0 | { |
974 | 0 | if (group = at[i].endpoint) |
975 | 0 | { |
976 | 0 | if (!( at[i].endpoint = nNewTgNumber[group] ) || nNextGroupNumber <= nNewTgNumber[group]) |
977 | 0 | { |
978 | 0 | ret = -1; /* program error */ |
979 | 0 | goto exit_function; |
980 | 0 | } |
981 | 0 | } |
982 | 0 | } |
983 | 0 | } |
984 | 4 | if (nNewTgNumber != nNewTgNumberStackArray) |
985 | 0 | { |
986 | 0 | inchi_free( nNewTgNumber ); |
987 | 0 | nNewTgNumber = nNewTgNumberStackArray; |
988 | 0 | } |
989 | 4 | if (nGroupNumber != nGroupNumberStackArray) |
990 | 0 | { |
991 | 0 | inchi_free( nGroupNumber ); |
992 | 0 | nGroupNumber = nGroupNumberStackArray; |
993 | 0 | } |
994 | 4 | if (nGroupNewNumber != nGroupNewNumberStackArray) |
995 | 0 | { |
996 | 0 | inchi_free( nGroupNewNumber ); |
997 | 0 | nGroupNewNumber = nGroupNewNumberStackArray; |
998 | 0 | } |
999 | 4 | if (!t_group_info->tGroupNumber) |
1000 | 0 | { |
1001 | 0 | t_group_info->tGroupNumber = (AT_NUMB *) inchi_malloc( 2 * max_num_t * sizeof( t_group_info->tGroupNumber[0] ) ); |
1002 | 0 | if (!t_group_info->tGroupNumber) |
1003 | 0 | { |
1004 | 0 | ret = -1; |
1005 | 0 | goto exit_function; |
1006 | 0 | } |
1007 | 0 | } |
1008 | | /* fill out t-group index 2004-02-27 */ |
1009 | 4 | memset( t_group_info->tGroupNumber, 0, 2 * max_num_t * sizeof( t_group_info->tGroupNumber[0] ) ); |
1010 | 8 | for (i = 0; i < num_t; i++) |
1011 | 4 | { |
1012 | 4 | if (t_group[i].nNumEndpoints && t_group[i].nGroupNumber) |
1013 | 4 | { |
1014 | 4 | t_group_info->tGroupNumber[t_group[i].nGroupNumber] = i + 1; |
1015 | 4 | } |
1016 | 4 | } |
1017 | | |
1018 | 4 | if (pBNS && ( pBNS->tot_st_cap == pBNS->tot_st_flow || ALWAYS_ADD_TG_ON_THE_FLY )) |
1019 | 4 | { |
1020 | 4 | T_GROUP_INFO tgi; |
1021 | 4 | int ret_bns; |
1022 | 4 | memset( &tgi, 0, sizeof( tgi ) ); |
1023 | 4 | tgi.num_t_groups = num_t; |
1024 | 4 | tgi.t_group = t_group; |
1025 | 4 | #if ( KETO_ENOL_TAUT == 1 ) |
1026 | 4 | tgi.bTautFlags |= ( t_group_info->bTautFlags & TG_FLAG_KETO_ENOL_TAUT ); /* needed in AddTGroups2BnStruct() */ |
1027 | 4 | #endif |
1028 | | /* reinitialize BN Structure */ |
1029 | 4 | ret_bns = ReInitBnStruct( pBNS, at, num_atoms, 0 ); |
1030 | 4 | if (IS_BNS_ERROR( ret_bns )) |
1031 | 0 | { |
1032 | 0 | return ret_bns; |
1033 | 0 | } |
1034 | 4 | if (*pBNS->pbTautFlags & TG_FLAG_MOVE_POS_CHARGES) |
1035 | 4 | { |
1036 | | /* set new charge groups */ |
1037 | 4 | ret_bns = AddCGroups2BnStruct( pCG, pBNS, at, num_atoms, cgi ); |
1038 | 4 | if (IS_BNS_ERROR( ret_bns )) |
1039 | 0 | { |
1040 | 0 | return ret_bns; |
1041 | 0 | } |
1042 | 4 | } |
1043 | | /* set new tautomeric groups */ |
1044 | 4 | ret_bns = AddTGroups2BnStruct( pCG, pBNS, at, num_atoms, &tgi ); |
1045 | 4 | if (IS_BNS_ERROR( ret_bns )) |
1046 | 0 | { |
1047 | 0 | return ret_bns; |
1048 | 0 | } |
1049 | 4 | } |
1050 | | |
1051 | 4 | *pnum_t = num_t; |
1052 | 4 | return difference; |
1053 | | |
1054 | 0 | exit_function: |
1055 | 0 | if (nNewTgNumber != nNewTgNumberStackArray) |
1056 | 0 | { |
1057 | 0 | inchi_free( nNewTgNumber ); |
1058 | 0 | } |
1059 | 0 | if (nGroupNumber != nGroupNumberStackArray) |
1060 | 0 | { |
1061 | 0 | inchi_free( nGroupNumber ); |
1062 | 0 | } |
1063 | 0 | if (nGroupNewNumber != nGroupNewNumberStackArray) |
1064 | 0 | { |
1065 | 0 | inchi_free( nGroupNewNumber ); |
1066 | 0 | } |
1067 | |
|
1068 | 0 | return ret; |
1069 | 4 | } |
1070 | | |
1071 | | |
1072 | | /**************************************************************************** |
1073 | | Change non-alternating and non-tautomeric bonds |
1074 | | (that is, single and double bonds) to tautomeric |
1075 | | ****************************************************************************/ |
1076 | | int SetTautomericBonds( inp_ATOM *at, |
1077 | | int nNumBondPos, |
1078 | | T_BONDPOS *BondPos ) |
1079 | 0 | { |
1080 | 0 | int k, n; |
1081 | 0 | for (k = n = 0; k < nNumBondPos; k++) |
1082 | 0 | { |
1083 | 0 | int neighbor_index = BondPos[k].neighbor_index; |
1084 | 0 | int center = BondPos[k].nAtomNumber; |
1085 | 0 | int bond_mark = at[center].bond_type[neighbor_index]; |
1086 | 0 | int bond_type = bond_mark & ~BOND_MARK_ALL; |
1087 | 0 | int neighbor; |
1088 | 0 | #if ( REPLACE_ALT_WITH_TAUT == 1 ) |
1089 | 0 | if (bond_type != BOND_TAUTOM) |
1090 | | #else |
1091 | | if (bond_type != BOND_ALTERN && bond_type != BOND_TAUTOM) |
1092 | | #endif |
1093 | 0 | { |
1094 | 0 | int ii; |
1095 | | /* change bond type to BOND_TAUTOM presering higher bits marks */ |
1096 | 0 | bond_type = ( bond_mark & BOND_MARK_ALL ) | BOND_TAUTOM; |
1097 | | /* change center-neighbor bond */ |
1098 | 0 | at[center].bond_type[neighbor_index] = bond_type; |
1099 | 0 | neighbor = at[center].neighbor[neighbor_index]; |
1100 | 0 | for (ii = 0; ii < at[neighbor].valence; ii++) |
1101 | 0 | { |
1102 | 0 | if (at[neighbor].neighbor[ii] == center) |
1103 | 0 | { |
1104 | | /* neighbor-center bond found */ |
1105 | 0 | at[neighbor].bond_type[ii] = bond_type; |
1106 | 0 | break; |
1107 | 0 | } |
1108 | 0 | } |
1109 | 0 | n++; |
1110 | 0 | } |
1111 | 0 | } |
1112 | |
|
1113 | 0 | return n; |
1114 | 0 | } |
1115 | | |
1116 | | |
1117 | | /****************************************************************************/ |
1118 | | int GetNeutralRepsIfNeeded( AT_NUMB *pri, |
1119 | | AT_NUMB *prj, |
1120 | | inp_ATOM *at, |
1121 | | int num_atoms, |
1122 | | T_ENDPOINT *EndPoint, |
1123 | | int nNumEndPoints, |
1124 | | C_GROUP_INFO *cgi ) |
1125 | 0 | { |
1126 | 0 | AT_NUMB ri = *pri; |
1127 | 0 | AT_NUMB rj = *prj; |
1128 | 0 | int i, k; |
1129 | 0 | AT_NUMB c_point, endpoint, r; |
1130 | |
|
1131 | 0 | if (( c_point = at[ri].c_point ) && |
1132 | 0 | ( at[rj].c_point == c_point ) && |
1133 | 0 | ( at[ri].charge == 1 || at[rj].charge == 1 ) && |
1134 | 0 | cgi && |
1135 | 0 | cgi->num_c_groups > 0) |
1136 | 0 | { |
1137 | | /* at[ri] and at[rj] belong to the same charge group, at least one is charged */ |
1138 | 0 | for (k = 0; k < cgi->num_c_groups; k++) /* MS VC++ 2008 reports unreachable code here ??? */ |
1139 | 0 | { |
1140 | 0 | if (cgi->c_group[k].nGroupNumber == c_point) |
1141 | 0 | { |
1142 | | /* cgi->c_group[k] is found to be this charge group */ |
1143 | 0 | if (cgi->c_group[k].num_CPoints - cgi->c_group[k].num[0] < 2) |
1144 | 0 | { |
1145 | | /* Only one neutral in the c-group: we will not be able to neutralize both */ |
1146 | | /* when looking for the alt path to discover the tautomerism. */ |
1147 | | /* Therefore we need to find a neutral t-group representative at[rj] */ |
1148 | 0 | if (endpoint = at[rj].endpoint) |
1149 | 0 | { |
1150 | 0 | for (i = 0; i < nNumEndPoints; i++) |
1151 | 0 | { |
1152 | 0 | if (( r = EndPoint[i].nAtomNumber ) == *prj) |
1153 | 0 | { |
1154 | 0 | continue; /* ignore at[*prj] */ |
1155 | 0 | } |
1156 | 0 | if (at[r].endpoint != endpoint) |
1157 | 0 | { |
1158 | 0 | continue; /* at[r] does not belong to the same t-group as at[*prj]; ignore the atom */ |
1159 | 0 | } |
1160 | 0 | if (!at[r].c_point) |
1161 | 0 | { |
1162 | 0 | rj = r; /* found a neutral t-group representative */ |
1163 | 0 | break; |
1164 | 0 | } |
1165 | 0 | if (at[r].c_point != c_point && c_point == at[rj].c_point) |
1166 | 0 | { |
1167 | | /* replace only once because of (c_point == at[rj].c_point) condition */ |
1168 | 0 | rj = r; |
1169 | 0 | } |
1170 | 0 | } |
1171 | 0 | if (rj == *prj /*&& at[ri].endpoint*/) |
1172 | 0 | { |
1173 | | /* !!! "&& at[ri].endpoint": only between 2 t-groups 2004-02-27; |
1174 | | the change disabled due to undiscovered yet possibility of ambiguity*/ |
1175 | | /* no replacement has been found in EndPoint[]; try all atoms in the t-group */ |
1176 | 0 | for (i = 0; i < num_atoms; i++) |
1177 | 0 | { |
1178 | 0 | if (at[i].endpoint != endpoint) |
1179 | 0 | { |
1180 | 0 | continue; |
1181 | 0 | } |
1182 | 0 | if (i == (int) *prj) |
1183 | 0 | { |
1184 | 0 | continue; |
1185 | 0 | } |
1186 | 0 | if (!at[i].c_point) |
1187 | 0 | { |
1188 | 0 | rj = (AT_NUMB) i; /* found neutral t-group representative */ |
1189 | 0 | break; |
1190 | 0 | } |
1191 | 0 | if (at[i].c_point != c_point && c_point == at[rj].c_point) |
1192 | 0 | { |
1193 | | /* replace only once */ |
1194 | 0 | rj = (AT_NUMB) i; |
1195 | 0 | } |
1196 | 0 | } |
1197 | 0 | } |
1198 | 0 | } |
1199 | | /* at[ri] */ |
1200 | 0 | if (endpoint = at[ri].endpoint) |
1201 | 0 | { |
1202 | 0 | for (i = 0; i < nNumEndPoints; i++) |
1203 | 0 | { |
1204 | 0 | if (( r = EndPoint[i].nAtomNumber ) == *pri) |
1205 | 0 | { |
1206 | 0 | continue; |
1207 | 0 | } |
1208 | 0 | if (at[r].endpoint != endpoint) |
1209 | 0 | { |
1210 | 0 | continue; |
1211 | 0 | } |
1212 | 0 | if (!at[r].c_point) |
1213 | 0 | { |
1214 | 0 | ri = r; /* found neutral t-group representative */ |
1215 | 0 | break; |
1216 | 0 | } |
1217 | 0 | if (at[r].c_point != c_point && c_point == at[ri].c_point && |
1218 | 0 | at[r].c_point != at[rj].c_point) |
1219 | 0 | { |
1220 | | /* replace only once */ |
1221 | 0 | ri = r; |
1222 | 0 | } |
1223 | 0 | } |
1224 | 0 | if (ri == *pri && at[rj].endpoint) |
1225 | 0 | { |
1226 | | /* !!! "&& at[rj].endpoint": only between 2 t-groups 2004-02-27; |
1227 | | the change disabled due to undiscovered yet possibility of ambiguity */ |
1228 | 0 | for (i = 0; i < num_atoms; i++) |
1229 | 0 | { |
1230 | 0 | if (at[i].endpoint != endpoint) |
1231 | 0 | { |
1232 | 0 | continue; |
1233 | 0 | } |
1234 | 0 | if (i == (int) *pri) |
1235 | 0 | { |
1236 | 0 | continue; |
1237 | 0 | } |
1238 | 0 | if (!at[i].c_point) |
1239 | 0 | { |
1240 | 0 | ri = (AT_NUMB) i; /* found neutral t-group representative */ |
1241 | 0 | break; |
1242 | 0 | } |
1243 | 0 | if (at[i].c_point != c_point && c_point == at[ri].c_point && |
1244 | 0 | at[i].c_point != at[rj].c_point) |
1245 | 0 | { |
1246 | | /* replace only once */ |
1247 | 0 | ri = (AT_NUMB) i; |
1248 | 0 | } |
1249 | 0 | } |
1250 | 0 | } |
1251 | 0 | } |
1252 | 0 | } |
1253 | 0 | } |
1254 | 0 | break; |
1255 | 0 | } |
1256 | 0 | *prj = rj; |
1257 | 0 | *pri = ri; |
1258 | 0 | } |
1259 | |
|
1260 | 0 | return 0; |
1261 | 0 | } |
1262 | | |
1263 | | |
1264 | | /****************************************************************************/ |
1265 | | int FindAccessibleEndPoints( CANON_GLOBALS *pCG, |
1266 | | T_ENDPOINT *EndPoint, |
1267 | | int *nNumEndPoints, |
1268 | | T_BONDPOS *BondPos, |
1269 | | int *nNumBondPos, |
1270 | | struct BalancedNetworkStructure *pBNS, |
1271 | | struct BalancedNetworkData *pBD, |
1272 | | inp_ATOM *at, |
1273 | | int num_atoms, |
1274 | | C_GROUP_INFO *cgi, |
1275 | | int taut_mode ) |
1276 | 0 | { |
1277 | 0 | AT_NUMB nTGroupRepresenative[MAXVAL], nTGroupEqu[MAXVAL], nTGEndPointNo[MAXVAL], ri, rj; |
1278 | 0 | AT_NUMB nCurTGroupNumber, nMaxTGroupNumber, nNumTgroupNumbers, nMaxEquNumber; |
1279 | 0 | int i, j, k, nNumDiffTGroupNumbers = 0, nNumFoundEqu, nErr; |
1280 | |
|
1281 | 0 | if (*nNumEndPoints != *nNumBondPos) |
1282 | 0 | return 0; |
1283 | | /* collect all group numbers. Fill EndPoint[i].nEquNumber */ |
1284 | 0 | for (i = 0; i < *nNumEndPoints; i++) |
1285 | 0 | { |
1286 | 0 | nCurTGroupNumber = EndPoint[i].nEquNumber = EndPoint[i].nGroupNumber; /* initial equivalence */ |
1287 | 0 | if (nCurTGroupNumber) |
1288 | 0 | { |
1289 | | /* found endpoint that already belongs to a t-group */ |
1290 | 0 | for (j = 0; j < nNumDiffTGroupNumbers; j++) |
1291 | 0 | { |
1292 | 0 | if (nTGroupEqu[j] == nCurTGroupNumber) |
1293 | 0 | { |
1294 | 0 | break; |
1295 | 0 | } |
1296 | 0 | } |
1297 | 0 | if (j == nNumDiffTGroupNumbers) |
1298 | 0 | { |
1299 | 0 | nTGroupRepresenative[nNumDiffTGroupNumbers] = EndPoint[i].nAtomNumber; |
1300 | 0 | nTGroupEqu[nNumDiffTGroupNumbers] = EndPoint[i].nGroupNumber; |
1301 | 0 | nTGEndPointNo[nNumDiffTGroupNumbers] = i; |
1302 | 0 | nNumDiffTGroupNumbers++; |
1303 | 0 | } |
1304 | 0 | } |
1305 | 0 | } |
1306 | | |
1307 | | /* check whether each pair belongs to the same t-group and establish the equivalence(s) */ |
1308 | 0 | for (i = 0, nNumFoundEqu = 0; i < nNumDiffTGroupNumbers; i++) |
1309 | 0 | { |
1310 | 0 | for (j = i + 1; j < nNumDiffTGroupNumbers; j++) |
1311 | 0 | { |
1312 | 0 | ri = nTGroupRepresenative[i]; |
1313 | 0 | rj = nTGroupRepresenative[j]; |
1314 | | /* both at[ri] and at[rj] are known to belong to tautomeric groups */ |
1315 | 0 | GetNeutralRepsIfNeeded( &ri, &rj, at, num_atoms, EndPoint, *nNumEndPoints, cgi ); |
1316 | 0 | nErr = bExistsAnyAltPath( pCG, pBNS, pBD, at, num_atoms, ri, rj, taut_mode ); |
1317 | 0 | if (IS_BNS_ERROR( nErr )) |
1318 | 0 | { |
1319 | 0 | return nErr; |
1320 | 0 | } |
1321 | 0 | if (0 == nErr) |
1322 | 0 | { |
1323 | 0 | continue; /* alt path between at[ri] and at[rj] not found */ |
1324 | 0 | } |
1325 | 0 | nCurTGroupNumber = inchi_min( nTGroupEqu[i], nTGroupEqu[j] ); |
1326 | 0 | nMaxTGroupNumber = inchi_max( nTGroupEqu[i], nTGroupEqu[j] ); |
1327 | 0 | for (k = 0; k < nNumDiffTGroupNumbers; k++) |
1328 | 0 | { |
1329 | 0 | if (nTGroupEqu[k] == nMaxTGroupNumber) |
1330 | 0 | { |
1331 | 0 | nTGroupEqu[k] = nCurTGroupNumber; |
1332 | 0 | nNumFoundEqu++; |
1333 | 0 | } |
1334 | 0 | } |
1335 | 0 | for (k = 0; k < *nNumEndPoints; k++) |
1336 | 0 | { |
1337 | 0 | if (EndPoint[k].nEquNumber == nMaxTGroupNumber) |
1338 | 0 | { |
1339 | 0 | EndPoint[k].nEquNumber = nCurTGroupNumber; |
1340 | 0 | } |
1341 | 0 | } |
1342 | 0 | } |
1343 | 0 | } |
1344 | | |
1345 | 0 | if (nNumFoundEqu) |
1346 | 0 | { |
1347 | | /* leave in only non-equivalent representatives */ |
1348 | 0 | for (i = 1, k = 0; i < nNumDiffTGroupNumbers; i++) |
1349 | 0 | { |
1350 | 0 | for (j = 0; j < i; j++) |
1351 | 0 | { |
1352 | 0 | if (nTGroupEqu[j] == nTGroupEqu[i]) |
1353 | 0 | { |
1354 | 0 | nTGroupEqu[i] = 0; /* i > j; mark equivalent for removal*/ |
1355 | 0 | break; |
1356 | 0 | } |
1357 | 0 | } |
1358 | 0 | } |
1359 | 0 | for (i = j = 0; i < nNumDiffTGroupNumbers; i++) |
1360 | 0 | { |
1361 | 0 | if (nTGroupEqu[i]) |
1362 | 0 | { |
1363 | 0 | if (i != j) |
1364 | 0 | { /* remove the marked */ |
1365 | 0 | nTGroupEqu[j] = nTGroupEqu[i]; |
1366 | 0 | nTGroupRepresenative[j] = nTGroupRepresenative[i]; |
1367 | 0 | nTGEndPointNo[j] = nTGEndPointNo[i]; |
1368 | 0 | } |
1369 | 0 | j++; |
1370 | 0 | } |
1371 | 0 | } |
1372 | 0 | nNumDiffTGroupNumbers = j; /* number of known t-group representatives */ |
1373 | 0 | } |
1374 | | |
1375 | | /* collect endpoints that have not been assigned to t-groups */ |
1376 | 0 | for (i = 0, j = nNumDiffTGroupNumbers; i < *nNumEndPoints; i++) |
1377 | 0 | { |
1378 | 0 | if (EndPoint[i].nEquNumber) |
1379 | 0 | { |
1380 | 0 | continue; |
1381 | 0 | } |
1382 | 0 | nTGroupEqu[j] = 0; |
1383 | 0 | nTGroupRepresenative[j] = EndPoint[i].nAtomNumber; |
1384 | 0 | nTGEndPointNo[j] = i; |
1385 | 0 | j++; |
1386 | 0 | } |
1387 | 0 | nNumTgroupNumbers = j; |
1388 | 0 | nMaxEquNumber = num_atoms + 1; /* impossible atom or t-group number */ |
1389 | | |
1390 | | /* check whether each pair belongs to the same group and establish the equivalence(s) */ |
1391 | 0 | for (i = 0, nNumFoundEqu = 0; i < nNumTgroupNumbers; i++) |
1392 | 0 | { |
1393 | 0 | for (j = i + 1; j < nNumTgroupNumbers; j++) |
1394 | 0 | { |
1395 | 0 | if (nTGroupEqu[i] != nTGroupEqu[j] && ( i >= nNumDiffTGroupNumbers || j >= nNumDiffTGroupNumbers ) || |
1396 | | /* equivalence of a t-group and a non-t-group atom */ |
1397 | 0 | !nTGroupEqu[i] && !nTGroupEqu[j] |
1398 | | /* equivalence of two non-t-group atoms */ |
1399 | 0 | ) |
1400 | 0 | { |
1401 | 0 | ri = nTGroupRepresenative[i]; |
1402 | 0 | rj = nTGroupRepresenative[j]; |
1403 | | /*------------------------------!!!--------------------------------------------- |
1404 | | Explanation why GetNeutralRepsIfNeeded() may need to be changed 2004-02-27 |
1405 | | The change has been disabled due to undiscovered yet possibility of ambiguity |
1406 | | to search for neutral only among EndPoint[] in case taut-not_taut pairs |
1407 | | |
1408 | | Counterexample: O=C-NH(+)=C-NH2 |
1409 | | 1 2 3 |
1410 | | |
1411 | | Has already been found: 2-3 (+)-charge exchange |
1412 | | 1-2 tautomerism (charge removed to 3) |
1413 | | Now testing: 2-3 tautomerism. If not commented out, |
1414 | | GetNeutralRepsIfNeeded() would replace 2-3 test with 1-3 test because: |
1415 | | o Charge group has only one neutral and both 2 and 3 belong to it, |
1416 | | therefore we cannot neutralize both; search for neutral representative; |
1417 | | o Since 1 and 2 belong to the same t-group and 1 is neutral, |
1418 | | test 1-3 instead of 2-3. |
1419 | | This breaks our condition: |
1420 | | Test tautomeric H movement only between neutral atoms. |
1421 | | -----------------------------------------------------------------------------*/ |
1422 | 0 | GetNeutralRepsIfNeeded( &ri, &rj, at, num_atoms, EndPoint, *nNumEndPoints, cgi ); |
1423 | |
|
1424 | 0 | nErr = bExistsAnyAltPath( pCG, pBNS, pBD, at, num_atoms, ri, rj, taut_mode ); |
1425 | 0 | if (IS_BNS_ERROR( nErr )) |
1426 | 0 | { |
1427 | 0 | return nErr; |
1428 | 0 | } |
1429 | 0 | if (nErr <= 0) |
1430 | 0 | { |
1431 | 0 | continue; |
1432 | 0 | } |
1433 | 0 | if (nTGroupEqu[i] && nTGroupEqu[j]) |
1434 | 0 | { |
1435 | | /* found equivalence of two t-groups; at least one of them must be a new one */ |
1436 | 0 | nCurTGroupNumber = inchi_min( nTGroupEqu[i], nTGroupEqu[j] ); |
1437 | 0 | nMaxTGroupNumber = inchi_max( nTGroupEqu[i], nTGroupEqu[j] ); |
1438 | 0 | for (k = 0; k < nNumTgroupNumbers; k++) |
1439 | 0 | { |
1440 | 0 | if (nTGroupEqu[k] == nMaxTGroupNumber) |
1441 | 0 | { |
1442 | 0 | nTGroupEqu[k] = nCurTGroupNumber; |
1443 | 0 | nNumFoundEqu++; |
1444 | 0 | } |
1445 | 0 | } |
1446 | 0 | for (k = 0; k < *nNumEndPoints; k++) |
1447 | 0 | { |
1448 | 0 | if (EndPoint[k].nEquNumber == nMaxTGroupNumber) |
1449 | 0 | { |
1450 | 0 | EndPoint[k].nEquNumber = nCurTGroupNumber; |
1451 | 0 | } |
1452 | 0 | } |
1453 | 0 | } |
1454 | 0 | else |
1455 | 0 | { |
1456 | 0 | if (nTGroupEqu[i]) |
1457 | 0 | { /* extend existing t-group */ |
1458 | 0 | nTGroupEqu[j] = nTGroupEqu[i]; |
1459 | 0 | EndPoint[nTGEndPointNo[j]].nEquNumber = nTGroupEqu[i]; |
1460 | 0 | } |
1461 | 0 | else |
1462 | 0 | { |
1463 | 0 | if (nTGroupEqu[j]) |
1464 | 0 | { /* extend existing t-group */ |
1465 | 0 | nTGroupEqu[i] = nTGroupEqu[j]; |
1466 | 0 | EndPoint[nTGEndPointNo[i]].nEquNumber = nTGroupEqu[j]; |
1467 | 0 | } |
1468 | 0 | else |
1469 | 0 | { /* establis a new t-group */ |
1470 | 0 | nTGroupEqu[i] = |
1471 | 0 | nTGroupEqu[j] = nMaxEquNumber; /* assign a fict. ID to establish equivalence */ |
1472 | 0 | EndPoint[nTGEndPointNo[i]].nEquNumber = |
1473 | 0 | EndPoint[nTGEndPointNo[j]].nEquNumber = nMaxEquNumber; |
1474 | 0 | nMaxEquNumber++; |
1475 | 0 | } |
1476 | 0 | } |
1477 | 0 | } |
1478 | 0 | } |
1479 | 0 | } |
1480 | 0 | } |
1481 | | |
1482 | | /* eliminate endpoints and bonds that do not belong to t-group(s) |
1483 | | (they have not been found connected by an alt path to any other endpoint) |
1484 | | */ |
1485 | 0 | for (i = 0, j = 0; i < *nNumEndPoints; i++) |
1486 | 0 | { |
1487 | 0 | if (EndPoint[i].nEquNumber) |
1488 | 0 | { |
1489 | 0 | #if ( IGNORE_SINGLE_ENDPOINTS == 1 ) /* 1-28-2003 */ |
1490 | 0 | for (k = 0, nNumFoundEqu = 0; k < *nNumEndPoints; k++) |
1491 | 0 | { |
1492 | 0 | nNumFoundEqu += ( EndPoint[i].nEquNumber == EndPoint[k].nEquNumber ); |
1493 | 0 | } |
1494 | 0 | if (nNumFoundEqu <= 1) |
1495 | 0 | { |
1496 | | /* one time it is equal to itself when i == k above */ |
1497 | | /* if EndPoint[i] is not "equivalent" to any other EndPoint then ignore it */ |
1498 | 0 | continue; |
1499 | 0 | } |
1500 | 0 | #endif |
1501 | 0 | if (i != j) |
1502 | 0 | { |
1503 | | /* save endpoints that are found to be connected to other endpoints by alt paths */ |
1504 | 0 | EndPoint[j] = EndPoint[i]; |
1505 | 0 | BondPos[j] = BondPos[i]; |
1506 | 0 | } |
1507 | 0 | j++; |
1508 | 0 | } |
1509 | 0 | } |
1510 | |
|
1511 | | #if ( IGNORE_SINGLE_ENDPOINTS != 1 ) /* 1-28-2003 */ |
1512 | | /* Do not allow a centerpoint to have only one tautomeric bond */ |
1513 | | /* Hack: we may have only one centerpoint */ |
1514 | | /* BondPos[*].nAtomNumber are centerpoints */ |
1515 | | if (j == 1) |
1516 | | { |
1517 | | /* check if there exist other centerpoint neighbors |
1518 | | * connected to it by another tautomeric-bond |
1519 | | */ |
1520 | | for (i = 0, k = 0; i < at[BondPos[0].nAtomNumber].valence; i++) |
1521 | | { |
1522 | | k += ( i != BondPos[0].neighbor_index && |
1523 | | BOND_TAUTOM == ( at[BondPos[0].nAtomNumber].bond_type[i] & ~BOND_MARK_ALL ) ); |
1524 | | } |
1525 | | if (!k) |
1526 | | { |
1527 | | j = 0; |
1528 | | } |
1529 | | } |
1530 | | #endif |
1531 | |
|
1532 | 0 | *nNumEndPoints = *nNumBondPos = j; |
1533 | |
|
1534 | 0 | return j; |
1535 | 0 | } |
1536 | | |
1537 | | |
1538 | | /****************************************************************************/ |
1539 | | /*#if ( MOVE_CHARGES == 1 ) */ /* { */ |
1540 | | /****************************************************************************/ |
1541 | | |
1542 | | /**********************************************/ |
1543 | | /* */ |
1544 | | /* definitions for positive ion recognition */ |
1545 | | /* */ |
1546 | | /**********************************************/ |
1547 | | |
1548 | | |
1549 | | /****************************************************************************/ |
1550 | | typedef struct tagChargeType |
1551 | | { |
1552 | | /* meaning see in bCanBeACPoint() */ |
1553 | | char elname[3]; |
1554 | | S_CHAR charge; |
1555 | | S_CHAR neutral_valence; |
1556 | | S_CHAR neutral_bonds_valence; /* valence of a neutral atom */ |
1557 | | S_CHAR cChangeValence; /* charge increases valence by this value */ |
1558 | | S_CHAR cChargeType; /* different types are treated separately */ |
1559 | | S_CHAR num_bonds; /* added 02-06-2005 */ |
1560 | | } CHARGE_TYPE; |
1561 | | |
1562 | | const CHARGE_TYPE CType[] = |
1563 | | { |
1564 | | { "N\0", 1, 3, 3, 1, 0, 0 }, |
1565 | | { "P\0", 1, 3, 3, 1, 1, 0 }, |
1566 | | #if ( ADD_MOVEABLE_O_PLUS == 1 ) |
1567 | | { "O\0", 1, 2, 2, 1, 2, 2 }, /* added 02-06-2005 */ |
1568 | | { "S\0", 1, 2, 2, 1, 3, 2 }, /* added 03-18-2005 */ |
1569 | | { "Se", 1, 2, 2, 1, 4, 2 }, /* added 03-18-2005 */ |
1570 | | { "Te", 1, 2, 2, 1, 5, 2 }, /* added 03-18-2005 */ |
1571 | | #endif |
1572 | | }; |
1573 | | |
1574 | | /* bits */ |
1575 | | |
1576 | 0 | #define C_SUBTYPE_CHARGED 0 |
1577 | 0 | #define C_SUBTYPE_p_DONOR 1 /* new */ |
1578 | 0 | #define C_SUBTYPE_p_ACCEPT 2 /* new */ |
1579 | 2 | #define C_SUBTYPE_H_ACCEPT 4 |
1580 | 0 | #define C_SUBTYPE_H_DONOR 8 |
1581 | 2 | #define C_SUBTYPE_NEUTRAL 16 |
1582 | | |
1583 | | /* make sure any C_SUBTYPE_CHARGED_... < any C_SUBTYPE_NEUTRAL_... */ |
1584 | | /* charged */ |
1585 | 0 | #define C_SUBTYPE_CHARGED_NON_TAUT (C_SUBTYPE_CHARGED) |
1586 | 0 | #define C_SUBTYPE_CHARGED_p_DONOR (C_SUBTYPE_CHARGED|C_SUBTYPE_p_DONOR) |
1587 | 0 | #define C_SUBTYPE_CHARGED_H_ACCEPT (C_SUBTYPE_CHARGED|C_SUBTYPE_H_ACCEPT) |
1588 | 0 | #define C_SUBTYPE_CHARGED_H_ACCEPT_p_DONOR (C_SUBTYPE_CHARGED|C_SUBTYPE_H_ACCEPT|C_SUBTYPE_p_DONOR) |
1589 | 0 | #define C_SUBTYPE_CHARGED_H_DONOR (C_SUBTYPE_CHARGED|C_SUBTYPE_H_DONOR |C_SUBTYPE_p_DONOR) |
1590 | | /* neutral */ |
1591 | 0 | #define C_SUBTYPE_NEUTRAL_NON_TAUT (C_SUBTYPE_NEUTRAL) |
1592 | 4 | #define C_SUBTYPE_NEUTRAL_H_ACCEPT (C_SUBTYPE_NEUTRAL|C_SUBTYPE_H_ACCEPT) |
1593 | 0 | #define C_SUBTYPE_NEUTRAL_H_ACCEPT_p_ACCEPT (C_SUBTYPE_NEUTRAL|C_SUBTYPE_H_ACCEPT|C_SUBTYPE_p_ACCEPT) |
1594 | 0 | #define C_SUBTYPE_NEUTRAL_H_DONOR (C_SUBTYPE_NEUTRAL|C_SUBTYPE_H_DONOR) |
1595 | | |
1596 | 25.7k | #define NUM_C_TYPES (int)(sizeof( CType )/sizeof(CType[0])) |
1597 | | |
1598 | | |
1599 | | /****************************************************************************/ |
1600 | | int bCanBeACPoint( inp_ATOM *at, |
1601 | | S_CHAR cCharge, |
1602 | | S_CHAR cChangeValence, |
1603 | | S_CHAR neutral_bonds_valence, |
1604 | | S_CHAR neutral_valence, |
1605 | | S_CHAR nEndpointValence, |
1606 | | S_CHAR *cChargeSubtype ) |
1607 | 480 | { |
1608 | 480 | int nChangeValence; |
1609 | 480 | int nNumBonds; |
1610 | 480 | int nBondsValence; |
1611 | 480 | int bNegCharge = ( at->charge == -1 ); /* add fict. bonds to (-) 2004-02-24*/ |
1612 | | |
1613 | 480 | if (at->charge == cCharge && at->valence == at->chem_bonds_valence && at->num_H) |
1614 | 0 | { |
1615 | | /* proton donors candidates >NH(+)-, >NH2(+), -NH3(+), >OH(+), -OH2(+) */ |
1616 | | /* charged, added p-transfer -- 01-28-2004 */ |
1617 | 0 | nChangeValence = at->charge * cChangeValence; /* +1 or -1; currently only +1 */ |
1618 | 0 | nBondsValence = at->chem_bonds_valence + at->num_H; |
1619 | 0 | if (nBondsValence == neutral_bonds_valence + nChangeValence && nEndpointValence) |
1620 | 0 | { |
1621 | 0 | *cChargeSubtype = C_SUBTYPE_CHARGED_p_DONOR; /* ignore Phosphorus p-donors for now */ |
1622 | 0 | } |
1623 | 0 | return 0; |
1624 | 0 | } |
1625 | | |
1626 | 480 | else |
1627 | 480 | { |
1628 | 480 | if (at->charge == cCharge && at->valence < at->chem_bonds_valence) |
1629 | 0 | { |
1630 | | /* the requirement at->valence < at->chem_bonds_valence rejects |
1631 | | candidates >NH(+)-, >NH2(+), -NH3(+), >N(+)<, >OH(+), -OH2(+), >O(+)- |
1632 | | Moveable charge requires double bonds; these ions have no double bonds |
1633 | | */ |
1634 | | |
1635 | | /* charged */ |
1636 | 0 | nChangeValence = at->charge * cChangeValence; /* +1 or -1; currently only +1 */ |
1637 | 0 | nBondsValence = at->chem_bonds_valence + at->num_H; |
1638 | 0 | nNumBonds = at->valence + at->num_H; |
1639 | 0 | if (nBondsValence == neutral_bonds_valence + nChangeValence) |
1640 | 0 | { |
1641 | | /* known valence */ |
1642 | 0 | if (nNumBonds == neutral_valence) |
1643 | 0 | { |
1644 | | /* non-tautomeric: >N(+)=, =O(+)- |
1645 | | possibly tautomeric donor: =NH(+)-, =NH2(+), =OH(+) */ |
1646 | 0 | if (at->valence == neutral_valence || !nEndpointValence) |
1647 | 0 | { |
1648 | | /* non-tautomeric: >N(+)=, =O(+)-; any suitable P+: >P(+)=, =PH(+)-, =PH2(+) */ |
1649 | 0 | *cChargeSubtype = C_SUBTYPE_CHARGED_NON_TAUT; |
1650 | 0 | } |
1651 | 0 | else |
1652 | 0 | { |
1653 | | /* possibly tautomeric donor: =NH(+)-, =NH2(+), =OH(+) */ |
1654 | 0 | *cChargeSubtype = C_SUBTYPE_CHARGED_H_DONOR; |
1655 | 0 | } |
1656 | 0 | return 1; |
1657 | 0 | } |
1658 | 0 | if (nNumBonds == neutral_valence - 1) |
1659 | 0 | { |
1660 | | /* possibly tutomeric acceptor: =N(+)=, #N(+)-, #NH(+), #O(+) */ |
1661 | 0 | if (nEndpointValence) |
1662 | 0 | { |
1663 | 0 | *cChargeSubtype = at->num_H ? C_SUBTYPE_CHARGED_H_ACCEPT_p_DONOR : C_SUBTYPE_CHARGED_H_ACCEPT; |
1664 | 0 | } |
1665 | 0 | else |
1666 | 0 | { |
1667 | | /* =P(+)=, #P(+)-, #PH(+) */ |
1668 | 0 | *cChargeSubtype = C_SUBTYPE_CHARGED_NON_TAUT; |
1669 | 0 | } |
1670 | 0 | return 1; /* charge type, charged */ |
1671 | 0 | } |
1672 | 0 | } |
1673 | 0 | } |
1674 | 480 | else |
1675 | 480 | { |
1676 | 480 | if (at->charge == 0 || bNegCharge) |
1677 | 480 | { |
1678 | | /* neutral atom or anion, all bonds are single */ |
1679 | 480 | nBondsValence = at->chem_bonds_valence + at->num_H + bNegCharge; /* add fict. bonds to (-) 2004-02-24*/ |
1680 | 480 | nNumBonds = at->valence + at->num_H + bNegCharge; /* add fict. bonds to (-) 2004-02-24*/ |
1681 | 480 | if (nBondsValence == neutral_bonds_valence) |
1682 | 2 | { |
1683 | 2 | if (nNumBonds == neutral_valence) |
1684 | 2 | { |
1685 | | /* only single bonds: >N-, >NH, -NH2, -O-, -OH, >P- >PH -PH2 */ |
1686 | | /* >N(-), -NH(-), -O(-). >P(-) -PH(-) */ |
1687 | 2 | { |
1688 | 2 | if (at->valence == neutral_valence || !nEndpointValence) |
1689 | 0 | { |
1690 | | /* >N-, -O-, any P(3 single bonds): >P- >PH -PH2 */ |
1691 | 0 | *cChargeSubtype = C_SUBTYPE_NEUTRAL_NON_TAUT; |
1692 | 0 | } |
1693 | 2 | else |
1694 | 2 | { |
1695 | 2 | if (at->valence < neutral_valence /*&& nEndpointValence */) |
1696 | 2 | { |
1697 | | /* num_H > 0: >NH -NH2 -OH */ |
1698 | | /* num_H = 0: none C_SUBTYPE_NEUTRAL_H_ACCEPT for now */ |
1699 | 2 | *cChargeSubtype = at->num_H ? C_SUBTYPE_NEUTRAL_H_DONOR : C_SUBTYPE_NEUTRAL_H_ACCEPT; |
1700 | 2 | } |
1701 | 0 | else |
1702 | 0 | { |
1703 | 0 | return 0; |
1704 | 0 | } |
1705 | 2 | } |
1706 | 2 | } |
1707 | 2 | return 1; /* charge type, neutral */ |
1708 | 2 | } |
1709 | 0 | if (nNumBonds == neutral_valence - 1) |
1710 | 0 | { |
1711 | | /* possibly tautomeric acceptor =N-, =NH, =O or non-taut =P-, =PH */ |
1712 | 0 | if (nEndpointValence) |
1713 | 0 | { |
1714 | | /* =N-, =NH, =O */ |
1715 | 0 | *cChargeSubtype = C_SUBTYPE_NEUTRAL_H_ACCEPT_p_ACCEPT; |
1716 | 0 | } |
1717 | 0 | else |
1718 | 0 | { |
1719 | | /* =P-, =PH */ |
1720 | 0 | *cChargeSubtype = C_SUBTYPE_NEUTRAL_NON_TAUT; |
1721 | 0 | } |
1722 | 0 | return 1; /* charge type, (+) => neutral */ |
1723 | 0 | } |
1724 | 0 | } |
1725 | 480 | } |
1726 | 480 | } |
1727 | 480 | } |
1728 | | |
1729 | 478 | return 0; |
1730 | 480 | } |
1731 | | |
1732 | | |
1733 | | /****************************************************************************/ |
1734 | | int GetChargeType( inp_ATOM *atom, int iat, S_CHAR *cChargeSubtype ) |
1735 | 3.68k | { |
1736 | 3.68k | int i, n; |
1737 | 3.68k | S_CHAR nEndpointValence; |
1738 | 3.68k | inp_ATOM *at = atom + iat; |
1739 | | |
1740 | 3.68k | *cChargeSubtype = 0; |
1741 | | /* ignore ion pairs and charges != 1 */ |
1742 | 3.68k | if (abs( at->charge ) == 1) |
1743 | 22 | { |
1744 | 56 | for (i = 0; i < at->valence; i++) |
1745 | 34 | { |
1746 | 34 | n = at->neighbor[i]; |
1747 | | /* allow negatively charged tautomeric neighbors 2004-02-26 */ |
1748 | 34 | if (abs( atom[n].charge + at->charge ) < abs( atom[n].charge - at->charge ) && !atom[n].endpoint) |
1749 | 0 | { |
1750 | 0 | return -1; /* charges have different signs */ |
1751 | 0 | } |
1752 | 34 | } |
1753 | 22 | } |
1754 | 3.66k | else |
1755 | 3.66k | { |
1756 | 3.66k | if (at->charge) |
1757 | 8 | { |
1758 | 8 | return -1; /* abs(charge) != 1 */ |
1759 | 8 | } |
1760 | 3.66k | } |
1761 | | |
1762 | | /* find candidates */ |
1763 | 25.7k | for (i = 0; i < NUM_C_TYPES; i++) |
1764 | 22.0k | { |
1765 | 22.0k | if (!strcmp( at->elname, CType[i].elname ) && |
1766 | 22.0k | ( !CType[i].num_bonds || CType[i].num_bonds == at->valence && at->nNumAtInRingSystem >= 5 )) |
1767 | 480 | { |
1768 | 480 | nEndpointValence = (S_CHAR) get_endpoint_valence( at->el_number ); |
1769 | 480 | if (bCanBeACPoint( at, CType[i].charge, CType[i].cChangeValence, CType[i].neutral_bonds_valence, |
1770 | 480 | CType[i].neutral_valence, nEndpointValence, cChargeSubtype )) |
1771 | 2 | { |
1772 | 2 | return CType[i].cChargeType; |
1773 | 2 | } |
1774 | 480 | } |
1775 | 22.0k | } |
1776 | | |
1777 | 3.67k | return -1; |
1778 | 3.67k | } |
1779 | | |
1780 | | |
1781 | | /****************************************************************************/ |
1782 | | int CmpCCandidates( const void *a1, const void *a2 ) |
1783 | 0 | { |
1784 | 0 | const C_CANDIDATE *c1 = (const C_CANDIDATE *) a1; |
1785 | 0 | const C_CANDIDATE *c2 = (const C_CANDIDATE *) a2; |
1786 | 0 | int ret; |
1787 | 0 | if (ret = (int) c1->type - (int) c2->type) |
1788 | 0 | { |
1789 | 0 | return ret; |
1790 | 0 | } |
1791 | 0 | if (ret = (int) c1->subtype - (int) c2->subtype) |
1792 | 0 | { |
1793 | 0 | return ret; |
1794 | 0 | } |
1795 | 0 | ret = (int) c1->atnumber - (int) c2->atnumber; |
1796 | |
|
1797 | 0 | return ret; |
1798 | 0 | } |
1799 | | |
1800 | | |
1801 | | /****************************************************************************/ |
1802 | | int RegisterCPoints( C_GROUP *c_group, |
1803 | | int *pnum_c, |
1804 | | int max_num_c, |
1805 | | T_GROUP_INFO *t_group_info, |
1806 | | int point1, |
1807 | | int point2, |
1808 | | int ctype, |
1809 | | inp_ATOM *at, |
1810 | | int num_atoms ) |
1811 | 0 | { |
1812 | 0 | int num_c = *pnum_c, i, i1, i2; |
1813 | 0 | AT_NUMB nGroupNumber = 0, nNewGroupNumber; |
1814 | |
|
1815 | 0 | if (at[point1].c_point == at[point2].c_point) |
1816 | 0 | { |
1817 | 0 | if (at[point1].c_point) |
1818 | 0 | { |
1819 | 0 | return 0; |
1820 | 0 | } |
1821 | 0 | memset( c_group + num_c, 0, sizeof( c_group[0] ) ); |
1822 | 0 | if (num_c < max_num_c) |
1823 | 0 | { |
1824 | 0 | c_group[num_c].num[0] = CHARGED_CPOINT( at, point1 ) + CHARGED_CPOINT( at, point2 ); |
1825 | 0 | c_group[num_c].num_CPoints += 2; |
1826 | 0 | c_group[num_c].cGroupType = ctype; |
1827 | | /* get next available c-group number */ |
1828 | 0 | for (i = 0; i < num_c; i++) |
1829 | 0 | { |
1830 | 0 | if (nGroupNumber < c_group[i].nGroupNumber) |
1831 | 0 | { |
1832 | 0 | nGroupNumber = c_group[i].nGroupNumber; |
1833 | 0 | } |
1834 | 0 | } |
1835 | 0 | nGroupNumber++; |
1836 | 0 | c_group[num_c].nGroupNumber = |
1837 | 0 | at[point1].c_point = |
1838 | 0 | at[point2].c_point = nGroupNumber; |
1839 | 0 | *pnum_c = num_c + 1; |
1840 | | /* count protons */ |
1841 | 0 | if (at[point1].num_H) |
1842 | 0 | { |
1843 | 0 | c_group[num_c].num[1] ++; |
1844 | 0 | } |
1845 | 0 | else |
1846 | 0 | { |
1847 | 0 | if (at[point2].num_H) |
1848 | 0 | { |
1849 | 0 | c_group[num_c].num[1] ++; |
1850 | 0 | } |
1851 | 0 | else |
1852 | 0 | { |
1853 | 0 | if (( at[point1].endpoint || at[point2].endpoint ) && t_group_info && t_group_info->t_group && t_group_info->num_t_groups) |
1854 | 0 | { |
1855 | | /* !!! add later !!! */ |
1856 | 0 | } |
1857 | 0 | } |
1858 | 0 | } |
1859 | 0 | return 1; |
1860 | 0 | } |
1861 | | |
1862 | 0 | return BNS_CPOINT_ERR; /* overflow */ |
1863 | 0 | } |
1864 | | |
1865 | 0 | if (at[point1].c_point > at[point2].c_point) |
1866 | 0 | { |
1867 | | /* make sure at[point1].c_point < at[point2].c_point */ |
1868 | 0 | i = point1; |
1869 | 0 | point1 = point2; |
1870 | 0 | point2 = i; |
1871 | 0 | } |
1872 | |
|
1873 | 0 | if (!at[point1].c_point) |
1874 | 0 | { |
1875 | | /* add a new c-endpoint to an existing c-group */ |
1876 | 0 | nGroupNumber = at[point2].c_point; |
1877 | 0 | for (i = 0; i < num_c; i++) |
1878 | 0 | { |
1879 | 0 | if (nGroupNumber == c_group[i].nGroupNumber) |
1880 | 0 | { |
1881 | 0 | at[point1].c_point = at[point2].c_point; |
1882 | 0 | c_group[i].num_CPoints++; |
1883 | 0 | c_group[i].num[0] += CHARGED_CPOINT( at, point1 ); |
1884 | 0 | return 1; |
1885 | 0 | } |
1886 | 0 | } |
1887 | 0 | return BNS_CPOINT_ERR; /* program error: c-group not found */ |
1888 | 0 | } |
1889 | 0 | else |
1890 | 0 | { |
1891 | | /* merge two c-groups */ |
1892 | 0 | nNewGroupNumber = at[point1].c_point; |
1893 | 0 | nGroupNumber = at[point2].c_point; |
1894 | 0 | for (i = 0, i1 = i2 = -1; i < num_c && ( i1 < 0 || i2 < 0 ); i++) |
1895 | 0 | { |
1896 | 0 | if (nNewGroupNumber == c_group[i].nGroupNumber) |
1897 | 0 | { |
1898 | 0 | i1 = i; |
1899 | 0 | continue; |
1900 | 0 | } |
1901 | 0 | if (nGroupNumber == c_group[i].nGroupNumber) |
1902 | 0 | { |
1903 | 0 | i2 = i; |
1904 | 0 | continue; |
1905 | 0 | } |
1906 | 0 | } |
1907 | 0 | if (i1 < 0 || i2 < 0) |
1908 | 0 | { |
1909 | 0 | return BNS_CPOINT_ERR; /* at least one not found */ |
1910 | 0 | } |
1911 | | |
1912 | 0 | c_group[i1].num[0] += c_group[i2].num[0]; |
1913 | 0 | c_group[i1].num_CPoints += c_group[i2].num_CPoints; |
1914 | 0 | num_c--; |
1915 | 0 | if (num_c > i2) |
1916 | 0 | { |
1917 | 0 | memmove( c_group + i2, c_group + i2 + 1, ( num_c - i2 ) * sizeof( c_group[0] ) ); |
1918 | 0 | } |
1919 | 0 | *pnum_c = num_c; |
1920 | | /* renumber c-groups */ |
1921 | 0 | for (i = 0; i < num_c; i++) |
1922 | 0 | { |
1923 | 0 | if (c_group[i].nGroupNumber > nGroupNumber) |
1924 | 0 | { |
1925 | 0 | c_group[i].nGroupNumber--; |
1926 | 0 | } |
1927 | 0 | } |
1928 | | /* renumber c-points */ |
1929 | 0 | for (i = 0; i < num_atoms; i++) |
1930 | 0 | { |
1931 | 0 | if (at[i].c_point > nGroupNumber) |
1932 | 0 | { |
1933 | 0 | at[i].c_point--; |
1934 | 0 | } |
1935 | 0 | else |
1936 | 0 | { |
1937 | 0 | if (at[i].c_point == nGroupNumber) |
1938 | 0 | { |
1939 | 0 | at[i].c_point = nNewGroupNumber; |
1940 | 0 | } |
1941 | 0 | } |
1942 | 0 | } |
1943 | 0 | return 1; |
1944 | 0 | } |
1945 | |
|
1946 | 0 | } |
1947 | | |
1948 | | |
1949 | | /****************************************************************************/ |
1950 | | int MarkChargeGroups( struct tagCANON_GLOBALS *pCG, |
1951 | | inp_ATOM *at, |
1952 | | int num_atoms, |
1953 | | C_GROUP_INFO *c_group_info, |
1954 | | T_GROUP_INFO *t_group_info, |
1955 | | struct BalancedNetworkStructure *pBNS, |
1956 | | struct BalancedNetworkData *pBD ) |
1957 | 110k | { |
1958 | 110k | int nNumChanges = 0; |
1959 | | |
1960 | 110k | if (c_group_info && c_group_info->c_candidate && c_group_info->max_num_candidates > 0) |
1961 | 806 | { |
1962 | 806 | int i, i1, i2, i3, j, num_tested; |
1963 | 806 | C_CANDIDATE *c_candidate = c_group_info->c_candidate; |
1964 | 806 | int nMaxNumCandidates = c_group_info->max_num_candidates; |
1965 | 806 | int nNumCandidates = c_group_info->num_candidates; |
1966 | 806 | S_CHAR c_type, c_subtype; |
1967 | 806 | int iat1, iat2, ret, nDelta; |
1968 | | |
1969 | 806 | if (nNumCandidates == -1) |
1970 | 0 | { |
1971 | 0 | nNumCandidates = 0; /* 2004-02-26 they could appear after t-group discovery */ |
1972 | | /*return 0;*/ |
1973 | 0 | } |
1974 | 806 | if (nNumCandidates == 0) |
1975 | 806 | { |
1976 | 4.49k | for (i = 0, nNumCandidates = 0; i < num_atoms; i++) |
1977 | 3.68k | { |
1978 | 3.68k | if (0 <= ( c_type = GetChargeType( at, i, &c_subtype ) )) |
1979 | 2 | { |
1980 | 2 | if (nNumCandidates >= nMaxNumCandidates) |
1981 | 0 | { |
1982 | 0 | return BNS_VERT_EDGE_OVFL; |
1983 | 0 | } |
1984 | 2 | c_candidate[nNumCandidates].atnumber = i; |
1985 | 2 | c_candidate[nNumCandidates].type = c_type; |
1986 | 2 | c_candidate[nNumCandidates].subtype = c_subtype; |
1987 | 2 | nNumCandidates++; |
1988 | 2 | } |
1989 | 3.68k | } |
1990 | 806 | if (nNumCandidates <= 1) |
1991 | 806 | { |
1992 | 806 | c_group_info->num_candidates = -1; /* no candidate exists */ |
1993 | 806 | return 0; |
1994 | 806 | } |
1995 | 806 | } |
1996 | | /* sorting keys: (1) atom type (N,P); (2) uncharged=16/charged=0; (3) other; |
1997 | | atom-charged-N .... i1 |
1998 | | ... |
1999 | | atom-charged-N |
2000 | | atom-neutral-N .... i2 |
2001 | | ... |
2002 | | atom-neutral-N |
2003 | | atom-charged-P .... i3 ... i1 |
2004 | | ... |
2005 | | atom-charged-P |
2006 | | atom-neutral-P ........... i2 |
2007 | | ... |
2008 | | atom-neutral-P |
2009 | | end. ........... i3 |
2010 | | */ |
2011 | | |
2012 | 0 | qsort( c_candidate, nNumCandidates, sizeof( c_candidate[0] ), CmpCCandidates ); |
2013 | |
|
2014 | 0 | i1 = 0; |
2015 | 0 | num_tested = 0; |
2016 | 0 | nDelta = 0; |
2017 | |
|
2018 | 0 | while (i1 < nNumCandidates) |
2019 | 0 | { |
2020 | | /* the the first charged candidate of a new atom type */ |
2021 | 0 | for (; i1 < nNumCandidates && ( c_candidate[i1].subtype & C_SUBTYPE_NEUTRAL ); i1++) |
2022 | 0 | { |
2023 | 0 | ; |
2024 | 0 | } |
2025 | 0 | if (i1 == nNumCandidates) |
2026 | 0 | { |
2027 | 0 | break; /* not found */ |
2028 | 0 | } |
2029 | | |
2030 | | /* bypass other charged candidates of the same atom type */ |
2031 | 0 | for (i2 = i1 + 1; i2 < nNumCandidates && |
2032 | 0 | c_candidate[i2].type == c_candidate[i1].type && |
2033 | 0 | !( c_candidate[i2].subtype & C_SUBTYPE_NEUTRAL ); |
2034 | 0 | i2++) |
2035 | 0 | { |
2036 | 0 | ; |
2037 | 0 | } |
2038 | 0 | if (i2 == nNumCandidates) |
2039 | 0 | { |
2040 | 0 | break; /* no neutral candidates */ |
2041 | 0 | } |
2042 | | |
2043 | | /* find next to the last neutral candidate of the same atom type */ |
2044 | 0 | for (i3 = i2; i3 < nNumCandidates && |
2045 | 0 | c_candidate[i3].type == c_candidate[i1].type; |
2046 | 0 | i3++) |
2047 | 0 | { |
2048 | 0 | ; |
2049 | 0 | } |
2050 | |
|
2051 | 0 | if (i3 == i2) |
2052 | 0 | { |
2053 | | /* no neutral candidates found */ |
2054 | 0 | if (i2 < nNumCandidates) |
2055 | 0 | { |
2056 | 0 | i1 = i3; |
2057 | 0 | continue; /* move to the next atom type */ |
2058 | 0 | } |
2059 | 0 | break; /* nothing more to do */ |
2060 | 0 | } |
2061 | | |
2062 | | /* found charged candidates: i1...i2-1; neutral candidates: i2...i3-1 */ |
2063 | 0 | for (i = i1; i < i2; i++) |
2064 | 0 | { |
2065 | 0 | iat1 = c_candidate[i].atnumber; |
2066 | 0 | for (j = i2; j < i3; j++) |
2067 | 0 | { |
2068 | | /* check alt path at[iat1]=-=-...-at[iat2]; at[iat1] is charged, at[iat2] is neutral */ |
2069 | 0 | num_tested++; |
2070 | 0 | iat2 = c_candidate[j].atnumber; |
2071 | 0 | if (at[iat1].c_point && at[iat1].c_point == at[iat2].c_point) |
2072 | 0 | { |
2073 | 0 | continue; |
2074 | 0 | } |
2075 | | |
2076 | 0 | ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, iat1, iat2, ALT_PATH_MODE_CHARGE ); |
2077 | |
|
2078 | 0 | if (IS_BNS_ERROR( ret )) |
2079 | 0 | { |
2080 | 0 | return ret; |
2081 | 0 | } |
2082 | 0 | if (ret & 1) |
2083 | 0 | { |
2084 | 0 | nDelta = ( ret & ~3 ) >> 2; |
2085 | 0 | nNumChanges += ( ret & 2 ); |
2086 | |
|
2087 | 0 | ret = RegisterCPoints( c_group_info->c_group, &c_group_info->num_c_groups, |
2088 | 0 | c_group_info->max_num_c_groups, t_group_info, |
2089 | 0 | iat1, iat2, c_candidate[i1].type, at, num_atoms ); |
2090 | |
|
2091 | 0 | if (IS_BNS_ERROR( ret )) |
2092 | 0 | { |
2093 | 0 | return ret; |
2094 | 0 | } |
2095 | 0 | if (nDelta) |
2096 | 0 | { |
2097 | 0 | goto quick_exit; |
2098 | 0 | } |
2099 | 0 | } |
2100 | 0 | } |
2101 | 0 | } |
2102 | 0 | i1 = i3; |
2103 | 0 | } |
2104 | | |
2105 | 0 | quick_exit: |
2106 | 0 | if (c_group_info->num_candidates == 0) |
2107 | 0 | { |
2108 | | /* first time: initialize */ |
2109 | 0 | c_group_info->num_candidates = num_tested ? nNumCandidates : -1; /* no candidate exists */ |
2110 | 0 | } |
2111 | 0 | } |
2112 | | |
2113 | 109k | return nNumChanges; |
2114 | 110k | } |
2115 | | |
2116 | | |
2117 | | /****************************************************************************/ |
2118 | | int GetSaltChargeType( inp_ATOM *at, |
2119 | | int at_no, |
2120 | | T_GROUP_INFO *t_group_info, |
2121 | | int *s_subtype ) |
2122 | 338k | { |
2123 | | /* |
2124 | | type (returned value): |
2125 | | -1 => ignore |
2126 | | 0 => oxygen |
2127 | | subtype: |
2128 | | 1 = SALT_DONOR_H => has H |
2129 | | 2 = SALT_DONOR_Neg => has (-) charge |
2130 | | 4 = SALT_ACCEPTOR => may be an acceptor of H or (-), but not necessarily |
2131 | | |
2132 | | O-atom should be: |
2133 | | - a terminal atom |
2134 | | - connected to unsaturated, uncharged, non-radical atom C that has chemical valence 4: |
2135 | | H-donors: =CH-OH, =C(-X)-OH |
2136 | | possible H-acceptors: -CH=O, >C=O |
2137 | | H-acceptors are true if O is tautomeric |
2138 | | */ |
2139 | | |
2140 | 338k | int iC, tg, i, type; |
2141 | | |
2142 | 338k | *s_subtype = 0; /* initialize the output */ |
2143 | | /* check whether it is a candidate */ |
2144 | 338k | if (at[at_no].valence != 1 || |
2145 | 338k | at[at_no].radical && at[at_no].radical != RADICAL_SINGLET || |
2146 | 338k | at[at_no].charge < -1 || |
2147 | 338k | at[at_no].charge > 0 && !at[at_no].c_point) |
2148 | 338k | { |
2149 | 338k | return -1; |
2150 | 338k | } |
2151 | | |
2152 | 30 | if (at[at_no].el_number == EL_NUMBER_O || |
2153 | 30 | at[at_no].el_number == EL_NUMBER_S || |
2154 | 30 | at[at_no].el_number == EL_NUMBER_SE || |
2155 | 30 | at[at_no].el_number == EL_NUMBER_TE) |
2156 | 30 | { |
2157 | 30 | type = 0; /* terminal oxygen atom, needs more to be checked... */ |
2158 | 30 | } |
2159 | 0 | else |
2160 | 0 | { |
2161 | 0 | type = -1; /* ignore this atom */ |
2162 | 0 | } |
2163 | | |
2164 | 30 | if (type < 0 || |
2165 | 30 | at[at_no].chem_bonds_valence + at[at_no].num_H != |
2166 | 30 | get_el_valence( at[at_no].el_number, at[at_no].charge, 0 )) |
2167 | 0 | { |
2168 | 0 | return -1; /* non-standard valence or not an oxygen */ |
2169 | 0 | } |
2170 | | |
2171 | 30 | iC = at[at_no].neighbor[0]; |
2172 | | |
2173 | 30 | #if ( SALT_WITH_PROTONS == 1 ) |
2174 | 30 | if (at[iC].el_number != EL_NUMBER_C || |
2175 | 30 | at[iC].chem_bonds_valence + at[iC].num_H != 4 || /* allow =C(H)-OH or -C(H)=O */ |
2176 | 30 | at[iC].charge || |
2177 | 30 | at[iC].radical && at[iC].radical != RADICAL_SINGLET || |
2178 | 30 | at[iC].valence == at[iC].chem_bonds_valence) |
2179 | 30 | { |
2180 | 30 | return -1; /* oxigen is connected to a wrong atom */ |
2181 | 30 | } |
2182 | | #else |
2183 | | if (at[iC].el_number != EL_NUMBER_C || |
2184 | | at[iC].num_H || |
2185 | | at[iC].chem_bonds_valence != 4 || /* allow only no H on C */ |
2186 | | at[iC].charge || |
2187 | | at[iC].radical && at[iC].radical != RADICAL_SINGLET || |
2188 | | at[iC].valence == at[iC].chem_bonds_valence) |
2189 | | { |
2190 | | return -1; /* oxigen is connected to a wrong atom */ |
2191 | | } |
2192 | | #endif |
2193 | 0 | if (( tg = at[at_no].endpoint ) && t_group_info && t_group_info->t_group) |
2194 | 0 | { |
2195 | | /* O-atom is in a tautomeric group */ |
2196 | 0 | for (i = 0; i < t_group_info->num_t_groups; i++) |
2197 | 0 | { |
2198 | 0 | if (tg == t_group_info->t_group[i].nGroupNumber) |
2199 | 0 | { |
2200 | | /* |
2201 | | t_group_info->t_group[i].num[0] = number of attached H-atoms and negative charges |
2202 | | t_group_info->t_group[i].num[1] = number of attached negative charges |
2203 | | */ |
2204 | 0 | if (t_group_info->t_group[i].num[0] > t_group_info->t_group[i].num[1]) |
2205 | 0 | { |
2206 | 0 | *s_subtype |= SALT_DONOR_H; /* has H */ |
2207 | 0 | } |
2208 | 0 | if (t_group_info->t_group[i].num[1]) |
2209 | 0 | { |
2210 | 0 | *s_subtype |= SALT_DONOR_Neg; /* has (-) */ |
2211 | 0 | } |
2212 | 0 | *s_subtype |= SALT_ACCEPTOR; /* there is always an acceptor in a t-group */ |
2213 | 0 | return type; |
2214 | 0 | } |
2215 | 0 | } |
2216 | 0 | return -1; /* error: t-group not found */ |
2217 | 0 | } |
2218 | | |
2219 | | /* O is not not in a tautomeric group */ |
2220 | | /* assume valence(O-) < valence(O) < valence(O+) */ |
2221 | 0 | if (at[at_no].charge == -1) |
2222 | 0 | { |
2223 | 0 | *s_subtype |= SALT_DONOR_Neg; /* has (-) */ |
2224 | 0 | } |
2225 | 0 | if (at[at_no].charge <= 0 && at[at_no].num_H) |
2226 | 0 | { |
2227 | 0 | *s_subtype |= SALT_DONOR_H; /* has H */ |
2228 | 0 | } |
2229 | 0 | if (at[at_no].charge == 0 && at[at_no].chem_bonds_valence == 2) |
2230 | 0 | { |
2231 | 0 | *s_subtype |= SALT_ACCEPTOR; |
2232 | 0 | } |
2233 | | /* since O cannot be a charge point, the following cannot happen: */ |
2234 | 0 | if (at[at_no].charge == 1 && at[at_no].c_point && at[at_no].chem_bonds_valence == 2 && at[at_no].num_H) |
2235 | 0 | { |
2236 | 0 | *s_subtype |= SALT_DONOR_H; /* has H */ |
2237 | 0 | } |
2238 | |
|
2239 | 0 | return type; |
2240 | 0 | } |
2241 | | |
2242 | | |
2243 | | /****************************************************************************/ |
2244 | | int bDoNotMergeNonTautAtom( inp_ATOM *at, int at_no ) |
2245 | 12 | { |
2246 | 12 | if (at[at_no].el_number == EL_NUMBER_N) |
2247 | 2 | { |
2248 | 2 | return 1; |
2249 | 2 | } |
2250 | 10 | return 0; |
2251 | 12 | } |
2252 | | |
2253 | | |
2254 | | /****************************************************************************/ |
2255 | | int GetOtherSaltChargeType( inp_ATOM *at, |
2256 | | int at_no, |
2257 | | T_GROUP_INFO *t_group_info, |
2258 | | int *s_subtype, |
2259 | | int bAccept_O ) |
2260 | 338k | { |
2261 | | /* |
2262 | | type (returned value): |
2263 | | -1 => ignore |
2264 | | 1 => not an oxygen |
2265 | | subtype: |
2266 | | 1 = SALT_DONOR_H => has H |
2267 | | 2 = SALT_DONOR_Neg => has (-) charge |
2268 | | 4 = SALT_ACCEPTOR => may be an acceptor of H or (-), but not necessarily |
2269 | | |
2270 | | the atom should be: |
2271 | | - a tautomeric endpoint atom |
2272 | | - connected to possible centerpoint atom |
2273 | | |
2274 | | another description of the atom searched here: |
2275 | | |
2276 | | any possibly tautomeric atom adjacent to a possibly centerpoint |
2277 | | that has at least one double bond (possibly if positively charged); |
2278 | | if eif.cAcceptor then the bond between the atom and the centerpoint must be possibly double |
2279 | | if eif.cAcceptor then the bond must be possibly single |
2280 | | Donors that belong to a t-group are also acceptors |
2281 | | */ |
2282 | | |
2283 | 338k | int tg, i, j, type, endpoint_valence, num_centerpoints, bond_type, centerpoint; |
2284 | 338k | ENDPOINT_INFO eif; |
2285 | | |
2286 | 338k | *s_subtype = 0; /* initialize the output */ |
2287 | 338k | if (!bAccept_O /* only N */ && |
2288 | 338k | ( at[at_no].el_number == EL_NUMBER_O || |
2289 | 0 | at[at_no].el_number == EL_NUMBER_S || |
2290 | 0 | at[at_no].el_number == EL_NUMBER_SE || |
2291 | 0 | at[at_no].el_number == EL_NUMBER_TE )) |
2292 | 0 | { |
2293 | 0 | return -1; /* we are not looking for oxygen here */ |
2294 | 0 | } |
2295 | | |
2296 | 338k | type = 1; |
2297 | 338k | if (!( endpoint_valence = nGetEndpointInfo( at, at_no, &eif ) )) |
2298 | 338k | { |
2299 | 338k | return -1; /* not a possible endpoint */ |
2300 | 338k | } |
2301 | 36 | else |
2302 | 36 | { |
2303 | | /* at[at_no] is not not in a tautomeric group; use eif previously filled out by nGetEndpointInfo */ |
2304 | | /* check whether there is adjacent atom-candidate for a centerpoint */ |
2305 | 36 | num_centerpoints = 0; |
2306 | 54 | for (j = 0; j < at[at_no].valence; j++) |
2307 | 42 | { |
2308 | 42 | bond_type = (int) at[at_no].bond_type[j] & BOND_TYPE_MASK; |
2309 | 42 | centerpoint = (int) at[at_no].neighbor[j]; /* a centerpoint candidate */ |
2310 | 42 | if (( eif.cAcceptor && ( bond_type == BOND_DOUBLE || |
2311 | 0 | bond_type == BOND_ALTERN || /* possibly double */ |
2312 | 0 | bond_type == BOND_ALT12NS || |
2313 | 0 | bond_type == BOND_TAUTOM ) || |
2314 | 42 | eif.cDonor && ( bond_type == BOND_SINGLE || |
2315 | 42 | bond_type == BOND_ALTERN || /* possibly single */ |
2316 | 42 | bond_type == BOND_ALT12NS || |
2317 | 42 | bond_type == BOND_TAUTOM ) ) && |
2318 | 42 | ( at[centerpoint].chem_bonds_valence > at[centerpoint].valence || |
2319 | | /* check for possible endpoint added 2004-02-24 */ |
2320 | 42 | at[centerpoint].chem_bonds_valence == at[centerpoint].valence && |
2321 | 18 | ( at[centerpoint].endpoint || at[centerpoint].c_point ) /* tautomerism or charge may increment at[centerpoint].chem_bonds_valence*/ ) && |
2322 | 42 | is_centerpoint_elem( at[centerpoint].el_number )) |
2323 | 24 | { |
2324 | 24 | num_centerpoints++; |
2325 | 24 | break; /* at least one possibly centerpoint neighbor has been found */ |
2326 | 24 | } |
2327 | 42 | } |
2328 | 36 | if (!num_centerpoints) |
2329 | 12 | { |
2330 | 12 | return -1; |
2331 | 12 | } |
2332 | | /* moved here from just after "type = 1;" line 2004-02-26 */ |
2333 | 24 | if (( tg = at[at_no].endpoint ) && t_group_info && t_group_info->t_group) |
2334 | 0 | { |
2335 | | /* atom is in a tautomeric group */ |
2336 | 0 | for (i = 0; i < t_group_info->num_t_groups; i++) |
2337 | 0 | { |
2338 | 0 | if (tg == t_group_info->t_group[i].nGroupNumber) |
2339 | 0 | { |
2340 | | /* |
2341 | | t_group_info->t_group[i].num[0] = number of attached H-atoms and negative charges |
2342 | | t_group_info->t_group[i].num[1] = number of attached negative charges |
2343 | | */ |
2344 | 0 | if (t_group_info->t_group[i].num[0] > t_group_info->t_group[i].num[1]) |
2345 | 0 | { |
2346 | 0 | *s_subtype |= SALT_DONOR_H; /* has H */ |
2347 | 0 | } |
2348 | 0 | if (t_group_info->t_group[i].num[1]) |
2349 | 0 | { |
2350 | 0 | *s_subtype |= SALT_DONOR_Neg; /* has (-) */ |
2351 | 0 | } |
2352 | 0 | *s_subtype |= SALT_ACCEPTOR; /* there is always an acceptor in a t-group */ |
2353 | 0 | return type; |
2354 | 0 | } |
2355 | 0 | } |
2356 | 0 | return -1; /* error: t-group not found */ |
2357 | 0 | } |
2358 | | |
2359 | 24 | if (eif.cAcceptor) |
2360 | 0 | { |
2361 | 0 | *s_subtype |= SALT_ACCEPTOR; |
2362 | 0 | } |
2363 | 24 | if (eif.cDonor) |
2364 | 24 | { |
2365 | 24 | if (at[at_no].charge == -1) |
2366 | 24 | { |
2367 | 24 | *s_subtype |= SALT_DONOR_Neg; /* has (-) */ |
2368 | 24 | } |
2369 | 24 | if (at[at_no].num_H) |
2370 | 0 | { |
2371 | 0 | *s_subtype |= SALT_DONOR_H; /* has H */ |
2372 | 0 | } |
2373 | 24 | } |
2374 | 24 | } |
2375 | | |
2376 | 24 | return type; |
2377 | 338k | } |
2378 | | |
2379 | | |
2380 | | /****************************************************************************/ |
2381 | | int GetOtherSaltType( inp_ATOM *at, int at_no, int *s_subtype ) |
2382 | 338k | { |
2383 | | /* |
2384 | | type (returned value): |
2385 | | -1 => ignore |
2386 | | 2 => found: SH |
2387 | | proton donor -CH2-SH, >CH-SH, >C< S(-) |
2388 | | proton acceptor -CH2-S(-), >CH-S(-), >C< |
2389 | | subtype: |
2390 | | 1 = SALT_DONOR_H => has H |
2391 | | 2 = SALT_DONOR_Neg => has (-) charge |
2392 | | 4 = SALT_ACCEPTOR => may be an acceptor of H or (-), but not necessarily |
2393 | | |
2394 | | non-O-atom should be: |
2395 | | - a tautomeric endpoint atom |
2396 | | - connected to possible middle point atom |
2397 | | */ |
2398 | | |
2399 | 338k | int type, endpoint_valence, bond_type, centerpoint; |
2400 | 338k | ENDPOINT_INFO eif; |
2401 | | |
2402 | 338k | if (at[at_no].valence != 1 || at[at_no].chem_bonds_valence != 1 || |
2403 | 338k | 1 != ( at[at_no].num_H == 1 ) + ( at[at_no].charge == -1 )) |
2404 | 338k | { |
2405 | 338k | return -1; |
2406 | 338k | } |
2407 | | |
2408 | 12 | *s_subtype = 0; /* initialize the output */ |
2409 | 12 | if (!( at[at_no].el_number == EL_NUMBER_S || |
2410 | 12 | at[at_no].el_number == EL_NUMBER_SE || |
2411 | 12 | at[at_no].el_number == EL_NUMBER_TE )) |
2412 | 12 | { |
2413 | 12 | return -1; /* we are not looking for oxygen here */ |
2414 | 12 | } |
2415 | | |
2416 | 0 | type = 2; /* non-tautomeric p-donor or acceptor: C-SH, C-S(-) */ |
2417 | |
|
2418 | 0 | if (!( endpoint_valence = nGetEndpointInfo( at, at_no, &eif ) ) || |
2419 | 0 | eif.cMoveableCharge && !at[at_no].c_point || !eif.cDonor || eif.cAcceptor) |
2420 | 0 | { |
2421 | 0 | return -1; /* not a possible -SH or -S(-) */ |
2422 | 0 | } |
2423 | 0 | else |
2424 | 0 | { |
2425 | | /* at[at_no] is not not in a tautomeric group; use eif previously filled out by nGetEndpointInfo */ |
2426 | | /* check whether there is adjacent atom-candidate for a centerpoint */ |
2427 | 0 | centerpoint = (int) at[at_no].neighbor[0]; |
2428 | 0 | bond_type = (int) at[at_no].bond_type[0] & BOND_TYPE_MASK; |
2429 | 0 | if (at[centerpoint].el_number != EL_NUMBER_C || |
2430 | 0 | at[centerpoint].charge || |
2431 | 0 | at[centerpoint].radical && at[centerpoint].radical != RADICAL_SINGLET || |
2432 | 0 | at[centerpoint].valence != at[centerpoint].chem_bonds_valence) |
2433 | 0 | { |
2434 | 0 | return -1; /* not a carbon with all single bonds */ |
2435 | 0 | } |
2436 | 0 | if (at[at_no].num_H == 1) |
2437 | 0 | { |
2438 | 0 | *s_subtype |= SALT_p_DONOR; |
2439 | 0 | } |
2440 | 0 | else |
2441 | 0 | { |
2442 | 0 | if (at[at_no].charge == -1) |
2443 | 0 | { |
2444 | 0 | *s_subtype |= SALT_p_ACCEPTOR; |
2445 | 0 | } |
2446 | 0 | else |
2447 | 0 | { |
2448 | 0 | return -1; |
2449 | 0 | } |
2450 | 0 | } |
2451 | 0 | } |
2452 | | |
2453 | 0 | return type; |
2454 | 0 | } |
2455 | | |
2456 | | |
2457 | | /************************************************************************************/ |
2458 | | /* new version: merge all, check alt paths, then unmerge unreachable O-atoms if any */ |
2459 | | /* Check for oxygen negative charge-H tautomerism (Salts) |
2460 | | allowed long-range tautomerism; more than one H or (-) can be moved, for example: |
2461 | | HO-C=C-O(-) O=C-C=O |
2462 | | / \ / \ |
2463 | | R R R R |
2464 | | | | => | | |
2465 | | R' R' R' R' |
2466 | | \ / \ / |
2467 | | O=C-C=O HO-C=C-O(-) |
2468 | | |
2469 | | To check: |
2470 | | |
2471 | | | | |
2472 | | -add all possible HO-C=, O=C, (-)O-C= (including all containing O t-groups) into one t-group; |
2473 | | -temporarily disconnect one of previously not belonging to any t-group O-atoms from the one t-group; |
2474 | | -find whether there is an alt path allowing H or (-) to migrate |
2475 | | from the temp. disconnected O to any one left in the group. |
2476 | | If the alt path does not exist then the temp. disconnected atom does not |
2477 | | participate in the H/(-) migrartion and it will be unmarked/unmerged. |
2478 | | */ |
2479 | | |
2480 | | |
2481 | | |
2482 | | /****************************************************************************/ |
2483 | | int comp_candidates( const void *a1, const void *a2 ) |
2484 | 0 | { |
2485 | 0 | const S_CANDIDATE *s1 = (const S_CANDIDATE *) a1; |
2486 | 0 | const S_CANDIDATE *s2 = (const S_CANDIDATE *) a2; |
2487 | 0 | int ret; |
2488 | 0 | if (s1->type >= 0 /* enabled < */ && s2->type < 0 /* disabled */) |
2489 | 0 | { |
2490 | 0 | return -1; /* enabled goes first */ |
2491 | 0 | } |
2492 | 0 | if (s1->type < 0 /* disabled > */ && s2->type >= 0 /* enabled */) |
2493 | 0 | { |
2494 | 0 | return 1; |
2495 | 0 | } |
2496 | 0 | if (s1->endpoint && !s2->endpoint) |
2497 | 0 | { |
2498 | 0 | return -1; /* tautomeric goes first; only tautomeric may be disabled */ |
2499 | 0 | } |
2500 | 0 | if (!s1->endpoint && s2->endpoint) |
2501 | 0 | { |
2502 | 0 | return 1; /* tautomeric goes first; only tautomeric may be disabled */ |
2503 | 0 | } |
2504 | 0 | if (s1->endpoint && s2->endpoint && ( ret = (int) s1->endpoint - (int) s2->endpoint )) |
2505 | 0 | { |
2506 | 0 | return ret; |
2507 | 0 | } |
2508 | | |
2509 | 0 | return (int) s1->atnumber - (int) s2->atnumber; |
2510 | 0 | } |
2511 | | |
2512 | | |
2513 | | /****************************************************************************/ |
2514 | | int MarkSaltChargeGroups2( CANON_GLOBALS *pCG, |
2515 | | inp_ATOM *at, |
2516 | | int num_atoms, |
2517 | | S_GROUP_INFO *s_group_info, |
2518 | | T_GROUP_INFO *t_group_info, |
2519 | | C_GROUP_INFO *c_group_info, |
2520 | | struct BalancedNetworkStructure *pBNS, |
2521 | | struct BalancedNetworkData *pBD ) |
2522 | 110k | { |
2523 | | /* BNS_EDGE_FORBIDDEN_TEMP */ |
2524 | 110k | #define ALT_PATH_FOUND (MAX_ATOMS+1) |
2525 | 110k | #define NO_ENDPOINT (MAX_ATOMS+2) /* the two defines must be different */ |
2526 | 110k | #define DISABLE_CANDIDATE 10 |
2527 | 110k | #define cPAIR(a,b) cPair[a+b*nNumLeftCandidates] |
2528 | 110k | #define ACCEPTOR_PAIR 1 |
2529 | 110k | #define DONOR_PAIR 2 |
2530 | | |
2531 | 110k | int nNumChanges = 0, nNumOtherChanges = 0, nNumAcidicChanges = 0, nTotNumChanges = 0; |
2532 | 110k | S_CHAR *cPair = NULL; |
2533 | 110k | T_ENDPOINT *EndPoint = NULL; |
2534 | 110k | if (s_group_info && s_group_info->s_candidate && s_group_info->max_num_candidates > 0) |
2535 | 110k | { |
2536 | 110k | int i, j, i1, j1; |
2537 | 110k | S_CANDIDATE *s_candidate = s_group_info->s_candidate; |
2538 | 110k | int nMaxNumCandidates = s_group_info->max_num_candidates; |
2539 | 110k | int nNumCandidates = s_group_info->num_candidates; |
2540 | 110k | int nNumOtherCandidates = s_group_info->num_other_candidates; |
2541 | 110k | int nNumPOnlyCandidates = s_group_info->num_p_only_candidates; |
2542 | 110k | int nNumLeftCandidates = 0; |
2543 | 110k | int nNumMarkedCandidates = 0; |
2544 | 110k | int s_type, s_subtype; |
2545 | 110k | int ret, nDelta; |
2546 | 110k | int bHardAddedRemovedProtons = t_group_info && ( t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT ); |
2547 | | |
2548 | 110k | int s_subtype_all = 0; |
2549 | 110k | int nDonorPairs, nAcceptorPairs, nCurDonorPairs, nCurAcceptorPairs, bAlreadyTested; |
2550 | | /* |
2551 | | ENDPOINT_INFO eif; |
2552 | | */ |
2553 | | |
2554 | 110k | #if ( IGNORE_TGROUP_WITHOUT_H == 1 ) |
2555 | 110k | int bTGroupHasNegativeChargesOnly = 1; |
2556 | 110k | #endif |
2557 | | /*return 0;*/ /* debug only */ |
2558 | | |
2559 | 110k | i1 = -1; |
2560 | | |
2561 | 110k | if (nNumCandidates <= -2 || !t_group_info || !t_group_info->t_group) |
2562 | 0 | { |
2563 | 0 | return 0; |
2564 | 0 | } |
2565 | | |
2566 | | /*************************************************************************/ |
2567 | | /* Find all candidates including those with differen s_type (other type) */ |
2568 | | /*************************************************************************/ |
2569 | 222k | for (i = 0, nNumCandidates = nNumOtherCandidates = nNumPOnlyCandidates = 0; i < num_atoms; i++) |
2570 | 112k | { |
2571 | 112k | if (0 == ( s_type = GetSaltChargeType( at, i, t_group_info, &s_subtype ) ) || |
2572 | | /* -C=O or =C-OH, O = S, Se, Te */ |
2573 | 112k | 1 == ( s_type = GetOtherSaltChargeType( at, i, t_group_info, &s_subtype, 1/* bAccept_O*/ ) ) || |
2574 | | /* =Z-MH or -Z=M, Z = centerpoint, M = endpoint, other than above */ |
2575 | 112k | 2 == ( s_type = GetOtherSaltType( at, i, &s_subtype ) ) || |
2576 | 112k | ( bHardAddedRemovedProtons && 4 == ( s_type = bIsHardRemHCandidate( at, i, &s_subtype ) ) ) |
2577 | | /* >C-SH, >C-S(-); S=S,Se,Te */ |
2578 | 112k | ) |
2579 | 8 | { |
2580 | | |
2581 | 8 | if (nNumCandidates >= nMaxNumCandidates) |
2582 | 0 | { |
2583 | 0 | return BNS_VERT_EDGE_OVFL; |
2584 | 0 | } |
2585 | 8 | s_candidate[nNumCandidates].atnumber = i; |
2586 | 8 | s_candidate[nNumCandidates].type = s_type; |
2587 | 8 | s_candidate[nNumCandidates].subtype = s_subtype; |
2588 | 8 | s_candidate[nNumCandidates].endpoint = at[i].endpoint; |
2589 | 8 | nNumCandidates++; |
2590 | 8 | nNumOtherCandidates += ( 1 == s_type ); |
2591 | 8 | s_subtype_all |= s_subtype; |
2592 | 8 | i1 = i; /* save a representative of a tautomeric group */ |
2593 | 8 | } |
2594 | 112k | } |
2595 | | |
2596 | 110k | if (nNumCandidates <= 1 || /* TG_FLAG_ALLOW_NO_NEGTV_O <=> CHARGED_SALTS_ONLY=0 */ |
2597 | 110k | !( s_subtype_all & SALT_ACCEPTOR ) || |
2598 | 110k | ( ( ( t_group_info->bTautFlags & TG_FLAG_ALLOW_NO_NEGTV_O ) || |
2599 | 0 | ( t_group_info->bTautFlagsDone & TG_FLAG_FOUND_SALT_CHARGES_DONE ) || |
2600 | 0 | ( t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT ) ) ? |
2601 | 0 | !( s_subtype_all & ( SALT_DONOR ) ) : |
2602 | 0 | ( !( s_subtype_all & SALT_DONOR_Neg ) || nNumOtherCandidates == nNumCandidates ) ) |
2603 | 110k | ) |
2604 | 110k | { |
2605 | 110k | s_group_info->num_candidates = 0; /* no candidate exists */ |
2606 | 110k | return 0; |
2607 | 110k | } |
2608 | | |
2609 | 0 | if (!( s_subtype_all & ( SALT_DONOR_Neg ) )) |
2610 | 0 | { |
2611 | 0 | t_group_info->bTautFlagsDone |= TG_FLAG_ALLOW_NO_NEGTV_O_DONE; |
2612 | 0 | } |
2613 | | |
2614 | | /************************************************************************************/ |
2615 | | /* Mark redundant candidates so that only one candidate from one t-group is left in */ |
2616 | | /************************************************************************************/ |
2617 | 0 | for (i = 0; i < nNumCandidates; i++) |
2618 | 0 | { |
2619 | 0 | if (2 == s_candidate[nNumCandidates].type) |
2620 | 0 | { |
2621 | 0 | s_candidate[i].type -= DISABLE_CANDIDATE; /* disable >C-SH candidates */ |
2622 | 0 | nNumLeftCandidates++; /* count rejected */ |
2623 | 0 | continue; |
2624 | 0 | } |
2625 | 0 | if (s_candidate[i].endpoint) |
2626 | 0 | { |
2627 | 0 | for (j = i - 1; 0 <= j; j--) |
2628 | 0 | { |
2629 | 0 | if (s_candidate[i].endpoint == s_candidate[j].endpoint) |
2630 | 0 | { |
2631 | 0 | s_candidate[i].type -= DISABLE_CANDIDATE; /* disable subsequent redundant */ |
2632 | 0 | nNumLeftCandidates++; /* count rejected */ |
2633 | 0 | break; |
2634 | 0 | } |
2635 | 0 | } |
2636 | 0 | } |
2637 | 0 | } |
2638 | |
|
2639 | 0 | nNumLeftCandidates = nNumCandidates - nNumLeftCandidates; /* subtract num. rejected from the total */ |
2640 | 0 | s_group_info->num_candidates = 0; /* reinit next time */ |
2641 | | /*********************************************************************/ |
2642 | | /* reorder so that all disabled are at the end, tautomeric are first */ |
2643 | | /*********************************************************************/ |
2644 | |
|
2645 | 0 | qsort( s_candidate, nNumCandidates, sizeof( s_candidate[0] ), comp_candidates ); |
2646 | |
|
2647 | 0 | cPair = (S_CHAR *) inchi_calloc( nNumLeftCandidates*nNumLeftCandidates, sizeof( cPair[0] ) ); |
2648 | 0 | if (!cPair) |
2649 | 0 | { |
2650 | | /*printf("BNS_OUT_OF_RAM-6\n");*/ |
2651 | 0 | nTotNumChanges = BNS_OUT_OF_RAM; |
2652 | 0 | goto quick_exit; |
2653 | 0 | } |
2654 | 0 | nDonorPairs = nAcceptorPairs = 0; |
2655 | | |
2656 | | /**********************************************************************/ |
2657 | | /* Find whether we have at least one donor pair and one acceptor pair */ |
2658 | | /**********************************************************************/ |
2659 | 0 | for (i = 0; i < nNumLeftCandidates; i++) |
2660 | 0 | { |
2661 | 0 | nCurDonorPairs = nCurAcceptorPairs = 0; |
2662 | 0 | for (j = 0; j <= i; j++) |
2663 | 0 | { |
2664 | 0 | if (i == j && !s_candidate[i].endpoint) |
2665 | 0 | { |
2666 | 0 | continue; /* same non-taut atom. However, success for i==j means * |
2667 | | * that the whole tautomeric group may donate or accept 2H */ |
2668 | 0 | } |
2669 | | /* check for acceptor pair */ |
2670 | 0 | if (( s_candidate[i].subtype & SALT_ACCEPTOR ) && ( s_candidate[j].subtype & SALT_ACCEPTOR ) && |
2671 | 0 | ( ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, s_candidate[i].atnumber, |
2672 | 0 | s_candidate[j].atnumber, ALT_PATH_MODE_ADD2H_TST ) )) |
2673 | 0 | { |
2674 | 0 | if (IS_BNS_ERROR( ret )) |
2675 | 0 | { |
2676 | 0 | nTotNumChanges = ret; |
2677 | 0 | goto quick_exit; |
2678 | 0 | } |
2679 | 0 | if (ret & 1) |
2680 | 0 | { |
2681 | 0 | nDelta = ( ret & ~3 ) >> 2; |
2682 | | /*nNumChanges += (ret & 2);*/ |
2683 | 0 | if (nDelta) |
2684 | 0 | { |
2685 | | /* alt path unleashed previously localized radicals and they annihilated */ |
2686 | 0 | nNumChanges = 0; |
2687 | 0 | nTotNumChanges = BNS_RADICAL_ERR; |
2688 | 0 | goto quick_exit; |
2689 | 0 | } |
2690 | 0 | cPAIR( i, j ) |= ACCEPTOR_PAIR; /* the result: mark the pair */ |
2691 | | /*cPAIR(j,i) |= ACCEPTOR_PAIR;*/ |
2692 | 0 | } |
2693 | 0 | } |
2694 | | /* Check for donor pair */ |
2695 | 0 | if (( s_candidate[i].subtype & SALT_DONOR ) && ( s_candidate[j].subtype & SALT_DONOR ) && |
2696 | 0 | ( ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, s_candidate[i].atnumber, |
2697 | 0 | s_candidate[j].atnumber, ALT_PATH_MODE_REM2H_TST ) )) |
2698 | 0 | { |
2699 | 0 | if (IS_BNS_ERROR( ret )) |
2700 | 0 | { |
2701 | 0 | nTotNumChanges = ret; |
2702 | 0 | goto quick_exit; |
2703 | 0 | } |
2704 | 0 | if (ret & 1) |
2705 | 0 | { |
2706 | 0 | nDelta = ( ret & ~3 ) >> 2; |
2707 | | /*nNumChanges += (ret & 2);*/ |
2708 | 0 | if (nDelta) |
2709 | 0 | { |
2710 | | /* Alt path unleashed previously localized radicals and they annihilated */ |
2711 | 0 | nNumChanges = 0; |
2712 | 0 | nTotNumChanges = BNS_RADICAL_ERR; |
2713 | 0 | goto quick_exit; |
2714 | 0 | } |
2715 | 0 | cPAIR( i, j ) |= DONOR_PAIR; /* the result: mark the pair */ |
2716 | | /*cPAIR(j,i) |= ACCEPTOR_PAIR;*/ |
2717 | 0 | } |
2718 | 0 | } |
2719 | | /* Since the results will be used later to change bonds, check only now */ |
2720 | | /* when both results for (i,j) have been obtained. */ |
2721 | 0 | if (cPAIR( i, j ) & ACCEPTOR_PAIR) |
2722 | 0 | { |
2723 | 0 | nCurAcceptorPairs++; |
2724 | 0 | if (nDonorPairs) |
2725 | 0 | { |
2726 | | /* Find donor pair (i1,j1) such that i!=i1, i!=j1, j!=i1, j!=j1 */ |
2727 | 0 | for (i1 = 0; i1 < i; i1++) |
2728 | 0 | { |
2729 | 0 | for (j1 = 0; j1 <= i1; j1++) |
2730 | 0 | { |
2731 | | /* Here always j1 < i && i1 < i therefore we do not compare i to i1 or j1 */ |
2732 | 0 | if (j1 != j && i1 != j && ( cPAIR( i1, j1 ) & DONOR_PAIR )) |
2733 | 0 | { |
2734 | | /* Both the donor and the acceptor pairs have been found */ |
2735 | 0 | goto bFound2Pairs; |
2736 | 0 | } |
2737 | 0 | } |
2738 | 0 | } |
2739 | 0 | } |
2740 | 0 | } |
2741 | 0 | if (cPAIR( i, j ) & DONOR_PAIR) |
2742 | 0 | { |
2743 | 0 | nCurDonorPairs++; |
2744 | 0 | if (nAcceptorPairs) |
2745 | 0 | { |
2746 | | /* Find acceptor pair (i1,j1) such that i!=i1, i!=j1, j!=i1, j!=j1 */ |
2747 | 0 | for (i1 = 0; i1 < i; i1++) |
2748 | 0 | { |
2749 | 0 | for (j1 = 0; j1 <= i1; j1++) |
2750 | 0 | { |
2751 | | /* Here always j1 < i && i1 < i therefore we do not compare i to i1 or j1 */ |
2752 | 0 | if (j1 != j && i1 != j && ( cPAIR( i1, j1 ) & ACCEPTOR_PAIR )) |
2753 | 0 | { |
2754 | | /* Both the donor and the acceptor pairs have been found */ |
2755 | 0 | goto bFound2Pairs; |
2756 | 0 | } |
2757 | 0 | } |
2758 | 0 | } |
2759 | 0 | } |
2760 | 0 | } |
2761 | 0 | } |
2762 | 0 | nDonorPairs += nCurDonorPairs; |
2763 | 0 | nAcceptorPairs += nCurAcceptorPairs; |
2764 | 0 | } |
2765 | | |
2766 | | /* Nothing has been found */ |
2767 | 0 | nNumChanges = 0; |
2768 | 0 | inchi_free( cPair ); |
2769 | 0 | cPair = NULL; |
2770 | 0 | goto quick_exit; |
2771 | | |
2772 | | |
2773 | | /* Both the donor and the acceptor pairs have been found */ |
2774 | 0 | bFound2Pairs: |
2775 | | /* first, try already found pairs */ |
2776 | 0 | i1 = i; |
2777 | 0 | j1 = j; |
2778 | | |
2779 | | /* Find all possible donor and acceptor pairs */ |
2780 | 0 | nNumMarkedCandidates = 0; |
2781 | 0 | for (i = 0; i < nNumLeftCandidates; i++) |
2782 | 0 | { |
2783 | 0 | nCurDonorPairs = nCurAcceptorPairs = 0; |
2784 | 0 | for (j = 0; j <= i; j++) |
2785 | 0 | { |
2786 | 0 | bAlreadyTested = ( i < i1 || i == i1 && j <= j1 ); |
2787 | 0 | if (bAlreadyTested && ( cPAIR( i, j ) & ACCEPTOR_PAIR ) || !bAlreadyTested) |
2788 | 0 | { |
2789 | | /* Checking for acceptor pair */ |
2790 | 0 | if (( s_candidate[i].subtype & SALT_ACCEPTOR ) && ( s_candidate[j].subtype & SALT_ACCEPTOR ) && |
2791 | 0 | ( ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, s_candidate[i].atnumber, |
2792 | 0 | s_candidate[j].atnumber, ALT_PATH_MODE_ADD2H_CHG ) )) |
2793 | 0 | { |
2794 | 0 | if (IS_BNS_ERROR( ret )) |
2795 | 0 | { |
2796 | 0 | nTotNumChanges = ret; |
2797 | 0 | goto quick_exit; |
2798 | 0 | } |
2799 | 0 | if (ret & 1) |
2800 | 0 | { |
2801 | 0 | nDelta = ( ret & ~3 ) >> 2; |
2802 | 0 | nNumChanges += ( ret & 2 ); |
2803 | 0 | if (nDelta) |
2804 | 0 | { |
2805 | | /* Alt path unleashed previously localized radicals and they annihilated */ |
2806 | 0 | nNumChanges = 0; |
2807 | 0 | nTotNumChanges = BNS_RADICAL_ERR; |
2808 | 0 | goto quick_exit; |
2809 | 0 | } |
2810 | 0 | cPAIR( i, j ) |= ACCEPTOR_PAIR; |
2811 | | /*cPAIR(j,i) |= ACCEPTOR_PAIR;*/ |
2812 | 0 | nCurAcceptorPairs += !bAlreadyTested; |
2813 | 0 | if (!( s_candidate[i].subtype & SALT_SELECTED )) |
2814 | 0 | { |
2815 | 0 | s_candidate[i].subtype |= SALT_SELECTED; |
2816 | 0 | nNumMarkedCandidates++; |
2817 | 0 | if (!s_candidate[i].endpoint && s_candidate[i].type) |
2818 | 0 | { |
2819 | 0 | nNumOtherChanges++; |
2820 | 0 | } |
2821 | 0 | else |
2822 | 0 | { |
2823 | 0 | nNumAcidicChanges++; |
2824 | 0 | } |
2825 | 0 | } |
2826 | 0 | if (!( s_candidate[j].subtype & SALT_SELECTED )) |
2827 | 0 | { |
2828 | 0 | s_candidate[j].subtype |= SALT_SELECTED; |
2829 | 0 | nNumMarkedCandidates++; |
2830 | 0 | if (!s_candidate[j].endpoint && s_candidate[j].type) |
2831 | 0 | { |
2832 | 0 | nNumOtherChanges++; |
2833 | 0 | } |
2834 | 0 | else |
2835 | 0 | { |
2836 | 0 | nNumAcidicChanges++; |
2837 | 0 | } |
2838 | 0 | } |
2839 | 0 | } |
2840 | 0 | } |
2841 | 0 | } |
2842 | | |
2843 | 0 | if (bAlreadyTested && ( cPAIR( i, j ) & DONOR_PAIR ) || !bAlreadyTested) |
2844 | 0 | { |
2845 | | /* Checking for donor pair */ |
2846 | 0 | if (( s_candidate[i].subtype & SALT_DONOR ) && ( s_candidate[j].subtype & SALT_DONOR ) && |
2847 | 0 | ( ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, s_candidate[i].atnumber, |
2848 | 0 | s_candidate[j].atnumber, ALT_PATH_MODE_REM2H_CHG ) )) |
2849 | 0 | { |
2850 | 0 | if (IS_BNS_ERROR( ret )) |
2851 | 0 | { |
2852 | 0 | nTotNumChanges = ret; |
2853 | 0 | goto quick_exit; |
2854 | 0 | } |
2855 | 0 | if (ret & 1) |
2856 | 0 | { |
2857 | 0 | nDelta = ( ret & ~3 ) >> 2; |
2858 | 0 | nNumChanges += ( ret & 2 ); |
2859 | 0 | if (nDelta) |
2860 | 0 | { |
2861 | | /* Alt path unleashed previously localized radicals and they annihilated */ |
2862 | 0 | nNumChanges = 0; |
2863 | 0 | nTotNumChanges = BNS_RADICAL_ERR; |
2864 | 0 | goto quick_exit; |
2865 | 0 | } |
2866 | 0 | cPAIR( i, j ) |= DONOR_PAIR; |
2867 | | /*cPAIR(j,i) |= ACCEPTOR_PAIR;*/ |
2868 | 0 | nCurDonorPairs += !bAlreadyTested; |
2869 | 0 | if (!( s_candidate[i].subtype & SALT_SELECTED )) |
2870 | 0 | { |
2871 | 0 | s_candidate[i].subtype |= SALT_SELECTED; |
2872 | 0 | nNumMarkedCandidates++; |
2873 | 0 | if (!s_candidate[i].endpoint && s_candidate[i].type) |
2874 | 0 | { |
2875 | 0 | nNumOtherChanges++; |
2876 | 0 | } |
2877 | 0 | else |
2878 | 0 | { |
2879 | 0 | nNumAcidicChanges++; |
2880 | 0 | } |
2881 | 0 | } |
2882 | 0 | if (!( s_candidate[j].subtype & SALT_SELECTED )) |
2883 | 0 | { |
2884 | 0 | s_candidate[j].subtype |= SALT_SELECTED; |
2885 | 0 | nNumMarkedCandidates++; |
2886 | 0 | if (!s_candidate[j].endpoint && s_candidate[j].type) |
2887 | 0 | { |
2888 | 0 | nNumOtherChanges++; |
2889 | 0 | } |
2890 | 0 | else |
2891 | 0 | { |
2892 | 0 | nNumAcidicChanges++; |
2893 | 0 | } |
2894 | 0 | } |
2895 | 0 | } |
2896 | 0 | } |
2897 | 0 | } |
2898 | 0 | } |
2899 | 0 | nDonorPairs += nCurDonorPairs; |
2900 | 0 | nAcceptorPairs += nCurAcceptorPairs; |
2901 | 0 | } |
2902 | | |
2903 | 0 | inchi_free( cPair ); |
2904 | 0 | cPair = NULL; |
2905 | |
|
2906 | 0 | if (nNumMarkedCandidates) |
2907 | 0 | { |
2908 | 0 | EndPoint = (T_ENDPOINT *) inchi_calloc( nNumMarkedCandidates, sizeof( EndPoint[0] ) ); |
2909 | 0 | if (!EndPoint) |
2910 | 0 | { |
2911 | | /*printf("BNS_OUT_OF_RAM-7\n");*/ |
2912 | 0 | nTotNumChanges = BNS_OUT_OF_RAM; |
2913 | 0 | goto quick_exit; |
2914 | 0 | } |
2915 | 0 | for (i = 0, j = 0; i < nNumLeftCandidates; i++) |
2916 | 0 | { |
2917 | 0 | if (s_candidate[i].subtype & SALT_SELECTED) |
2918 | 0 | { |
2919 | 0 | s_candidate[i].subtype ^= SALT_SELECTED; /* remove the flag */ |
2920 | 0 | if (j < nNumMarkedCandidates) |
2921 | 0 | { |
2922 | 0 | i1 = s_candidate[i].atnumber; /* save a representative of the t-group to be created */ |
2923 | 0 | AddEndPoint( EndPoint + j, at, i1 ); |
2924 | 0 | } |
2925 | 0 | j++; |
2926 | 0 | } |
2927 | 0 | } |
2928 | 0 | if (j != nNumMarkedCandidates) |
2929 | 0 | { |
2930 | 0 | nTotNumChanges = BNS_PROGRAM_ERR; |
2931 | 0 | goto quick_exit; |
2932 | 0 | } |
2933 | | |
2934 | | /* Merge all marked atoms and their t-groups into one t-group */ |
2935 | 0 | ret = RegisterEndPoints( pCG, t_group_info, EndPoint, |
2936 | 0 | nNumMarkedCandidates, at, |
2937 | 0 | num_atoms, c_group_info, pBNS ); |
2938 | |
|
2939 | 0 | if (ret == -1) |
2940 | 0 | { |
2941 | 0 | ret = BNS_PROGRAM_ERR; |
2942 | 0 | } |
2943 | 0 | if (ret < 0) |
2944 | 0 | { |
2945 | 0 | nTotNumChanges = ret; |
2946 | 0 | goto quick_exit; |
2947 | 0 | } |
2948 | 0 | nTotNumChanges += ( ret > 0 ); |
2949 | 0 | inchi_free( EndPoint ); |
2950 | 0 | EndPoint = NULL; |
2951 | |
|
2952 | 0 | if (nNumMarkedCandidates) |
2953 | 0 | { |
2954 | 0 | for (i = nNumLeftCandidates; i < nNumCandidates; i++) |
2955 | 0 | { |
2956 | 0 | s_candidate[i].type += DISABLE_CANDIDATE; |
2957 | 0 | j1 = s_candidate[i].atnumber; |
2958 | 0 | if (at[j1].endpoint == at[i1].endpoint) |
2959 | 0 | { |
2960 | 0 | if (!s_candidate[i].endpoint && s_candidate[i].type) |
2961 | 0 | { |
2962 | 0 | nNumOtherChanges++; |
2963 | 0 | } |
2964 | 0 | else |
2965 | 0 | { |
2966 | 0 | nNumAcidicChanges++; |
2967 | 0 | } |
2968 | 0 | } |
2969 | 0 | } |
2970 | 0 | } |
2971 | 0 | else |
2972 | 0 | { |
2973 | 0 | for (i = nNumLeftCandidates; i < nNumCandidates; i++) |
2974 | 0 | { |
2975 | 0 | s_candidate[i].type += DISABLE_CANDIDATE; |
2976 | 0 | } |
2977 | 0 | } |
2978 | | |
2979 | | /* Find whether the new t-group have any movable H */ |
2980 | 0 | for (i = 0, bTGroupHasNegativeChargesOnly = 0; i < t_group_info->num_t_groups; i++) |
2981 | 0 | { |
2982 | 0 | if (t_group_info->t_group[i].nGroupNumber == at[i1].endpoint && |
2983 | 0 | t_group_info->t_group[i].num[0] == t_group_info->t_group[i].num[1]) |
2984 | 0 | { |
2985 | 0 | bTGroupHasNegativeChargesOnly = 1; |
2986 | 0 | break; |
2987 | 0 | } |
2988 | 0 | } |
2989 | 0 | } |
2990 | 0 | nTotNumChanges = ( nTotNumChanges > 0 ); |
2991 | |
|
2992 | 0 | #if ( IGNORE_TGROUP_WITHOUT_H == 1 ) |
2993 | 0 | if (nTotNumChanges && bTGroupHasNegativeChargesOnly) |
2994 | 0 | { |
2995 | 0 | nTotNumChanges = 2; /* means no moveable H has been affected */ |
2996 | 0 | } |
2997 | 0 | #endif |
2998 | 0 | } |
2999 | | |
3000 | 0 | quick_exit: |
3001 | 0 | if (nNumOtherChanges && nTotNumChanges == 1) |
3002 | 0 | { |
3003 | 0 | nTotNumChanges = 5; /* not only acidic atoms merged */ |
3004 | 0 | } |
3005 | 0 | if (cPair) |
3006 | 0 | { |
3007 | 0 | inchi_free( cPair ); |
3008 | | /*cPair = NULL;*/ |
3009 | 0 | } |
3010 | 0 | if (EndPoint) |
3011 | 0 | { |
3012 | 0 | inchi_free( EndPoint ); |
3013 | | /*EndPoint = NULL;*/ |
3014 | 0 | } |
3015 | |
|
3016 | 0 | return nTotNumChanges; /* 0=>no changes, 1=>new salt tautomerism found, 2=>only new charge tautomerism found */ |
3017 | 110k | #undef ALT_PATH_FOUND |
3018 | 110k | #undef NO_ENDPOINT |
3019 | 110k | } |
3020 | | |
3021 | | |
3022 | | /****************************************************************************/ |
3023 | | /* regular one-path version: find alt paths then merge */ |
3024 | | /* Check for oxygen negative charge-H tautomerism (Salts) |
3025 | | allowed long-range tautomerism; only one H or (-) can be moved, for example: |
3026 | | HO-C=X-Y=Z-...-C=O => O=C-X=Y-Z=...=C-OH |
3027 | | */ |
3028 | | |
3029 | | #if ( SALT_WITH_PROTONS == 1 ) |
3030 | | |
3031 | | #define MAX_LOCAL_TGNUM 0 /* was 32; disable since it has not been used */ |
3032 | | |
3033 | | #if ( MAX_LOCAL_TGNUM > 0 ) |
3034 | | typedef struct tagTGroupData { |
3035 | | S_SHORT nGroupNumber; /* t-group number from t_group_info->t_group->nGroupNumber */ |
3036 | | S_SHORT nGroupIndex; /* TGroupData[nGroupNumber]nGroupIndex = index of t_group in t_group_info */ |
3037 | | S_SHORT nDonorM; /* number of endpoint-donors that have negative charge (Minus) */ |
3038 | | S_SHORT nDonorH; /* number of endpoint-donors that have only H */ |
3039 | | S_SHORT nAccepM; /* number of endpoint-acceptors that have negative charge (Minus) */ |
3040 | | S_SHORT nAccepH; /* number of endpoint-acceptors that have H and no negative charge */ |
3041 | | S_SHORT nAccep0; /* number of endpoint-acceptors that have no H and no negative charge */ |
3042 | | S_SHORT nDonorA; /* number of acidic endpoint-donors */ |
3043 | | S_SHORT nAccepS; /* number of acidic endpoint-acceptors */ |
3044 | | } TGroupData; |
3045 | | #endif |
3046 | | |
3047 | | |
3048 | | /****************************************************************************/ |
3049 | | int MarkSaltChargeGroups( CANON_GLOBALS *pCG, |
3050 | | inp_ATOM *at, |
3051 | | int num_atoms, |
3052 | | S_GROUP_INFO *s_group_info, |
3053 | | T_GROUP_INFO *t_group_info, |
3054 | | C_GROUP_INFO *c_group_info, |
3055 | | struct BalancedNetworkStructure *pBNS, |
3056 | | struct BalancedNetworkData *pBD ) |
3057 | 110k | { |
3058 | | |
3059 | 110k | int nNumChanges = 0, nTotNumChanges = 0; |
3060 | 110k | if (s_group_info && s_group_info->s_candidate && s_group_info->max_num_candidates > 0) |
3061 | 110k | { |
3062 | 110k | int i, i1, i2, j, j1, j2, jj, ii1, ii2, jj1, jj2, /*k,*/ num_tested; |
3063 | 110k | S_CANDIDATE *s_candidate = s_group_info->s_candidate; |
3064 | 110k | int nMaxNumCandidates = s_group_info->max_num_candidates; |
3065 | 110k | int nNumCandidates = s_group_info->num_candidates; |
3066 | 110k | int nNumOtherCandidates = s_group_info->num_other_candidates; |
3067 | 110k | int nNumPOnlyCandidates = s_group_info->num_p_only_candidates; |
3068 | 110k | int s_type, s_subtype; |
3069 | 110k | int ret, nDelta, /*nMobile,*/ err = 0; |
3070 | 110k | int s_subtype_all = 0; |
3071 | 110k | int nGroupNumber; |
3072 | 110k | T_ENDPOINT EndPoint[2]; |
3073 | | #if ( MAX_LOCAL_TGNUM > 0 ) |
3074 | | TGroupData tgData[MAX_LOCAL_TGNUM]; |
3075 | | TGroupData *ptgData = tgData; |
3076 | | #endif |
3077 | 110k | int cond1 = 0, cond2a = 0, cond2b = 0, cond2c = 0, cond2 = 0; |
3078 | | |
3079 | 110k | if (nNumCandidates <= -1 || !t_group_info || !t_group_info->t_group) |
3080 | 0 | { |
3081 | 0 | return 0; |
3082 | 0 | } |
3083 | | |
3084 | | /* count t-groups */ |
3085 | 110k | for (i = 0, nGroupNumber = 0; i < t_group_info->num_t_groups; i++) |
3086 | 0 | { |
3087 | 0 | if (nGroupNumber < t_group_info->t_group[i].nGroupNumber) |
3088 | 0 | { |
3089 | 0 | nGroupNumber = t_group_info->t_group[i].nGroupNumber; /* max. t-group number */ |
3090 | 0 | } |
3091 | 0 | } |
3092 | | #if ( MAX_LOCAL_TGNUM > 0 ) |
3093 | | /* prepare memory */ |
3094 | | if (nGroupNumber >= MAX_LOCAL_TGNUM) |
3095 | | { |
3096 | | if (!( ptgData = (TGroupData*) inchi_calloc( nGroupNumber + 1, sizeof( TGroupData ) ) )) |
3097 | | { |
3098 | | err = BNS_OUT_OF_RAM; |
3099 | | goto quick_exit; |
3100 | | } |
3101 | | } |
3102 | | else |
3103 | | { |
3104 | | memset( ptgData, 0, sizeof( tgData ) ); |
3105 | | } |
3106 | | ptgData[0].nGroupIndex = -1; /* data for non-tautomeric atoms */ |
3107 | | for (i = 0, nGroupNumber = 0; i < t_group_info->num_t_groups; i++) |
3108 | | { |
3109 | | if (nGroupNumber = t_group_info->t_group[i].nGroupNumber) |
3110 | | { |
3111 | | ptgData[nGroupNumber].nGroupIndex = i; |
3112 | | ptgData[i].nGroupNumber = nGroupNumber; |
3113 | | } |
3114 | | } |
3115 | | #endif |
3116 | 110k | nNumCandidates = 0; /* always recalculate 2004-03-22 */ |
3117 | 110k | num_tested = 0; |
3118 | | |
3119 | 110k | if (nNumCandidates == 0) |
3120 | 110k | { |
3121 | 222k | for (i = 0, nNumCandidates = nNumOtherCandidates = nNumPOnlyCandidates = 0; i < num_atoms; i++) |
3122 | 112k | { |
3123 | 112k | if (0 == ( s_type = GetSaltChargeType( at, i, t_group_info, &s_subtype ) ) || |
3124 | | /* -C=O or =C-OH, O = S, Se, Te */ |
3125 | 112k | #if ( INCL_NON_SALT_CANDIDATATES == 1 ) |
3126 | 112k | 1 == ( s_type = GetOtherSaltChargeType( at, i, t_group_info, &s_subtype, 1 ) ) || |
3127 | | /* =Z-MH or -Z=M, Z = centerpoint, M = endpoint, other than above */ |
3128 | 112k | #endif |
3129 | 112k | 2 == ( s_type = GetOtherSaltType( at, i, &s_subtype ) ) |
3130 | | /* >C-SH, >C-S(-); S=S,Se,Te */ |
3131 | 112k | ) |
3132 | 8 | { |
3133 | | |
3134 | 8 | if (nNumCandidates >= nMaxNumCandidates) |
3135 | 0 | { |
3136 | 0 | err = BNS_VERT_EDGE_OVFL; |
3137 | 0 | goto quick_exit; |
3138 | 0 | } |
3139 | 8 | s_candidate[nNumCandidates].atnumber = i; |
3140 | 8 | s_candidate[nNumCandidates].type = s_type; |
3141 | 8 | s_candidate[nNumCandidates].subtype = s_subtype; |
3142 | 8 | s_candidate[nNumCandidates].endpoint = at[i].endpoint; |
3143 | 8 | nNumCandidates++; |
3144 | 8 | nNumOtherCandidates += ( 1 == s_type ); |
3145 | 8 | nNumPOnlyCandidates += ( 2 == s_type ); |
3146 | 8 | s_subtype_all |= s_subtype; |
3147 | | /*i1 = i;*/ /* save a representative of a tautomeric group */ |
3148 | 8 | } |
3149 | 112k | } /* for */ |
3150 | | |
3151 | | /* changes: TG_FLAG_ALLOW_NO_NEGTV_O replaced CHARGED_SALTS_ONLY==0 */ |
3152 | 110k | cond1 = s_subtype_all & SALT_ACCEPTOR; |
3153 | 110k | cond2a = t_group_info->bTautFlags & TG_FLAG_ALLOW_NO_NEGTV_O; |
3154 | 110k | cond2b = t_group_info->bTautFlagsDone & TG_FLAG_FOUND_SALT_CHARGES_DONE; |
3155 | 110k | cond2c = t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT; |
3156 | 110k | if (cond2a || cond2b || cond2c) |
3157 | 0 | { |
3158 | 0 | cond2 = !( s_subtype_all & ( SALT_DONOR_Neg | SALT_DONOR_H ) ); |
3159 | 0 | } |
3160 | 110k | else |
3161 | 110k | { |
3162 | 110k | cond2 = !( s_subtype_all & SALT_DONOR_Neg ) || nNumOtherCandidates == nNumCandidates; |
3163 | 110k | } |
3164 | 110k | if (nNumCandidates <= 1 || !cond1 || cond2 |
3165 | | /*( |
3166 | | ( cond2a || cond2b || cond2c ) |
3167 | | ? !(s_subtype_all & (SALT_DONOR_Neg | SALT_DONOR_H)) |
3168 | | : ( !(s_subtype_all & SALT_DONOR_Neg) || nNumOtherCandidates==nNumCandidates) ) */ |
3169 | 110k | ) |
3170 | 110k | { |
3171 | 110k | s_group_info->num_candidates = -1; /* no candidate exists */ |
3172 | 110k | goto quick_exit; |
3173 | 110k | } |
3174 | 0 | if (!( s_subtype_all & ( SALT_DONOR_Neg ) )) |
3175 | 0 | { |
3176 | 0 | t_group_info->bTautFlagsDone |= TG_FLAG_ALLOW_NO_NEGTV_O_DONE; |
3177 | 0 | } |
3178 | 0 | } |
3179 | 0 | else |
3180 | 0 | { |
3181 | 0 | for (i = 0; i < nNumCandidates; i++) |
3182 | 0 | { |
3183 | 0 | i1 = s_candidate[i].atnumber; |
3184 | 0 | if (0 <= ( s_type = GetSaltChargeType( at, i1, t_group_info, &s_subtype ) ) |
3185 | 0 | #if ( INCL_NON_SALT_CANDIDATATES == 1 ) |
3186 | 0 | || 0 < ( s_type = GetOtherSaltChargeType( at, i1, t_group_info, &s_subtype, 1 /* bAccept_O*/ ) ) |
3187 | 0 | #endif |
3188 | 0 | ) |
3189 | 0 | { |
3190 | 0 | s_candidate[nNumCandidates].type = s_type; |
3191 | 0 | s_candidate[nNumCandidates].subtype = s_subtype; |
3192 | 0 | s_candidate[nNumCandidates].endpoint = at[i1].endpoint; |
3193 | 0 | } |
3194 | 0 | } |
3195 | 0 | } |
3196 | | |
3197 | | /* Look for alt paths connecting: |
3198 | | SALT_DONOR_Neg to SALT_ACCEPTOR : long distance migration of negative charges |
3199 | | SALT_DONOR_H to SALT_ACCEPTOR : long distance migration of H-atoms |
3200 | | */ |
3201 | 0 | do |
3202 | 0 | { |
3203 | 0 | nNumChanges = 0; |
3204 | 0 | for (i1 = 0; i1 < nNumCandidates; i1++) |
3205 | 0 | { |
3206 | 0 | j1 = s_candidate[i1].atnumber; |
3207 | 0 | for (i2 = i1 + 1; i2 < nNumCandidates; i2++) |
3208 | 0 | { |
3209 | | /* prev. approach: do not test if both candidates are not "salt-type". Disabled 2004-03-18 |
3210 | | if ( s_candidate[i1].type && s_candidate[i2].type ) |
3211 | | continue; |
3212 | | */ |
3213 | 0 | j2 = s_candidate[i2].atnumber; |
3214 | 0 | if (at[j1].endpoint && at[j1].endpoint == at[j2].endpoint) |
3215 | 0 | { |
3216 | 0 | continue; |
3217 | 0 | } |
3218 | 0 | for (j = 0; j < 2; j++) |
3219 | 0 | { |
3220 | 0 | if (j) |
3221 | 0 | { |
3222 | 0 | ii1 = i2; /* candidate 1 (donor) ordering number */ |
3223 | 0 | ii2 = i1; /* candidate 2 (acceptor) ordering number */ |
3224 | 0 | jj1 = j2; /* candidate 1 (donor) atom number */ |
3225 | 0 | jj2 = j1; /* candidate 2 (acceptor) atom number */ |
3226 | 0 | } |
3227 | 0 | else |
3228 | 0 | { |
3229 | | /* transposition */ |
3230 | 0 | ii1 = i1; /* candidate 1 (donor) ordering number */ |
3231 | 0 | ii2 = i2; /* candidate 2 (acceptor) ordering number */ |
3232 | 0 | jj1 = j1; /* candidate 1 (donor) atom number */ |
3233 | 0 | jj2 = j2; /* candidate 2 (acceptor) atom number */ |
3234 | 0 | } |
3235 | |
|
3236 | 0 | if (( s_candidate[ii1].subtype & ( SALT_DONOR_Neg | SALT_DONOR_H ) ) && |
3237 | 0 | ( s_candidate[ii2].subtype & SALT_ACCEPTOR )) |
3238 | 0 | { |
3239 | | /****printf("\nChkpt @ %-s:%-d ", __FILE__,__LINE__); fflush(stdout);*/ |
3240 | |
|
3241 | 0 | ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, jj2, jj1, ALT_PATH_MODE_4_SALT ); |
3242 | |
|
3243 | 0 | num_tested++; |
3244 | 0 | if (IS_BNS_ERROR( ret )) |
3245 | 0 | { |
3246 | 0 | err = ret; |
3247 | 0 | goto quick_exit; |
3248 | 0 | } |
3249 | | |
3250 | 0 | if (ret & 1) |
3251 | 0 | { |
3252 | 0 | nDelta = ( ret & ~3 ) >> 2; |
3253 | 0 | nNumChanges += ( ret & 2 ); |
3254 | 0 | for (i = 0; i < 2; i++) |
3255 | 0 | { |
3256 | 0 | jj = i ? jj2 : jj1; |
3257 | 0 | AddEndPoint( EndPoint + i, at, jj ); |
3258 | 0 | } |
3259 | | |
3260 | | /* add/merge taut groups and reinit pBNS in the fly */ |
3261 | 0 | ret = RegisterEndPoints( pCG, t_group_info, |
3262 | 0 | EndPoint, 2, at, |
3263 | 0 | num_atoms, c_group_info, |
3264 | 0 | pBNS ); |
3265 | 0 | if (ret == -1) |
3266 | 0 | { |
3267 | 0 | ret = BNS_PROGRAM_ERR; |
3268 | 0 | } |
3269 | 0 | if (ret < 0) |
3270 | 0 | { |
3271 | 0 | err = ret; |
3272 | 0 | goto quick_exit; |
3273 | 0 | } |
3274 | 0 | if (nDelta) |
3275 | 0 | { |
3276 | 0 | err = BNS_RADICAL_ERR; |
3277 | 0 | goto quick_exit; |
3278 | 0 | } |
3279 | 0 | nNumChanges += ( ret > 0 ); |
3280 | 0 | break; /* avoid redundant repetition */ |
3281 | 0 | } |
3282 | 0 | } |
3283 | 0 | } |
3284 | 0 | } |
3285 | 0 | } |
3286 | 0 | nTotNumChanges += nNumChanges; |
3287 | 0 | } while (num_tested && nNumChanges); |
3288 | | |
3289 | 110k | quick_exit: |
3290 | 110k | if (!err) |
3291 | 110k | { |
3292 | 110k | nTotNumChanges += nNumChanges; /* nNumChanges != 0 only in case of 'goto quick_exit' */ |
3293 | 110k | if (s_group_info->num_candidates == 0) |
3294 | 0 | { |
3295 | | /* first time: initialize */ |
3296 | 0 | s_group_info->num_candidates = num_tested ? nNumCandidates : -1; /* no candidate exists */ |
3297 | 0 | } |
3298 | 110k | } |
3299 | 0 | else |
3300 | 0 | { |
3301 | 0 | nTotNumChanges = err; |
3302 | 0 | } |
3303 | | #if ( MAX_LOCAL_TGNUM > 0 ) |
3304 | | if (ptgData != tgData) |
3305 | | { |
3306 | | inchi_free( ptgData ); |
3307 | | } |
3308 | | #endif |
3309 | 110k | } |
3310 | | |
3311 | 110k | return nTotNumChanges; |
3312 | 110k | } |
3313 | | |
3314 | | |
3315 | | #else |
3316 | | |
3317 | | |
3318 | | /********************************************************************************************************/ |
3319 | | int MarkSaltChargeGroups( CANON_GLOBALS *pCG, inp_ATOM *at, int num_atoms, S_GROUP_INFO *s_group_info, |
3320 | | T_GROUP_INFO *t_group_info, C_GROUP_INFO *c_group_info, |
3321 | | struct BalancedNetworkStructure *pBNS, struct BalancedNetworkData *pBD ) |
3322 | | { |
3323 | | |
3324 | | int nNumChanges = 0, nTotNumChanges = 0; |
3325 | | if (s_group_info && s_group_info->s_candidate && s_group_info->max_num_candidates > 0) |
3326 | | { |
3327 | | int i, i1, i2, j, j1, j2, jj, ii1, ii2, jj1, jj2, k, num_tested; |
3328 | | S_CANDIDATE *s_candidate = s_group_info->s_candidate; |
3329 | | int nMaxNumCandidates = s_group_info->max_num_candidates; |
3330 | | int nNumCandidates = s_group_info->num_candidates; |
3331 | | int nNumOtherCandidates = s_group_info->num_other_candidates; |
3332 | | int s_type, s_subtype; |
3333 | | int ret, nDelta, nMobile; |
3334 | | int s_subtype_all = 0; |
3335 | | T_ENDPOINT EndPoint[2]; |
3336 | | |
3337 | | if (nNumCandidates <= -1 || !t_group_info || !t_group_info->t_group) |
3338 | | { |
3339 | | return 0; |
3340 | | } |
3341 | | else |
3342 | | if (nNumCandidates == 0) |
3343 | | { |
3344 | | for (i = 0, nNumCandidates = nNumOtherCandidates = 0; i < num_atoms; i++) |
3345 | | { |
3346 | | if (0 <= ( s_type = GetSaltChargeType( at, i, t_group_info, &s_subtype ) )) |
3347 | | { |
3348 | | if (nNumCandidates >= nMaxNumCandidates) |
3349 | | { |
3350 | | return BNS_VERT_EDGE_OVFL; |
3351 | | } |
3352 | | s_candidate[nNumCandidates].atnumber = i; |
3353 | | s_candidate[nNumCandidates].type = s_type; |
3354 | | s_candidate[nNumCandidates].subtype = s_subtype; |
3355 | | s_candidate[nNumCandidates].endpoint = at[i].endpoint; |
3356 | | nNumCandidates++; |
3357 | | s_subtype_all |= s_subtype; |
3358 | | /*i1 = i;*/ /* save a representative of a tautomeric group */ |
3359 | | } |
3360 | | #if ( INCL_NON_SALT_CANDIDATATES == 1 ) |
3361 | | else /* new */ |
3362 | | if (0 < ( s_type = GetOtherSaltChargeType( at, i, t_group_info, &s_subtype, 1 /* bAccept_O*/ ) )) |
3363 | | { |
3364 | | if (nNumCandidates >= nMaxNumCandidates) |
3365 | | { |
3366 | | return BNS_VERT_EDGE_OVFL; |
3367 | | } |
3368 | | s_candidate[nNumCandidates].atnumber = i; |
3369 | | s_candidate[nNumCandidates].type = s_type; |
3370 | | s_candidate[nNumCandidates].subtype = s_subtype; |
3371 | | s_candidate[nNumCandidates].endpoint = at[i].endpoint; |
3372 | | nNumCandidates++; |
3373 | | nNumOtherCandidates++; |
3374 | | s_subtype_all |= s_subtype; |
3375 | | } |
3376 | | #endif |
3377 | | } |
3378 | | |
3379 | | /* changes: TG_FLAG_ALLOW_NO_NEGTV_O replaced CHARGED_SALTS_ONLY==0 */ |
3380 | | if (nNumCandidates <= 1 || nNumOtherCandidates == nNumCandidates || |
3381 | | ( ( t_group_info->bTautFlags & TG_FLAG_ALLOW_NO_NEGTV_O ) ? |
3382 | | !( s_subtype_all & ( SALT_DONOR_Neg | SALT_DONOR_H ) ) : |
3383 | | !( s_subtype_all & SALT_DONOR_Neg ) ) || |
3384 | | !( s_subtype_all & SALT_ACCEPTOR )) |
3385 | | { |
3386 | | s_group_info->num_candidates = -1; /* no candidate exists */ |
3387 | | return 0; |
3388 | | } |
3389 | | if (!( s_subtype_all & ( SALT_DONOR_Neg ) )) |
3390 | | { |
3391 | | t_group_info->bTautFlagsDone |= TG_FLAG_ALLOW_NO_NEGTV_O_DONE; |
3392 | | } |
3393 | | } |
3394 | | else |
3395 | | { |
3396 | | for (i = 0; i < nNumCandidates; i++) |
3397 | | { |
3398 | | i1 = s_candidate[i].atnumber; |
3399 | | if (0 <= ( s_type = GetSaltChargeType( at, i1, t_group_info, &s_subtype ) ) |
3400 | | #if ( INCL_NON_SALT_CANDIDATATES == 1 ) |
3401 | | || 0 < ( s_type = GetOtherSaltChargeType( at, i1, t_group_info, &s_subtype, 1 /* bAccept_O*/ ) ) |
3402 | | #endif |
3403 | | ) |
3404 | | { |
3405 | | s_candidate[nNumCandidates].type = s_type; |
3406 | | s_candidate[nNumCandidates].subtype = s_subtype; |
3407 | | s_candidate[nNumCandidates].endpoint = at[i1].endpoint; |
3408 | | } |
3409 | | } |
3410 | | } |
3411 | | /* Look for alt paths connecting: |
3412 | | SALT_DONOR_Neg to SALT_ACCEPTOR : long distance migration of negative charges |
3413 | | SALT_DONOR_H to SALT_ACCEPTOR : long distance migration of H-atoms |
3414 | | */ |
3415 | | num_tested = 0; |
3416 | | do |
3417 | | { |
3418 | | nNumChanges = 0; |
3419 | | for (i1 = 0; i1 < nNumCandidates; i1++) |
3420 | | { |
3421 | | j1 = s_candidate[i1].atnumber; |
3422 | | for (i2 = i1 + 1; i2 < nNumCandidates; i2++) |
3423 | | { |
3424 | | if (s_candidate[i1].type && s_candidate[i2].type) |
3425 | | continue; /* both candidates are not "salt-type" */ |
3426 | | j2 = s_candidate[i2].atnumber; |
3427 | | if (at[j1].endpoint && at[j1].endpoint == at[j2].endpoint) |
3428 | | { |
3429 | | continue; |
3430 | | } |
3431 | | for (j = 0; j < 2; j++) |
3432 | | { |
3433 | | if (j) |
3434 | | { |
3435 | | ii1 = i2; /* candidate 1 (donor) ordering number */ |
3436 | | ii2 = i1; /* candidate 2 (acceptor) ordering number */ |
3437 | | jj1 = j2; /* candidate 1 (donor) atom number */ |
3438 | | jj2 = j1; /* candidate 2 (acceptor) atom number */ |
3439 | | } |
3440 | | else |
3441 | | { /* transposition */ |
3442 | | ii1 = i1; /* candidate 1 (donor) ordering number */ |
3443 | | ii2 = i2; /* candidate 2 (acceptor) ordering number */ |
3444 | | jj1 = j1; /* candidate 1 (donor) atom number */ |
3445 | | jj2 = j2; /* candidate 2 (acceptor) atom number */ |
3446 | | } |
3447 | | |
3448 | | if (( s_candidate[ii1].subtype & ( SALT_DONOR_Neg | SALT_DONOR_H ) ) && |
3449 | | ( s_candidate[ii2].subtype & SALT_ACCEPTOR )) |
3450 | | { |
3451 | | ret = bExistsAltPath( pCG, pBNS, pBD, NULL, at, num_atoms, jj2, jj1, ALT_PATH_MODE_4_SALT ); |
3452 | | num_tested++; |
3453 | | if (IS_BNS_ERROR( ret )) |
3454 | | { |
3455 | | return ret; |
3456 | | } |
3457 | | if (ret & 1) |
3458 | | { |
3459 | | nDelta = ( ret & ~3 ) >> 2; |
3460 | | nNumChanges += ( ret & 2 ); |
3461 | | for (i = 0; i < 2; i++) |
3462 | | { |
3463 | | jj = i ? jj2 : jj1; |
3464 | | EndPoint[i].nAtomNumber = jj; |
3465 | | EndPoint[i].nEquNumber = 0; |
3466 | | EndPoint[i].nGroupNumber = at[jj].endpoint; |
3467 | | if (at[jj].endpoint) |
3468 | | { |
3469 | | memset( EndPoint[i].num, 0, sizeof( EndPoint[i].num ) ); |
3470 | | } |
3471 | | else |
3472 | | { |
3473 | | AddAtom2num( EndPoint[i].num, at, jj, 2 ); /* fill out */ |
3474 | | AddAtom2DA( EndPoint[i].num_DA, at, jj, 2 ); |
3475 | | /* |
3476 | | nMobile = EndPoint[i].num[1] = (at[jj].charge == -1); |
3477 | | nMobile = EndPoint[i].num[0] = at[jj].num_H + nMobile; |
3478 | | for ( k = 0; k < T_NUM_ISOTOPIC; k ++ ) { |
3479 | | EndPoint[i].num[T_NUM_NO_ISOTOPIC+k] = at[jj].num_iso_H[NUM_H_ISOTOPES-k-1]; |
3480 | | } |
3481 | | */ |
3482 | | } |
3483 | | } |
3484 | | /* add/merge taut groups and reinit pBNS */ |
3485 | | ret = RegisterEndPoints( pCG, t_group_info, |
3486 | | EndPoint, 2, at, num_atoms, c_group_info, pBNS ); |
3487 | | if (ret < 0) |
3488 | | { |
3489 | | return ret; |
3490 | | } |
3491 | | nNumChanges += ( ret > 0 ); |
3492 | | if (nDelta) |
3493 | | { |
3494 | | goto quick_exit; |
3495 | | } |
3496 | | break; /* avoid redundant repetition */ |
3497 | | } |
3498 | | } |
3499 | | } |
3500 | | } |
3501 | | } |
3502 | | nTotNumChanges += nNumChanges; |
3503 | | } while (num_tested && nNumChanges); |
3504 | | |
3505 | | quick_exit: |
3506 | | nTotNumChanges += nNumChanges; /* nNumChanges != 0 only in case of 'goto quick_exit' */ |
3507 | | if (s_group_info->num_candidates == 0) |
3508 | | { |
3509 | | /* first time: initialize */ |
3510 | | s_group_info->num_candidates = num_tested ? nNumCandidates : -1; /* no candidate exists */ |
3511 | | } |
3512 | | } |
3513 | | return nTotNumChanges; |
3514 | | } |
3515 | | #endif |
3516 | | |
3517 | | |
3518 | | /****************************************************************************/ |
3519 | | int MergeSaltTautGroups( CANON_GLOBALS *pCG, |
3520 | | inp_ATOM *at, |
3521 | | int num_atoms, |
3522 | | S_GROUP_INFO *s_group_info, |
3523 | | T_GROUP_INFO *t_group_info, |
3524 | | C_GROUP_INFO *c_group_info, |
3525 | | struct BalancedNetworkStructure *pBNS ) |
3526 | 110k | { |
3527 | | /* Count candidates to be connected: exclude pure donors that do not belong to any t-group */ |
3528 | 110k | AT_NUMB nCurTGroupNumber; |
3529 | 110k | int i, j, /*k,*/ ret, iat, /*nMobile,*/ nMinNumEndpoints; |
3530 | 110k | int s_subtype_all, s_subtype_taut; |
3531 | 110k | int nMaxNumCandidates, nNumCandidates, nNumCandidates2; |
3532 | 110k | T_ENDPOINT EndPointStackArray[MAX_STACK_ARRAY_LEN]; /* will be reallocated if too short */ |
3533 | 110k | T_ENDPOINT *EndPoint = EndPointStackArray; |
3534 | | |
3535 | | |
3536 | 110k | if (!s_group_info || !s_group_info->s_candidate || /*s_group_info->num_candidates <= 0 ||*/ |
3537 | 110k | !t_group_info || !t_group_info->t_group || !c_group_info) |
3538 | 0 | { |
3539 | 0 | return 0; |
3540 | 0 | } |
3541 | 110k | nMinNumEndpoints = 0; |
3542 | 110k | nMaxNumCandidates = s_group_info->max_num_candidates; |
3543 | 110k | nCurTGroupNumber = MAX_ATOMS; /* impossible t-group number */ |
3544 | 110k | s_subtype_all = s_subtype_taut = 0; |
3545 | | |
3546 | | /* Collect tautomeric acidic O and previously non-tautomeric C-OH, C-SH, C-O(-), C-S(-) */ |
3547 | | /* find whether previously found tautomeric atoms have both mobile H and (-) */ |
3548 | 110k | if (1 || ( s_group_info->num_candidates < 0 )) |
3549 | 110k | { |
3550 | | /* Can be only -O(-) and -OH */ |
3551 | 110k | int s_type, s_subtype; |
3552 | 110k | S_CANDIDATE *s_candidate = s_group_info->s_candidate; |
3553 | 222k | for (i = 0, nNumCandidates = nNumCandidates2 = 0; i < num_atoms; i++) |
3554 | 112k | { |
3555 | 112k | s_subtype = 0; |
3556 | 112k | if (0 == ( s_type = GetSaltChargeType( at, i, t_group_info, &s_subtype ) ) || |
3557 | | /* -C=O or =C-OH, O = S, Se, Te */ |
3558 | | |
3559 | | /*(t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT) &&*/ |
3560 | 112k | 1 == ( s_type = GetOtherSaltChargeType( at, i, t_group_info, &s_subtype, 1/* bAccept_O*/ ) ) || |
3561 | | /* =Z-MH or -Z=M, Z = centerpoint, M = endpoint, other than above. M may be N */ |
3562 | | |
3563 | 112k | 2 == ( s_type = GetOtherSaltType( at, i, &s_subtype ) ) || |
3564 | | /* >C-SH, >C-S(-); S=S,Se,Te */ |
3565 | | |
3566 | | /* other proton donor or acceptor */ |
3567 | 112k | bHasAcidicHydrogen( at, i ) && ( ( s_type = 3 ), ( s_subtype = SALT_p_DONOR ) ) || |
3568 | 112k | bHasAcidicMinus( at, i ) && ( ( s_type = 3 ), ( s_subtype = SALT_p_ACCEPTOR ) ) |
3569 | 112k | ) |
3570 | 12 | { |
3571 | | |
3572 | 12 | if (nNumCandidates >= nMaxNumCandidates) |
3573 | 0 | { |
3574 | 0 | return BNS_VERT_EDGE_OVFL; |
3575 | 0 | } |
3576 | 12 | if (at[i].endpoint) |
3577 | 0 | { |
3578 | 0 | s_subtype_taut |= s_subtype; |
3579 | 0 | } |
3580 | 12 | else |
3581 | 12 | { |
3582 | 12 | if (bDoNotMergeNonTautAtom( at, i )) |
3583 | 2 | { |
3584 | 2 | continue; /* ignore non-tautomeric N */ |
3585 | 2 | } |
3586 | 12 | } |
3587 | 10 | if (!( s_subtype & SALT_DONOR_ALL ) || |
3588 | 10 | ( s_subtype & SALT_ACCEPTOR ) && !at[i].endpoint) |
3589 | 0 | { |
3590 | 0 | continue; /* do not include non-taut acceptors like -C=O */ |
3591 | 0 | } |
3592 | 10 | s_candidate[nNumCandidates].atnumber = i; |
3593 | 10 | s_candidate[nNumCandidates].type = s_type; |
3594 | 10 | s_candidate[nNumCandidates].subtype = s_subtype; |
3595 | 10 | s_candidate[nNumCandidates].endpoint = at[i].endpoint; |
3596 | 10 | nNumCandidates++; |
3597 | 10 | s_subtype_all |= s_subtype; |
3598 | 10 | } |
3599 | 112k | } |
3600 | | |
3601 | | /* |
3602 | | Forced merging occurs upon: |
3603 | | =========================== |
3604 | | (t_group_info->bTautFlags & TG_FLAG_ALLOW_NO_NEGTV_O) or |
3605 | | (t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT) |
3606 | | |
3607 | | |
3608 | | Allow forced merging in cases: |
3609 | | {t-groups} (H, (-)} {H, (-), t-groups} |
3610 | | |
3611 | | |
3612 | | Normal salt merging in cases: |
3613 | | (H, (-)} {H, (-), t-groups}, |
3614 | | |
3615 | | Cannot merge H into t-groups if no (-) is present |
3616 | | */ |
3617 | 110k | if (( t_group_info->bTautFlags & TG_FLAG_ALLOW_NO_NEGTV_O ) || |
3618 | 110k | ( t_group_info->bTautFlagsDone & TG_FLAG_FOUND_SALT_CHARGES_DONE ) || |
3619 | 110k | ( t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT )) |
3620 | 0 | { |
3621 | | /* Force merge even though no negative charges are present */ |
3622 | 0 | if (nNumCandidates <= 1 || |
3623 | 0 | ( !( s_subtype_all & SALT_DONOR_Neg2 ) || !( s_subtype_all & SALT_DONOR_H2 ) ) && |
3624 | 0 | !t_group_info->num_t_groups) |
3625 | 0 | { |
3626 | 0 | s_group_info->num_candidates = -1; /* no candidate exists */ |
3627 | 0 | return 0; |
3628 | 0 | } |
3629 | 0 | } |
3630 | 110k | else |
3631 | 110k | { |
3632 | | /* normal salt mode: merge if both -XH and -X(-) are present */ |
3633 | 110k | if (nNumCandidates <= 1 || |
3634 | 110k | ( !( s_subtype_all & SALT_DONOR_Neg2 ) || !( s_subtype_all & SALT_DONOR_H2 ) )) |
3635 | 110k | { |
3636 | 110k | s_group_info->num_candidates = -1; /* no candidate exists */ |
3637 | 110k | return 0; |
3638 | 110k | } |
3639 | 110k | } |
3640 | | /* -- old code -- |
3641 | | if ( nNumCandidates <= 1 || |
3642 | | (((t_group_info->bTautFlags & TG_FLAG_ALLOW_NO_NEGTV_O) || |
3643 | | (t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT)) ? |
3644 | | !(s_subtype_all & SALT_DONOR_ALL): |
3645 | | !(s_subtype_all & SALT_DONOR_Neg2) |
3646 | | ) |
3647 | | ) { |
3648 | | s_group_info->num_candidates = -1; |
3649 | | return 0; |
3650 | | } |
3651 | | */ |
3652 | 4 | if (!( s_subtype_all & ( SALT_DONOR_Neg2 ) )) |
3653 | 0 | { |
3654 | 0 | t_group_info->bTautFlagsDone |= TG_FLAG_ALLOW_NO_NEGTV_O_DONE; |
3655 | 0 | } |
3656 | 4 | s_group_info->num_candidates = nNumCandidates; |
3657 | 4 | } |
3658 | | |
3659 | 12 | for (i = 0; i < s_group_info->num_candidates; i++) |
3660 | 8 | { |
3661 | 8 | iat = s_group_info->s_candidate[i].atnumber; |
3662 | 8 | if (( s_group_info->s_candidate[i].subtype & SALT_ACCEPTOR ) && !at[iat].endpoint) |
3663 | 0 | { |
3664 | 0 | continue; /* should not happen */ |
3665 | 0 | } |
3666 | 8 | s_subtype_all |= s_group_info->s_candidate[i].subtype; |
3667 | 8 | if (at[iat].endpoint != nCurTGroupNumber || !at[iat].endpoint) |
3668 | 8 | { |
3669 | 8 | nMinNumEndpoints++; |
3670 | 8 | } |
3671 | 8 | nCurTGroupNumber = (int) at[iat].endpoint; |
3672 | 8 | } |
3673 | 4 | if (nMinNumEndpoints <= 1) |
3674 | 0 | { |
3675 | 0 | return 0; /* too few endpoints */ |
3676 | 0 | } |
3677 | | |
3678 | | /* Make sure we have enough memory */ |
3679 | 4 | if (nMinNumEndpoints > MAX_STACK_ARRAY_LEN) |
3680 | 0 | { |
3681 | 0 | if (!( EndPoint = (T_ENDPOINT *) inchi_calloc( nMinNumEndpoints, sizeof( EndPoint[0] ) ) )) |
3682 | 0 | { |
3683 | | /*printf("BNS_OUT_OF_RAM-8\n");*/ |
3684 | 0 | return BNS_OUT_OF_RAM; |
3685 | 0 | } |
3686 | 0 | } |
3687 | | |
3688 | 4 | nCurTGroupNumber = MAX_ATOMS; /* impossible t-group number */ |
3689 | 12 | for (i = j = 0; i < s_group_info->num_candidates; i++) |
3690 | 8 | { |
3691 | 8 | iat = s_group_info->s_candidate[i].atnumber; |
3692 | 8 | if (s_group_info->s_candidate[i].subtype == SALT_ACCEPTOR && !at[iat].endpoint) |
3693 | 0 | { |
3694 | 0 | continue; |
3695 | 0 | } |
3696 | 8 | if (at[iat].endpoint != nCurTGroupNumber || !at[iat].endpoint) |
3697 | 8 | { |
3698 | 8 | AddEndPoint( EndPoint + j, at, iat ); |
3699 | 8 | j++; |
3700 | 8 | } |
3701 | 8 | nCurTGroupNumber = (int) at[iat].endpoint; |
3702 | 8 | } |
3703 | | |
3704 | 4 | ret = RegisterEndPoints( pCG, t_group_info, EndPoint, j, at, |
3705 | 4 | num_atoms, c_group_info, pBNS ); |
3706 | | |
3707 | 4 | if (ret == -1) |
3708 | 0 | { |
3709 | 0 | ret = BNS_PROGRAM_ERR; |
3710 | 0 | } |
3711 | | |
3712 | 4 | if (EndPoint != EndPointStackArray) |
3713 | 0 | { |
3714 | 0 | inchi_free( EndPoint ); |
3715 | 0 | } |
3716 | | |
3717 | 4 | return ret; |
3718 | 4 | } |
3719 | | |
3720 | | |
3721 | | /****************************************************************************/ |
3722 | | int MakeIsotopicHGroup( inp_ATOM *at, |
3723 | | int num_atoms, |
3724 | | S_GROUP_INFO *s_group_info, |
3725 | | T_GROUP_INFO *t_group_info ) |
3726 | 45 | { |
3727 | | /* All tautomeric atoms and all possible H+ donors and acceptors that have H */ |
3728 | 45 | int i, j, k, n, bHasH, tg, nError = 0; |
3729 | 45 | int s_subtype_all, s_subtype_taut; |
3730 | 45 | int nMaxNumCandidates, nNumCandidates, nNumNonTautCandidates; |
3731 | | |
3732 | | |
3733 | 45 | if (!s_group_info || !s_group_info->s_candidate || /*s_group_info->num_candidates <= 0 ||*/ |
3734 | 45 | !t_group_info || !t_group_info->t_group) |
3735 | 0 | { |
3736 | 0 | return 0; |
3737 | 0 | } |
3738 | 45 | nMaxNumCandidates = s_group_info->max_num_candidates; |
3739 | 45 | s_subtype_all = s_subtype_taut = 0; |
3740 | 45 | memset( t_group_info->num_iso_H, 0, sizeof( t_group_info->num_iso_H ) ); |
3741 | 45 | if (1 || ( s_group_info->num_candidates < 0 )) |
3742 | 45 | { |
3743 | 45 | int s_type, s_subtype; |
3744 | 45 | S_CANDIDATE *s_candidate = s_group_info->s_candidate; |
3745 | 1.35k | for (i = 0, nNumCandidates = nNumNonTautCandidates = 0; i < num_atoms; i++) |
3746 | 1.31k | { |
3747 | 1.31k | s_subtype = 0; |
3748 | 1.31k | s_type = 0; |
3749 | 1.31k | if (at[i].endpoint) |
3750 | 4 | { |
3751 | 4 | if (( tg = t_group_info->tGroupNumber[at[i].endpoint] ) && |
3752 | 4 | at[i].endpoint == t_group_info->t_group[tg -= 1].nGroupNumber) |
3753 | 4 | { |
3754 | 4 | bHasH = (int) t_group_info->t_group[tg].num[0] - (int) t_group_info->t_group[tg].num[1]; |
3755 | 4 | } |
3756 | 0 | else |
3757 | 0 | { |
3758 | 0 | nError = BNS_PROGRAM_ERR; |
3759 | 0 | break; |
3760 | 0 | } |
3761 | 4 | } |
3762 | 1.31k | else |
3763 | 1.31k | { |
3764 | 1.31k | bHasH = (int) at[i].num_H; |
3765 | 1.31k | } |
3766 | 1.31k | if (bHasH && at[i].endpoint || /* tautomeric atoms */ |
3767 | | /* non-tautomeric heteroatoms that |
3768 | | (a) have H and |
3769 | | (b) may be donors of H |
3770 | | therefore may exchange isotopic-non-isotopic H */ |
3771 | 1.31k | bHasH && |
3772 | 1.31k | ( 0 == ( s_type = GetSaltChargeType( at, i, t_group_info, &s_subtype ) ) || |
3773 | | /* -C=O or =C-OH, O = S, Se, Te */ |
3774 | | |
3775 | | /*(t_group_info->tni.bNormalizationFlags & FLAG_FORCE_SALT_TAUT) &&*/ |
3776 | 0 | 1 == ( s_type = GetOtherSaltChargeType( at, i, t_group_info, &s_subtype, 1/* bAccept_O*/ ) ) || |
3777 | | /* =Z-MH or -Z=M, Z = centerpoint, M = endpoint, other than above. M may be N */ |
3778 | |
|
3779 | 0 | 2 == ( s_type = GetOtherSaltType( at, i, &s_subtype ) ) || |
3780 | | /* >C-SH, >C-S(-); S=S,Se,Te */ |
3781 | | |
3782 | | /* other proton donor or acceptor */ |
3783 | 0 | bHasAcidicHydrogen( at, i ) && ( ( s_type = 3 ), ( s_subtype = SALT_p_DONOR ) ) || |
3784 | 0 | bHasAcidicMinus( at, i ) && ( ( s_type = 3 ), ( s_subtype = SALT_p_ACCEPTOR ) ) || |
3785 | 0 | bHasOtherExchangableH( at, i ) && ( ( s_type = 3 ), ( s_subtype = SALT_DONOR_H ) ) ) |
3786 | | |
3787 | 1.31k | ) |
3788 | 4 | { |
3789 | 4 | if (nNumCandidates >= nMaxNumCandidates) |
3790 | 0 | { |
3791 | 0 | return BNS_VERT_EDGE_OVFL; |
3792 | 0 | } |
3793 | 4 | s_candidate[nNumCandidates].atnumber = i; |
3794 | 4 | s_candidate[nNumCandidates].type = s_type; |
3795 | 4 | s_candidate[nNumCandidates].subtype = s_subtype; |
3796 | 4 | s_candidate[nNumCandidates].endpoint = at[i].endpoint; |
3797 | 4 | nNumCandidates++; |
3798 | 4 | nNumNonTautCandidates += !at[i].endpoint; |
3799 | 4 | s_subtype_all |= s_subtype; |
3800 | 4 | } |
3801 | 1.31k | } |
3802 | | |
3803 | 45 | if (nError) |
3804 | 0 | { |
3805 | 0 | return nError; |
3806 | 0 | } |
3807 | | |
3808 | 45 | if (nNumCandidates > 0) |
3809 | 2 | { |
3810 | 2 | t_group_info->nIsotopicEndpointAtomNumber = (AT_NUMB *) inchi_calloc( nNumNonTautCandidates + 1, sizeof( t_group_info->nIsotopicEndpointAtomNumber[0] ) ); |
3811 | 2 | t_group_info->nIsotopicEndpointAtomNumber[0] = nNumNonTautCandidates; |
3812 | 6 | for (i = 0, n = 1; i < nNumCandidates; i++) |
3813 | 4 | { |
3814 | 4 | k = s_candidate[i].atnumber; |
3815 | 4 | if (!at[k].endpoint) |
3816 | 0 | { |
3817 | 0 | t_group_info->nIsotopicEndpointAtomNumber[n++] = k; |
3818 | 0 | } |
3819 | 16 | for (j = 0; j < NUM_H_ISOTOPES; j++) |
3820 | 12 | { |
3821 | 12 | t_group_info->num_iso_H[j] += at[k].num_iso_H[j]; |
3822 | 12 | } |
3823 | 4 | at[k].cFlags |= AT_FLAG_ISO_H_POINT; |
3824 | 4 | } |
3825 | 2 | t_group_info->nNumIsotopicEndpoints = nNumNonTautCandidates + 1; |
3826 | 2 | } |
3827 | 45 | } |
3828 | | |
3829 | 45 | return nNumCandidates; |
3830 | 45 | } |
3831 | | |
3832 | | |
3833 | | /*#else*/ /* } DISCONNECT_SALTS == 0 */ |
3834 | | |
3835 | | |
3836 | | /********************************************************************************** |
3837 | | Charges and tautomeric endpoints (N only) |
3838 | | ********************************************************************************** |
3839 | | |
3840 | | H = number of possibly moveable hydrogen atoms |
3841 | | C = possibly moveable positive charge |
3842 | | |
3843 | | - = single bond |
3844 | | = = double bond |
3845 | | # = triple bond |
3846 | | |
3847 | | +-----------------------------------------------------------------------------+ |
3848 | | |ca-| H | edges to t- | 1 bond | 2 bonds | 3 bonds *) | |
3849 | | |se | C | and c-groups | (valence) | (valence) | (valence) | |
3850 | | | # | | (edges flow) | | | | |
3851 | | +---|------+---------------+----------------+----------------+----------------| |
3852 | | | 1 | H=0 | -- (1) | =NH (3) | =N- (3) | >N- (3) | |
3853 | | | | C=0 | == | | | | |
3854 | | +---|------+---------------+----------------+----------------+----------------| |
3855 | | | 2 | H=1 | == (2) | -NH2 (3) | -NH- (3) | none | |
3856 | | | | C=0 | == | | | | |
3857 | | +---|------+---------------+----------------+----------------+----------------| |
3858 | | | 3 | H=0 | -- (0) | #NH(+) (4) | =N(+)= (4) +)| >N(+)= (4) | |
3859 | | | | C=1 | -- | (prohibited | | | |
3860 | | | | | | by edge cap) | | | |
3861 | | +---|------+---------------+----------------+----------------+----------------| |
3862 | | | 4 | H=1 | == (1) | =NH2(+) (4) +)| =NH(+)- (4) +)| >NH(+)- (4) | |
3863 | | | | C=1 | -- | | | | |
3864 | | +---+-------------------------------------------------------------------------+ |
3865 | | |
3866 | | *) Cannot be a tautomeric endpoint |
3867 | | |
3868 | | +) The three charged types of atoms [=N(+)=, =NH(+)-, =NH2(+)] should be |
3869 | | checked for possible H-tautomerism. Other types in the marked by *) |
3870 | | column should not be checked as long as H(+) exchange is not considered |
3871 | | tautomeric. |
3872 | | |
3873 | | Other possibilities: -NH3(+) >NH2(+) >N(+)< cannot be H-tautomeric endpoints. |
3874 | | |
3875 | | Case #1 (H=0, C=0) and #4 (H=1,C=0) is indistinguishable from the |
3876 | | viewpoint of edges flow and capacities except for flow from N to (+) vertex. |
3877 | | |
3878 | | Without taking precautions H(+) can be transferred |
3879 | | |
3880 | | from =NH2(+) to =NH, |
3881 | | from =NH(+)- to =N-, |
3882 | | from >NH(+)- to >N- |
3883 | | |
3884 | | or to any other appropriate atom that has a lone electron pair and bonds |
3885 | | will not change. In this case no bond must be marked as tautomeric. |
3886 | | |
3887 | | For this reason before attempting to transfer H from one endpoint to |
3888 | | another the charges on the two atoms should be set to zero by |
3889 | | forcing zero flow from each of atoms to the (+)-vertices if the |
3890 | | atoms belong to a c-group. |
3891 | | |
3892 | | **********************************************************************************/ |
3893 | | |
3894 | | |
3895 | | /**************************************************************************** |
3896 | | Mark Tautomer Groups: |
3897 | | do not identify positively charged N as endpoints for now |
3898 | | ****************************************************************************/ |
3899 | | int MarkTautomerGroups( CANON_GLOBALS *pCG, |
3900 | | inp_ATOM *at, |
3901 | | int num_atoms, |
3902 | | T_GROUP_INFO *t_group_info, |
3903 | | C_GROUP_INFO *c_group_info, |
3904 | | struct BalancedNetworkStructure *pBNS, |
3905 | | struct BalancedNetworkData *pBD ) |
3906 | 110k | { |
3907 | 110k | int i, j, k, m, endpoint_valence, centerpoint, endpoint, bond_type, nMobile, num_changes = 0, tot_changes = 0; |
3908 | 110k | T_ENDPOINT EndPoint[MAXVAL]; |
3909 | 110k | T_BONDPOS BondPos[MAXVAL]; |
3910 | 110k | AT_NUMB nGroupNumber; |
3911 | 110k | int bDiffGroups; |
3912 | 110k | int nNumEndPoints, nNumBondPos, nNumPossibleMobile; |
3913 | 110k | int bTautBond, bNonTautBond, bAltBond; |
3914 | 110k | int nNumDonor, nNumAcceptor, bPossiblyEndpoint; |
3915 | 110k | T_GROUP *t_group; |
3916 | 110k | int *pnum_t, max_num_t, bIgnoreIsotopic; |
3917 | 110k | ENDPOINT_INFO eif1, eif2; |
3918 | 110k | int nErr = 0; |
3919 | 110k | #define ALLOWED_EDGE(PBNS, IAT,IBOND) ( !(PBNS) || !(PBNS)->edge || !(PBNS)->vert || !(PBNS)->edge[(PBNS)->vert[IAT].iedge[IBOND]].forbidden) |
3920 | 110k | #define ACTUAL_ORDER(PBNS, IAT,IBOND, BTYPE) ( ((PBNS) && (PBNS)->edge && (PBNS)->vert &&\ |
3921 | 110k | ((BTYPE)==BOND_ALT_123 || (BTYPE)==BOND_ALT_13 || (BTYPE)==BOND_ALT_23))? (PBNS)->edge[(PBNS)->vert[IAT].iedge[IBOND]].flow+BOND_TYPE_SINGLE:(BTYPE)) |
3922 | | |
3923 | | |
3924 | 110k | if (!t_group_info || !( t_group_info->bTautFlags & TG_FLAG_TEST_TAUT__ATOMS )) |
3925 | 3 | { |
3926 | 3 | return 0; |
3927 | 3 | } |
3928 | | |
3929 | | /* Initial t_group allocation */ |
3930 | 110k | if (!t_group_info->t_group && !t_group_info->max_num_t_groups) |
3931 | 110k | { |
3932 | 110k | INCHI_MODE bTautFlags = t_group_info->bTautFlags; /* save initial setting */ |
3933 | 110k | INCHI_MODE bTautFlagsDone = t_group_info->bTautFlagsDone; /* save previous findings, if any */ |
3934 | 110k | TNI tni = t_group_info->tni; |
3935 | 110k | AT_NUMB *tGroupNumber = t_group_info->tGroupNumber; |
3936 | 110k | bIgnoreIsotopic = t_group_info->bIgnoreIsotopic; |
3937 | 110k | memset( t_group_info, 0, sizeof( *t_group_info ) ); |
3938 | 110k | t_group_info->bIgnoreIsotopic = bIgnoreIsotopic; /* restore initial setting */ |
3939 | 110k | t_group_info->bTautFlags = bTautFlags; |
3940 | 110k | t_group_info->bTautFlagsDone = bTautFlagsDone; |
3941 | 110k | t_group_info->tni = tni; |
3942 | 110k | t_group_info->tGroupNumber = tGroupNumber; |
3943 | 110k | t_group_info->max_num_t_groups = num_atoms / 2 + 1; /* upper limit */ |
3944 | 110k | if (!( t_group_info->t_group = (T_GROUP*) inchi_calloc( t_group_info->max_num_t_groups, sizeof( t_group[0] ) ) )) |
3945 | 0 | { |
3946 | 0 | return ( t_group_info->max_num_t_groups = -1 ); /* failed, out of RAM */ |
3947 | 0 | } |
3948 | 110k | } |
3949 | | |
3950 | | /* Check if t_group_info exists */ |
3951 | 110k | if (!t_group_info->t_group || !t_group_info->max_num_t_groups) |
3952 | 0 | { |
3953 | 0 | return 0; |
3954 | 0 | } |
3955 | | |
3956 | 110k | if (0 > t_group_info->max_num_t_groups) |
3957 | 0 | { |
3958 | 0 | return t_group_info->max_num_t_groups; |
3959 | 0 | } |
3960 | | |
3961 | 110k | pnum_t = &t_group_info->num_t_groups; /* number of found tautomer endpoint groups */ |
3962 | 110k | t_group = t_group_info->t_group; |
3963 | 110k | max_num_t = t_group_info->max_num_t_groups; |
3964 | 110k | bIgnoreIsotopic = t_group_info->bIgnoreIsotopic; |
3965 | | |
3966 | | /* 1-3 tautomers */ |
3967 | 222k | for (i = 0; i < num_atoms; i++) |
3968 | 112k | { |
3969 | | /* Find possible endpoint Z = at[i] */ |
3970 | 112k | if (endpoint_valence = nGetEndpointInfo( at, i, &eif1 )) |
3971 | 12 | { |
3972 | | /* 1st endpoint candidate found. Find centerpoint candidate */ |
3973 | 26 | for (j = 0; j < at[i].valence; j++) |
3974 | 14 | { |
3975 | 14 | bond_type = (int) at[i].bond_type[j] & ~BOND_MARK_ALL; |
3976 | | #if ( FIX_BOND23_IN_TAUT == 1 ) |
3977 | | bond_type = ACTUAL_ORDER( pBNS, i, j, bond_type ); |
3978 | | #endif |
3979 | 14 | centerpoint = (int) at[i].neighbor[j]; /* a centerpoint candidate */ |
3980 | 14 | if (( bond_type == BOND_DOUBLE || |
3981 | 14 | bond_type == BOND_ALTERN || |
3982 | 14 | bond_type == BOND_ALT12NS || |
3983 | 14 | bond_type == BOND_TAUTOM ) && is_centerpoint_elem( at[centerpoint].el_number ) |
3984 | 14 | && ALLOWED_EDGE( pBNS, i, j ) |
3985 | 14 | ) |
3986 | 0 | { |
3987 | | /* Test a centerpoint candidate. */ |
3988 | | /* find all endpoints including at[i] and store them into EndPoint[] */ |
3989 | 0 | nNumPossibleMobile = 0; |
3990 | 0 | nGroupNumber = (AT_NUMB) num_atoms; /* greater than any tautomeric group number */ |
3991 | 0 | bDiffGroups = -1; /* ignore the first difference */ |
3992 | 0 | nNumDonor = nNumAcceptor = 0; |
3993 | 0 | for (k = 0, nNumEndPoints = 0, nNumBondPos = 0; k < at[centerpoint].valence; k++) |
3994 | 0 | { |
3995 | 0 | endpoint = at[centerpoint].neighbor[k]; /* endpoint candidate */ |
3996 | 0 | bond_type = (int) at[centerpoint].bond_type[k] & ~BOND_MARK_ALL; |
3997 | | #if ( FIX_BOND23_IN_TAUT == 1 ) |
3998 | | bond_type = ACTUAL_ORDER( pBNS, centerpoint, k, bond_type ); |
3999 | | #endif |
4000 | 0 | bTautBond = |
4001 | 0 | bNonTautBond = |
4002 | 0 | bAltBond = |
4003 | 0 | bPossiblyEndpoint = 0; |
4004 | 0 | if (!ALLOWED_EDGE( pBNS, centerpoint, k )) |
4005 | 0 | { |
4006 | 0 | continue; |
4007 | 0 | } |
4008 | 0 | else |
4009 | 0 | { |
4010 | 0 | if (bond_type == BOND_ALTERN || bond_type == BOND_ALT12NS || bond_type == BOND_TAUTOM) |
4011 | 0 | { |
4012 | 0 | bTautBond = 1; |
4013 | 0 | #if ( REPLACE_ALT_WITH_TAUT == 1 ) |
4014 | 0 | bAltBond = ( bond_type == BOND_ALTERN || bond_type == BOND_ALT12NS ); |
4015 | 0 | #endif |
4016 | 0 | } |
4017 | 0 | else |
4018 | 0 | { |
4019 | 0 | if (bond_type == BOND_SINGLE || bond_type == BOND_DOUBLE) |
4020 | 0 | { |
4021 | 0 | bNonTautBond = 1; |
4022 | 0 | } |
4023 | 0 | else |
4024 | 0 | { |
4025 | 0 | continue; |
4026 | 0 | } |
4027 | 0 | } |
4028 | 0 | } |
4029 | | |
4030 | 0 | if (!( endpoint_valence = nGetEndpointInfo( at, endpoint, &eif1 ) )) |
4031 | 0 | continue; /* not an endpoint element or can't have mobile groups */ |
4032 | | /* save information about the found possible tautomeric endpoint */ |
4033 | | /* 2 = T_NUM_NO_ISOTOPIC non-isotopic values */ |
4034 | 0 | nMobile = |
4035 | 0 | AddAtom2num( EndPoint[nNumEndPoints].num, at, endpoint, 2 ); /* fill out */ |
4036 | 0 | AddAtom2DA( EndPoint[nNumEndPoints].num_DA, at, endpoint, 2 ); |
4037 | | /* --- why is isitopic info missing ? -- see below |
4038 | | nMobile = EndPoint[nNumEndPoints].num[1] = (at[endpoint].charge == -1); |
4039 | | nMobile = EndPoint[nNumEndPoints].num[0] = at[endpoint].num_H + nMobile; |
4040 | | */ |
4041 | 0 | if (bNonTautBond) |
4042 | 0 | { |
4043 | 0 | m = ( bond_type == BOND_SINGLE && ( nMobile || at[endpoint].endpoint ) ); |
4044 | 0 | nNumDonor += m; |
4045 | 0 | bPossiblyEndpoint += m; |
4046 | 0 | m = ( bond_type == BOND_DOUBLE ); |
4047 | 0 | nNumAcceptor += m; |
4048 | 0 | bPossiblyEndpoint += m; |
4049 | 0 | } |
4050 | 0 | else |
4051 | 0 | { |
4052 | | /* Tautomeric or alternating bond */ |
4053 | 0 | m = ( 0 != at[endpoint].endpoint || eif1.cDonor ); |
4054 | 0 | nNumDonor += m; |
4055 | 0 | bPossiblyEndpoint += m; |
4056 | 0 | m = ( at[endpoint].endpoint || |
4057 | 0 | eif1.cNeutralBondsValence > at[endpoint].valence ); |
4058 | 0 | nNumAcceptor += m; |
4059 | 0 | bPossiblyEndpoint += m; |
4060 | 0 | } |
4061 | 0 | if (!bPossiblyEndpoint) |
4062 | 0 | { |
4063 | 0 | continue; |
4064 | 0 | } |
4065 | 0 | EndPoint[nNumEndPoints].nGroupNumber = at[endpoint].endpoint; /* =0 if it is an endpoint for the 1st time */ |
4066 | 0 | EndPoint[nNumEndPoints].nEquNumber = 0; |
4067 | 0 | EndPoint[nNumEndPoints].nAtomNumber = (AT_NUMB) endpoint; |
4068 | 0 | if (nGroupNumber != at[endpoint].endpoint) |
4069 | 0 | { |
4070 | 0 | bDiffGroups++; |
4071 | 0 | nGroupNumber = at[endpoint].endpoint; |
4072 | 0 | } |
4073 | | |
4074 | | /* save positions of all, not only possibly tautomeric bonds */ |
4075 | | #if ( REPLACE_ALT_WITH_TAUT != 1 ) |
4076 | | if (bNonTautBond || bAltBond) |
4077 | | { |
4078 | | #endif |
4079 | 0 | BondPos[nNumBondPos].nAtomNumber = (AT_NUMB) centerpoint; |
4080 | 0 | BondPos[nNumBondPos].neighbor_index = (AT_NUMB) k; /* bond ordering number; used to change bonds to tautomeric only */ |
4081 | 0 | nNumBondPos++; |
4082 | | #if ( REPLACE_ALT_WITH_TAUT != 1 ) |
4083 | | } |
4084 | | #endif |
4085 | | /* mobile group is possible if (a) the endpoint has a mobile group or */ |
4086 | | /* (b) the centerpoint is adjacent to another endpoint */ |
4087 | 0 | nNumPossibleMobile += ( nMobile > 0 || at[endpoint].endpoint ); |
4088 | 0 | nNumEndPoints++; |
4089 | | /*printf("Found %d %d %d %d\n", centerpoint+1, at[centerpoint].el_number, endpoint+1, at[endpoint].el_number);*/ |
4090 | 0 | } |
4091 | 0 | if (nNumEndPoints > 1 && nNumPossibleMobile && nNumDonor && nNumAcceptor) |
4092 | 0 | { |
4093 | | /* |
4094 | | * a tautomeric group has been found |
4095 | | * |
4096 | | * at this point: |
4097 | | * nGroupNumber = 0 if all endpoints belong to a newly discovered tautomeric group |
4098 | | * bDiffGroups > 0 if at least 2 tautomeric groups are to be merged (one of them can be new) |
4099 | | * case (nGroupNumber != 0 && bDiffGroups = 0 ) ignored because all endpoints belong to the same known t-group |
4100 | | * case (nGroupNumber != 0 && bDiffGroups < 0 ) cannot happen |
4101 | | */ |
4102 | |
|
4103 | 0 | nErr = FindAccessibleEndPoints( pCG, EndPoint, |
4104 | 0 | &nNumEndPoints, |
4105 | 0 | BondPos, &nNumBondPos, |
4106 | 0 | pBNS, pBD, at, |
4107 | 0 | num_atoms, c_group_info, |
4108 | 0 | ALT_PATH_MODE_TAUTOM ); |
4109 | |
|
4110 | 0 | if (IS_BNS_ERROR( nErr )) |
4111 | 0 | { |
4112 | 0 | return nErr; |
4113 | 0 | } |
4114 | 0 | nErr = 0; |
4115 | 0 | if (nNumEndPoints > 0) { |
4116 | 0 | if (!nGroupNumber || bDiffGroups > 0) { |
4117 | |
|
4118 | 0 | num_changes = RegisterEndPoints(pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS); |
4119 | 0 | if (num_changes == -1) { |
4120 | 0 | nErr = CT_TAUCOUNT_ERR; |
4121 | 0 | } |
4122 | 0 | if (num_changes < 0) { |
4123 | 0 | nErr = num_changes; |
4124 | 0 | } |
4125 | 0 | if (nErr) |
4126 | 0 | goto exit_function; |
4127 | 0 | tot_changes += (num_changes>0); |
4128 | 0 | } |
4129 | 0 | if (nNumBondPos > 0) { |
4130 | | /* some of the bonds have not been marked as tautomeric yet */ |
4131 | 0 | num_changes = SetTautomericBonds(at, nNumBondPos, BondPos); |
4132 | 0 | tot_changes += (num_changes>0); |
4133 | 0 | } |
4134 | 0 | } |
4135 | 0 | } |
4136 | 0 | } |
4137 | 14 | } |
4138 | 12 | } |
4139 | 112k | } |
4140 | | |
4141 | | |
4142 | 110k | #if ( KETO_ENOL_TAUT == 1 ) /***** post v.1 feature *****/ |
4143 | 110k | if (t_group_info->bTautFlags & TG_FLAG_KETO_ENOL_TAUT) |
4144 | 0 | { |
4145 | | /* 1,3 keto-enol tautomerism */ |
4146 | 0 | for (i = 0; i < num_atoms; i++) |
4147 | 0 | { |
4148 | | /* Find possible endpoint Z = at[i] */ |
4149 | 0 | if (endpoint_valence = nGetEndpointInfo_KET( at, i, &eif1 )) |
4150 | 0 | { |
4151 | | /* 1st endpoint candidate found. Find centerpoint candidate */ |
4152 | 0 | for (j = 0; j < at[i].valence; j++) |
4153 | 0 | { |
4154 | 0 | bond_type = (int) at[i].bond_type[j] & ~BOND_MARK_ALL; |
4155 | | #if ( FIX_BOND23_IN_TAUT == 1 ) |
4156 | | bond_type = ACTUAL_ORDER( pBNS, i, j, bond_type ); |
4157 | | #endif |
4158 | 0 | centerpoint = (int) at[i].neighbor[j]; /* a centerpoint candidate */ |
4159 | 0 | if (( bond_type == BOND_DOUBLE || |
4160 | 0 | bond_type == BOND_ALTERN || |
4161 | 0 | bond_type == BOND_ALT12NS || |
4162 | 0 | bond_type == BOND_TAUTOM ) && |
4163 | 0 | is_centerpoint_elem_KET( at[centerpoint].el_number ) && |
4164 | 0 | !at[centerpoint].charge && !at[centerpoint].radical && |
4165 | | /* only normal carbon is allowed */ |
4166 | 0 | 4 == at[centerpoint].chem_bonds_valence + at[centerpoint].num_H |
4167 | 0 | && ALLOWED_EDGE( pBNS, i, j )) |
4168 | 0 | { |
4169 | 0 | int num_O = 0; |
4170 | 0 | int num_C = 0; |
4171 | | /* Test a centerpoint candidate. */ |
4172 | | /* find all endpoints including at[i] and store them into EndPoint[] */ |
4173 | 0 | nNumPossibleMobile = 0; |
4174 | 0 | nGroupNumber = (AT_NUMB) num_atoms; /* greater than any tautomeric group number */ |
4175 | 0 | bDiffGroups = -1; /* ignore the first difference */ |
4176 | 0 | nNumDonor = nNumAcceptor = 0; |
4177 | 0 | for (k = 0, nNumEndPoints = 0, nNumBondPos = 0; k < at[centerpoint].valence; k++) |
4178 | 0 | { |
4179 | 0 | endpoint = at[centerpoint].neighbor[k]; /* endpoint candidate */ |
4180 | 0 | bond_type = (int) at[centerpoint].bond_type[k] & ~BOND_MARK_ALL; |
4181 | | #if ( FIX_BOND23_IN_TAUT == 1 ) |
4182 | | bond_type = ACTUAL_ORDER( pBNS, centerpoint, k, bond_type ); |
4183 | | #endif |
4184 | 0 | bTautBond = |
4185 | 0 | bNonTautBond = |
4186 | 0 | bAltBond = |
4187 | 0 | bPossiblyEndpoint = 0; |
4188 | 0 | if (!ALLOWED_EDGE( pBNS, centerpoint, k )) |
4189 | 0 | { |
4190 | 0 | continue; |
4191 | 0 | } |
4192 | 0 | else |
4193 | 0 | { |
4194 | 0 | if (bond_type == BOND_ALTERN || bond_type == BOND_ALT12NS || bond_type == BOND_TAUTOM) |
4195 | 0 | { |
4196 | 0 | bTautBond = 1; |
4197 | 0 | #if ( REPLACE_ALT_WITH_TAUT == 1 ) |
4198 | 0 | bAltBond = ( bond_type == BOND_ALTERN || bond_type == BOND_ALT12NS ); |
4199 | 0 | #endif |
4200 | 0 | } |
4201 | 0 | else |
4202 | 0 | { |
4203 | 0 | if (bond_type == BOND_SINGLE || bond_type == BOND_DOUBLE) |
4204 | 0 | { |
4205 | 0 | bNonTautBond = 1; |
4206 | 0 | } |
4207 | 0 | else |
4208 | 0 | { |
4209 | 0 | continue; |
4210 | 0 | } |
4211 | 0 | } |
4212 | 0 | } |
4213 | | |
4214 | 0 | if (!( endpoint_valence = nGetEndpointInfo_KET( at, endpoint, &eif2 ) )) |
4215 | 0 | { |
4216 | 0 | continue; |
4217 | 0 | } |
4218 | | /* |
4219 | | if ( 3 != eif1.cKetoEnolCode + eif2.cKetoEnolCode && endpoint != i ) |
4220 | | continue; |
4221 | | */ |
4222 | | /* save information about the found possible tautomeric endpoint */ |
4223 | | /* 2 = T_NUM_NO_ISOTOPIC non-isotopic values */ |
4224 | 0 | nMobile = |
4225 | 0 | AddAtom2num( EndPoint[nNumEndPoints].num, at, endpoint, 2 ); /* fill out */ |
4226 | 0 | AddAtom2DA( EndPoint[nNumEndPoints].num_DA, at, endpoint, 2 ); |
4227 | | /* --- why is isitopic info missing ? -- see below |
4228 | | nMobile = EndPoint[nNumEndPoints].num[1] = (at[endpoint].charge == -1); |
4229 | | nMobile = EndPoint[nNumEndPoints].num[0] = at[endpoint].num_H + nMobile; |
4230 | | */ |
4231 | 0 | if (bNonTautBond) |
4232 | 0 | { |
4233 | 0 | m = ( bond_type == BOND_SINGLE && ( nMobile || at[endpoint].endpoint ) ); |
4234 | 0 | nNumDonor += m; |
4235 | 0 | bPossiblyEndpoint += m; |
4236 | 0 | m = ( bond_type == BOND_DOUBLE ); |
4237 | 0 | nNumAcceptor += m; |
4238 | 0 | bPossiblyEndpoint += m; |
4239 | 0 | } |
4240 | 0 | else |
4241 | 0 | { |
4242 | | /* Tautomeric or alternating bond */ |
4243 | 0 | m = ( 0 != at[endpoint].endpoint || eif1.cDonor ); |
4244 | 0 | nNumDonor += m; |
4245 | 0 | bPossiblyEndpoint += m; |
4246 | 0 | m = ( at[endpoint].endpoint || |
4247 | 0 | eif1.cNeutralBondsValence > at[endpoint].valence ); |
4248 | 0 | nNumAcceptor += m; |
4249 | 0 | bPossiblyEndpoint += m; |
4250 | 0 | } |
4251 | 0 | if (!bPossiblyEndpoint) |
4252 | 0 | { |
4253 | 0 | continue; |
4254 | 0 | } |
4255 | | |
4256 | 0 | num_O += ( endpoint_valence == 2 ); |
4257 | 0 | num_C += ( endpoint_valence == 4 ); |
4258 | |
|
4259 | 0 | EndPoint[nNumEndPoints].nGroupNumber = at[endpoint].endpoint; /* =0 if it is an endpoint for the 1st time */ |
4260 | 0 | EndPoint[nNumEndPoints].nEquNumber = 0; |
4261 | 0 | EndPoint[nNumEndPoints].nAtomNumber = (AT_NUMB) endpoint; |
4262 | 0 | if (nGroupNumber != at[endpoint].endpoint) |
4263 | 0 | { |
4264 | 0 | bDiffGroups++; |
4265 | 0 | nGroupNumber = at[endpoint].endpoint; |
4266 | 0 | } |
4267 | | |
4268 | | /* Save positions of all, not only possibly tautomeric bonds */ |
4269 | | #if ( REPLACE_ALT_WITH_TAUT != 1 ) |
4270 | | if (bNonTautBond || bAltBond) |
4271 | | { |
4272 | | #endif |
4273 | 0 | BondPos[nNumBondPos].nAtomNumber = (AT_NUMB) centerpoint; |
4274 | 0 | BondPos[nNumBondPos].neighbor_index = (AT_NUMB) k; /* bond ordering number; used to change bonds to tautomeric only */ |
4275 | 0 | nNumBondPos++; |
4276 | | #if ( REPLACE_ALT_WITH_TAUT != 1 ) |
4277 | | } |
4278 | | #endif |
4279 | | /* Mobile group is possible if (a) the endpoint has a mobile group or */ |
4280 | | /* (b) the centerpoint is adjacent to another endpoint */ |
4281 | 0 | nNumPossibleMobile += ( nMobile > 0 || at[endpoint].endpoint ); |
4282 | 0 | nNumEndPoints++; |
4283 | 0 | } |
4284 | 0 | if (nNumEndPoints > 1 && nNumPossibleMobile && |
4285 | 0 | nNumDonor && nNumAcceptor && |
4286 | 0 | num_O == 1 && num_C) |
4287 | 0 | { |
4288 | | /* |
4289 | | * A tautomeric group has been found |
4290 | | * |
4291 | | * at this point: |
4292 | | * nGroupNumber = 0 if all endpoints belong to a newly discovered tautomeric group |
4293 | | * bDiffGroups > 0 if at least 2 tautomeric groups are to be merged (one of them can be new) |
4294 | | * case (nGroupNumber != 0 && bDiffGroups = 0 ) ignored because all endpoints belong to the same known t-group |
4295 | | * case (nGroupNumber != 0 && bDiffGroups < 0 ) cannot happen |
4296 | | */ |
4297 | |
|
4298 | 0 | nErr = FindAccessibleEndPoints( pCG, EndPoint, &nNumEndPoints, BondPos, &nNumBondPos, |
4299 | 0 | pBNS, pBD, at, num_atoms, c_group_info, ALT_PATH_MODE_TAUTOM_KET ); |
4300 | |
|
4301 | 0 | if (IS_BNS_ERROR( nErr )) |
4302 | 0 | { |
4303 | 0 | return nErr; |
4304 | 0 | } |
4305 | 0 | nErr = 0; |
4306 | |
|
4307 | 0 | if (nNumEndPoints > 0) |
4308 | 0 | { |
4309 | 0 | if (!nGroupNumber || bDiffGroups > 0) |
4310 | 0 | { |
4311 | |
|
4312 | 0 | num_changes = RegisterEndPoints( pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS ); |
4313 | 0 | if (num_changes == -1) |
4314 | 0 | { |
4315 | 0 | nErr = CT_TAUCOUNT_ERR; |
4316 | 0 | } |
4317 | 0 | if (num_changes < 0) |
4318 | 0 | { |
4319 | 0 | nErr = num_changes; |
4320 | 0 | } |
4321 | 0 | if (nErr) |
4322 | 0 | { |
4323 | 0 | goto exit_function; |
4324 | 0 | } |
4325 | 0 | tot_changes += ( num_changes > 0 ); |
4326 | 0 | } |
4327 | 0 | if (nNumBondPos > 0) |
4328 | 0 | { |
4329 | | /* Some of the bonds have not been marked as tautomeric yet */ |
4330 | 0 | num_changes = SetTautomericBonds( at, nNumBondPos, BondPos ); |
4331 | 0 | tot_changes += ( num_changes > 0 ); |
4332 | 0 | } |
4333 | 0 | } |
4334 | 0 | } |
4335 | 0 | } |
4336 | 0 | } |
4337 | 0 | } |
4338 | 0 | } |
4339 | 0 | } |
4340 | 110k | #endif /* KETO_ENOL_TAUT */ |
4341 | | |
4342 | 110k | #if ( TAUT_OTHER == 1 ) /* { */ |
4343 | 110k | if (!tot_changes) |
4344 | 110k | { |
4345 | 110k | #define MAX_ALT_PATH_LEN 8 |
4346 | 110k | int nMaxLenDfsPath = MAX_ALT_PATH_LEN; |
4347 | 110k | int i1, i2; |
4348 | 110k | AT_RANK *nDfsPathPos = (AT_RANK *) inchi_calloc( num_atoms, sizeof( nDfsPathPos[0] ) ); |
4349 | 110k | DFS_PATH DfsPath[MAX_ALT_PATH_LEN]; |
4350 | 110k | int ret; |
4351 | | |
4352 | 110k | if (!nDfsPathPos || !DfsPath) |
4353 | 0 | { |
4354 | 0 | tot_changes = CT_OUT_OF_RAM; /* <BRKPT> */ |
4355 | 0 | goto free_memory; |
4356 | 0 | } |
4357 | | |
4358 | 110k | #if ( TAUT_15_NON_RING == 1 ) /***** post v.1 feature *****/ |
4359 | 110k | if (t_group_info->bTautFlags & TG_FLAG_1_5_TAUT) |
4360 | 0 | { |
4361 | | /* 1,5 tautomerism; one of the endpoints should no be on a ring */ |
4362 | | /* |
4363 | | O OH O |
4364 | | || | || |
4365 | | A--pos- A--pos- A--pos- |
4366 | | / sib- // sib- ? / sib- |
4367 | | C ly C ly CH ly |
4368 | | \\ a <--> \ a <--> \ a |
4369 | | B--ring B--ring B--ring |
4370 | | | || || |
4371 | | NH N N |
4372 | | |
4373 | | Note: few recent modifications now allow the terminal N be in a ring, too |
4374 | | */ |
4375 | 0 | for (i1 = 0; i1 < num_atoms; i1++) |
4376 | 0 | { |
4377 | | /* Find possible endpoint Z = at[i1] */ |
4378 | 0 | if (!( endpoint_valence = nGetEndpointInfo( at, i1, &eif1 ) ) /*|| |
4379 | 0 | at[i1].nNumAtInRingSystem > 1*/) |
4380 | 0 | { |
4381 | 0 | continue; /* not a possibly endpoint */ |
4382 | 0 | } |
4383 | | |
4384 | 0 | if (1) |
4385 | 0 | { |
4386 | 0 | nNumEndPoints = 0; |
4387 | 0 | nNumBondPos = 0; |
4388 | |
|
4389 | 0 | ret = nGet15TautInAltPath( pCG, at, i1, nDfsPathPos, |
4390 | 0 | DfsPath, nMaxLenDfsPath, |
4391 | 0 | EndPoint, sizeof( EndPoint ) / sizeof( EndPoint[0] ), |
4392 | 0 | BondPos, sizeof( BondPos ) / sizeof( BondPos[0] ), |
4393 | 0 | &nNumEndPoints, &nNumBondPos, |
4394 | 0 | pBNS, pBD, num_atoms ); |
4395 | |
|
4396 | 0 | if (ret > 0) |
4397 | 0 | { |
4398 | 0 | if (nNumEndPoints) |
4399 | 0 | { |
4400 | 0 | num_changes = RegisterEndPoints( pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS ); |
4401 | 0 | if (num_changes == -1) |
4402 | 0 | { |
4403 | 0 | nErr = CT_TAUCOUNT_ERR; |
4404 | 0 | } |
4405 | 0 | if (num_changes < 0) |
4406 | 0 | { |
4407 | 0 | nErr = num_changes; |
4408 | 0 | } |
4409 | 0 | if (nErr) |
4410 | 0 | { |
4411 | 0 | goto free_memory; |
4412 | 0 | } |
4413 | 0 | tot_changes += ( num_changes > 0 ); |
4414 | 0 | } |
4415 | 0 | if (nNumBondPos) |
4416 | 0 | { |
4417 | 0 | tot_changes += ( 0 < SetTautomericBonds( at, nNumBondPos, BondPos ) ); |
4418 | 0 | } |
4419 | 0 | } |
4420 | 0 | else |
4421 | 0 | { |
4422 | 0 | if (IS_BNS_ERROR( ret )) |
4423 | 0 | { |
4424 | 0 | nErr = ret; |
4425 | 0 | goto free_memory; |
4426 | 0 | } |
4427 | 0 | } |
4428 | 0 | } |
4429 | 0 | } |
4430 | 0 | } |
4431 | 110k | #endif |
4432 | | |
4433 | 110k | #if ( TAUT_4PYRIDINOL_RINGS == 1 ) |
4434 | | /* 6-member rings */ |
4435 | | /* |
4436 | | O OH OH |
4437 | | || | | |
4438 | | / \ // \ / \\ |
4439 | | || || <--> | || <--> || | |
4440 | | \ / \\ / \ // |
4441 | | NH N N |
4442 | | */ |
4443 | 222k | for (i1 = 0; i1 < num_atoms; i1++) |
4444 | 112k | { |
4445 | | /* Find possible endpoint Z = at[i1] */ |
4446 | 112k | if (3 != ( endpoint_valence = nGetEndpointInfo( at, i1, &eif1 ) ) || |
4447 | 112k | 2 != at[i1].valence) |
4448 | 112k | { |
4449 | 112k | continue; /* not a nitrogen atom or a wrong valence */ |
4450 | 112k | } |
4451 | | |
4452 | 2 | if (at[i1].nNumAtInRingSystem >= 6) |
4453 | 0 | { |
4454 | 0 | nNumEndPoints = 0; |
4455 | 0 | nNumBondPos = 0; |
4456 | |
|
4457 | 0 | ret = nGet15TautIn6MembAltRing( pCG, at, i1, nDfsPathPos, |
4458 | 0 | DfsPath, nMaxLenDfsPath, EndPoint, |
4459 | 0 | sizeof( EndPoint ) / sizeof( EndPoint[0] ), |
4460 | 0 | BondPos, |
4461 | 0 | sizeof( BondPos ) / sizeof( BondPos[0] ), |
4462 | 0 | &nNumEndPoints, &nNumBondPos, |
4463 | 0 | pBNS, pBD, num_atoms ); |
4464 | 0 | if (ret > 0) |
4465 | 0 | { |
4466 | 0 | if (nNumEndPoints) |
4467 | 0 | { |
4468 | 0 | num_changes = RegisterEndPoints( pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS ); |
4469 | 0 | if (num_changes == -1) |
4470 | 0 | { |
4471 | 0 | nErr = CT_TAUCOUNT_ERR; |
4472 | 0 | } |
4473 | 0 | if (num_changes < 0) |
4474 | 0 | { |
4475 | 0 | nErr = num_changes; |
4476 | 0 | } |
4477 | 0 | if (nErr) |
4478 | 0 | { |
4479 | 0 | goto free_memory; |
4480 | 0 | } |
4481 | 0 | tot_changes += ( num_changes > 0 ); |
4482 | 0 | } |
4483 | 0 | if (nNumBondPos) |
4484 | 0 | { |
4485 | 0 | tot_changes += ( 0 < SetTautomericBonds( at, nNumBondPos, BondPos ) ); |
4486 | 0 | } |
4487 | 0 | } |
4488 | 0 | else |
4489 | 0 | { |
4490 | 0 | if (IS_BNS_ERROR( ret )) |
4491 | 0 | { |
4492 | 0 | nErr = ret; |
4493 | 0 | goto free_memory; |
4494 | 0 | } |
4495 | 0 | } |
4496 | 0 | } |
4497 | 2 | } |
4498 | 110k | #endif /* TAUT_4PYRIDINOL_RINGS */ |
4499 | | |
4500 | 110k | #if ( TAUT_PYRAZOLE_RINGS == 1 ) |
4501 | | /* 5-member rings: |
4502 | | |
4503 | | Z Z |
4504 | | / \\ // \ |
4505 | | X Y <--> X Y |
4506 | | \\ / \ // |
4507 | | N--NH HN--N |
4508 | | |
4509 | | ^ ^ |
4510 | | search for these NH |
4511 | | */ |
4512 | | /* 5-member rings (pyrazole derivatives): look for the neighboring N */ |
4513 | 222k | for (i1 = 0; i1 < num_atoms; i1++) |
4514 | 112k | { |
4515 | 112k | if (2 == at[i1].valence && |
4516 | 112k | at[i1].nNumAtInRingSystem >= 5 && |
4517 | 112k | 3 == ( endpoint_valence = nGetEndpointInfo( at, i1, &eif1 ) ) |
4518 | 112k | ) |
4519 | 0 | { |
4520 | 0 | nMobile = at[i1].num_H + ( at[i1].charge == -1 ); |
4521 | 0 | for (j = 0; j < at[i1].valence; j++) |
4522 | 0 | { |
4523 | 0 | int nMobile2, endpoint_valence2; |
4524 | 0 | i2 = at[i1].neighbor[j]; |
4525 | | |
4526 | | /* may be important */ |
4527 | 0 | if (i2 >= i1) |
4528 | 0 | { |
4529 | 0 | continue; /* do not try same pair 2 times */ |
4530 | 0 | } |
4531 | | |
4532 | 0 | if (at[i2].nRingSystem != at[i1].nRingSystem) |
4533 | 0 | { |
4534 | 0 | continue; |
4535 | 0 | } |
4536 | | |
4537 | 0 | bond_type = ( at[i1].bond_type[j] & ~BOND_MARK_ALL ); |
4538 | 0 | if (bond_type != BOND_SINGLE && |
4539 | 0 | bond_type != BOND_TAUTOM && |
4540 | 0 | bond_type != BOND_ALT12NS && |
4541 | 0 | bond_type != BOND_ALTERN || /* added 1-15-2002 */ |
4542 | 0 | 2 != at[i2].valence || |
4543 | 0 | 3 != ( endpoint_valence2 = nGetEndpointInfo( at, i2, &eif2 ) )) |
4544 | 0 | { |
4545 | 0 | continue; /* not a nitrogen atom or a wrong valence or not a single bond */ |
4546 | 0 | } |
4547 | 0 | nMobile2 = at[i2].num_H + ( at[i2].charge == -1 ); /* number of mobile groups */ |
4548 | | #if ( TAUT_IGNORE_EQL_ENDPOINTS == 1 ) |
4549 | | if (at[i1].endpoint && at[i1].endpoint == at[i2].endpoint) |
4550 | | { |
4551 | | continue; /* atoms already belong to the same t-group */ |
4552 | | } |
4553 | | #endif |
4554 | 0 | if (!at[i1].endpoint && !at[i2].endpoint && 1 != nMobile + nMobile2) |
4555 | 0 | { |
4556 | 0 | continue; |
4557 | 0 | } |
4558 | | |
4559 | 0 | ret = nGet12TautIn5MembAltRing( pCG, at, i1, j, nDfsPathPos, |
4560 | 0 | DfsPath, nMaxLenDfsPath, |
4561 | 0 | EndPoint, |
4562 | 0 | sizeof( EndPoint ) / sizeof( EndPoint[0] ), |
4563 | 0 | BondPos, |
4564 | 0 | sizeof( BondPos ) / sizeof( BondPos[0] ), |
4565 | 0 | &nNumEndPoints, &nNumBondPos, |
4566 | 0 | pBNS, pBD, num_atoms ); |
4567 | 0 | if (ret > 0) |
4568 | 0 | { |
4569 | 0 | if (nNumEndPoints) |
4570 | 0 | { |
4571 | 0 | num_changes = RegisterEndPoints( pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS ); |
4572 | 0 | if (num_changes == -1) |
4573 | 0 | { |
4574 | 0 | nErr = CT_TAUCOUNT_ERR; |
4575 | 0 | } |
4576 | 0 | if (num_changes < 0) |
4577 | 0 | { |
4578 | 0 | nErr = num_changes; |
4579 | 0 | } |
4580 | 0 | if (nErr) |
4581 | 0 | goto free_memory; |
4582 | 0 | tot_changes += ( num_changes > 0 ); |
4583 | 0 | } |
4584 | 0 | if (nNumBondPos) |
4585 | 0 | { |
4586 | 0 | tot_changes += ( 0 < SetTautomericBonds( at, nNumBondPos, BondPos ) ); |
4587 | 0 | } |
4588 | 0 | } |
4589 | 0 | else |
4590 | 0 | { |
4591 | 0 | if (IS_BNS_ERROR( ret )) |
4592 | 0 | { |
4593 | 0 | nErr = ret; |
4594 | 0 | goto free_memory; |
4595 | 0 | } |
4596 | 0 | } |
4597 | 0 | } |
4598 | 0 | } |
4599 | 112k | } |
4600 | 110k | #endif /* TAUT_PYRAZOLE_RINGS */ |
4601 | | |
4602 | 110k | #if ( TAUT_TROPOLONE_7 == 1 || TAUT_TROPOLONE_5 == 1 ) /* { */ |
4603 | | /******************************************************** |
4604 | | * A B |
4605 | | * | || |
4606 | | * 7-member rings (tropolones): look for M=Q--R--ZH, |
4607 | | * ^ ^ ^ ^ |
4608 | | * endpoint1 i1 i2 endpoint2 |
4609 | | * where A-Q-R=B belong to a 7-member alt. (except Q-R bond) ring: ..=A-(Q-R)=B-.. |
4610 | | * Bond Q-R should be single or tautomeric or alternating |
4611 | | * M=Q and R-ZH should be chain (non-ring) bonds |
4612 | | * Same for 5-member rings |
4613 | | */ |
4614 | 222k | for (i1 = 0; i1 < num_atoms; i1++) |
4615 | 112k | { |
4616 | 112k | if (at[i1].nNumAtInRingSystem >= |
4617 | 112k | #if ( TAUT_TROPOLONE_5 == 1 ) |
4618 | 112k | 5 |
4619 | | #else |
4620 | | 7 |
4621 | | #endif |
4622 | 112k | && |
4623 | 112k | bIsCenterPointStrict( at, i1 ) && |
4624 | 112k | #if ( TAUT_RINGS_ATTACH_CHAIN == 1 ) |
4625 | 112k | at[i1].bCutVertex && |
4626 | 112k | #endif |
4627 | 112k | at[i1].valence == 3 && !at[i1].endpoint) |
4628 | 0 | { |
4629 | 0 | int nMobile1, endpoint1, endpoint1_valence, bond_type1; |
4630 | 0 | int nMobile2, endpoint2, endpoint2_valence, bond_type2; |
4631 | 0 | for (j = 0; j < at[i1].valence; j++) |
4632 | 0 | { |
4633 | 0 | i2 = at[i1].neighbor[j]; |
4634 | | /* may be important |
4635 | | if ( i2 > i1 ) |
4636 | | continue; |
4637 | | do not try same pair 2 times |
4638 | | */ |
4639 | 0 | if (at[i2].nRingSystem != at[i1].nRingSystem || |
4640 | 0 | !bIsCenterPointStrict( at, i2 ) || |
4641 | 0 | #if ( TAUT_RINGS_ATTACH_CHAIN == 1 ) |
4642 | 0 | !at[i2].bCutVertex || |
4643 | 0 | #endif |
4644 | 0 | at[i2].valence != 3 || at[i2].endpoint) |
4645 | 0 | { |
4646 | 0 | continue; |
4647 | 0 | } |
4648 | 0 | bond_type = ( at[i1].bond_type[j] & ~BOND_MARK_ALL ); |
4649 | 0 | if (bond_type != BOND_SINGLE && |
4650 | 0 | bond_type != BOND_TAUTOM && |
4651 | 0 | bond_type != BOND_ALT12NS && |
4652 | 0 | bond_type != BOND_ALTERN) |
4653 | 0 | { |
4654 | 0 | continue; /* not a single bond between Q-R */ |
4655 | 0 | } |
4656 | | |
4657 | | /* Find endpoints */ |
4658 | 0 | for (k = 0; k < at[i1].valence; k++) |
4659 | 0 | { |
4660 | 0 | endpoint1 = at[i1].neighbor[k]; |
4661 | 0 | if (endpoint1 == i2) |
4662 | 0 | { |
4663 | 0 | continue; /* j == k */ |
4664 | 0 | } |
4665 | 0 | if (!( endpoint1_valence = nGetEndpointInfo( at, endpoint1, &eif1 ) )) |
4666 | 0 | { |
4667 | 0 | continue; /* not an endpoint1 element or can't have mobile groups */ |
4668 | 0 | } |
4669 | 0 | #if ( TAUT_RINGS_ATTACH_CHAIN == 1 ) |
4670 | 0 | if (at[endpoint1].nRingSystem == at[i1].nRingSystem) |
4671 | 0 | { |
4672 | 0 | continue; |
4673 | 0 | } |
4674 | 0 | #endif |
4675 | 0 | nMobile1 = at[endpoint1].num_H + ( at[endpoint1].charge == -1 ); /* number of mobile groups */ |
4676 | 0 | if (nMobile1 + at[endpoint1].chem_bonds_valence != endpoint1_valence) |
4677 | 0 | continue; /* abnormal endpoint1 valence; ignore. */ |
4678 | 0 | bond_type1 = ( at[i1].bond_type[k] & ~BOND_MARK_ALL ); |
4679 | |
|
4680 | 0 | if (bond_type1 != BOND_SINGLE && |
4681 | 0 | bond_type1 != BOND_DOUBLE && |
4682 | 0 | bond_type1 != BOND_TAUTOM && |
4683 | 0 | bond_type1 != BOND_ALT12NS && |
4684 | 0 | bond_type1 != BOND_ALTERN) |
4685 | 0 | { |
4686 | 0 | continue; |
4687 | 0 | } |
4688 | 0 | for (m = 0; m < at[i2].valence; m++) |
4689 | 0 | { |
4690 | 0 | endpoint2 = at[i2].neighbor[m]; |
4691 | 0 | if (endpoint2 == i1) |
4692 | 0 | { |
4693 | 0 | continue; |
4694 | 0 | } |
4695 | 0 | if (!( endpoint2_valence = nGetEndpointInfo( at, endpoint2, &eif2 ) )) |
4696 | 0 | { |
4697 | 0 | continue; /* not an endpoint2 element or can't have mobile groups */ |
4698 | 0 | } |
4699 | 0 | #if ( TAUT_RINGS_ATTACH_CHAIN == 1 ) |
4700 | 0 | if (at[endpoint2].nRingSystem == at[i2].nRingSystem) |
4701 | 0 | { |
4702 | 0 | continue; |
4703 | 0 | } |
4704 | 0 | #endif |
4705 | 0 | nMobile2 = at[endpoint2].num_H + ( at[endpoint2].charge == -1 ); /* number of mobile groups */ |
4706 | 0 | bond_type2 = ( at[i2].bond_type[m] & ~BOND_MARK_ALL ); |
4707 | |
|
4708 | 0 | if (bond_type2 != BOND_SINGLE && |
4709 | 0 | bond_type2 != BOND_DOUBLE && |
4710 | 0 | bond_type2 != BOND_TAUTOM && |
4711 | 0 | bond_type2 != BOND_ALT12NS && |
4712 | 0 | bond_type2 != BOND_ALTERN) |
4713 | 0 | { |
4714 | 0 | continue; |
4715 | 0 | } |
4716 | | |
4717 | | /* Final test for possible tautomerism */ |
4718 | 0 | nMobile = 0; |
4719 | |
|
4720 | 0 | if (ALLOWED_EDGE( pBNS, i1, k ) && ALLOWED_EDGE( pBNS, i2, m )) |
4721 | 0 | { |
4722 | | /* Can mobile group move from 1 to 2? */ |
4723 | 0 | nMobile += ( at[endpoint1].endpoint || nMobile1 ) && /* from endpoint1 */ |
4724 | 0 | ( bond_type1 != BOND_DOUBLE ) && |
4725 | |
|
4726 | 0 | ( at[endpoint2].endpoint || /* to endpoint2 */ |
4727 | 0 | eif2.cNeutralBondsValence > at[endpoint2].valence ) && |
4728 | 0 | ( bond_type2 != BOND_SINGLE ); |
4729 | | |
4730 | | /* Can mobile group move from 2 to 1? */ |
4731 | 0 | nMobile += ( at[endpoint2].endpoint || nMobile2 ) && /* from endpoint2 */ |
4732 | 0 | ( bond_type2 != BOND_DOUBLE ) && /*changed from BOND_SINGLE 2004-02-26 */ |
4733 | |
|
4734 | 0 | ( at[endpoint1].endpoint || /* to endpoint1 */ |
4735 | 0 | eif1.cNeutralBondsValence > at[endpoint1].valence ) && |
4736 | 0 | ( bond_type1 != BOND_SINGLE ); |
4737 | 0 | } |
4738 | 0 | if (!nMobile) |
4739 | 0 | { |
4740 | 0 | continue; |
4741 | 0 | } |
4742 | | |
4743 | 0 | if (bond_type1 == bond_type2 && |
4744 | 0 | ( bond_type1 == BOND_SINGLE || bond_type1 == BOND_DOUBLE )) |
4745 | 0 | { |
4746 | 0 | continue; |
4747 | 0 | } |
4748 | | |
4749 | | /* -- old -- |
4750 | | if ( !at[endpoint1].endpoint && !at[endpoint2].endpoint && 1 != nMobile1 + nMobile2 ) |
4751 | | continue; |
4752 | | */ |
4753 | | /* -- new -- |
4754 | | |
4755 | | if ( !at[endpoint1].endpoint && !at[endpoint2].endpoint ) { |
4756 | | if ( !(bond_type1 == BOND_SINGLE || bond_type1 == BOND_DOUBLE) || |
4757 | | !(bond_type2 == BOND_SINGLE || bond_type2 == BOND_DOUBLE) ) { |
4758 | | // at this point bond_type1 != bond_type2 |
4759 | | continue; |
4760 | | } |
4761 | | if ( bond_type1 == BOND_SINGLE && !nMobile1 || |
4762 | | bond_type2 == BOND_SINGLE && !nMobile2 || |
4763 | | 0 == nMobile1 + nMobile2 ) { |
4764 | | continue; |
4765 | | } |
4766 | | } |
4767 | | */ |
4768 | | |
4769 | 0 | #if ( TAUT_TROPOLONE_7 == 1 ) |
4770 | 0 | if (at[i1].nNumAtInRingSystem >= 7) |
4771 | 0 | { |
4772 | |
|
4773 | 0 | ret = nGet14TautIn7MembAltRing( pCG, at, i1, |
4774 | 0 | j, k, m, nDfsPathPos, |
4775 | 0 | DfsPath, nMaxLenDfsPath, |
4776 | 0 | EndPoint, |
4777 | 0 | sizeof( EndPoint ) / sizeof( EndPoint[0] ), |
4778 | 0 | BondPos, |
4779 | 0 | sizeof( BondPos ) / sizeof( BondPos[0] ), |
4780 | 0 | &nNumEndPoints, &nNumBondPos, |
4781 | 0 | pBNS, pBD, num_atoms ); |
4782 | 0 | if (ret > 0) |
4783 | 0 | { |
4784 | 0 | if (nNumEndPoints) |
4785 | 0 | { |
4786 | 0 | num_changes = RegisterEndPoints( pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS ); |
4787 | 0 | if (num_changes == -1) |
4788 | 0 | { |
4789 | 0 | nErr = CT_TAUCOUNT_ERR; |
4790 | 0 | } |
4791 | 0 | if (num_changes < 0) |
4792 | 0 | { |
4793 | 0 | nErr = num_changes; |
4794 | 0 | } |
4795 | 0 | if (nErr) |
4796 | 0 | { |
4797 | 0 | goto free_memory; |
4798 | 0 | } |
4799 | 0 | tot_changes += ( num_changes > 0 ); |
4800 | 0 | } |
4801 | 0 | if (nNumBondPos) |
4802 | 0 | { |
4803 | 0 | tot_changes += ( 0 < SetTautomericBonds( at, nNumBondPos, BondPos ) ); |
4804 | 0 | } |
4805 | 0 | } |
4806 | 0 | else |
4807 | 0 | { |
4808 | 0 | if (IS_BNS_ERROR( ret )) |
4809 | 0 | { |
4810 | 0 | nErr = ret; |
4811 | 0 | goto free_memory; |
4812 | 0 | } |
4813 | 0 | } |
4814 | 0 | } |
4815 | 0 | #endif |
4816 | | |
4817 | 0 | #if ( TAUT_TROPOLONE_5 == 1 ) |
4818 | 0 | if (at[i1].nNumAtInRingSystem >= 5) |
4819 | 0 | { |
4820 | |
|
4821 | 0 | ret = nGet14TautIn5MembAltRing( pCG, at, i1, j, k, m, nDfsPathPos, |
4822 | 0 | DfsPath, nMaxLenDfsPath, |
4823 | 0 | EndPoint, sizeof( EndPoint ) / sizeof( EndPoint[0] ), |
4824 | 0 | BondPos, sizeof( BondPos ) / sizeof( BondPos[0] ), |
4825 | 0 | &nNumEndPoints, &nNumBondPos, |
4826 | 0 | pBNS, pBD, num_atoms ); |
4827 | 0 | if (ret > 0) |
4828 | 0 | { |
4829 | 0 | if (nNumEndPoints) |
4830 | 0 | { |
4831 | 0 | num_changes = RegisterEndPoints( pCG, t_group_info, EndPoint, nNumEndPoints, at, num_atoms, c_group_info, pBNS ); |
4832 | 0 | if (num_changes == -1) |
4833 | 0 | { |
4834 | 0 | nErr = CT_TAUCOUNT_ERR; |
4835 | 0 | } |
4836 | 0 | if (num_changes < 0) |
4837 | 0 | { |
4838 | 0 | nErr = num_changes; |
4839 | 0 | } |
4840 | 0 | if (nErr) |
4841 | 0 | { |
4842 | 0 | goto free_memory; |
4843 | 0 | } |
4844 | 0 | tot_changes += ( num_changes > 0 ); |
4845 | 0 | } |
4846 | 0 | if (nNumBondPos) |
4847 | 0 | { |
4848 | 0 | tot_changes += ( 0 < SetTautomericBonds( at, nNumBondPos, BondPos ) ); |
4849 | 0 | } |
4850 | 0 | } |
4851 | 0 | else |
4852 | 0 | { |
4853 | 0 | if (IS_BNS_ERROR( ret )) |
4854 | 0 | { |
4855 | 0 | nErr = ret; |
4856 | 0 | goto free_memory; |
4857 | 0 | } |
4858 | 0 | } |
4859 | 0 | } |
4860 | 0 | #endif |
4861 | 0 | } |
4862 | 0 | } |
4863 | 0 | } |
4864 | 0 | } |
4865 | 112k | } |
4866 | 110k | #endif /* } TAUT_TROPOLONE */ |
4867 | | |
4868 | 110k | free_memory: |
4869 | 110k | if (nDfsPathPos) |
4870 | 110k | { |
4871 | 110k | inchi_free( nDfsPathPos ); |
4872 | 110k | } |
4873 | 110k | #undef MAX_ALT_PATH_LEN |
4874 | 110k | } |
4875 | 110k | #endif /* } FIND_RING_SYSTEMS */ |
4876 | | |
4877 | 110k | exit_function: |
4878 | | |
4879 | 110k | return nErr < 0 ? nErr : tot_changes; |
4880 | 110k | } |
4881 | | |
4882 | | |
4883 | | /****************************************************************************/ |
4884 | | int free_t_group_info( T_GROUP_INFO *t_group_info ) |
4885 | 384k | { |
4886 | 384k | if (t_group_info) |
4887 | 384k | { |
4888 | 384k | if (t_group_info->t_group) |
4889 | 110k | { |
4890 | 110k | inchi_free( t_group_info->t_group ); |
4891 | 110k | } |
4892 | 384k | if (t_group_info->nEndpointAtomNumber) |
4893 | 47 | { |
4894 | 47 | inchi_free( t_group_info->nEndpointAtomNumber ); |
4895 | 47 | } |
4896 | 384k | if (t_group_info->tGroupNumber) |
4897 | 47 | { |
4898 | 47 | inchi_free( t_group_info->tGroupNumber ); |
4899 | 47 | } |
4900 | 384k | if (t_group_info->nIsotopicEndpointAtomNumber) |
4901 | 4 | { |
4902 | 4 | inchi_free( t_group_info->nIsotopicEndpointAtomNumber ); |
4903 | 4 | } |
4904 | 384k | memset( t_group_info, 0, sizeof( *t_group_info ) ); |
4905 | 384k | } |
4906 | | |
4907 | 384k | return 0; |
4908 | 384k | } |
4909 | | |
4910 | | |
4911 | | /****************************************************************************/ |
4912 | | int make_a_copy_of_t_group_info( T_GROUP_INFO *t_group_info, |
4913 | | T_GROUP_INFO *t_group_info_orig ) |
4914 | 5 | { |
4915 | 5 | int err = 0, len; |
4916 | 5 | free_t_group_info( t_group_info ); |
4917 | 5 | if (t_group_info_orig && t_group_info) |
4918 | 5 | { |
4919 | 5 | if (( len = t_group_info_orig->max_num_t_groups ) > 0) |
4920 | 5 | { |
4921 | 5 | if (t_group_info->t_group = |
4922 | 5 | (T_GROUP*) inchi_malloc( len * sizeof( t_group_info->t_group[0] ) )) |
4923 | 5 | { |
4924 | 5 | memcpy( t_group_info->t_group, |
4925 | 5 | t_group_info_orig->t_group, |
4926 | 5 | len * sizeof( t_group_info->t_group[0] ) ); |
4927 | 5 | } |
4928 | 0 | else |
4929 | 0 | { |
4930 | 0 | err++; |
4931 | 0 | } |
4932 | 5 | } |
4933 | 5 | if (( len = t_group_info_orig->nNumEndpoints ) > 0) |
4934 | 4 | { |
4935 | 4 | if (t_group_info->nEndpointAtomNumber = |
4936 | 4 | (AT_NUMB*) inchi_malloc( len * sizeof( t_group_info->nEndpointAtomNumber[0] ) )) |
4937 | 4 | { |
4938 | 4 | memcpy( t_group_info->nEndpointAtomNumber, |
4939 | 4 | t_group_info_orig->nEndpointAtomNumber, |
4940 | 4 | len * sizeof( t_group_info->nEndpointAtomNumber[0] ) ); |
4941 | 4 | } |
4942 | 0 | else |
4943 | 0 | { |
4944 | 0 | err++; |
4945 | 0 | } |
4946 | 4 | } |
4947 | 5 | if (( len = t_group_info_orig->num_t_groups ) > 0) |
4948 | 4 | { |
4949 | 4 | if (t_group_info->tGroupNumber = |
4950 | 4 | (AT_NUMB*) inchi_malloc( len * TGSO_TOTAL_LEN * sizeof( t_group_info->tGroupNumber[0] ) )) |
4951 | 4 | { |
4952 | 4 | memcpy( t_group_info->tGroupNumber, |
4953 | 4 | t_group_info_orig->tGroupNumber, |
4954 | 4 | len * TGSO_TOTAL_LEN * sizeof( t_group_info->tGroupNumber[0] ) ); |
4955 | 4 | } |
4956 | 0 | else |
4957 | 0 | { |
4958 | 0 | err++; |
4959 | 0 | } |
4960 | 4 | } |
4961 | 5 | if (( len = t_group_info_orig->nNumIsotopicEndpoints ) > 0) |
4962 | 2 | { |
4963 | 2 | if (t_group_info->nIsotopicEndpointAtomNumber = |
4964 | 2 | (AT_NUMB*) inchi_malloc( len * sizeof( t_group_info->nIsotopicEndpointAtomNumber[0] ) )) |
4965 | 2 | { |
4966 | 2 | memcpy( t_group_info->nIsotopicEndpointAtomNumber, |
4967 | 2 | t_group_info_orig->nIsotopicEndpointAtomNumber, |
4968 | 2 | len * sizeof( t_group_info->nIsotopicEndpointAtomNumber[0] ) ); |
4969 | 2 | } |
4970 | 0 | else |
4971 | 0 | { |
4972 | 0 | err++; |
4973 | 0 | } |
4974 | 2 | } |
4975 | 5 | if (!err) |
4976 | 5 | { |
4977 | 5 | t_group_info->nNumEndpoints = t_group_info_orig->nNumEndpoints; |
4978 | 5 | t_group_info->num_t_groups = t_group_info_orig->num_t_groups; |
4979 | 5 | t_group_info->max_num_t_groups = t_group_info_orig->max_num_t_groups; |
4980 | 5 | t_group_info->bIgnoreIsotopic = t_group_info_orig->bIgnoreIsotopic; |
4981 | 5 | t_group_info->nNumIsotopicEndpoints = t_group_info_orig->nNumIsotopicEndpoints; |
4982 | 5 | t_group_info->tni = t_group_info_orig->tni; |
4983 | | /* |
4984 | | t_group_info->nNumRemovedExplicitH = t_group_info_orig->nNumRemovedExplicitH; |
4985 | | t_group_info->nNumRemovedProtons = t_group_info_orig->nNumRemovedProtons; |
4986 | | t_group_info->bNormalizationFlags = t_group_info_orig->bNormalizationFlags; |
4987 | | */ |
4988 | | /* |
4989 | | t_group_info->bHardAddedRemovedProtons = t_group_info_orig->bHardAddedRemovedProtons; |
4990 | | t_group_info->bSimpleAddedRemovedProtons = t_group_info_orig->bSimpleAddedRemovedProtons; |
4991 | | t_group_info->nNumCanceledCharges = t_group_info_orig->nNumCanceledCharges; |
4992 | | */ |
4993 | 5 | } |
4994 | 5 | t_group_info->bTautFlags = t_group_info_orig->bTautFlags; |
4995 | 5 | t_group_info->bTautFlagsDone = t_group_info_orig->bTautFlagsDone; |
4996 | 5 | } |
4997 | | |
4998 | 5 | return err; |
4999 | 5 | } |
5000 | | |
5001 | | |
5002 | | /**************************************************************************** |
5003 | | Set tautomer group isotopic sort keys |
5004 | | ****************************************************************************/ |
5005 | | int set_tautomer_iso_sort_keys( T_GROUP_INFO *t_group_info ) |
5006 | 5 | { |
5007 | 5 | T_GROUP *t_group; |
5008 | 5 | T_GROUP_ISOWT Mult = 1; |
5009 | 5 | int i, j, num_t_groups, num_iso_t_groups = 0; |
5010 | | |
5011 | 5 | if (!t_group_info || !( t_group = t_group_info->t_group ) || |
5012 | 5 | 0 >= ( num_t_groups = t_group_info->num_t_groups ) || t_group_info->nNumIsotopicEndpoints) |
5013 | 3 | { |
5014 | 3 | return 0; |
5015 | 3 | } |
5016 | | |
5017 | 4 | for (i = 0; i < num_t_groups; i++) |
5018 | 2 | { |
5019 | 2 | t_group[i].iWeight = 0; |
5020 | 2 | j = T_NUM_ISOTOPIC - 1; |
5021 | 2 | Mult = 1; |
5022 | 2 | do |
5023 | 6 | { |
5024 | 6 | t_group[i].iWeight += Mult * (T_GROUP_ISOWT) t_group[i].num[T_NUM_NO_ISOTOPIC + j]; |
5025 | 6 | } while (--j >= 0 && ( Mult *= T_GROUP_ISOWT_MULT )); |
5026 | | |
5027 | 2 | num_iso_t_groups += ( t_group[i].iWeight != 0 ); |
5028 | 2 | } |
5029 | | |
5030 | 2 | return num_iso_t_groups; |
5031 | 5 | } |
5032 | | |
5033 | | |
5034 | | /**************************************************************************** |
5035 | | Fill t_group_info with information necessary to fill out tautomer part |
5036 | | of the linear connection table record. |
5037 | | |
5038 | | Note: on input, t_group_info should contain information created by MarkTautomerGroups() |
5039 | | No previous t_group_info adjustment due to throwing out disconnected parts of |
5040 | | the chemical structure is needed. |
5041 | | |
5042 | | Note2: throws out t_groups containing negative charges only (IGNORE_TGROUP_WITHOUT_H==1) |
5043 | | (leave their tautomeric bonds unchanged) |
5044 | | Note3: removes negative charges from other tautomeric groups |
5045 | | and adjust counts of mobile atoms if permitted (REMOVE_TGROUP_CHARGE==1) |
5046 | | ****************************************************************************/ |
5047 | | int CountTautomerGroups( sp_ATOM *at, |
5048 | | int num_atoms, |
5049 | | T_GROUP_INFO *t_group_info ) |
5050 | 110k | { |
5051 | 110k | int i, j, ret = 0, nNumEndpoints, max_t_group, num_groups_noH; |
5052 | | |
5053 | 110k | AT_NUMB nGroupNumber, nNewGroupNumber, *nCurrEndpointAtNoPos = NULL; |
5054 | | |
5055 | 110k | T_GROUP *t_group; |
5056 | 110k | int num_t; |
5057 | | /* int bIgnoreIsotopic, max_num_t; */ |
5058 | 110k | AT_NUMB *nTautomerGroupNumber = NULL; |
5059 | 110k | AT_NUMB *nEndpointAtomNumber = NULL; |
5060 | 110k | AT_NUMB *tGroupNumber = NULL; |
5061 | | |
5062 | 110k | if (!t_group_info || !t_group_info->t_group || 0 >= t_group_info->max_num_t_groups) |
5063 | 0 | { |
5064 | 0 | return 0; /* empty t-groups */ |
5065 | 0 | } |
5066 | 110k | num_t = t_group_info->num_t_groups; |
5067 | 110k | t_group = t_group_info->t_group; |
5068 | | /* |
5069 | | max_num_t = t_group_info->max_num_t_groups; |
5070 | | bIgnoreIsotopic = t_group_info->bIgnoreIsotopic; |
5071 | | */ |
5072 | 110k | num_groups_noH = 0; |
5073 | | |
5074 | | /* the following 2 arrays are to be rebuilt here */ |
5075 | 110k | if (t_group_info->nEndpointAtomNumber) |
5076 | 0 | { |
5077 | 0 | inchi_free( t_group_info->nEndpointAtomNumber ); |
5078 | 0 | t_group_info->nEndpointAtomNumber = NULL; |
5079 | 0 | } |
5080 | 110k | if (t_group_info->tGroupNumber) |
5081 | 0 | { |
5082 | 0 | inchi_free( t_group_info->tGroupNumber ); |
5083 | 0 | t_group_info->tGroupNumber = NULL; |
5084 | 0 | } |
5085 | | |
5086 | | /* Find max_t_group */ |
5087 | 110k | for (i = 0, max_t_group = 0; i < t_group_info->num_t_groups; i++) |
5088 | 4 | { |
5089 | 4 | if (max_t_group < t_group[i].nGroupNumber) |
5090 | 4 | max_t_group = t_group[i].nGroupNumber; |
5091 | 4 | } |
5092 | | /* Allocate memory for temp storage of numbers of endpoints */ |
5093 | 110k | if (max_t_group && |
5094 | 110k | !( nTautomerGroupNumber = (AT_NUMB*) inchi_calloc( max_t_group + 1, sizeof( nTautomerGroupNumber[0] ) ) /*temp*/ )) |
5095 | 0 | { |
5096 | 0 | goto err_exit_function; /* program error: out of RAM */ /* <BRKPT> */ |
5097 | 0 | } |
5098 | | |
5099 | | /* Count endpoints for each tautomer group */ |
5100 | 222k | for (i = 0, nNumEndpoints = 0; i < num_atoms; i++) |
5101 | 112k | { |
5102 | 112k | if (( j = at[i].endpoint ) == 0) |
5103 | 112k | { |
5104 | 112k | continue; |
5105 | 112k | } |
5106 | 8 | if (j > max_t_group) /* debug only */ |
5107 | 0 | { |
5108 | 0 | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5109 | 0 | } |
5110 | 8 | nTautomerGroupNumber[j] ++; |
5111 | 8 | nNumEndpoints++; |
5112 | 8 | } |
5113 | | |
5114 | 110k | if (!nNumEndpoints) |
5115 | 110k | { |
5116 | 110k | goto exit_function; /* not a tautomer */ |
5117 | 110k | } |
5118 | | |
5119 | | /* allocate temporary array */ |
5120 | 4 | if (!( nEndpointAtomNumber = (AT_NUMB*) inchi_calloc( nNumEndpoints, sizeof( nEndpointAtomNumber[0] ) ) ) || |
5121 | 4 | !( nCurrEndpointAtNoPos = (AT_NUMB*) inchi_calloc( num_t, sizeof( nCurrEndpointAtNoPos[0] ) ) /*temp*/ )) |
5122 | 0 | { |
5123 | 0 | goto err_exit_function; /* program error: out of RAM */ /* <BRKPT> */ |
5124 | 0 | } |
5125 | | |
5126 | | /* |
5127 | | * Remove missing endpoints from t_group. Since only one |
5128 | | * disconnected part is processed, some endpoints groups may have disappeared. |
5129 | | * Mark t_groups containing charges only for subsequent removal |
5130 | | */ |
5131 | 8 | for (i = 0, nNewGroupNumber = 0; i < num_t; /*i ++*/) |
5132 | 4 | { |
5133 | 4 | int bNoH = 0, nNumH; |
5134 | 4 | nGroupNumber = t_group[i].nGroupNumber; |
5135 | 8 | for (j = 1, nNumH = t_group[i].num[0]; j < T_NUM_NO_ISOTOPIC; j++) |
5136 | 4 | { |
5137 | 4 | nNumH -= (int) t_group[i].num[j]; |
5138 | 4 | } |
5139 | 4 | if (t_group[i].nNumEndpoints != nTautomerGroupNumber[(int) nGroupNumber] |
5140 | 4 | #if ( IGNORE_TGROUP_WITHOUT_H == 1 ) |
5141 | 4 | || ( bNoH = ( t_group[i].num[0] == t_group[i].num[1] ) ) /* only for (H,-) t-groups; (+) t-groups are not removed */ |
5142 | 4 | #endif |
5143 | 4 | ) |
5144 | 0 | { |
5145 | 0 | if (!nTautomerGroupNumber[(int) nGroupNumber] || bNoH) |
5146 | 0 | { |
5147 | | /* The group belongs to another disconnected part of the structure or has only charges */ |
5148 | | /* Remove the group */ |
5149 | 0 | num_t--; |
5150 | 0 | if (i < num_t) |
5151 | 0 | memmove( t_group + i, t_group + i + 1, ( num_t - i ) * sizeof( t_group[0] ) ); |
5152 | 0 | if (bNoH) |
5153 | 0 | { |
5154 | | /* Group contains no mobile hydrogen atoms, only charges. Prepare to remove it. */ |
5155 | 0 | nTautomerGroupNumber[(int) nGroupNumber] = 0; |
5156 | 0 | num_groups_noH++; |
5157 | 0 | } |
5158 | | /*i --;*/ |
5159 | 0 | } |
5160 | 0 | else |
5161 | 0 | { |
5162 | | /* Different number of endpoints */ |
5163 | 0 | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5164 | 0 | } |
5165 | 0 | } |
5166 | 4 | else |
5167 | 4 | { |
5168 | | /* Renumber t_group and prepare to renumber at[i].endpoint */ |
5169 | 4 | nTautomerGroupNumber[(int) nGroupNumber] = |
5170 | 4 | t_group[i].nGroupNumber = ++nNewGroupNumber; /* = i+1 */ |
5171 | | /* get first group atom orig. number position in the nEndpointAtomNumber[] */ |
5172 | | /* and in the tautomer endpoint canon numbers part of the connection table */ |
5173 | 4 | t_group[i].nFirstEndpointAtNoPos = nCurrEndpointAtNoPos[i] = |
5174 | 4 | i ? ( t_group[i - 1].nFirstEndpointAtNoPos + t_group[i - 1].nNumEndpoints ) : 0; |
5175 | 4 | t_group[i].num[0] = nNumH; |
5176 | | #if ( REMOVE_TGROUP_CHARGE == 1 ) |
5177 | | t_group[i].num[1] = 0; /* remove only (-) charges */ |
5178 | | #endif |
5179 | | /* -- wrong condition. Disabled. |
5180 | | if ( t_group[i].nGroupNumber != i + 1 ) { // for debug only |
5181 | | goto err_exit_function; // program error |
5182 | | } |
5183 | | */ |
5184 | 4 | i++; |
5185 | 4 | } |
5186 | 4 | } |
5187 | 4 | if (num_t != nNewGroupNumber) |
5188 | 0 | { |
5189 | | /* for debug only */ |
5190 | 0 | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5191 | 0 | } |
5192 | | |
5193 | | /* Check if any tautomer group was left */ |
5194 | 4 | if (!nNewGroupNumber) |
5195 | 0 | { |
5196 | 0 | if (!num_groups_noH) |
5197 | 0 | { |
5198 | 0 | goto err_exit_function; /* program error: not a tautomer */ /* <BRKPT> */ |
5199 | 0 | } |
5200 | 0 | else |
5201 | 0 | { |
5202 | 0 | goto exit_function; |
5203 | 0 | } |
5204 | 0 | } |
5205 | | |
5206 | | /* |
5207 | | * An array for tautomer group sorting later, at the time of storing Connection Table |
5208 | | * Later the sorting consists out of 2 steps: |
5209 | | * 1) Sort t_group[i].nNumEndpoints endpoint atom ranks within each endpoint group |
5210 | | * starting from t_group[i].nFirstEndpointAtNoPos; i = 0..t_group_info->num_t_groups-1 |
5211 | | * 2) Sort the groups indexes t_group_info->tGroupNumber[] |
5212 | | */ |
5213 | 4 | if (!( tGroupNumber = |
5214 | 4 | (AT_NUMB*) inchi_calloc( nNewGroupNumber*TGSO_TOTAL_LEN, sizeof( tGroupNumber[0] ) ) )) |
5215 | 0 | { |
5216 | 0 | goto err_exit_function; /* out of RAM */ |
5217 | 0 | } |
5218 | 8 | for (i = 0; i < nNewGroupNumber; i++) |
5219 | 4 | { |
5220 | 4 | tGroupNumber[i] = (AT_NUMB) i; /* initialization: original t_group number = (at[i]->endpoint-1) */ |
5221 | 4 | } |
5222 | | /* |
5223 | | * Renumber endpoint atoms and save their orig. atom |
5224 | | * numbers for filling out the tautomer part of the LinearCT. |
5225 | | * nCurrEndpointAtNoPos[j] is an index of the atom number in the nEndpointAtomNumber[] |
5226 | | */ |
5227 | 24 | for (i = 0; i < num_atoms; i++) |
5228 | 20 | { |
5229 | 20 | if (j = (int) at[i].endpoint) |
5230 | 8 | { |
5231 | 8 | j = (int) ( at[i].endpoint = nTautomerGroupNumber[j] ) - 1; /* new t_group number */ |
5232 | 8 | if (j >= 0) |
5233 | 8 | { |
5234 | | /* j=-1 in case of no mobile hydrogen atoms (charges only), group being removed */ |
5235 | 8 | if (nCurrEndpointAtNoPos[j] >= /* debug only */ |
5236 | 8 | t_group[j].nFirstEndpointAtNoPos + t_group[j].nNumEndpoints) |
5237 | 0 | { |
5238 | 0 | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5239 | 0 | } |
5240 | 8 | nEndpointAtomNumber[(int) nCurrEndpointAtNoPos[j] ++] = (AT_NUMB) i; |
5241 | 8 | } |
5242 | 0 | else |
5243 | 0 | { |
5244 | 0 | nNumEndpoints--; /* endpoint has been removed */ |
5245 | 0 | } |
5246 | 8 | } |
5247 | 20 | } |
5248 | 4 | t_group_info->num_t_groups = nNewGroupNumber; |
5249 | 4 | t_group_info->nNumEndpoints = nNumEndpoints; |
5250 | 4 | t_group_info->nEndpointAtomNumber = nEndpointAtomNumber; |
5251 | 4 | t_group_info->tGroupNumber = tGroupNumber; /* only the 1st segment filled */ |
5252 | 4 | inchi_free( nTautomerGroupNumber ); |
5253 | 4 | inchi_free( nCurrEndpointAtNoPos ); |
5254 | 4 | return nNumEndpoints + T_GROUP_HDR_LEN * nNewGroupNumber + 1; /* nLenLinearCTTautomer */ |
5255 | | |
5256 | 0 | err_exit_function: |
5257 | 0 | ret = CT_TAUCOUNT_ERR; |
5258 | |
|
5259 | 110k | exit_function: |
5260 | | |
5261 | | /* release allocated memory; set "no tautomeric group" */ |
5262 | 110k | if (nEndpointAtomNumber) |
5263 | 0 | { |
5264 | 0 | inchi_free( nEndpointAtomNumber ); |
5265 | 0 | } |
5266 | 110k | if (nTautomerGroupNumber) |
5267 | 0 | { |
5268 | 0 | inchi_free( nTautomerGroupNumber ); |
5269 | 0 | } |
5270 | 110k | if (tGroupNumber) |
5271 | 0 | { |
5272 | 0 | inchi_free( tGroupNumber ); |
5273 | 0 | } |
5274 | 110k | if (nCurrEndpointAtNoPos) |
5275 | 0 | { |
5276 | 0 | inchi_free( nCurrEndpointAtNoPos ); |
5277 | 0 | } |
5278 | 110k | t_group_info->nNumEndpoints = 0; |
5279 | 110k | t_group_info->num_t_groups = 0; |
5280 | 110k | if (!ret && ( ( t_group_info->tni.bNormalizationFlags & FLAG_NORM_CONSIDER_TAUT ) || |
5281 | 110k | t_group_info->nNumIsotopicEndpoints > 1 && ( t_group_info->bTautFlagsDone & ( TG_FLAG_FOUND_ISOTOPIC_H_DONE | TG_FLAG_FOUND_ISOTOPIC_ATOM_DONE ) ) )) |
5282 | 1 | { |
5283 | 1 | ret = 1; /* only protons have been (re)moved or neitralization happened */ |
5284 | 1 | } |
5285 | | |
5286 | 110k | return ret; |
5287 | 0 | } |
5288 | | |
5289 | | |
5290 | | #if ( READ_INCHI_STRING == 1 ) |
5291 | | #if ( INCLUDE_NORMALIZATION_ENTRY_POINT == 1 ) |
5292 | | |
5293 | | |
5294 | | /****************************************************************************/ |
5295 | | int CountTautomerGroupsInpAt( inp_ATOM *at, |
5296 | | int num_atoms, |
5297 | | T_GROUP_INFO *t_group_info ) |
5298 | | { |
5299 | | int i, j, ret = 0, nNumEndpoints, max_t_group, num_groups_noH; |
5300 | | |
5301 | | AT_NUMB nGroupNumber, nNewGroupNumber, *nCurrEndpointAtNoPos = NULL; |
5302 | | |
5303 | | T_GROUP *t_group; |
5304 | | int num_t; |
5305 | | /* int bIgnoreIsotopic, max_num_t; */ |
5306 | | AT_NUMB *nTautomerGroupNumber = NULL; |
5307 | | AT_NUMB *nEndpointAtomNumber = NULL; |
5308 | | AT_NUMB *tGroupNumber = NULL; |
5309 | | |
5310 | | if (!t_group_info || !t_group_info->t_group || 0 >= t_group_info->max_num_t_groups) |
5311 | | { |
5312 | | return 0; /* empty t-groups */ |
5313 | | } |
5314 | | num_t = t_group_info->num_t_groups; |
5315 | | t_group = t_group_info->t_group; |
5316 | | /* |
5317 | | max_num_t = t_group_info->max_num_t_groups; |
5318 | | bIgnoreIsotopic = t_group_info->bIgnoreIsotopic; |
5319 | | */ |
5320 | | num_groups_noH = 0; |
5321 | | |
5322 | | /* the following 2 arrays are to be rebuilt here */ |
5323 | | if (t_group_info->nEndpointAtomNumber) |
5324 | | { |
5325 | | inchi_free( t_group_info->nEndpointAtomNumber ); |
5326 | | t_group_info->nEndpointAtomNumber = NULL; |
5327 | | } |
5328 | | if (t_group_info->tGroupNumber) |
5329 | | { |
5330 | | inchi_free( t_group_info->tGroupNumber ); |
5331 | | t_group_info->tGroupNumber = NULL; |
5332 | | } |
5333 | | /* find max_t_group */ |
5334 | | for (i = 0, max_t_group = 0; i < t_group_info->num_t_groups; i++) |
5335 | | { |
5336 | | if (max_t_group < t_group[i].nGroupNumber) |
5337 | | max_t_group = t_group[i].nGroupNumber; |
5338 | | } |
5339 | | /* allocate memory for temp storage of numbers of endpoints */ |
5340 | | if (max_t_group && |
5341 | | !( nTautomerGroupNumber = (AT_NUMB*) inchi_calloc( max_t_group + 1, sizeof( nTautomerGroupNumber[0] ) ) /*temp*/ )) |
5342 | | { |
5343 | | goto err_exit_function; /* program error: out of RAM */ /* <BRKPT> */ |
5344 | | } |
5345 | | |
5346 | | /* count endpoints for each tautomer group */ |
5347 | | for (i = 0, nNumEndpoints = 0; i < num_atoms; i++) |
5348 | | { |
5349 | | if (( j = at[i].endpoint ) == 0) |
5350 | | continue; |
5351 | | if (j > max_t_group) /* debug only */ |
5352 | | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5353 | | nTautomerGroupNumber[j] ++; |
5354 | | nNumEndpoints++; |
5355 | | } |
5356 | | |
5357 | | if (!nNumEndpoints) |
5358 | | { |
5359 | | goto exit_function; /* not a tautomer */ |
5360 | | } |
5361 | | |
5362 | | /* allocate temporary array */ |
5363 | | if (!( nEndpointAtomNumber = (AT_NUMB*) inchi_calloc( nNumEndpoints, sizeof( nEndpointAtomNumber[0] ) ) ) || |
5364 | | !( nCurrEndpointAtNoPos = (AT_NUMB*) inchi_calloc( num_t, sizeof( nCurrEndpointAtNoPos[0] ) ) /*temp*/ )) |
5365 | | { |
5366 | | goto err_exit_function; /* program error: out of RAM */ /* <BRKPT> */ |
5367 | | } |
5368 | | /* |
5369 | | * Remove missing endpoints from t_group. Since only one |
5370 | | * disconnected part is processed, some endpoints groups may have disappeared. |
5371 | | * Mark t_groups containing charges only for subsequent removal |
5372 | | */ |
5373 | | for (i = 0, nNewGroupNumber = 0; i < num_t; /*i ++*/) |
5374 | | { |
5375 | | int bNoH = 0, nNumH; |
5376 | | nGroupNumber = t_group[i].nGroupNumber; |
5377 | | for (j = 1, nNumH = t_group[i].num[0]; j < T_NUM_NO_ISOTOPIC; j++) |
5378 | | { |
5379 | | nNumH -= (int) t_group[i].num[j]; |
5380 | | } |
5381 | | if (t_group[i].nNumEndpoints != nTautomerGroupNumber[(int) nGroupNumber] |
5382 | | #if ( IGNORE_TGROUP_WITHOUT_H == 1 ) |
5383 | | || ( bNoH = ( t_group[i].num[0] == t_group[i].num[1] ) ) /* only for (H,-) t-groups; (+) t-groups are not removed */ |
5384 | | #endif |
5385 | | ) |
5386 | | { |
5387 | | if (!nTautomerGroupNumber[(int) nGroupNumber] || bNoH) |
5388 | | { |
5389 | | /* the group belongs to another disconnected part of the structure or has only charges */ |
5390 | | /* Remove the group */ |
5391 | | num_t--; |
5392 | | if (i < num_t) |
5393 | | memmove( t_group + i, t_group + i + 1, ( num_t - i ) * sizeof( t_group[0] ) ); |
5394 | | if (bNoH) |
5395 | | { |
5396 | | /* group contains no mobile hydrogen atoms, only charges. Prepare to remove it. */ |
5397 | | nTautomerGroupNumber[(int) nGroupNumber] = 0; |
5398 | | num_groups_noH++; |
5399 | | } |
5400 | | /*i --;*/ |
5401 | | } |
5402 | | else |
5403 | | { |
5404 | | /* different number of endpoints */ |
5405 | | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5406 | | } |
5407 | | } |
5408 | | else |
5409 | | { |
5410 | | /* renumber t_group and prepare to renumber at[i].endpoint */ |
5411 | | nTautomerGroupNumber[(int) nGroupNumber] = |
5412 | | t_group[i].nGroupNumber = ++nNewGroupNumber; /* = i+1 */ |
5413 | | /* get first group atom orig. number position in the nEndpointAtomNumber[] */ |
5414 | | /* and in the tautomer endpoint canon numbers part of the connection table */ |
5415 | | t_group[i].nFirstEndpointAtNoPos = nCurrEndpointAtNoPos[i] = |
5416 | | i ? ( t_group[i - 1].nFirstEndpointAtNoPos + t_group[i - 1].nNumEndpoints ) : 0; |
5417 | | t_group[i].num[0] = nNumH; |
5418 | | #if ( REMOVE_TGROUP_CHARGE == 1 ) |
5419 | | t_group[i].num[1] = 0; /* remove only (-) charges */ |
5420 | | #endif |
5421 | | /* -- wrong condition. Disabled. |
5422 | | if ( t_group[i].nGroupNumber != i + 1 ) { // for debug only |
5423 | | goto err_exit_function; // program error |
5424 | | } |
5425 | | */ |
5426 | | i++; |
5427 | | } |
5428 | | } |
5429 | | if (num_t != nNewGroupNumber) |
5430 | | { /* for debug only */ |
5431 | | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5432 | | } |
5433 | | |
5434 | | /* check if any tautomer group was left */ |
5435 | | if (!nNewGroupNumber) |
5436 | | { |
5437 | | if (!num_groups_noH) |
5438 | | goto err_exit_function; /* program error: not a tautomer */ /* <BRKPT> */ |
5439 | | else |
5440 | | goto exit_function; |
5441 | | } |
5442 | | /* |
5443 | | * an array for tautomer group sorting later, at the time of storing Connection Table |
5444 | | * Later the sorting consists out of 2 steps: |
5445 | | * 1) Sort t_group[i].nNumEndpoints endpoint atom ranks within each endpoint group |
5446 | | * starting from t_group[i].nFirstEndpointAtNoPos; i = 0..t_group_info->num_t_groups-1 |
5447 | | * 2) Sort the groups indexes t_group_info->tGroupNumber[] |
5448 | | */ |
5449 | | if (!( tGroupNumber = |
5450 | | (AT_NUMB*) inchi_calloc( nNewGroupNumber*TGSO_TOTAL_LEN, sizeof( tGroupNumber[0] ) ) )) |
5451 | | { |
5452 | | goto err_exit_function; /* out of RAM */ |
5453 | | } |
5454 | | for (i = 0; i < nNewGroupNumber; i++) |
5455 | | { |
5456 | | tGroupNumber[i] = (AT_NUMB) i; /* initialization: original t_group number = (at[i]->endpoint-1) */ |
5457 | | } |
5458 | | /* |
5459 | | * renumber endpoint atoms and save their orig. atom |
5460 | | * numbers for filling out the tautomer part of the LinearCT. |
5461 | | * nCurrEndpointAtNoPos[j] is an index of the atom number in the nEndpointAtomNumber[] |
5462 | | */ |
5463 | | for (i = 0; i < num_atoms; i++) |
5464 | | { |
5465 | | if (j = (int) at[i].endpoint) |
5466 | | { |
5467 | | j = (int) ( at[i].endpoint = nTautomerGroupNumber[j] ) - 1; /* new t_group number */ |
5468 | | if (j >= 0) |
5469 | | { /* j=-1 in case of no mobile hydrogen atoms (charges only), group being removed */ |
5470 | | if (nCurrEndpointAtNoPos[j] >= /* debug only */ |
5471 | | t_group[j].nFirstEndpointAtNoPos + t_group[j].nNumEndpoints) |
5472 | | { |
5473 | | goto err_exit_function; /* program error */ /* <BRKPT> */ |
5474 | | } |
5475 | | nEndpointAtomNumber[(int) nCurrEndpointAtNoPos[j] ++] = (AT_NUMB) i; |
5476 | | } |
5477 | | else |
5478 | | { |
5479 | | nNumEndpoints--; /* endpoint has been removed */ |
5480 | | } |
5481 | | } |
5482 | | } |
5483 | | t_group_info->num_t_groups = nNewGroupNumber; |
5484 | | t_group_info->nNumEndpoints = nNumEndpoints; |
5485 | | t_group_info->nEndpointAtomNumber = nEndpointAtomNumber; |
5486 | | t_group_info->tGroupNumber = tGroupNumber; /* only the 1st segment filled */ |
5487 | | inchi_free( nTautomerGroupNumber ); |
5488 | | inchi_free( nCurrEndpointAtNoPos ); |
5489 | | return nNumEndpoints + T_GROUP_HDR_LEN * nNewGroupNumber + 1; /* nLenLinearCTTautomer */ |
5490 | | |
5491 | | err_exit_function: |
5492 | | ret = CT_TAUCOUNT_ERR; |
5493 | | exit_function: |
5494 | | /* release allocated memory; set "no tautomeric group" */ |
5495 | | if (nEndpointAtomNumber) |
5496 | | { |
5497 | | inchi_free( nEndpointAtomNumber ); |
5498 | | } |
5499 | | if (nTautomerGroupNumber) |
5500 | | { |
5501 | | inchi_free( nTautomerGroupNumber ); |
5502 | | } |
5503 | | if (tGroupNumber) |
5504 | | { |
5505 | | inchi_free( tGroupNumber ); |
5506 | | } |
5507 | | if (nCurrEndpointAtNoPos) |
5508 | | { |
5509 | | inchi_free( nCurrEndpointAtNoPos ); |
5510 | | } |
5511 | | t_group_info->nNumEndpoints = 0; |
5512 | | t_group_info->num_t_groups = 0; |
5513 | | if (!ret && ( ( t_group_info->tni.bNormalizationFlags & FLAG_NORM_CONSIDER_TAUT ) || |
5514 | | t_group_info->nNumIsotopicEndpoints > 1 && ( t_group_info->bTautFlagsDone & ( TG_FLAG_FOUND_ISOTOPIC_H_DONE | TG_FLAG_FOUND_ISOTOPIC_ATOM_DONE ) ) )) |
5515 | | { |
5516 | | ret = 1; /* only protons have been (re)moved or neitralization happened */ |
5517 | | } |
5518 | | |
5519 | | return ret; |
5520 | | } |
5521 | | #endif |
5522 | | #endif |
5523 | | |
5524 | | |
5525 | | /**************************************************************************** |
5526 | | Tautomers: Compare for sorting |
5527 | | |
5528 | | Compare for sorting Ranks only |
5529 | | Globals: pn_tRankForSort |
5530 | | ****************************************************************************/ |
5531 | | int CompRankTautomer( const void* a1, const void* a2, void *p ) |
5532 | 4 | { |
5533 | 4 | AT_RANK *pn_tRankForSort = (AT_RANK *) p; |
5534 | 4 | int ret = (int) pn_tRankForSort[(int) ( *(const AT_RANK*) a1 )] - |
5535 | 4 | (int) pn_tRankForSort[(int) ( *(const AT_RANK*) a2 )]; |
5536 | | |
5537 | 4 | return ret; |
5538 | 4 | } |
5539 | | |
5540 | | |
5541 | | /****************************************************************************/ |
5542 | | int SortTautomerGroupsAndEndpoints( CANON_GLOBALS *pCG, |
5543 | | T_GROUP_INFO *t_group_info, |
5544 | | int num_atoms, |
5545 | | int num_at_tg, |
5546 | | AT_RANK *nRank ) |
5547 | 110k | { |
5548 | 110k | int i, nFirstEndpointAtNoPos, nNumEndpoints; |
5549 | 110k | AT_RANK *pn_tRankForSort; |
5550 | 110k | AT_NUMB *nEndpointAtomNumber; |
5551 | 110k | int num_t_groups = num_at_tg - num_atoms; |
5552 | 110k | T_GROUP *t_group = NULL; |
5553 | | |
5554 | | /* Check if sorting is required */ |
5555 | 110k | if (num_t_groups <= 0 || t_group_info->nNumEndpoints < 2) |
5556 | 110k | { |
5557 | 110k | return 0; /* no tautomer data */ |
5558 | 110k | } |
5559 | | |
5560 | 4 | t_group = t_group_info->t_group; |
5561 | | |
5562 | | /* Sort endpoints within the groups */ |
5563 | 8 | for (i = 0; i < num_t_groups; i++) |
5564 | 4 | { |
5565 | 4 | if (t_group[i].nNumEndpoints < 2) |
5566 | 0 | { |
5567 | 0 | continue; /* program error; should not happen */ /* <BRKPT> */ |
5568 | 0 | } |
5569 | | |
5570 | | /* Set globals for sorting */ |
5571 | 4 | nFirstEndpointAtNoPos = t_group[i].nFirstEndpointAtNoPos; |
5572 | 4 | nNumEndpoints = t_group[i].nNumEndpoints; |
5573 | 4 | if (nNumEndpoints + nFirstEndpointAtNoPos > t_group_info->nNumEndpoints) |
5574 | 0 | { |
5575 | | /* for debug only */ |
5576 | 0 | return CT_TAUCOUNT_ERR; /* program error */ /* <BRKPT> */ |
5577 | 0 | } |
5578 | 4 | nEndpointAtomNumber = t_group_info->nEndpointAtomNumber + (int) nFirstEndpointAtNoPos; |
5579 | 4 | pn_tRankForSort = nRank; |
5580 | | |
5581 | 4 | insertions_sort( pn_tRankForSort, nEndpointAtomNumber, nNumEndpoints, |
5582 | 4 | sizeof( nEndpointAtomNumber[0] ), CompRankTautomer ); |
5583 | 4 | } |
5584 | | |
5585 | | /* Sort the tautomeric groups according to their ranks only |
5586 | | (that is, ignoring the isotopic composition of the mobile groups and ranks of the endpoints) */ |
5587 | 4 | if (t_group_info->num_t_groups > 1) |
5588 | 0 | { |
5589 | | /* Set globals for sorting */ |
5590 | | /* a hack: the ranks of all tautomeric groups are */ |
5591 | | /* located at nRank[num_atoms..num_at_tg-1] */ |
5592 | 0 | pn_tRankForSort = nRank + num_atoms; |
5593 | | /* Sort */ |
5594 | | /* ordering numbers to sort : t_group_info->tGroupNumber; */ |
5595 | |
|
5596 | 0 | insertions_sort( pn_tRankForSort, t_group_info->tGroupNumber, |
5597 | 0 | num_t_groups, sizeof( t_group_info->tGroupNumber[0] ), |
5598 | 0 | CompRankTautomer ); |
5599 | 0 | } |
5600 | | |
5601 | 4 | return t_group_info->num_t_groups; |
5602 | 4 | } |